Download An Earth-System Prediction Initiative for the 21st Century

Science Foundation for Climate, Weather and
Environmental Prediction and Services in 21st Century
Courtesy of NASA Earth Science Enterprise
A Compendium of papers submitted for Publication in
the Bulletin of the American Meteorological Society (AMS)
WMO Research Department
1.
An Earth-System Prediction Initiative for the 21st Century
2.
Addressing the Complexity of the Earth System
3.
Climate Prediction from Weeks to Decades in the 21st Century: Towards a
New Generation of World Climate Research and Computing Facilities
4.
A Unified Modeling Approach to Climate System Prediction
5.
Toward A Seamless Process for the Prediction of Weather and Climate: the
advancement of sub-seasonal to seasonal prediction
6.
The Multi-scale Organization of Tropical Convection and its Interaction with
the Global Circulation: Year of Tropical Convection (YOTC)
An Earth-System Prediction Initiative for the 21st Century
Melvyn Shapiro 1 , Jagadish Shukla2, Michel Béland3, John Church4, Kevin Trenberth5, Randall
Dole6, Richard Anthes7, Brian Hoskins8, Guy Brasseur9, Mike Wallace10, Gordon McBean11,
Antonio Busalacchi12, Ghassem Asrar13, David Rogers14, Gilbert Brunet15, Leonard Barrie16,
David Parsons17, David Burridge18, Tetsuo Nakazawa19, Martin Miller20, Philippe Bougeault21,
Zoltan Toth22, Gerald Meehl23, Mitch Moncrieff24, Hervé Le Treut25, Alberto Troccoli26, Tim
Palmer27, Jochem Marotzke28, John Mitchell29, Adrian Simmons30, Brian Mills31, Øystein
Hov32, Haraldur Olafsson33, Takeshi Enomoto34, Jim Caughey 35, Louis Uccellini36 and Arlene
Lang37.
Capsule: The foundation for an international initiative to accelerate advances in
knowledge, prediction, use and value of weather, climate and Earth-system prediction
information.
1
Cooperative Institute for Environmental Sciences, University of Colorado, Boulder CO, USA,
Geophysical Institute, University of Bergen, Bergen, Norway, National Center for Atmospheric Research
(NCAR), Boulder CO, USA; 2Professor, George Mason University (GMU), and President, Institute of
Global Environment and Society (IGES), USA; 3President, WMO Commission on Atmospheric Sciences
(CAS) at Environment Canada's Atmospheric Science and Technology Directorate, Montreal, Canada;
4
Centre for Australian Weather and Climate Research, a partnership between the Bureau of Meteorology
and CSIRO, Hobart, SIRO, Australia; 5Chair, World Climate Research Progamme (WCRP) Observations
and Assimilation Panel at the National Center for Atmospheric Research (NCAR), Boulder CO, USA;
6
NOAA Earth System Research Laboratory, Boulder CO, USA; 7President of the University Corporation
for Atmospheric Research, Boulder CO, USA; 7Head, Dynamical Processes Group, Dept. of
Meteorology, Univ. of Reading, UK; 9Director, Earth and Sun Systems Laboratory and Associate
Director of NCAR, Boulder CO, USA; 10Professor, Dept. of Meteorology, Univ. of Washington, Seattle
WA, USA; 11Professor, Institute for Catastrophic Loss Reduction at University of Western Ontario,
London, Ontario, Canada; 12Chair, WCRP Joint Scientific Committee, Earth System Science
Interdisciplinary Center (ESSIC), Univ. of Maryland USA; 13Director, WCRP, Geneva, Switzerland;
14
President, Health and Climate Foundation, Washington, D.C., USA; 15Chair, WMO-WWRP Joint
Steering Committee and Research Director of Environment, Canada; 16Director, WMO-WWRP, Geneva;
17
Chief, WMO-WWRP, Geneva; 18 WMO, Geneva, Switzerland; 19Chair, Asian Regional Committee for
THORPEX, Meteorological Research Institute, Japan Meteorological Agency, Japan; 20Chair, WMOCAS/WCRP Working Group on Numerical Experimentation (WGNE) at ECMWF, Reading, UK 21Head
of Research, MeteoFrance, Toulouse, France ; 21NOAA/NWS/NCEP, Environmental Modeling Center,
Camp Springs, MD, USA; 22Chair, GFS-TIGG Working Group, NOAA/NCEP Environmental Modeling
Center, USA; 23NCAR, USA; 24Co-chair, Year Of Tropical Convection (YOTC), NCAR, Boulder CO,
USA; 25Directeur du Laboratoire de Météorologie Dynamique, France; 26; Environmental Systems
Science Centre, University of Reading, Reading, UK; 27Head Probability Forecasting and Diagnostics
Division, ECMWF, UK; 28Director, Max Planck Institute for Meteorology, DE; 29Met. Office, UK;
30
Chair GCOS/WCRP Atmospheric Observation Panel for Climate and ECMWF; 31Adaptation and
Impacts Research Division, Environment Canada; 32Director of Research, Norwegian Meteorological
Institute, Oslo, Norway; 33University of Iceland and University of Bergen, Norway, 34Earth Simulator
Center and, Japan Agency for Marine-Earth Science and Technology, Japan; 35Scientific Officer, THe
Observing-system Research and Predictability Experiment (WWRP-THORPEX), Geneva; 36Director,
NOAA/NWS/National Centers for Environmental Modeling, USA, 37; NCAR ,Boulder CO, USA.
1
1. Introduction
We stand at the threshold of accelerating advances in the prediction of high-impact
weather and climate, and the complex interaction between the physical-biologicalchemical Earth system 2 and global societies (NRC 2007, 2008). This opportunity stems
from notable achievements in monitoring and predicting weather hazards, climate
variability and change, and the use of this information. As examples, regional predictions
on spatial scales of a few kilometers and time scales from minuets to days, provide timely
and accurate warnings of tornados, flash-flood rainstorms, disruptive snow and ice
storms, excessive river flows, coastal storm surges, hurricane track and intensity, and airquality emergencies. Global 5-day forecasts have an accuracy comparable to 2-day
forecasts of 25 years ago (Fig. 1), with an increasing ability to identify the possibility of
extreme local weather 7 to 10 days in advance. Seasonal forecasts provide useful
information on the development of El Niño/La Niña and their influence on regional
weather, such as shifts in the North Pacific storm track. Assessments and projections of
global temperature, sea level, ice and precipitation over decades to centuries were critical
in establishing human influences on climate change and provided the scientific
underpinning for international action to reduce greenhouse gas and aerosol emissions;
Meehl et al. 2007. Risk, impact, adaptation, mitigation and assessment models are
increasingly important in exposing vulnerabilities and evaluating the outcomes of
different decisions. These accomplishments are among the most significant scientific,
technological and societal achievements of the 20th century.
The necessity and opportunity to capitalize on these achievements was recognized by the
World Meteorological Organization (WMO), World Climate Research Program (WCRP)
and International Geosphere-Biosphere Programme (IGBP), whose representatives
introduced the foundation for an Earth-system Prediction Initiative at the Group on Earth
Observations (GEO) Summit, 2007, Cape Town, South Africa (Shapiro et al. 2007). The
Initiative would provide the international framework for revolutionary advances in Earthsystem prediction and help managers of disaster response, public health and food
security, water supply, energy, and environmental policy minimize the impacts of
extreme weather and climate. This endeavor is as challenging as the International Space
Station, Genome Project, and Hubble Telescope, with a high societal benefits-to-cost
ratio. It will contribute to international programs for advancing observing systems and
prediction, such as the Global Climate Observing System (GCOS), WCRP, World
Weather Research Programme (WWRP), and hence to the Global Earth Observation
System of Systems (GEOSS). It will build upon the International Council for Science
(ICSU), International Ocean Commission (IOC), WMO, World Health Organization
(WHO) and GEO, to coordinate the effort across the weather, climate, Earth-system,
natural-hazards, and socioeconomic communities.
2
The Earth-system encompasses the atmosphere and its chemical composition, the oceans, land/sea-ice and other
cryosphere components; the land-surface, including surface hydrology and wetlands, lakes and human activities. On short-time
scales, it includes phenomena that result from the interaction between one or more components, such as ocean waves and storm
surges. On longer time scales (e.g., climate), the terrestrial and ocean ecosystems, including the carbon and nitrogen cycles and
slowly varying cryosphere components (e.g., the large continental ice sheets and permafrost) are also part of the Earth-system.
2
Fig. 1: Evolution of forecast skill for the extratropical northern and southern
hemispheres, January 1980 to August 2008. Anomaly-correlation coefficients of 3, 5, 7
and 10-day ECMWF 500-mb height forecasts are plotted as 12-month running means.
Shading shows differences in scores between hemispheres at the forecast ranges
indicated (updated from Simmons and Hollingsworth 2002).
This article introduces the rationale for this Initiative, while the companion papers in this
issue by Brunet et al. 2009, Moncrieff et al. 2009, Shukla et al. 2009 and Nobre et al.
2009, present key research aspects and institutional perspectives.
2. Rationale
Hurricane Katrina, the deadly 2003 European heat wave, the multi-decadal drought south
of the Sahel and the unprecedented wildfires in Australia in 2009 attest to the
vulnerability of modern society and the environment to adverse weather and climate. In
fact, 75 percent of natural disasters are triggered by extreme weather and climate 3 .
Effective mitigation and adaptation to this vulnerability requires accurate prediction at
global, regional and local scales. Today’s observation and prediction systems address
specific needs of environmentally sensitive sectors and resources, including energy,
water, human health, transportation, agriculture, fisheries, leisure, ecosystems,
biodiversity, and national security. Probabilistic predictions provide quantitative
measures of the likelihood of occurrence and severity of different outcomes, including
potentially catastrophic extremes. These capabilities yield substantial benefits, because
3
http://www.unisdr.org/eng/media-room/facts-sheets/2008-disasters-in-numbers-ISDR-CRED.pdf
3
they enable rational decisions that reduce human, economic and environmental losses,
and also maximize economic opportunities. The opportunities include selection of
optimal trade routes, energy allocation, crop selection, and pollutant emission mitigation
strategies. These achievements and opportunities are the culmination of investments by
governments, international agencies and other stakeholders in underpinning science and
technology and their transition to operational services. These investments greatly
expanded our capability to observe the atmosphere, oceans, land and cryosphere,
including bio-geochemical properties. High-performance computers, global
communications, and improvements in numerical methods, have been crucial to advances
in predictions and applications. Investments in research have advanced knowledge of
atmospheric and oceanic predictability and provided insight into the requirements for
addressing the influence of climate variability and change on regional high-impact
weather. Based on the achievements, we foresee the potential to respond more effectively
and realize even greater benefit from these and future investments (NRC 2007).
The vulnerability of society and the environment to regional weather extremes within a
changing climate and projections that these extremes will increase in coming decades
(Meehl et al. 2007; CCSP, 2008) increases the urgency for advancing mitigation and
adaptation capabilities. We must better inform climate-change mitigation policies, in
particular, carbon sequestration and its interaction with other biogeochemical cycles like
reactive nitrogen. This requires introduction and more accurate representation of key
processes in climate models, such as the Earth/ocean methane release and glacial and seaice melting. The present suites of climate models are limited in their capability to provide
the required level of detailed regional information. It is also imperative that we reduce the
uncertainty of climate-change predictions and scenarios. There is a corresponding
necessity to advance predictive skill and diversify forecast applications on hourly-toseasonal time scales. Accelerated progress will optimize policy and risk management
decisions valued at billions of dollars. The potential to accelerate further advances
depends critically on cooperation across many nations, international organizations and
scientific disciplines.
If we are to address these fundamental challenges, it is essential to bridge individual
disciplinary boundaries and move toward a more comprehensive Earth-system approach,
with predictive capability that considers interactions between different system
components and across time scales. The connectivity between climate, weather, landsurface properties, biochemical processes, human health and ecosystems is exemplified
by the regional-to-global impact of Saharan sand and dust storms; Fig. 2. These storms,
and those of the other great deserts, e.g., the Gobi of China, are a major source of
tropospheric aerosols that represent a significant component in the atmospheric regionalto-global radiative balance, thereby affecting climate e.g., Kaufman, et al. 2002; Li, et al.
2004. Variations in seasonal to sub-seasonal climate circulations, such as the negative
phase of the North Atlantic Oscillation (NAO), deflect the North Atlantic storm track
southward into the Mediterranean and North Africa, where individual cyclones become
catalysts for massive sand and dust events, such in Fig. 2. Cyclone interactions with
mountains of North Africa contribute to localized enhancements of the desert winds
through mountain waves over the Atlas and Haggar ranges, and enhanced surface wind
4
around the smaller Tibesti and Ajir mountains; Todd, et al, 2005. In addition, the large
amplitude diurnal temperature cycle over the Sahara modulates the depth of the planetary
boundary layer and hence the diurnal concentration of sand and dust; Schepanski, et al.
2007. When transported westward over the Atlantic, Saharan dust may affect
development of tropical cyclones (Evan et al 2006) and replenish nutrients in the soils of
the Amazon, which is critical to the sustainability of the rain forest. Northward transport
is a major source of aerosols to Europe impacting visibility, health and local weather,
e.g., temperature and precipitation. Saharan dust provides nutrients to ocean biota, whose
concentrations modulate the opacity of the upper ocean and its radiative characteristics
and the oceanic uptake of carbon dioxide, providing another feedback to the climate
system. Saharan aerosols are under investigation as major contributors to central African
meningitis epidemics that can affect 250,000 people each year. In short, this is not
simply a weather problem or climate problem, or an atmosphere, ocean, land surface,
chemistry or biology problem. It is all of these and more. Comprehensive understanding
of such events and their consequences requires consideration of the interactions among
the different Earth-system components, including interactions with and implications for
humankind.
MOROCCO
MAURITANIA
MALI
Fig. 2: MODIS satellite view of an extreme Saharan sand and dust event on 6 March
2004.
The possibility of "climate surprises", i.e., unexpected rapid climate changes outside of
current climate models projections, presents another important example of the necessity
5
for developing a comprehensive Earth-system approach. Anticipating rapid climate
change is a major scientific challenge and a central concern for decision makers, with
potentially enormous implications for society and the environment. Possible triggers for
rapid changes include relatively fast processes in the Earth system, such as changes in
biology through enhanced methane release associated with melting permafrost; changes
in sea ice extent, with the annual minimum Arctic sea-ice extent having declined
approximately 30 percent over the last 29 years and changes in hydrological processes;
NRC 2002 and CCSP 2008. Rapid changes typically involve feedbacks among different
components of the Earth system, necessitating more comprehensive and integrative
approaches to address this challenge. Reducing the likelihood of future climate surprises
requires improved observations and monitoring of Earth-system components and their
interactions, advances in understanding the causes for rapid changes, and substantial
improvements in modeling, including incorporation of processes not present in current
generation climate models; CCSP 2008. Because of the rapid nature of component
interactions, joint studies of weather, climate and earth-system phenomena are very
informative as to possible responses and feedbacks among the different system
components, as in the case of Saharan dust storms.
3. A holistic approach
Earth-system sciences, mitigation and adaptation strategies require a suite of diagnostic
and prediction models integrated over all spatial and temporal scales; Dole 2008, Palmer
et al. 2008, Brunet et al. 2009, Hurrell et al. 2009 and Meehl et al. 2009. This holistic
approach is spatially and temporally continuous, spanning highly-localized cloud
systems to global circulations, from minutes to millennia; linking mesoscale weather life
cycles and climate variability and change. It is integrated across the disciplines of
physics, mathematics, chemistry, social and decision sciences, and their Earth-system
elements. It requires coordination and support across academic institutions, government
research and service agencies, private enterprise providers, hazard risk-reduction and
adaptation government agencies, and humanitarian organizations. It bridges political
boundaries from municipalities to nations to the world. Within this framework,
socioeconomic and environmental requirements play a leading role in the design and
implementation of a new generation of science-based global to regional early warning
and planning systems.
In order for weather, climate and Earth-system information to have timely and beneficial
impacts, it must have the following elements: content with accuracy and precision in
space and time, and relevance of product information to the users, including a quantified
estimate of its accuracy, and probability of occurrence of particular events; distribution
of products on spatial and temporal scales sufficient for action; communication with
product formats that users can comprehend and interpret; integration of information into
user decision-support systems; recognition by users that the information has value; and
response to the information. These elements are links along a chain of action. If any link
is weak or broken, then the impact and the value of the information will be diminished.
Socioeconomic research and its applications can identify ways to strengthen these links
6
and lead to the development of new methods for enhancing the use and value of weather,
climate, Earth-system, and socioeconomic information.
4. Core elements
The Earth-system Prediction Initiative will provide the basis for more skilful predictions
of weather, climate and Earth-system processes. This includes their interactions, past,
present and future variability and change, and known confidence. It will involve
advanced observing capabilities and data assimilation, diagnostic and prediction models,
and address ways in which predictions inform and affect decision-making frameworks.
The research and tools required to achieve these objectives will include:
Climate and weather observations to monitor the properties and evolution of the Earth
system, as well as changes in forcing of the system. The observations must satisfy the
climate principles promoted by GCOS for the design/implementation of observations as
an element of GEOSS. They must be anchored with benchmark measurements and strive
for accuracy, noting that continuity is a primary requirement for reliable detection of
global-to-regional anomalies. This requires stemming the ongoing decline in surface and
upper-air global observing networks and the development and implementation of a new
generation of in-situ and space-based systems to meet the ever increasing observational
demands of the prediction and early -warning systems of today and in future generations.
A comprehensive coordinated observing system is the backbone of this Earth information
system; Trenberth 2008.
High-resolution global and regional data-assimilation and analysis systems that
integrate observations of the atmosphere, ocean, land and ice from space-based, aircraft,
and ground-based observing platforms across the full suite of Earth system modeling
components. The Global Monitoring for Environmental and Security (GMES) initiative
(Hollingsworth et al. 2008), Fig. 3, exemplifies a step in this direction. This process
provides the basis for the analysis and reanalysis (as systems improve and more historical
data are recovered) of the long-term observational record that is required for monitoring
and assessing past events and change, including their socioeconomic impacts. It provides
the basis for the analysis of present conditions for initializing new predictions and for
verifying earlier predictions. An important aspect linking all these core elements is the
performance of Observing System Simulation Experiments (OSSEs) to assess the role
and value of new and future observations and measurement platforms in support of Earthsystem monitoring and prediction.
7
Fig. 3: Air pollutant forecast for Europe prepared by the German Aerospace Research
Establishment-German Remote Sensing Data Center (DLR-DFD) as part of the GMES
Service Element PROMOTE funded by the European Space Agency. It shows the 72h ensemble forecast for surface-level particulate matter (PM10) for 7 October, 2007.
Advanced prediction models that capture the complex interaction among the
atmosphere, oceans, and other components of the Earth system and the human dimension.
These models must have sufficient resolution to faithfully represent the multi-scale
processes of the Earth-system appropriate for the spatial scale of the applications (Figs. 4,
5); see Brunet et al. 2009 and Shukla et al. 2009; Nobre et al. 2009, this issue. These
models will incorporate increasingly diverse components of the Earth system. For
example, the next IPCC assessment will include, for the first time, models with
interactive carbon cycles. To address challenges like the potential for rapid changes in sea
level beyond current IPCC projections, future models will incorporate dynamical ice
sheet components to better assess the potential for accelerated ice loss. Human
interactions with the system will be modeled as well; Nobre, et al. 2009, this issue.
Addressing needs for increased model resolution and complexity will require substantial
increases in computing capacity beyond current generation global NWP and/or IPCC
models, as discussed in more detail below.
8
Fig. 4: Left panel, global cloud distribution in a 320-km coarse-resolution climate
simulation experiment. Right panel, same, but for a 20-km resolution simulation with
the same model, comparable in resolution to the most advanced operational weather
forecast models of today (Moncrieff et al. 2009). The proposed Initiative will provide
high-resolution climate models that capture the properties of regional high-impact
weather events, such as tropical cyclones; heat waves; sand and dust storms,
associated within multi-decadal climate projections of climate variability and change
(Courtesy of Shintaro Karahawa, Earth Simulator Center/JAMSTEC).
Fig. 5: The capability of high-resolution regional forecast models to predict the intensity
of high-impact weather events, such as the 29 August 2005 landfall of hurricane
Katrina, over New Orleans, Louisiana, USA, (Davis et al. 2008). NCAR/WRF
simulation of Hurricane Katrina precipitation radar reflectivity computed 3-days before
landfall (left), compared with radar observations of the actual landfall (right). Within the
next decade, global data-assimilation and deterministic and ensemble medium-range to
seasonal prediction systems will advance to such high-resolution capabilities.
9
It is essential that numerical experimentation with high-resolution weather, climate and
complex Earth-system models, with established fidelity and skill, provide scientificallybased assessments of the global-to-regional impact of engineering hypotheses (Fig. 6),
for controlling climate variability and change prior to their design and implementation, as
discussed in Keith 2001; Angel 2006; Crutzen 2006; Latham et al. 2008; Robock 2008 .
Fig. 6: Schematic representation of various climate-engineering proposals, from Keith
2001.
Attribution and diagnostic studies to determine causes of past and current conditions
including extreme events, such as severe droughts, heat waves, and cyclone frequency,
intensity and tracks. Distinguishing between natural variability and human-forced longterm trends is vital to informing both adaptation and mitigation decisions. Establishing
attribution requires good estimates of different process forcing, including ensemble
model experiments of sufficient size to determine the most likely cause or causes for
features of interest at a specified level of confidence. Attribution studies address the
predictability of the system, the odds of the observed anomalies and extremes being
realized, and the confidence that can be placed in the predictions. This element includes
performance assessment and analysis of detailed numerical simulations of weather,
climate and Earth-system processes conducted under the WWRP, WCRP-THORPEX,
WGNE, GAW, ICSU, IOC and WMO-sponsored programs.
Modern information and decision systems that provide timely, user-friendly, issuetargeted production of weather, ocean, climate and Earth-system information crucial in
assisting decision-making processes for risk reduction and adapting to weather and
climate events, developing mitigation policies and achieving sustainable development.
This necessitates engagement of the users of environmental information to assess and
incorporate their requirements, see, e.g., ICSU 2008.
Heightened multi-disciplinary collaborations and applications that encourage
scientists in natural sciences to interact with colleagues from health, economic, water,
agriculture, energy, food, and policy disciplines. Recent examples are the Meningitis
10
Environmental Risk Information Technologies (MERIT) project4 and the UCAR Africa
Initiative 5 on tropical health/climate/weather linkages. Collaborations such as the WCRP
and WWRP-THORPEX Year of Tropical Convection (YOTC) are engaging the weather
and climate communities to improve the prediction of the multi-scale organization of
tropical precipitating cloud systems (Fig. 7), see Brunet, et al. 2009 and Moncrieff et al.
2009, this issue. The 2007-2008 International Polar Year (IPY) 6 included a
multidisciplinary approach to marine biology issues, ocean and atmospheric physics,
chemistry, and social sciences. These recent collaborations provide a template for future
efforts.
Data and forecast archives to provide an internationally-coordinated Earth-information
system (see NRC 2003) to store and manage data and enable access to global-to-regional
operational and historical analyses and forecasts of weather, air quality, climate, other
Earth-system components and socioeconomic impacts, commensurate with the highest
resolution achievable, given near-term observational and computational constraints. The
archive should facilitate advanced analysis and visualizations of observed and predicted
events, including assessment of the effectiveness of mitigation and adaptation actions.
Fig. 7: Multi-scale tropical convective organization associated with a Madden-Julian
Oscillation (MJO) over the Indian Ocean on 2 May 2002 (left panel). By 9 May 2002
(right panel), the MJO propagated eastwards over Indonesia and spawned twin tropical
cyclones in its wake, leading to flooding rains and hurricane-force winds over northern
Madagascar, and heavy precipitation over Yemen. The twin tropical cyclones illustrate
high-impact organised weather events directly associated with large-scale convective
organisation and equatorial waves (see, Moncrieff et al. 2009, this issue).
High-performance computing (HPC) is crucial to accelerate advances in prediction
models and their socioeconomic applications. These models are links in a chain,
4
http://www.hc-foundation.org
http://www.africa.ucar.edu/index.html
6
http://www.ipy.org/
5
11
involving access to billions of time-dependent observations that fuel data-assimilation
systems, which in turn provide the initial conditions to a series of prediction and
projection models. The scope ranges from minutes to hours for severe weather events, to
decades and centuries for climate change scenarios and ecosystem impacts. Advanced
HPC will facilitate high-resolution ensemble prediction systems that include many
hundreds of possible predictions and projections. These models will incorporate more
realistic processes and a high degree of Earth-system complexity, including two-way
interactions between society and predicted responses. This requires dedicated facilities
with sustained speeds well beyond that of the most advanced computers of today, but
which are achievable within the next 10-20 years, see Shukla et al. 2009, this issue.
Realizing the full research and operational benefits of HPC will require advanced data
processing and visualization methods, user-friendly high-speed and broad-bandwidth
communication and common data formats. It will include training of scientists in the use
of advanced computer systems and integrated data-distribution systems that facilitate
access to most information in near real time. It will also require the development of
advanced numerical methods and software systems to enable modelling systems to fully
exploit the capacity of future generations of supercomputers.
5. Deliverables
The proposed Earth-system Prediction Initiative will:
•
Provide improved monitoring and projections of multi-decadal, global-toregional climate change, including, statistics of projected changes in extreme
regional weather events and the full range of socioeconomic and environmental
outcomes, including assumptions, confidence and uncertainty. This will
provide a science-based foundation for long-term adaptation and mitigation
strategies and sustainable development of global societies.
•
Accelerate advances short-term weather forecasts, and predictions of seasonal
through inter-annual regional high-impact weather, such as the frequency,
intensity and tracks winter storms and tropical cyclones, and the onset and
cessation of regional heat and cold waves, droughts and floods. This will
contribute to mitigation and adaptation strategies required for management of
energy, water, food production, health, the economy, emergency response and
the support of other environmental and socioeconomic strategic planning. It
places a high priority on early-warning systems to reduce vulnerability to
famine, water shortage, pestilence and disease in developing nations and
promote social and economic development.
•
Assess the responses and possible unintended consequences of climateengineering hypotheses for controlling climate variability and change through
experimentation with high-resolution weather, climate and complex Earthsystem models.
12
•
Contribute to the establishment and further development of emerging national
and international climate services.
6. The Way Forward
In order to succeed, the Earth System Prediction Initiative requires:
Interdisciplinary research and applications: This demands intellectual respect across
the scientific disciplines; collaboration between those who excel in their own field; joint
proposal development guided by multi-agency and multi-national environmental and
socioeconomic priorities; long term commitments from scientists, supporting agencies
and stakeholders. It will benefit from the establishment of national and international
research centers (see Shukla et al. 2009b, this issue) with access to sufficient computing
capability, and scientific and technical resources to develop and implement research
objectives on the scale envisioned, including drawing upon the broad expertise in
universities, private industry, governments and other research centers.
Excitement and challenge: The endeavour must not only motivate the scientific
community, but communicate the importance and the character of its multi-generational
impacts and benefits. Most importantly, it must capture the hearts and inspire the minds
of young scientists.
Global and regional engagement: The nature and complexity of the scientific and
socioeconomic challenges, and the resources, intellect, technology, and infrastructure
requires partnerships for the successful development and implementation of the Initiative.
In this context, the success of the Global Atmospheric Research Program, GARP ( Döös
2004; Uppala et al. 2004) provides a template for the successful outcome of global
engagement. GARP was the catalyst for great advances in knowledge, global
observations and prediction of the coupled atmosphere-ocean circulations in the second
half of the 20th century. The ECMWF pooled scientific and management capabilities of
the Europe to advance medium-range deterministic and probabilistic (ensemble) forecast
skill, and the fundamentals of data assimilation, to a level of success that would not have
been achievable if undertaken by a single nation.
The Initiative requires champions to inform governments and other stakeholders of the
urgency of supporting the effort, recognizing that such enterprises take years to succeed.
In this vein, international organizations and agencies are currently coordinating activities
that embrace components of the proposed Initiative. These activities include:
•
GEO as an international coordination framework across disciplines, and
observational, prediction, and information systems, that will advocate for
advancing climate, weather, water and Earth-system prediction. GEO aims to
implement the GEO 2009-2011 Work Plan, and several tasks relevant to this
13
Initiative, e.g. Task CL-09-01: Environmental Information for Decisionmaking, Risk Management and Adaptation 7 .
•
The World Meteorological Organization, sponsor of the World Weather
Research Programme (WWRP), the Global Atmosphere Watch (GAW) air
chemistry research programme and co-sponsor of GCOS and the World Climate
Research Programme (WCRP) are implementing the concepts outlined in this
article. In June 2008, the Executive Council of WMO commissioned a Task
Team On Research Aspects Of An Enhanced Climate, Weather, Water And
Environmental Prediction Framework 8 to prepare a strategy focusing on
strengthening prediction research and related scientific assessments in support
of enhanced climate, weather, water and environmental services in the next
decade for consideration by the National Hydrological and Meteorological
Services and their research partners. The WWRP and WCRP joint scientific
committees are coordinating these recommendations linking global weather and
climate research.
•
The engagement of ICSU through its co-sponsorship of GCOS and WCRP and
its academic constituency.
7. The Grand Challenge
The proposed Initiative will require unprecedented international collaboration and good
will, making the global scope of the problem inescapable. All nations stand to benefit
from its success. As nations, we have collaborated to advance global observing systems,
weather forecasting, climate prediction, communication networks, and emergency
preparedness and response. We must now extend this collaboration to embrace the full
Earth-system and the next frontier of socioeconomic and environmental applications of
our science. Our community and its supporting organizations are poised for the
discoveries ahead, and the opportunity to make its knowledge available to users and
decision makers. This is our grand challenge. Addressing this challenge will provide
deliverables that are at the forefront of meeting the needs of global societies and the
destiny of our planet.
7
8
http://www.earthobservations.org/documents.shtml
http://www.wmo.int/ecrtt
14
8. References
Angel, R., 2006: Feasibility of cooling the earth with a cloud of small spacecraft near the inner La15
Grange-point (l1), Proc. Natl. Acad. Sci., 103, 17184–17189.
Crutzen, P. J., 2006: Albedo enhancement by stratospheric sulphur injections: A contribution to resolve
a policy dilemma? Climatic Change, 77, 211–219.
Brunet, G., M. A. Shapiro, B. Hoskins, M. Moncrieff, R. Dole, G. N. Kiladis, B. Kirtman, A. Lorenc, B.
Mills, R. Morss, S. Polavarapu, D. Rogers, 2009: Toward a seamless process for the prediction of
weather and climate: the advancement of sub-seasonal to seasonal prediction. Bull. Amer. Met., this
issue.
Davis, C., W. Wang, S. S. Chen, Y. Chen, K. Corbosiero, M. Demera, J. Dudia, G. Holland, J. Klemp,
J. Michalakes, H. Reeves, R. Rotuno, C. Snyder, and Q. Xiao, 2008: Prediction of land falling
hurricanes with the Advanced Hurricane WRF Model. Mon. Wea. Rev., 126, 1990-2005.
Dole, R. M., 2008: Linking weather and climate. Amer. Met. Soc., Meteor. Mono., 33, No. 55, 297-
348.
Döös, B. R., 2004: The First GARP Global Experiment – a model of international cooperation. WMO
Bulletin, 53, No.3, 199-206.
Evan, A. T., J. Dunion, J. A. Foley, A. K. Heidinger, and C. S. Velden, 2006: New evidence for a
relationship between Atlantic tropical cyclone activity and African dust outbreaks, Geophys. Res. Lett.,
33, L19813.
ICSU, International Council of Science, 2008: A Science plan for Integrated Research on Disaster
Risk: Addressing the challenge of natural and human-induced environmental hazards. ISBN 978-0930357-66-5
Hollingsworth, A., and coauthors, 2008: Toward a monitoring and forecasting system for atmospheric
composition: The GEMS Project. Bull. Amer. Meteor. Soc., 89, 1147–1164.
Hurrell, J., G.A. Meehl, D. Bader, T. Delworth, B. Kirtman and B. Wielicki, 2009: Climate system
prediction, Bull. Amer. Met. Soc., in review.
Keith, D. W., 2001: Geoengineering. Nature, 409, 420.
Latham, J., and coauthors, 2008: Global temperature stabilization via controlled albedo enhancement of
low-level maritime clouds. Philos. Trans. Roy. Soc. A, 366(1882), 3969–3987.
Li F., A. M. Vogelmann, and V. Ramanathan, 2004: Saharan dust aerosol radiative forcing
measured from space. J. Climate, 17, 2558–2571.
Meehl, G. A., and co-authors, 2007: Global Climate Projections. In: Climate Change 2007: The
Physical Science Basis. Contribution of Working Group I to the Fourth Assessment Report of the
Intergovernmental Panel on Climate Change [Solomon, S., D. Qin, M. Manning, Z. Chen, M. Marquis,
K.B. Averyt, M. Tignor and H.L. Miller (eds.)]. Cambridge University Press, Cambridge, United
Kingdom and New York, NY, USA, 747-845.
15
Meehl, G.A., L. Goddard, J. Murphy, R.J. Stouffer, G. Boer, G. Danabasoglu, K. Dixon, M.A.
Giorgetta, A. Greene, E. Hawkins, G. Hegerl, D. Karoly, N. Keenlyside, M. Kimoto, B.
Kirtman, A. Navarra, R. Pulwarty, D. Smith, D. Stammer, and T. Stockdale, 2009: Decadal
prediction: Can it be skillful? Bull. Amer. Meteorol. Soc., accepted.
Moncrieff, M., D. Waliser, M.A. Shapiro, 2009: The mesoscale organization of tropical convection and
its interaction with the global circulation: the Year of Tropical convection (YOTC). Bull. Amer. Met.
Soc., this issue.
Schiermeier, Q, 2009: International Polar Year: In from the cold, 25 February 2009. |Nature
457, 1072-1077, doi:10.1038/4571074a.
Nobre, C., K. Noone, S. Seitzinger, M. Lahsen, J. Ometto, G. Brasseur, M. Stadford-Smith, and K.
Hibbard, 2009: The International Geochemical Biological Programme (IGBP) Earth-system
perspective, Bull Amer. Met. Soc., this issue.
NRC, 2003, Satellite observations of the Earth’s environment: Accelerating the transition of research to
operations, National Academies Press, Washington D.C., 163pp.
NRC, 2007: Earth science and applications from Space-National Imperatives for the next decade and
beyond. National Academies Press, Washington, D.C., 428 pp.
NRC, 2008: Earth Observations from Space-The first 50 years of scientific achievements. National
Academies Press, Washington, D.C., 129 pp.
Palmer, T. N., F. J. Doblas-Reyes, A. Weisheimer, and M. J. Rodwell, 2008: Toward seamless
prediction: Calibration of climate change projections using seasonal forecasts. Bull. Amer. Meteor. Soc.,
89, 459–470.
Robock, A., 2008: 20 reasons why geoengineering may be a bad idea. Bull. Atomic Scientists, 64, 2, 1418.
Schepanski, K., I. Tegen, B. Laurent, B. Heinold, and A. Macke, 2007, A new Saharan dust source
activation frequency map derived from MSG-SEVIRI IR-channels, Geophys. Res. Lett., 34, L18803.
Shapiro, M. A., and coauthors, 2007: The socio-economic and environmental benefits of a revolution in
weather, climate and Earth-system analysis and prediction. The Full Picture, Group on Earth
Observations, Tudor Rose, pp 137-139.
Shukla J, R. Hagedorn, B. Hoskins, J. Kinter, J. Marotzke, M. Miller, T. Palmer, J. Slingo, 2009a:
Revolution in Climate Prediction is Both Necessary and Possible: A Declaration at the World
Modelling Summit for Climate Prediction. Bull. Amer. Meteor. Soc., 90, 16-19.
Shukla, J., T.N. Palmer, R. Hagedorn, B. Hoskins, J. Kinter, J. Marotzke, M. Miller, J. Slingo, 2009b:
Climate Prediction from Weeks to Decades in the 21st Century: Towards a New Generation of World
Climate Research and Computing Facilities. Bull. Amer. Meteor. Soc., this issue.
Simmons, A.J. and A. Hollingsworth, 2002: Some aspects of the improvement in skill of numerical
weather prediction. Quart. J. Roy. Meteor. Soc., 128, 647-677.
16
Trenberth, K. E., 2008: Observational needs for climate prediction and adaptation. WMO Bulletin, 57
(1) 17-21.
Todd, M.C., R. Washington, S. Raghavan, G. Lizcano, and P. Knippertz, 2008: Regional Model
Simulations of the Bodélé Low-Level Jet of Northern Chad during the Bodélé Dust Experiment
(BoDEx 2005). J. Climate, 21, 995–1012.
U.S. Committee for the Global Atmospheric Research Program, Division of Physical Sciences, National
Research Council, 1969: National Academy of Sciences, Washington, D. C., 79pp.
Uppala, S., A. J. Simmons and P. Källberg, 2004: Global numerical weather prediction – an outcome of
FGGE and a quantum leap for meteorology. WMO Bulletin, 53, No. 3, 207-212.
Yoshioka, M., N.M. Mahowald, A.J. Conley, W.D. Collins, D.W. Fillmore, C.S. Zender, and
D.B. Coleman, 2007: Impact of Desert Dust Radiative Forcing on Sahel Precipitation: Relative
Importance of Dust Compared to Sea Surface Temperature Variations, Vegetation Changes,
and Greenhouse Gas Warming. J. Climate, 20, 1445–1467.
Appendix 1: Acronyms
CAS: Commission for Atmospheric Science
CIRES: Co-operative Institute for Research in Environmental Science
CliC: Climate and the Cryosphere project of WCRP
CLIVAR, Climate Variability and Predictability project of WCRP
COPES: Coordinated Observation and Prediction of the Earth System
ECMWF: European Centre for Medium-range Weather Forecasts
GAW: Global Atmospheric Watch
GCOS: Global Climate Observing System
GEO: Group on Earth Observations
GEOSS: Global Earth Observation System of Systems
GEWEX: Global Energy and Water cycle Experiment project of the WCRP
GMES: Global Monitoring for Environment and Security
HPC: High Performance Computing
ICSU: International Council for Science
IGBP: International Geosphere-Biosphere Programme
ILEAPS: Integrated Land Ecosystem-Atmosphere Processes Study
IOC: International Organizing Committee
IPCC: Intergovernmental Panel on Climate Change
IPY: International Polar Year
JAMSTEC: Japan Agency for Marine-Earth Science and Technology
MERIT: Meningitis Environmental Risk Information Technologies
NCAR: National Center for Atmosphere Research
NCEP: National Centers for Environmental Prediction
NHMS: National Hydrological and Meteorological Service
NOAA: National Oceanographic and Atmospheric Administration
NRC: National Research Council
NWP: Numerical Weather Prediction
OSSEs: Observing System Simulation Experiments
PROMOTE: PROtocol MOniToring for the GMES Service Element
17
SOLAS: Surface Ocean-Lower Atmospheric Study
SPARC: Stratospheric Processes And their Role in Climate project of WCRP
THORPEX: THe Observing-system, Research, and Predictability EXperiment
UCAR: University Corporation for Atmospheric Research
WCRP: World Climate Research Programme
WGNE: Working Group on Numerical Experimentation
WHO: World Health Organization
WMO: World Meteorological Organization
WWRP: World Weather Research Programme
WRF: Weather Research and Forecast model
YOTC: Year of Tropical Convection
18
Addressing the Complexity of the Earth System
Carlos A. Nobre1, Guy P. Brasseur2, Melvyn A. Shapiro3, Myanna Lahsen4, Gilbert
Brunet5, Antonio J. Busalacchi6, Kathy Hibbard7, Sybil Seitzinger8, Kevin Noone9, and
Jean P. B. Ometto4
1
National Institute for Space Research (INPE), Brazil; 2National Center for Atmospheric
Research (NCAR), Boulder, CO, USA; 3NOAA-CIRES and NCAR, Boulder, CO, USA
and Geophysical Institute, University of Bergen, Norway; 4National Institute for Space
Research (INPE), Brazil; 5Meteorological Research Division of Environment Canada,
Montreal, Canada; 6Earth System Science Interdisciplinary Center, University of
Maryland, USA; 7IGBP/Analysis, Integration and Modeling of the Earth System
(AIMES) Office, Boulder, CO, USA; 8International Geosphere-Biosphere Programme
(IGBP), Stockholm, Sweden; 9University of Stockholm, Sweden.
Submitted to the Bulletin of the American Meteorological Society - BAMS
August 2009
Capsule: A prediction system that integrates physical, biogeochemical and societal
processes in a unified Earth system framework and aspires to account for interactions
between weather, climate, biogeochemistry and society.
Abstract
This paper addresses the role of the Earth-system biosphere and highlights the complex:
interactions between the biosphere and the atmosphere in the Amazon Basin, and the
impacts of changes in nitrogen cycling, ocean chemistry, and land use. It introduces three
important requirements for accelerating the development and use of Earth system
information. The first is to develop Earth-system analysis and prediction models that
account for physical, chemical and biological processes, and their interactions in the
coupled atmosphere-ocean-land-ice system. The development of these models requires
partnerships between academia, national research centers, and operational prediction
facilities, and will build upon the accomplishments in weather and climate prediction
systems. They will highlight the regional aspects of global environmental change, and
include modules for water resources, agriculture, forestry, energy, air quality, health, etc.
The second requirement is to model interactions between humans and the weatherclimate-biogeochemical system. The will introduce novel methodologies that account for
societal drivers, impacts and feedbacks. This is a challenging endeavor requiring creative
solutions and some compromises because human behavior cannot yet be fully represented
within the framework of present-day physical prediction systems.
1. Introduction
It is increasingly evident that the consequences of human activities have interfered with
the natural forces regulating the Earth system, potentially endangering the system’s longterm sustainability. There is evidence that these activities have altered atmospheric
composition, climate, and terrestrial, coastal and marine ecosystems, and are responsible
for increased emissions of greenhouse gases and aerosols, as well as, e.g., land-use
changes and the detrimental impacts of nitrogen-fertilizer production and its applications.
There is a risk that human influences on natural processes will exceed the limits for
human and environmental security. Species and ecosystems are threatened by habitat
loss, over-harvesting, air, water and soil pollution and climate change. In some cases, the
interactions of species and ecosystems with weather, water and climate systems further
aggravate the effects. Immediate and concerted actions are required to mitigate and adapt
to recent and anticipated changes.
Humans benefited from intervening in the natural evolution of the Earth system. These
interventions include energy production, industrialization and urbanization, water
irrigation, food production, health improvement, etc. Many of these benefits have been
unequally distributed amongst human societies and some have come at considerable cost
to the environment and humans, e.g., those associated with coal strip mining,
deforestation, and loss of air and water quality. Not all of these costs are fully recognized
within the political and economic frameworks that guide development decisions,
including how these compounding costs will burden future generations. Estimates of the
value of the services that ecosystems provide and recognition that these services have no
substitutes suggest the need for identification and pursuit of more robust and sustainable
development pathways. Sustainable development requires adapting to changing
conditions without compromising the wellbeing of current and future generations.
Earth-system science assesses and seeks to predict the natural and human-driven
processes, including their interactions on the evolution and habitability of the planet. Its
goal is to develop and deliver knowledge to guide effective responses to global and
regional hazards and changes. This requires recognizing that the Earth system is
interactive, involving its atmosphere, oceans, ice, land, biogeochemical, and human
components, and that the Earth system can be advertently and inadvertently perturbed by
human activities,
with both positive and negative consequences. It is essential to
accelerate the development and implementation of advanced global-to-regional
monitoring and prediction systems that provide quantitative information for effective
mitigation of and adaptation to global environmental change. As noted in Shapiro et al.
(2010), effectively predicting the evolution of the full Earth system in a way that
embraces the next frontier of socio-economic and environmental applications demands
international commitment and coordination.
This paper proposes the development of a prediction system that integrates physical,
biogeochemical and societal processes in a unified Earth system framework. Section 2
introduces the role of the biosphere. Section 3 discusses the development of Earth-system
prediction models. Section 4 presents future directions for accounting for the complexity
of the Earth system in sufficient detail and yielding useful projections and predictions of
its future states, on the basis of a range of different scenarios and assumptions.
2. The Role of the Biosphere
The biosphere is the “life zone” of the Earth system. It embraces all living things,
including their two-way interactions with the geophysical and biological elements within
the lithosphere (solid Earth), the hydrosphere and the atmosphere. Until recently, the
biosphere was primarily studied within the context of geophysical influences, with lesser
attention to the feedbacks of biospheric processes on weather and climate. There exist
many active biogeochemical feedback systems within the Earth system, several of which
exhibit highly nonlinear behavior. Abrupt changes can be triggered by both natural and
human activities. There can be “tipping points” of the Earth system that society may not
want to transgress (Steffen, et al., 2003; Lenton et al., 2008; Rockström et al., 2009).
Examples of planetary-to-regional interactions linking human activities, ecosystems,
biogeochemical cycles and the physical weather/climate system include:
Biosphere-Atmosphere Interactions in the Amazon Basin: Modeling experiments
suggest that there may exist an vegetation-climate bi-stability in the Amazon basin in
which two stable states emerge: a state dominated by tropical forests and a second state
subsuming a mix of tropical forests and open savannas (Oyama and Nobre 2003 and
Salazar, et al. 2007). This finding is consistent with paleo-climatic evidence of savanna
replacing parts of the tropical forest in the Holocene (Mayle and Power, 2008). The
current energy and water balance within the Amazon basin is driven by rainforest
vegetation albedo and evapotranspiration. There is the potential that external forcing by
global climate variability and change or by land cover change (deforestation) (Sampaio et
al., 2007) could transition the stable state of forests into the stable state of forests and
savannas (Nobre and Borma, 2009), resulting in a radically altered energy balance of the
region. Surface energy and hydrology balances modulate the strength and location of
organized convective cloud systems over the Amazon rainforest, which in turn modulate
the strength and location of the Inter-Tropical Convergence Zone (ITCZ). The associated
changes in the tropical convective heating and momentum flux can alter the intensity and
location of the northern and southern hemispheric subtropical jet stream, potentially
affecting sub-tropical and mid-latitude weather patterns.
Model simulations show that anthropogenic changes in the Amazon basin biosphere, e.g.,
deforestation, affect surface temperature and precipitation as far away as North America,
the African continent, and the Himalayan region, that in turn influence the African and
Asian monsoon circulations (Nobre et al., 1991; Gedney and Valdes, 2000; Werth and
Avissar, 2002). In this way, changes in one region can reverberate throughout the entire
Earth system.
Changes in Nutrient (Nitrogen) Cycling: As recently as the 1960s, nitrogen was
primarily controlled by natural processes, and biological N2-fixation in supported up to 2
billion people on Earth.. Today, anthropogenically produced reactive nitrogen, i.e.,
industrial N2-fixation for fertilizer production, biological N2-fixation associated with
leguminous crops and NOx emissions from fossil fuel combustion exceeds natural
sources to terrestrial ecosystems (Galloway et al. 2008). Anthropogenic reactive nitrogen
has ledto massive increases in food production. The mobility of nitrogen compounds can
also produce unanticipated and detrimental consequences substantial distances from the
original emission location, at several geographical scales. Nitrogen compounds travel
through the Earth’s atmosphere and aquatic systems, and can be transported by humans in
food and animal feed by international commerce.
The exponential increase in anthropogenic nitrogen is a significant factor in climate
change, through direct effects of aerosol-N, NOx and N2O on climate and through
linkages between the N and C cycles. Furthermore, excess N contributes to loss of
terrestrial biodiversity, and causes a variety of pollution problems, including increasing
surface-level ozone and eutrophication and related effects, in freshwater- and marine
ecosystems, including anoxic “dead zones” in coastal marine systems. Atmospheric
transport of anthropogenic nitrogen accounts for approximately one third of the open
ocean’s external (non-recycled) nitrogen supply and up to three percent of the annual new
marine biological production (Duce et al. 2008). By contrast, the shortage of nitrogenbased fertilizer in many developing countries has contributed to food insecurity and
continuing degradation of land fertility (Vitouseck et al, 2009). It is a major priority to
optimize the use of nitrogen to promote human sustainability, while minimizing its
harmful environmental, including climate, impacts.
Earth system models need to
incorporate the multiplicity of effects of disturbances in the nitrogen cycle at regional and
global scales on land, atmosphere, and marine processes, including the associated
feedbacks between the latter. This includes the coupling of N, C and other elemental
cycles in global climate models.
Changes in Ocean Biosphere-Atmosphere Interactions:
Physical and biological
processes in the oceans have important feedbacks on climate and CO2 fluxes globally.
For example, the amplitude and phasing of the El Niño-Southern Oscillation (ENSO)
from year to year in the equatorial Pacific Ocean affects seasonal climate, including
rainfall and temperature, and terrestrial ecosystem CO2 fluxes in many areas of the world.
El Niño episodes lead to less precipitation and higher temperatures in northern Brazil and
equatorial Africa resulting in suppression of vegetation and enhanced soil decomposition
thus increasing CO2 release to the atmosphere (Qian et al. 2008). Such changes in
temperature and precipitation related to ENSO have societal impacts on agricultural
production, water availability, and energy demand, among others, which affect terrestrial
CO2 fluxes. Biological processes in the oceans affect climate through the production of
volatile organic compounds e.g., dimethyl sulfide and isoprene, which are oxidized in the
atmosphere forming secondary organic aerosols, which in turn modify the chemical
composition and size distribution of marine cloud condensation nuclei, thus affecting
cloudiness and decreasing the radiative flux to the Earth’s surface (Charlson et al. 1987;
Meskhidze and Nenes 2006). As noted by Shapiro et al. 2009, the nutrient fertilization of
the ocean by sand and dust mobilized by synoptic and mesoscale weather systems over
the Sahara, Gobi and Australian deserts, has a dramatic influence on the ocean biology,
which feeds back into ocean radiative and carbon cycle processes.
The oceans provide an extremely valuable function by removing CO2 from the
atmosphere. The oceans removes approximately 25% of the atmospheric carbon dioxide
(CO2) released by human through dissolution of CO2 into sea water and the uptake of
carbon by marine organisms. However, when CO2 dissolves in the oceans, it increases
the acidity (decreases pH) of the surface water, which has a detrimental effect on some
marine organisms. If the pH decreases sufficiently, the calcium carbonate produced by
some marine organisms to make their solid shells (aragonite, a meta-stable form of
calcium carbonate) becomes soluble. Through this process, parts of the southern ocean
could become corrosive to aragonite by as early as 2050-2060 (Orr, et al., 2005), leading
to substantial changes in ecosystem composition and dynamics. Ocean acidification
threatens the health of coral reefs, as well as the bio-diverse and very important
ecosystems associated with them (Guinotte et al., 2003). Many other marine organisms
(e.g., foraminifera, pteropods) that form aragonite could be severely impacted,
presumably with effects up the food chain, leading to negative consequences for humans
dependent on fish and marine products as a primary source of protein.
Changes in land use and land cover: For thousands of years, humans have modified
land for food, water, habitat and the pursuit of happiness (Redman 1999; Thomas 1956).
Current rates, extent, and intensity of land use and land cover changes are driving
unprecedented changes in ecosystems and environmental processes at local, regional and
global scales (Turner et al., 2007). Recent widespread human transformation of once
highly diverse natural ecosystems into relatively species-poor managed ecosystems
places great pressure on environments, potentially altering ecological functions and life
support services that are vital to the well-being of human societies (Steffen et al., 2004).
Reductions in biodiversity can alter both the magnitude and stability of ecosystem
processes and human livelihoods (Constanza et al., 2007). Processes that influence plant
productivity, soil moisture and fertility, water quality, atmospheric composition and many
other local and global environmental conditions become forcings and responses to
weather, and climate change that ultimately affect human welfare and the health of the
planet (Steffen et al., 2004).
3. Earth-System Prediction Models
Earth-System Models (ESM) are being developed at different levels of complexity in
order to advance knowledge of mechanisms and their interactions that determine the
evolution of the Earth system at the global and regional scales. These models are testbeds for simulations of the coupled natural-human system to address issues beyond
physical weather and climate systems. They provide a predictive capability for, e.g.,
marine and terrestrial ecosystems, surface hydrology, the needs of different
socioeconomic sectors, such as agriculture, industry, energy, health, etc. ESMs are also
suited to assess the value of observations within a predictive context through data
assimilation approaches applied to the coupled Earth System. They can identify and
optimize the observational systems required for sustained monitoring and improved
prediction from days to decades.
A key challenge is the incorporation of human interactions within for next-generation
ESMs to provide the coupling between physical weather/climate models, impact models
and integrated assessment models.
Figure 1 is a conceptualization of the diverse
elements of a unified ESM system, and the important links between existing models and
observations of the atmospheric, ocean and land processes. Such tools will play a key role
in assessing risks and identifying potential societal hazards as well as opportunities.
TowardsOperational Earth System Monitoring, Assimilation
and Prediction Systems
Atmosphere
Models
The Earth System
Unifying the Models
Climate / Weather
Models
The Predictive
Earth System
Hydrology
Process
Models
Ocean
Models
Land
Surface
Models
Natural Hazard
Prediction
Terrestrial
Biosphere
Models
Megaflops
Gigaflops
Teraflops
2000
Petaflops
2010
Fig. 1: A conceptualization of the development of a complex Earth-system analysis and
prediction system for assimilating observations, and providing experimental and operational
forecasts for the state of the atmosphere, ocean and land. The investments in high-capacity
computers are critical in order to facilitate combining the diverse interacting components at high
spatial resolution and so provide a dynamical view of the complex evolution of the Earth system
and of its coupled interaction with human activities.
Figure 2 presents a conceptual modelling system that couples the physical climate system
with biogeochemical cycles (greenhouse gases), atmospheric chemistry, aerosol
microphysics, ecosystem dynamics and hydrology in an environment affected by
anthropogenic emissions, land-use change and fires, urbanization and perturbations in
land hydrology through activities such as damming and irrigation. It encompasses key
physical, biological and chemical interactions. The introduction of such complex
processes in Earth-system prediction systems is a challenge that can only be met if key
processes are simultaneously studied in the laboratory and through field experiments and
observational programs to assess their representations in the models. It is also well
recognized that weather processes have significant up-scale effects on the evolution of
climate processes in the Earth system, and vice-versa (Brunet et. al, 2010). Many of the
envisaged ESM applications will address prediction problems that will encompass time
scales from few hours to decades. Hence ESM development will necessitate an
unprecedented multi-disciplinary collaborative effort between the climate, weather and
biosphere scientific communities.
Fig. 2: Interactions between the physical climate, greenhouse gases, aserosols, gas-phase
atmospheric chemistry, ecosystem dynamics, land-use and the water system. These interactions
have to be included in the formulation of Earth system models (Source: Peter Cox, personal
communication).
A noteworthy accomplishment in Earth system predictions on short time scales (days) is
the implementation of the European GEMS/MACC Project coordinated by the European
Center for Medium-range Weather Forecast (ECMWF). ECMWF is now issuing global
and regional daily predictions of the atmospheric composition 1 . Such products will help
1
http://gems.ecmwf.int/
reduce the vulnerability of populations to air pollution, particularly in urbanized areas. It
provides information on the short- and longer-term evolution of biogeochemical cycles
under a changing climate, such as the carbon cycle, for instance. An example of the
predicted global distribution of a reactive gas such as carbon monoxide (CO) is shown in
Figure 3. In the case of carbon dioxide (CO2), observations from the Atmospheric
Infrared Sounder (AIRS) are assimilated by ECMWF in their model. Global budgets of
carbon dioxide have been derived by inverse modeling. Figure 3 also shows the monthly
average surface exchanges of CO2 estimated for July 2005. The atmospheric transport has
been computed by the general circulation model of the Laboratoire de Météorologie
Dynamique at a resolution of 3.75×2.5 deg2 (longitude-latitude), nudged to the ECMWF
analyzed winds. The CO2 atmospheric reanalysis is in the form of 6-hourly, twodimensional mean concentration fields in the free troposphere, where the AIRS CO2
weighting function peaks.
Fig. 3: Top: Predicted global distribution of carbon monoxide surface mixing ratio [ppb] by
ECMWF (EU-funded GEMS Project) with assimilation of space observations. Bottom: Monthly
mean exchange surface flux of carbon [gC m-2 day-1] derived from atmospheric CO2 observations
by the Atmospheric Infrared Sounder (AIRS) and atmospheric transport calculated using winds
from the ECMWF reanalysis.
An example of a regional Earth system model (Figure 4) is the Chesapeake Bay Forecast
System (CBFS) developed at the University of Maryland’s Earth System Science
Interdisciplinary Center. The CBFS provides decision support to enable policy makers,
urban development planners, natural resource managers and planners, as well as a variety
of private users, to base their decisions on a convenient information system. The
Chesapeake Bay is undergoing climate and environmental change, and will continue to
undergo such change over the next 50 years, due to its exposure to sea-level rise, past and
continuing changes in land use and land cover change, fishing pressure, and increasing
population density. A paramount question facing decision makers in the region is how to
prepare and adapt to the anticipated change in the coming decades. The CBFS is based on
downscaled coupled land-atmosphere-ocean-ecosystem models, combined with remotelysensed observations. It provides integrated Earth system analyses and prediction
capabilities for the Chesapeake Bay and its watershed, with products designed to address
user needs at time scales from days to decades.
Fig. 4: The Chesapeake Bay Forecast System Decision Support Interface depicting the land
cover/land use types at 30 m resolution incorporated into the coupled ocean-atmopshere-landecosystem model of the Chesapaeake Bay watershed (left), changes to nitrogen loading to the
Bay if all run off from poultry farms is remediated by 2018 (middle), and present nitrogen
concentrations in Chesapeake Bay (right).
Current coupled weather-climate-Earth system models highlight natural processes, and
treat human actions as external forcing parameters. A major task for the next decade is to
develop methodologies by which qualitative information associated with human activities
can be includedr in prediction models. Natural and social science knowledge and
approaches must be joined to address a range of environmental. For instance, to improve
climate projections and their relevance, it is important to account for relevant social
dynamics in coupled biogeochemical cycle climate models. A possible path forward is to
couple reduced-form GCMs (models of lesser complexity) with social dynamics in a nonstructural way, creating a range of defined scenarios through integration of information
about possible pathways, related to energy sources, consumption, and land use, for
instance, with what is known about the role of path-dependency, human psychology, and
other factors bearing on behavioral tendencies.
Earth-system models of intermediate complexity provide a starting point for the
development of models that couple natural and social knowledge, and allow for the
development of new methodologies. While such models do not simulate in detail all the
processes involved, they can provide helpful heuristics allowing societies to anticipate
and consider in greater depth the trade-offs between various energy paths, among other
things. Figure 5 shows a conceptual view of a modeling framework that seeks to account
not only for the influences of human actions on the natural systems – as done in classic
approaches – but also for the impacts of environmental goods and services on human
welfare. Important feedbacks in the Earth system occur through societal responses.
Fig. 5: A coupled human-environmental modeling framework that accounts not only for
the influences of human actions on the natural systems, but also for the impacts of
environmental good and services on human welfare.
4. Summary and Future Directions
This paper introduced three important requirements for accelerating the development and
use of Earth system information. The first is to develop Earth system analysis and
prediction models that account for multi-scale weather and climate processes, chemical
and biological processes, and their interactions in the coupled atmosphere-ocean-land-ice
system. These models require partnerships between academia, national research centers,
and operational analysis and prediction facilities. They will build upon accomplishments
in weather and climate predictions. The second is to model the interactions between
humans and the weather-climate-biogeochemical system. This includes a global modeling
framework that incorporates all aspects of physics, biology and chemistry and will
highlight the regional aspects of global change, and include modules for water resources,
agriculture, forestry, energy, air quality, health, etc. The third is to introduce appropriate
methodologies that account for societal drivers, impacts and feedbacks. This is a major
endeavor because the laws governing human behavior are not easily represented within
the framework of present-day physical prediction systems.
The development of comprehensive high-resolution weather and climate models (Brunet,
et al., 2010; Moncrieff et al., 2010; Shapiro et al., 2010; Shukla et al., 2010) is critical to
realization of the full benefit of the Earth system models. To successfully meet these new
challenges, the support of teams that are both disciplinary and interdisciplinary, national
and international, will be essential, as will the availability of high performance computing
facilities. On this point, the last sixty-year track record of the science of weather and
climate prediction provides a solid base for confidence.
The current awareness of the state of the biosphere and availability of environmental
resources and services drive demand for better ESM simulations. The expectation is that
societies will benefit from sector- and user-specific information about such things as air
quality and availability and quality of water. ESMs and their predictions can be useful as
a basis for improved management of the Earth system, including its natural resources, at
regional and global scales. The cost of the issues calculated by ESMs, combined with the
cost of taking action, can inform deliberation as to possible solutions (for instance, in the
areas of adaptation, mitigation and geo-engineering). Despite the progress in ESMs and
the vast increase of available observations, there are still important gaps in our
knowledge of the Earth system. These gaps reduce the usefulness of the model output for
policy processes, and suggest the need for major advances in monitoring, analysis and
prediction systems, combined with accelerated research to address unresolved processes
and feedbacks in the Earth system.
5. References
Brunet, G., M. A. Shapiro, B. Hoskins, M. Moncrieff, R. Dole, G. N. Kiladis, B. Kirtman,
A. Lorenc, B. Mills, R. Morss, S. Polavarapu, D. Rogers, J. Schaake, and J. Shukla,
2009: Toward a seamless process for the prediction of weather and climate: the
advancement of sub-seasonal to seasonal prediction. Bull. Amer. Met., this issue.
Charlson, R.J., J.E. Lovelock, M.O. Andreae and S.G. Warren. 1987. Oceanic
phytoplankton, atmospheric sulphur, cloud albedo and climate. Nature 326: 655-661.
Costanza, R., L. J Graumlich, W. Steffen, C. Crumley, J. Dearing, , K. Hibbard, R.
Leemans, C. Redman, and D. Schimel, 2007: Sustainability or collapse: What can we
learn from integrating the history of humans and the rest of nature?, Ambio,
36[7],522-517.
Duce, R. A., J. LaRoche, K. Altieri, K. R. Arrigo, A. R. Baker, D. G. Capone, S. Cornell,
F. Dentener, J. Galloway, R. S. Ganeshram, R. J. Geider, T. Jickells, M. M. Kuypers,
R. Langlois, P. S. Liss, S. M. Liu, J. J. Middelburg, C. M. Moore, S. Nickovic, A.
Oschlies, T. Pedersen, J. Prospero, R. Schlitzer, S. Seitzinger, L. L. Sorensen, M.
Uematsu, O. Ulloa, M. Voss, B. Ward, and L. Zamora, 2008: Impacts of Atmospheric
Anthropogenic Nitrogen on the Open Ocean, Science 320: 893-897.
Galloway, J.N., A.R. Townsend, J.W. Erisman, M. Bekunda, Z. Cai, J.R. Freney, L.A.
Martinelli, S.P. Seitzinger, and M.A. Sutton. 2008. Transformation of the nitrogen
cycle: Recent trends, questions, and potential solutions. Science. 320:889-892.
Gedney, N and P.J. Valdes, 2000: The effect of Amazonian deforestation on the northern
hemisphere circulation and climate, Geophys. Res. Lett., 27, 12753-12758.
Guinotte, J. M., R. W. Buddemeier, and J. A. Kleypas, 2003: Future coral reef habitat
marginality: temporal and spaeial effects of climate change in the Pacific basin. Coral
Reefs, 22, 551-558.
Lenton, T. M., H. Held, E. Kriegler, J. W. Hall, W. Lucht, S. Rahmstorf, and H. J.
Schellnhuber, 2008: Inaugural Article: Tipping elements in the Earth's climate system.
Proceedings of the National Academy of Sciences, 105, 1786-1793.
Mayle, F. E. & Power, M. J. 2008 Impact of a drier Early- Mid-Holocene climate upon
Amazonian forests. Phil. Trans. R. Soc. B 363, 1829–1838. (doi:10.1098/rstb.
2007.0019).
Meskhidze, N. and A. Nenes. 2006: Phytoplankton and cloudiness in the Southern
Ocean. Science 314: 1419-1423.
Moncrieff, M., D. Waliser, M.A. Shapiro, 2009: The mesoscale organization of tropical
convection and its interaction with the global circulation: the Year of Tropical
convection (YOTC). Bull. Amer. Met. Soc., this issue.
Nobre, C. A. ; Sellers, P. ; Shukla, J., 1991: Amazonian Deforestation and Regional
Climate Change. Journal of Climate, v. 4, n. 10, p. 957-988k.
Nobre, C. A. and L. Borma, 2009: ‘Tipping Points’ of the Amazon Forest. Current
Opinion on Environmental Sustainability (in press).
Orr, J. C., V. J. Fabry, O. Aumont, L. Bopp, S. C. Doney, R. A. Feely, A. Gnanadesikan,
N. Gruber, A. Ishida, F. Joos, R. M. Key, K. Lindsay, E. Maier-Reimer, R. Matear, P.
Monfray, A. Mouchet, R. G. Najjar, G.-K. Plattner, K. B. Rodgers, C. L. Sabine, J. L.
Sarmiento, R. Schlitzer, R. D. Slater, I. J. Totterdell, M.-F. Weirig, Y. Yamanaka, and
A. Yool, 2005: Anthropogenic ocean acidification over the twenty-first century and its
impact on calcifying organisms. Nature, 437, 681-686.
Oyama, M. D. and C. A. Nobre, 2003: A new climate-vegetation equilibrium state for
tropical South America. Geophys. Res. Lett., 30, 2199.
Qian, H., J. Renu, and Z. Ning. 2008. Response of the terrestrial carbon cycle to the El
Niño-Southern Oscillation. Tellus 60B, 537-550.
Redman CL, 1999. Human Impact on Ancient Environments. Univ of Arizona Press,
Tucson, AZ.
Rockström, J. et al., 2009. Planetary Boundaries: Exploring the safe operating space for
humanity in the Anthropocene. Nature (in press).
Shapiro, M. A., and others, 2010: An Earth-System Prediction Initiative for the 21st
Century, Bull. Amer. Meteor. Soc., this issue.
Shukla, J., T.N. Palmer, R. Hagedorn, B. Hoskins, J. Kinter, J. Marotzke, M. Miller, J.
Slingo, 2009b: Climate Prediction from Weeks to Decades in the 21st Century:
Towards a New Generation of World Climate Research and Computing Facilities.
Bull. Amer. Meteor. Soc., this issue.
Salazar, L. F., C. A. Nobre, and M. D. Oyama, 2007: Climate change consequences on
the biome distribution in tropical South America. Geophys. Res. Lett., 34, L09708.
Sampaio G, Nobre C, Costa MH, Satyamurty P, Soares-Filho BS, Cardoso M, 2007:
Regional climate change over eastern Amazonia caused by pasture and soybean
cropland expansion. Geophys.l Res. Lett., 34 (L17709).
Snyder, P. K., C. Delire, and J. A. Foley, 2004: Evaluating the influence of different
vegetation biomes on the global climate. Climate Dynamics, 23, 279-302.
Steffen, W., A. Sanderson, P. Tyson, J. Jäger, P. A. Matson, B. Moore III, F. Oldfield, K.
Richardson, H. J. Schellnhuber, B. L. Turner II, and R. Wasson, Eds., 2004: Global
change and the earth system: a planet under pressure. Global change - the IGBP
series, Springer-Verlag, 336 pp.
Thomas WM, Jr, ed., 1956: Man’s Role in Changing the Face of the Earth. Univ of
Chicago Press, Chicago.
Turner et al., 2007: Land Change Science Special Feature: The emergence of land change
science for global environmental change and sustainability. Proceedings of the
National Academy of Sciences, vol. 105, issue 7, pp. 2751-2751.
Vitousek, P. M., R. Naylor, T. Crews, M. B. David, L. E. Drinkwater, E. Holland, P. J.
Johnes, J. Katzenberger, L. A. Martinelli, P. A. Matson, G. Nziguheba, D. Ojima, C.
A. Palm, G. P. Robertson,P. A. Sanchez,A. R. Townsend,F. S. Zhang, 2009: Nutrient
Imbalances in Agricultural Development. Science, 340, 1519-1520
Werth, D. and R. Avissar, 2002: The local and global effects of Amazon deforestation, J.
Geophys. Res., 107, D20, 8087, doi:10.1029/2001JD000717.
Climate Prediction from Weeks to Decades in the 21st Century:
Towards a New Generation of World Climate Research and
Computing Facilities
J. Shukla, T.N. Palmer, R. Hagedorn, B. Hoskins, J. Kinter, J. Marotzke, M. Miller, J. Slingo
Abstract
The impending threat of global climate change and its regional manifestations are
among the most important and urgent problems facing humanity. Society needs accurate
and reliable estimates of changes in the probability of regional weather variations to
develop science-based adaptation and mitigation strategies. Recent advances in weather
prediction and in our understanding and ability to model the climate system suggest that it
is both necessary and possible to revolutionize climate prediction to meet these societal
needs. However, the scientific workforce and the computational capability required to
bring about such a revolution is not available in any single nation.
Motivated by the success of internationally-funded infrastructure in other areas of
science, this paper argues that, because of the complexity of the climate system, and
because the regional manifestations of climate change are mainly through changes in the
statistics of regional weather variations, the scientific and computational requirements to
predict its behavior reliably are so enormous that the nations of the world should create a
small number of multi-national research and high performance computing facilities
dedicated to the grand challenges of predicting climate change on both global and regional
scales over the coming decades. Such facilities will play a key role in the development of
next generation climate models, build global capacity, nurture a highly trained workforce,
and engage the global user community, policymakers and stakeholders.
We recommend the creation of a small number of highly connected multi-national
facilities with computer capability at each facility of about 20 petaflop in the near future
and about 200 petaflop by the end of the next decade. Each facility should have sufficient
scientific workforce to develop and maintain the modeling software and data analysis
infrastructure. Such facilities will enable future IPCC assessment to be made using about
10 km resolution climate models; seasonal and decadal predictions using 3-5 km cloud
system resolving atmosphere models and eddy revolving ocean models; and fundamental
research on weather-climate interactions using about 1 km resolution models. Each facility
should have enabling infrastructure including hardware, software and data analysis
support, and scientific capacity to interact with the national centers and other visitors. This
-1-
will enable the climate community to provide society with climate predictions which are
based on our best knowledge of science and the most advanced technology.
1. Why do we need to revolutionize climate predictions?
Weather and climate conditions are undisputedly major factors for the well-being and
development of society, impacting on all scales from individual lives to global economies
(Sachs, 2008). Human societies have flourished by adapting to and taking advantage of
current climate conditions. However, it has been estimated that during the past 25 years,
weather-related disasters have caused more than 600,000 fatalities and $1.3 trillion of
economic losses. This demonstrates humanity’s dependence on favorable weather and
climate conditions, and reminds us of the fragility of the relationship between climate and
society.
Extensive research by the world’s climate community assessed by the
Intergovernmental Panel on Climate Change (IPCC), has established with a high degree of
confidence that humans are affecting Earth’s climate. It is now recognized by scientists and
world leaders that climate change is perhaps the gravest of the many threats that humanity
will have to face in the coming century. Considering the increasing frequency of extreme
weather and climate events together with our enhanced vulnerability to weather and
climate hazards, caused by rapid economic and population growth, humanitarian and
economic losses will continue to rise. As the Stern Report on the economics of climate
change has emphasized, (Stern, 2007), climate change is a trillion-dollar problem: while
cutting emissions of greenhouse gases to mitigate climate change could cost the world
economy by as much as 1% of its GDP, the effects of inaction will be many times greater
than this.
IPCC reports have alerted society to the risk the world faces from climate change,
and climate-change related policies are beginning to be formulated by governments around
the world. However, formulating cost-effective and responsible mitigation and adaptation
strategies requires reliable answers to specific questions about the details of climate
change. How far can greenhouse-gas concentrations rise before dangerous climate changes
are inevitable? Since greenhouse-gas concentrations are already at unprecedented levels,
what level of investment is needed for society to adapt to inevitable climate changes? How
will climate change at the regional level not just in terms of temperature, but other key
variables such as precipitation and storminess? For example, what is needed to ensure that
society in regions at risk of increased levels of drought will have adequate water supplies,
while those in regions at risk of increased levels of precipitation are sufficiently protected
against flooding? Questions like: “Is New Orleans sustainable?” or “When should we
begin planning to re-accommodate the hundreds of millions of climate refugees from lowlying areas of the globe, left homeless if the Greenland and West Antarctic ice sheets begin
to disintegrate?” will have to be answered. What if society is unable to substantially reduce
CO2 emissions in the coming centuries? There are centuries of coal ready to mine and
-2-
many experts feel it will be difficult to prevent developing countries from burning it,
especially if their governments’ prime aspiration is to improve the standard of living of its
citizens to be comparable to those of the developed countries. If climate change proves to
be as severe as some estimates suggest, we may be forced to consider radical measures to
offset global warming. Such measures might not only be expensive, but also potentially
dangerous to the Earth system.
Answering such questions will require detailed quantitative predictions of climate
change for the coming century on both global and regional scales, and with sufficient detail
to provide statistics on hazardous weather events. In view of the critical role that climate
prediction will play in determining governmental strategies worldwide on climate
mitigation and adaptation and the enormous economic impact the decisions will have on
society, it is imperative that the predictions have a high level of reliability. Indeed there
are several features of climate change for which the current IPCC models show robust
levels of consistency. However, large disagreement among the individual model
predictions which form the basis of the IPCC reports, points to the fact that there remains
considerable uncertainty in the magnitude of future climate change, especially on the
regional scale. Furthermore, the inability of current climate models to describe accurately
key modes of variability of climate indicates common model deficiencies. This makes the
long-term prediction of regional climate even more uncertain.
At the World Modelling Summit for Climate Prediction (Shukla et. al., 2009),
attended by a large number of scientists from many countries (http://wcrp.wmo.int), it was
declared that the scientific community currently lacks the appropriate tools (models and
computers) to make the step change in climate prediction required to give fully trustworthy
answers to the questions posed by society. On the other hand, there is universal agreement
among scientists and policy makers that sustainable development of the world’s societies
will require science-based adaptation strategies. There is also a near-universal agreement
among scientists that climate models must be run at a sufficiently high resolution to be able
to accurately represent the key regional climate processes in the atmosphere, and in the
oceans. But the constraint of limited climate-dedicated computing power at the national,
institutional or university level has meant that contemporary climate models have to resort
to highly truncated representations of the mathematical laws of physics. Currently,
scientists have to over-simplify key physical processes due to lack of sufficient computing
power needed to represent them adequately. Hurricanes are good examples to illustrate this
point. It is perhaps not widely known that, unlike the daily weather forecasts, the forecasts
of the number and strength of hurricanes in the coming season issued by national weather
services are not based on prediction methods which use the laws of physics. Instead they
use empirical methods based on the past performance of certain pre-selected indicators
(e.g. Bell et al. 2008). Current global climate models used to predict climate change
likewise do not resolve hurricanes. Since the eyewall of a hurricane is about 10-20 km
-3-
wide, it is impossible for climate models with a spatial resolution of 200 km to describe
hurricanes realistically. To interpret possible hurricane behavior from the output of these
models, various proxies for hurricanes are used and conclusions are drawn about
hurricanes in a future climate based on the behavior of these proxies.
From both simple theory and by inspection of diagnostics of model-generated
energy spectra, the scales reasonably described by a numerical model of the atmosphere or
ocean are many times larger than the nominal grid length. This means that typical climate
model resolution (with 200 km grids in the atmosphere and 100 km grids in the ocean)
cannot represent many sub-synoptic scale systems at all, and only poorly represent many
smaller baroclinic features. The atmospheric storm tracks in these models underestimate
the number of storms actually observed, and the statistics of mid-latitude blocking are
poorly simulated (Jung et al., 2007). Use of much higher resolutions can improve both the
description of important structures within synoptic weather systems, further refine the
blocking skill and provide opportunities to capture the true intensity of the highly energetic
mesoscale systems both in the tropics and higher latitudes.
It is often said that high-resolution regional models, widely used to provide climate
impact assessments at the regional and local level, can be used to provide information
including, for example, the changing frequency and intensity of hurricanes This is a
questionable concept in the context of climate change (WCRP, 2007). Regional
downscaling cannot capture the global telecommunications of regional variations and
cannot improve this aspect of regional prediction. For example, the effects of increases in
CO2 are felt by hurricanes through changes in the large-scale circulation of atmosphere and
oceans, and the precursors of hurricanes are often formed far away from the regions
ultimately affected. It is well known that the frequency of Atlantic hurricanes is strongly
influenced by Pacific Ocean temperatures. It can be argued that only a global-high
resolution climate model can estimate the changes in the frequency and intensity of
hurricanes, and indeed other hazardous and high impact weather events, in a changing
climate. We believe it is unacceptable that society should have to make major decisions
about the habitability of hurricane-affected coastal areas using information from climate
models which cannot resolve hurricanes; especially when scientific knowledge and
technology now exist which would allow more informed decisions to be made.
The requirement to increase resolution of climate models is based on considerable
experience with weather and short term climate prediction where predictive skill has
steadily improved during the past 30 years as models use higher resolution and better
representations of physical processes. The experience by Japanese research groups to
simulate global weather by use of a most powerful supercomputer, the Earth Simulator,
with horizontal resolution of 3.5-10km has established the scientific basis for very highresolution models.In the last 20 years or so, operational weather centers have made several
significant horizontal resolution changes (with concomitant changes in vertical resolution).
-4-
Each change to higher resolution improves accuracy in a) the representation of basic
components such as orography and land/sea definition, b) synoptic and sub-synoptic
systems, c) weather features and parameters such as fronts, cloud and rain bands, jets etc.,
and d) assimilating observations, both space-based and surface-based. Each change also
contributes significantly to the long-term positive trends in objective forecast skill
measures which are seen in the major numerical weather prediction (NWP) centers’
performance statistics. Furthermore there is little doubt that further refinements in
resolution, will continue these improvements.
In addition to the intrinsic merits of running climate models at a resolution
comparable with that of NWP models, the continual confrontation of an NWP model with
observations can provide important constraints on much longer timescale climate
predictions, when the same model is used in climate mode. That is, an ability to apply the
insights and constraints of numerical weather prediction to climate is one of the key
motivations for the development of so-called seamless prediction systems, as first
discussed by Palmer and Webster (1995). Examples of the use of seamless prediction
techniques to calibrate probabilistic climate change projections have been given by Palmer
et al (2008).
Short-range (limited-area) forecast models at ultra-high resolutions are giving some
encouraging results using grid lengths of a few kilometers, without parameterizing deep
convection (e.g. Fudeyasu et al. 2008). Since the features reasonably resolved by a discrete
model have scales more than five times the grid length, such grids are still far from
resolving individual deep convective clouds. Nevertheless, these models better resolve
macrostructures such as downdraft outflows and other mesoscale processes, likely leading
to the perceived benefits. Much more realistic kinetic energy spectra are also captured with
grids of 10 km or less.
However, higher resolution alone is not sufficient to improve simulation of current
climate. Considerable research effort, including the design and implementation of physical
parameterizations, is required to improve the simulation of climate variability using very
high-resolution models. Explicitly resolving deep-convective clouds, which requires
designing and utilizing global climate models with grid resolutions of the order of 1 km,
poses many research issues, and the computational factors challenge the whole algorithmic
design. Exstensive research needs to be done to ensure that once the grid resolution is high
enough (~1 km) to resolve the convective cloud systems, the rapid growth of the
uncertainties in these unpredictable cloud systems do not overwhelm the predictable largest
scale flow.
The requirement to represent properly the processes of deep convective cloud
systems and hurricane rain bands in the atmosphere, and energetic eddies in the oceans, are
only some of several reasons why climate prediction modeling must have substantially
more powerful and dedicated computing and research infrastructure than is currently
-5-
available at the individual institute or university level. Other requirement for substantial
increases in available computer power and enhanced research efforts are summarized as
follows.
•
For predictions on the century timescale, models must represent more than just the
physical climate system; they must also represent a range of biogeochemical
processes, so that the concentrations of radiatively-active gases can be predicted.
Presently, representation of such biogeochemical processes in climate models is
crude, again at least in part because of limitations of computer power.
•
A key prognostic variable for any prediction is uncertainty. Prediction uncertainty
can be estimated by making an ensemble of forecasts, varying uncertain aspects of
the initial conditions, the model equations, and other input fields such as
greenhouse gas concentrations - a process known as ensemble forecasting, now
well established in numerical weather forecasting and seasonal climate prediction.
Due to lack of sufficient scientific and computational capacity, it is currently not
feasible to sample uncertainty adequately with models of sufficient resolution or
complexity.
•
Climate models must be tested by running them over past paleoclimatic epochs, for
example to assess whether they can simulate past glacial/interglacial cycles, and
some of the abrupt climate changes known to have occurred during past glacial
periods. Currently there are insufficient computer resources to run contemporary
climate models over periods much longer than a few centuries.
•
The ingestion of Earth observations into climate models requires the development
of what is known as data assimilation capability. Operational data assimilation is an
exceptionally complex computational process, allowing sophisticated validation
and initialization of climate models, currently impossible in most climate institutes
due to the lack of computer resources. A sustained research effort to develop
assimilation techniques for the total climate system (land, ocean, cryosphere, etc.)
is needed. Consequently, the full value of space-based observations is not being
realized because space-based observations are typically made on scales much
smaller than the resolution of current climate models.
Together with the problem of insufficient resolution, the above considerations
collectively argue for a very substantial augmentation of computer power and research
effort, needed to make reliable, quantitative predictions of regional climate change. For this
reason the participants of the World Modelling Summit concluded that it is time to
accelerate progress in climate modelling in order to be able to fulfill society’s needs and
expectations from climate science.
-6-
2. What is necessary to fulfill society’s expectations?
It is somewhat ironic that the science community uses climate models with insufficient and
inadequate spatial resolution, not because of a lack of knowledge of physics and dynamics
of climate, but because of the lack of powerful computers, and the lack of a critical mass of
qualified scientists working together as a team to build a new generation of weather and
climate models. The scientific basis and knowledge for building such models, and the
evidence for improving predictions using models with higher resolution and better
representation of physical processes has been available for more than 20 years. However,
the underlying infrastructure, which enables climate models to simulate and predict the
current and future climate, exists only in national laboratories. Because current
computational infrastructures are funded for only short periods (typically 5 years), no
single modeling center in the world has been able to plan and acquire the required
supercomputing resources and the critical mass of scientists to build and run climate
models with cloud system resolving atmosphere, eddy resolving oceans, and landscape
resolving land surfaces for predictions over the next century.
2.1 High Performance Computing Requirements
The operational NWP centers have been able to increase the spatial resolution of global
weather forecast models from about 100 km in 1990 to about 20 km today. This has been
possible because of about a 1000-fold increase over the past 25 years in the sustained speed
of computers from about 1 Gigaflop (1 gigaflop = 109 floating point operations per second)
to about 1 Teraflop (1 teraflop = 1012 floating point operations per second). For example,
the computer at the European Centre for Medium-range Weather Forecasts (ECMWF),
with a sustained speed of about 2 Teraflops, is currently able to produce 10-day weather
forecasts in about 20 minutes with a global atmosphere model of about 20 km resolution.
If, for argument’s sake, we were to run the ECMWF weather forecast model as a climate
model coupled to an equally complex and computationally demanding global ocean model,
both with about 4 km resolution, achieving a century of simulation each month, the
sustained speed of the computers would have to increase another 1000-fold to about 2
Petaflops (1 petaflop = 1015 floating point operations per second). This does not take into
consideration the need for additional computing power to include more complexity
(biogeochemistry etc.) which is required to advance from physical climate system models
to physical - chemical - biological Earth system models, and larger ensembles.
The global atmospheric model currently with the highest spatial resolution of about
3.5 km is the NICAM model being run on the Earth Simulator in Japan (Satoh et al. 2008).
With NICAM sustaining 7.4 Teraflops on the Earth Simulator, 24 hours of simulation takes
about 6 hours. As in the example of the ECMWF model, running NICAM at 3.5 km
resolution as a climate change model fully coupled to a global ocean model and integrated
-7-
for a century per month, the capability has to be increased by a factor of 250, which is
similarly about 2 Petaflops sustained.
How can this huge increase in computing capability be achieved? The
manufacturers of high-performance computers predict that within three years, a peak
computing performance of about 20 Petaflops will be achievable (e.g., http://www03.ibm.com/press/us/en/pressrelease/26599.wss). Within the decade following the
introduction of multi-petaflop machines, peak speeds of 200 Petaflops might be achievable,
giving sustained performance of up to 20 Petaflops on typical fluid dynamical codes. Large
domain limited area models with 1 km grids and climate simulations with 3-10 km grids,
with much improved (possibly stochastic-dynamic) parameterizations, are likely to be the
best way forward given the computing potential of the next decade, leading to improved
simulations through better parameterizations, better physical understanding of phenomena,
better appreciation of the shortcomings of current methods and better basic dynamics of the
building blocks of climate.
The ‘grand challenge’ of running global models at 1 km resolution should be
tackled. It is unlikely to be possible to make production runs at this resolution in the next
10 years, but this should not prevent extensive research experimentation on basic weatherclimate interactions at cloud-system permitting resolution.
2.2 Infrastructure Requirements
Using very high performance computing power efficiently requires adequate infrastructure
and human resources. First of all, current climate model software is not designed for the
massively parallel architecture of multi-petaflop machines. This implies that major
recoding will be needed to fully exploit the opportunities offered by available machines.
Such a task can only be handled if a critical number of scientists work together in a
dedicated team. That is, a sufficient number of computing specialists and climate scientists
have to be gathered and motivated to work together to achieve this challenging goal.
Secondly, the data volumes produced by the proposed new generation climate
models will increase by orders of magnitudes. This will pose enormous technical
challenges of archiving, interrogating, analyzing and visualizing the data produced by these
high resolution simulations. Efficient and optimum utilization of the outputs from such
high resolution models by a large number of scientists worldwide will require development
of novel techniques and solutions, requiring the collaboration of many dedicated technical
and scientific experts. Experience shows that the success of major international
collaborations critically depends on the development and adherence to universally accepted
standards and methods. Resources can only be efficiently used if appropriate tools are
available to exploit their potential. Therefore, it is imperative that the facilities proposed
-8-
here comprise both computing and human resources. Only the two components together
will ensure the success of this new endeavor.
3. How can we realize the necessary facilities?
Climate models have traditionally been developed by individual government institutes and
university departments around the world. However, over the last few decades it has
become a well-established tenet that many of the most scientifically and technically
challenging problems can only be tackled by investment in the type of infrastructure
unaffordable at the national level. These days, much cutting-edge scientific research is
achieved through international collaboration. For example, countries have come together to
fund the development of high-energy particle accelerators to probe the nature of matter on
the smallest scales, and powerful telescopes to probe the structure of the universe on the
largest scales. Recently, several countries announced the creation of a single jointly-funded
laboratory called ITER (International Thermonuclear Experimental Reactor) to create
sustainable nuclear fusion on Earth (http://www.iter.org/index.htm). The type of facility
enabled by such multi-national funding would not have been possible at the national level,
and, as a result, neither would the resulting scientific breakthroughs. The success of
international investment in scientific and technical infrastructure is also apparent in
weather and seasonal climate prediction, as illustrated by the European Centre for
Medium-Range Weather Forecasts and the International Research Institute for Climate and
Society.
The notion of funding science infrastructure at the international level is not new.
CERN
(Hermann
et.
al,
1987)
and
the
Hubble
Telescope
(http://en.wikipedia.org/wiki/Hubble_telescope) are outstanding successful examples of
internationally funded infrastructure, and the ITER nuclear fusion facility may prove to be
another. Indeed, the Consultative Group on International Agricultural Research (CGIAR)
is credited with launching the Green revolution leading to sustainable food security and
poverty reduction in the world (Evenson and Gollin, 2003). We assert that the climate
prediction community needs the sort of international infrastructure that particle physicists,
nuclear fusion researchers, astronomers and others now take for granted. We propose to
take climate prediction science into the 21st century through the creation of a small number
of international research and high-performance computing facilities dedicated entirely to
the problem of climate research and prediction. More specifically, we propose facilities
where global climate models can be run over century timescales with resolutions which
allow the key climatic processes, involving the organized deep-convective cloud system in
the atmosphere, and energetic eddies in the oceans to be explicitly resolved. Such machines
will not be available at the laboratory level, and are unlikely to be available at national
levels, in the coming decade. However, we argue that concerted international funding
would stimulate the computing industry to produce the sort of machine required in the
-9-
coming 5-10 years. As in particle physics, astronomy, and nuclear fusion physics, the sort
of facility which would enable breakthroughs in climate research, would be unaffordable at
the national level.
Climate change is a global problem and therefore it is reasonable that the solution to
the problem of providing governments and society with reliable predictions of climate
change also lies at the global level. Such facilities would draw on the strengths of existing
weather and climate prediction centers. The cost of such a single facility (order 1 billion
dollars for scientific workforce and powerful computers over 5 years), if shared among the
nations of the world with contributions in proportion to national GDP, would be affordable
at the national level.
At present, a huge quantity of space-based observations are acquired at a cost of tens
of billions of dollars annually. The spatial resolution at which satellites provide data (1-10
km) is so much finer than the resolution of models (50-200 km), and the research and
computing infrastructure is so inadequate, that only a small fraction of space-based
observations is being assimilated in forecast models. A research and supercomputing
facility with sustained Petaflop computing capability and a critical mass of scientists will
also accelerate progress in data assimilation (an important component for multi-decadal
forecasts) and prediction of the total climate system.
Another key focus of these international facilities will be to facilitate very highresolution global forecasts on timescales of a few decades. Decadal prediction combines
both the need for accurate initial conditions, central to weather forecasting, and the need
for scenarios of greenhouse gases, central to centennial climate prediction. A main focus
for decadal and multi-decadal prediction will be modes of climate variability. For example,
in the coming decades, can we make useful predictions about changes in the Atlantic
Meridional Overturning Circulation? Is El Niño likely to become more or less prevalent,
more or less intense? Decadal and multi-decadal predictions will, for example, provide
regional forecasts of precipitation change for the first half of the 21st Century, which are
key inputs to infrastructure investment (e.g. in terms of reservoir capacity or flood defense
capability) to adapt to predicted climate change.
Such multi-petaflop computational facilities, dedicated to climate simulation and
prediction, will be beyond the financial means of individual institutes and universities.
Indeed it now appears likely that the multi-petaflop machines in the USA, Europe and
Japan will be funded at the national level, for multi-purpose scientific applications. There
are indeed many demands on very high-performance computing: nanotechnology, biology,
astrophysics, high-energy physics and so on. In this case, climate applications would only
receive a small share of time on such national-level computing infrastructure. Moreover,
the machine architecture may not necessarily be optimal, so sharing time on such
computers with multiple scientific applications would certainly not provide society with
the best possible climate predictions. Establishing a small number of dedicated computing
- 10 -
facilities at the international level will allow climate predictions, at both global and
regional level, to be made using models whose realism is commensurate with the need to
guide policy on limiting carbon emissions, and to provide the basis for science-based
adaptation strategies for guiding infrastructure decisions to adapt to regional climate
change. Funding for such multi-national facilities should not and cannot compete with the
funding for national centers. In fact, each national center must be an important and integral
part of these facilities, and funding for the national centers should also be enhanced.
Considering the global humanitarian aspects of the problem, foundations and international
organizations should be viewed as possible funders. The need for accurate global climate
predictions in a changing climate will remain for several decades and such multi-national
facilities should not be subject to year-to-year funding uncertainties. Private corporations,
wealthy individuals and foundations should be approached to create an endowment fund so
that sustained funding can be guaranteed.
The return on investments in enhancing existing climate modelling activities, and
creating major international research and computing facilities is expected to be quite high.
Such facilities would almost certainly enable the accuracy and reliability of predictions of
global weather and short-term climate variations to be improved. Such a research and
computing facility will help national weather services to improve the day-to-day weather
forecasts of extreme, high-impact weather events and provide invaluable information to
agencies responsible for management of weather-related disasters. The new generation of
models will have an improved representation of the statistics of daily weather, and
therefore, will have the potential to provide better predictions of regional climate variations
on seasonal time scales. Even modest improvements in prediction of seasonal and interannual variations will provide huge socio-economic benefits to global societies by proper
application of climate forecasts to the agriculture, water resources, health and energy
sectors of national and international economies.
Based on these estimates, we propose that governments, the computing industry and
the science community work together toward the common goal of establishing a small
number of highly connected international research and computing facilities with sustained
computer capability of at least 20 Petaflops within five years, and at least 200 Petaflops by
the end of the next decade. This two-step approach would enable us to achieve the more
immediate goal of running global climate models for multi-decadal climate predictions and
IPCC projections at a resolution of about 10 km. This will also enable seasonal and
decadal prediction experiments with climate models at a resolution of 3-5 km. In parallel to
that, these facilities should be used to address the many research questions related to the
grand challenge of running global climate models at a much higher resolution of about 1
km. These are, of course, current goals. By the time individual weather services might
have acquired such computers for routine weather prediction, the international facilities
will be producing climate predictions using computers of a much higher capability.
- 11 -
4. Summary
By recognizing the threat of climate change and starting to formulate adaptation and
mitigation strategies, society now requires reliable, quantitative predictions of regional
climate change, not just temperature, but of precipitation, storminess and other relevant
variables. Since current climate systems models are not able to provide predictions with
adequate accuracy and detail, climate prediction needs to be revolutionized to be able to
fulfill society’s expectations. To accelerate progress in understanding the mechanisms of
climate variations, and predicting future climate change with reliable estimates of
uncertainties, the existing national facilities should be enhanced, and in addition, a small
number of multi-national research and computing facilities dedicated to climate prediction,
with resources and capabilities beyond those of a single national effort, should be
established. We recommend the creation of a small number of highly connected multinational facilities with computer capability for each facility of at least 20 Petaflop in the
near future and 200 Petaflops by the end of the next decade.
A major benefit of internationally coordinated climate prediction facilities will be to
provide science-based adaptation strategies to the policymakers in a changing climate.
Leaders of many nations will be faced with extremely serious questions, for example: Will
the Gulf Coast remain habitable? Should New Orleans be rebuilt? Should Holland be
building higher dikes? How do we relocate hundreds of millions of climate refugees? Are
countries like Bangladesh and Mauritius sustainable? These questions are so profound and
implications of decisions in response to these questions are so enormous that policymakers
must base their adaptation strategies on the most reliable information that the current state
of the science can provide.
The key goal of climate research is to provide society with the best possible estimates
of the predictability of climate. It is inevitable that some day the societal demand for
policy-relevant climate predictions will be so great that the most advanced technology and
the best available talent will be brought to bear to address this great challenge. We are
proposing that the time to begin that process has now arrived!
- 12 -
References:
Bell, G. D., and Co-authors 2008: The 2007 Atlantic Hurricane Season: A Climate
Perspective. State of the Climate in 2007. A. M. Waple and J. H. Lawrimore, Eds.
Bull. Amer. Meteor. Soc., 89, S1-S78.
Evenson, R. E. and D. Gollin, 2003: Assessing the Impact of the Green Revolution, 1960
to 2000. Science, 300, 758.
Fudeyasu, H., Y. Wang, M. Satoh, T. Nasuno, H. Miura, and W. Yanase, 2008: Global
cloud-system-resolving model NICAM successfully simulated the life cycles of two
real tropical cyclones. Geophys. Res. Lett., 35, doi:10.1029/2008GL036003.
Hermann, A., J. Krige, L. Belloni, D. Pestre, and U. Mersits, 1987: History of CERN.
North Holland.
Jung, T., S. K. Gulev, I. Rudeva, V. Soloviov, 2007: Sensitivity of extratropical cyclone
characteristics to horizontal resolution in the ECMWF model. Quart. J. Roy.
Meteor. Soc., 132, 1839 – 1857.
Palmer. T.N. and P.J. Webster, 1995: Towards a unified approach to climate and weather
prediction. Global Change. EUR 15158 427pp European Commission. ISBN 92826-7757-5.
Palmer, T.N., F.-J. Doblas-Reyes, M.Rodwell, A.Weisheimer, 2008: Towards seamless
prediction: calibration of climate change forecasts using seasonal prediction. Bull.
Amer. Meteor. Soc., 89, 459-470.
Rodwell, M and T.N. Palmer, 2007: Using numerical weather prediction to assess climate
models. Quart. J. Roy. Meteor. Soc., 133, 129-146.
Sachs, J.D., 2008: Common Wealth, Economies for a Crowded Planet. Allen Lane,
Penguin Books.
Satoh, M., T. Nasuno, H. Miura, H. Tomita, S. Iga, and Y. Takayabu (2008b): Precipitation
statistics comparison between global cloud resolving simulation with NICAM and
TRMM PR data. In High Resolution Numerical Modelling of the Atmosphere and
Ocean, W. Ohfuchi and K. Hamilton, eds. (Springer, New York), 99 – 112.
Shukla J, R. Hagedorn, B. Hoskins, J. Kinter, J. Marotzke, M. Miller, T. Palmer, J. Slingo,
2009: Revolution in Climate Prediction is Both Necessary and Possible: A
Declaration at the World Modelling Summit for Climate Prediction. Bull. Amer.
Meteor. Soc., 90, 16-19.
- 13 -
Stern, N., 2007: The economics of climate change: the Stern review Cambridge University
Press (Cambridge, UK), 692 p.
WCRP, 2007: Report of the Twenty-eighth session of the Joint Scientific Committee,
Zanzibar, Tanzania, 26-30 March 2007. WMO/TD-No. 1395, pp. 173-176.
- 14 -
AMERICAN
METEOROLOGICAL
SOCIETY
Bulletin of the American Meteorological Society
EARLY ONLINE RELEASE
This is a preliminary PDF of the author-produced
manuscript that has been peer-reviewed and
accepted for publication. Since it is being posted so
soon after acceptance, it has not yet been
copyedited, formatted, or processed by AMS
Publications. This preliminary version of the
manuscript may be downloaded, distributed, and
cited, but please be aware that there will be visual
differences and possibly some content differences
between this version and the final published version
The DOI for this manuscript is doi:
10.1175/2009BAMS2752.1
The final published version of this manuscript will replace the
preliminary version at the above DOI once it is available.
© 2009 American Meteorological Society
A Unified Modeling Approach to Climate System
Prediction
James Hurrell*1, Gerald A. Meehl1, David Bader2, Thomas L. Delworth3, Ben Kirtman4,
and Bruce Wielicki5
1
National Center for Atmospheric Research, Boulder, CO
2
Lawrence Livermore National Laboratory, Livermore, CA
3
Geophysical Fluid Dynamics Laboratory, NOAA, Princeton, NJ
4
Rosenstiel School of Marine and Atmospheric Science, University of Miami, Miami, FL and
Center for Ocean-Land-Atmosphere Studies, Calverton, MD
5
NASA Langley Research Center, Hampton, VA
Bulletin of the American Meteorological Society
Revised: 17 February, 2009
Final Revision: 26 June 2009
*
Corresponding author: [email protected]
National Center for Atmospheric Research╬
Climate Analysis Section
P.O. Box 3000
Boulder, CO 80307-3000
Phone: 303-497-1383
Fax: 303-497-1333
╬
The National Center for Atmospheric Research is sponsored by the National Science Foundation
1
Capsule Summary
Demand for more accurate predictions of regional climate necessitates a unified modeling
approach explicitly recognizing that many processes are common to predictions across
time scales. Applying and testing models with this approach has many benefits.
2
Abstract
There is a new perspective of a continuum of prediction problems, with a blurring of the
distinction between short-term predictions and long-term climate projections. At the heart
of this new perspective is the realization that all climate system predictions, regardless of
time scale, share common processes and mechanisms; moreover, interactions across time
and space scales are fundamental to the climate system itself. Further, just as seasonal to
interannual predictions start from an estimate of the state of the climate system, there is a
growing realization that decadal and longer term climate predictions could be initialized
with estimates of the current observed state of the atmosphere, oceans, cryosphere, and
land surface. Even though the prediction problem itself is seamless, the best practical
approach to it may be described as unified: models aimed at different time scales and
phenomena may have large commonality but place emphasis on different aspects of the
system. The potential benefits of this commonality are significant and include improved
predictions on all time scales and stronger collaboration and shared knowledge,
infrastructure and technical capabilities among those in the weather and climate
prediction communities.
3
1. Introduction
The global coupled atmosphere-ocean-land-cryosphere system exhibits a wide range
of physical and dynamical phenomena with associated physical, biological and chemical
feedbacks that collectively result in a continuum of temporal and spatial variability. The
traditional boundaries between weather and climate are, therefore, somewhat artificial.
The large scale climate, for instance, determines the environment for microscale (1 km or
less) and mesoscale (several km to several hundred km) processes that govern weather
and local climate, and these small scale processes likely have significant impacts on the
evolution of the large scale circulation (Fig. 1, derived from Meehl et al. 2001).
The accurate representation of this continuum of variability in numerical models is,
consequently, a challenging but essential goal. Fundamental barriers to advancing
weather and climate prediction on time scales from days to years, as well as longstanding systematic errors in weather and climate models, are partly attributable to our
limited understanding and capability to simulate the complex, multi-scale interactions
intrinsic to atmospheric, oceanic and cryospheric fluid motions.
The purpose of this paper is to identify some of the research questions and challenges
that are raised by the movement towards a more unified modeling framework that
provides for the hierarchical treatment of forecast and climate phenomena that span a
wide range of space and time scales. This has sometimes been referred to as the
“seamless prediction” of weather and climate (WCRP 2005; Palmer et al. 2008;
Shapiro et al. 2009; Brunet et al. 2009). The central unifying theme is that all climate
system predictions, regardless of time scale, share processes and mechanisms that
consequently could benefit from initialization of coupled general circulation models with
4
best estimates of the observed state of the climate (e.g., Smith et al. 2007;
Keenlyside et al. 2008; Pohlmann et al. 2009). But what is the best method of
initialization, given biases in models that make observations possibly incompatible with
the model climate state, and how can predictions best be performed and verified?
Hurricane prediction, for example, has traditionally been regarded as a short-term
weather prediction from an initialized atmospheric model. However, hurricanes generate
a cold wake as they churn up the ocean and not only extract considerable amounts of heat
through evaporative cooling but also mix heat down into the thermocline (e.g., Emanuel
2001, 2006; Trenberth and Fasullo 2007; Korty et al. 2008). Feedback from the cold
wake is now thought to be important to improving the forecast accuracy of intensity and
track, and the heat and fresh water fluxes could contribute to multi-decadal variability in
the Atlantic Ocean climate system (e.g. Hu and Meehl 2009).
Hence, hurricane
forecasting is a short-term coupled problem as well as a longer term climate problem
requiring not only an initialized atmospheric model, but also the initialization of a model
of the ocean and its heat content.
2. Scale interactions and climate system predictions
Scale interactions, both spatial and temporal, are the dominant feature of all aspects
of atmospheric and oceanic prediction. The hope is that predictions will improve as
models begin to explicitly resolve processes on ever finer spatial scales. Weather and
climate predictions, consequently, have been major drivers for higher resolution models
requiring advanced numerical and physical techniques and for sophisticated computing
systems.
5
State-of-the-art weather forecasting is carried out using Atmospheric General
Circulation Models (AGCMs) that have traditionally been forced with sea surface
temperature (SST) anomalies observed at some initial time, but are then projected and
damped toward climatological conditions as the integrations proceed out to typically 1014 days. On these time scales, dynamical interactions of the atmosphere with other
climate system components were generally thought to be unimportant and, therefore,
have typically not been included.
For decadal to centennial predictions, the radiative forcings and coupled interactions
and feedbacks among the climate system components are critical. Usually, these coupled
model integrations are initialized from an arbitrary and relatively stable climate state
obtained from a several-century control (without external forcing) integration. Such
coupled “Atmosphere-Ocean General Circulation Models” (AOGCMs) typically include
components of the atmosphere, ocean, land surface and sea ice.
These two time scales address two distinct scientific problems. For a weather forecast
on the scale of days, deterministic time evolution of individual synoptic systems must be
forecast as an initial value problem and effects of longer-term coupled processes, such as
the Meridional Overturning Circulation (MOC) in the ocean, are small. For climate time
scales of seasonal and beyond, statistics of the collections of weather systems are of
interest and are crucial to the fidelity of the climate simulation and/or prediction, but the
deterministic time evolution of the weather systems cannot be predicted.
For seasonal predictions, coupled air-sea interactions are especially important, but it
is an open question whether the prediction of an El Niño event depends critically on
aspects of the climate system that evolve on oven longer timescales, such as the MOC or
6
the state of the Pacific Decadal Oscillation (PDO). For even longer time scales, however,
interactions of the atmosphere with not only the ocean, but also the sea ice, land, snow
cover, land ice, and fresh water reservoirs become very important. Biogeochemistry and
interactive vegetation, and external effects, such as changes in solar activity, volcanic
eruptions, and human influences, all influence the evolution of the climate system.
While the validity of the assumptions made in designing and conducting numerical
experiments must be evaluated in the context of the problem being studied, a more
unified approach explicitly recognizes the importance of processes and mechanisms
shared across the time and space scales, and the potential benefit of greater convergence
of methods used in weather and climate forecasting, in particular with regards to
initialization of the climate system.
The El Niño/Southern Oscillation (ENSO) phenomenon, for example, can now be
predicted with some skill with an initialized state of the atmosphere and at least an upper
ocean model of the tropical Pacific, but profound gaps in our prediction abilities remain.
Large systematic errors in the coupled models mean that: (i) the coupled model mean
state does not agree with the observed mean state with sufficient fidelity; and (ii) the
space-time evolution of the simulated climate anomalies is not sufficiently realistic.
Historically, these two problems have been addressed from semi-empirical
perspectives. The first approach is to improve the individual physical parameterizations
in the component models (e.g., Toniazzo et al. 2008). A specific example (Fig. 2) is the
improvement in the simulation of ENSO by the Community Climate System Model
(CCSM; Collins et al. 2006) after improvements to the parameterization of deep
convection in the atmospheric model component (Neale et al. 2008). The second
7
approach has centered on how best to use imperfect models to make predictions: for
example, through calibration analysis (Rodwell and Palmer 2007; Palmer et al. 2008), by
utilizing a multi-model ensemble, or through stochastic-dynamic parameterization (e.g.,
Palmer et al., 2009; see Section 4d).
Another relevant consideration is that current climate models have been limited to
relatively coarse resolution compared to numerical weather prediction (NWP) models.
The coarse resolution limits the accurate simulation of atmospheric (e.g., the MaddenJulian Oscillation (MJO) and synoptic weather systems) and oceanic (e.g., tropical
instability waves) dynamics and, thus, their interactions with climate. A way forward is to
better resolve the weather-climate link (e.g., Palmer et al. 2008), but how best to
represent the important missing elements of the simulation of day-to-day weather in
climate models?
The typical assumption for sub-grid scale parameterization is to assume that the
statistics of sub-grid scale processes can be parameterized in terms of the grid scale
variables. However, in many cases this assumption may be seriously flawed. An
alternative strategy has been to reduce the grid size of the model and resolve more of the
motions explicitly as in NWP (e.g., Shapiro and Thorpe 2004), but this approach has been
limited, so far, by available computing power. The history of climate prediction has been
marked by compromise between model resolution, the inclusion of additional processes,
the length and number of simulations, and available computing resources. Global climate
predictions would certainly benefit from running AOGCMs at resolutions near or
matching current NWP models (Shapiro et al. 2009), but it has not yet been feasible to
marshal the considerable computer resources necessary (e.g., Shukla 2009).
8
3. Improving climate models
a. Upscaling research
The climate research community is beginning to use higher resolution (~50 km)
models for the decadal prediction problem (e.g., Meehl et al. 2009), but global modeling
frameworks that resolve mesoscale processes are needed to improve our understanding of
the multi-scale interactions in the coupled system, identify those of greatest importance,
and document their effects on climate. Ultimately, such basic research will help
determine how to better represent small-scale processes in relatively coarse resolution
Earth System Models (ESMs). We refer to the impacts of small-scale processes on larger
scales as “upscaling”.
There is a wide range of upscale interactions to be considered. Current
parameterization schemes do not adequately handle the mesoscale organization of
convection, which is a critical missing link in the scale interaction process (e.g.,
Moncrieff et al. 2007; 2009). The limited representation of convection and cloud
processes is likely a major factor in the inadequate simulation of tropical oscillations
(Fig. 3). Cloud and convective processes also appear to play a role in the well known
double Inter-Tropical Convergence Zone (ITCZ) bias issue (e.g., Fig. 4, top), though
coupled processes involving a systematically intense equatorial cold tongue in the ocean
also likely contribute to this persistent systematic error (Randall et al. 2007).
Uncertainty in the representation of clouds (on all scales) is also a major influence in
the response of the climate system to changes in radiative forcing. Improved simulation
of cloud processes in the Multi-scale Modeling Framework (MMF; Randall et al. 2003),
which embeds two-dimensional cloud resolving physics within three-dimensional
9
weather scale physics, has shown improved MJO variability and reduced the bias in
Kelvin wave propagation (Fig. 3; see Khairoutdinov et al. 2008).
Another scale interaction problem is the challenge in modeling the Subtropical
Eastern Boundary (STEB) regimes off the coasts of Southwest Africa, Peru-EcuadorChile, and Baja-Southern California. These regimes are marked by marine stratus,
equatorward alongshore winds, and ocean upwelling. Large and Danabasoglu (2006)
suggest that better resolution of these features produce not only a better simulation of the
regional climate, but also effects that propagate and strongly influence the large-scale
climate system, reducing rainfall biases across the tropical oceans (Fig. 4, bottom).
Other examples of “hot spots” with significant upscaled effects include the monsoon
regions of India and Tibet and Central and South America where steep topographical
gradients and mesoscale processes such as low-level jets and mesoscale convective
complexes play an important role in the water and energy budgets locally and remotely
(e.g., Webster 2006). Over the Maritime Continent, Lorenz and Jacob (2005) presented a
study of two-way coupling using global and regional models and demonstrated large and
positive impacts on the tropospheric temperature and large scale circulation in the global
climate simulation.
Clearly, addressing these errors is critical to climate prediction on all time scales.
Therefore, there is a strong need to develop pilot projects to demonstrate the
methodologies and impacts of multi-scale interactions on the regional and global climate.
While numerical models and techniques will be central to this effort, so too will be
sophisticated theoretical and physical research to both understand and specify the critical
10
interactions. Significant increases to computing resources to facilitate explicit simulation
of smaller-scale processes and their interactions with the larger scale will be essential.
b. Value of testing models on all time scales
A paradigm has long been that it is not essential to get all of the details of weather
correct as long as their statistically averaged effects on the climate system are adequately
captured. A key question is whether the rectification effects of small scale and high
frequency weather events can be adequately captured if details are not explicitly
represented. Water resources are a case in point as they rely on good predictions of
precipitation. This means not only precipitation amount but also precipitation intensity,
frequency, duration and type (snow versus rain). The character of precipitation affects
runoff and flooding, and thus soil moisture and stream flow.
The diurnal and annual cycles provide excellent tests for model evaluation. Model
response to these well-known climate forcings can provide crucial insights on a host of
important physical processes. For example, the diurnal cycle is strongest in summer over
land and affects the timing, location and intensity of precipitation events. Models
typically have onset of precipitation that is too early in the day and with insufficient
intensity compared with observations, demonstrating the need to improve boundary layer
and convective processes in models (e.g., Trenberth et al. 2003; Trenberth 2008a). The
annual cycle is an obvious strong test for measuring the response of a model to a major
climate forcing, albeit one that affects only those parts of the climate system capable of
responding on such a short time scale. Interannual variability, such as how well models
simulate ENSO, provides another necessary but not sufficient test of models. These tests
11
highlight the shortcomings and help identify steps to be taken to build confidence in
models (WCRP 2008).
4. Prediction across scales
a. Effect of initial conditions
For weather prediction, detailed analyses of the observed state of the atmosphere are
required but uncertainties in the initial state grow rapidly over several days. Other
components of the climate system are typically fixed as observed. For climate predictions,
the initial state of the atmosphere is less critical, and states separated by a day or so can
be substituted. However, the initial states of other climate system components become
vital. For predictions of a season to a year or so, the SSTs, sea ice extent and upper ocean
heat content, soil moisture, snow cover, and state of surface vegetation over land are all
important. Such initial value predictions are already operational for forecasting El Niño,
and extensions to the global oceans are under way. For the decadal prediction problem,
increased information throughout the ocean could be essential (Smith et al. 2007;
Trenberth 2008b; Meehl et al. 2009; Shukla 2009). Initial conditions for the global ocean
could conceivably be provided by existing ocean data assimilation exercises. However,
hindcast predictions for the 20th century, which are desirable to test models, are severely
hampered by poor salinity reconstructions prior to the early 2000’s when ARGO floats
began to provide much better depictions of temperature and salinity in the upper 2000 m
of the near-global ocean. Challenging research tasks are to develop optimal methods for
initializing climate model predictions with the current observational network and
identifying an optimal set of ocean observations to use for initializing climate predictions
(Meehl et al. 2009).
12
The mass, extent, thickness, and state of sea ice and snow cover are vital at high
latitudes. The states of soil moisture and surface vegetation are especially important in
understanding and predicting warm season precipitation and temperature anomalies along
with other aspects of the land surface, but are difficult to quantify. Any information on
systematic changes to the atmosphere (especially its composition and influences from
volcanic eruptions) as well as external forcings, such as from changes in the sun, are also
needed; otherwise these are specified as fixed at climatological average values. The
errors induced by incorrect initial conditions should become less apparent as the
simulations evolve as systematic “boundary” and external influences become more
important, but could still be evident through the course of the simulations.
A good rule of thumb for prediction is that an upper bound on predictability
corresponds approximately to one lifecycle of the phenomenon being considered. Hence
one could hope to predict a single convective element, cyclone wave, MJO cycle, ENSO
warm event, or fluctuation of the Atlantic MOC over its lifecycle, but not the second
generation event. This rule of thumb is consistent with the climate system being a chaotic
dynamical system with limited predictability. Additional predictability, however, could
arise from the slowly evolving components of the climate system.
The pathways leading from high frequency processes to low frequency phenomena,
however, may progressively involve more aspects of the climate system. For example,
convection associated with the MJO needs the ocean mixed layer to be accurately
specified in the initial state. Thus it follows that the MJO influence on ENSO needs an
accurate depiction of the initial state of the Southern Oscillation and the thermocline
slope across the equatorial Pacific. A unified modeling approach to climate system
13
prediction, in principle, lets all of these interactions occur as they do in nature. If the
models fall short, one can track how and learn why.
b. Effect of systematic errors
Another significant obstacle is the systematic errors present in current AOGCMs.
Some of these errors, such as the double ITCZ (Fig. 4, top), are very persistent and have
been present in multiple generations of coupled models. One approach to addressing
such errors is to vary the parameters in various physical parameterizations within the
range of uncertainty based on observations in an effort to reduce the known biases and to
form an ensemble of the uncertainty. A second approach is to improve the models so that
they more accurately simulate the phenomena in question. This can occur through
enhanced resolution, improved knowledge of the relevant physics from observations,
improvements in the parameterizations of unresolved physics, and numerical
experimentation to better understand existing parameterizations.
Efforts to reduce the systematic errors are crucial, since biases in the mean state could
affect a climate model’s climate sensitivity (the response to altered radiative forcing) and,
thus, its utility as a predictive tool. Quantifying the effects of systematic errors is difficult
because of the highly non-linear nature of the climate system. One promising approach,
at least for the atmospheric component, is to run it in NWP hindcast mode and observe
the biases as they develop (Phillips et al. 2004).
To understand the implication of systematic errors on forecast skill, it is important to
note how coupled forecasts are initialized. Due to the limitations of both observational
ocean data and computer resources, one way to initialize a coupled model is to start with
initial states determined separately for the atmosphere and ocean (e.g., coupling an
14
atmospheric initial state to an ocean re-analysis product). However, the sub-surface
ocean thermal state associated with the ocean initial condition is likely significantly
different than the climate of the free running coupled model. As a consequence, at
forecast initialization, the coupled model rapidly adjusts away from the observed climate
estimate towards the coupled model climate that is itself a product of its own systematic
errors. This adjustment in the tropics is primarily accomplished via Kelvin waves, which
ultimately lead to an erroneous SST response 2-4 months into the forecast evolution. This
is often referred to as an “initialization shock” or “coupling shock”. One approach to
address coupling shock is through “anomaly initialization” (Schneider et al. 1999; see
also Smith et al. 2007; Keenlyside et al. 2008; Pohlmann et al. 2009). In this approach,
models are initialized with observed anomalies added to the model climate, rather than
initialized with observed values, and the model climate is removed to obtain forecast
anomalies.
Ultimately, the solution to this problem is to improve the simulation of the coupled
modes of the climate system. For example, preliminary results with the NOAA climate
forecast system (CFS) indicate that a higher horizontal resolution model has more
irregularity of tropical eastern Pacific SST associated with ENSO, and the amplitude of
the SST variability is in better agreement with observed estimates. Atmospheric model
resolution experiments conducted with the Italian SINTEX coupled model also indicate
significant improvements in simulated ENSO periodicity with increasing atmospheric
model resolution (Navarra et al. 2008). However, as shown in Fig. 2, improvements to
the parameterization of deep atmospheric convection have also led to a better simulation
of ENSO frequency in the CCSM (Neale et al. 2008), and Toniazzo et al. (2008)
15
demonstrate the sensitivity of the simulation of ENSO in a version of the Hadley Centre
coupled model to perturbed atmospheric parameters. Therefore, improvements in model
fidelity with increasing resolution are likely part of the solution, but not the entire answer.
Active research efforts on how to initialize the coupled modes of the coupled models,
given that they do not agree with those of nature (Zhang et al. 2007a), recognize that the
best state estimate for the individual component models may not be best for coupled
forecasts. Much of the research focuses on how to identify the slow manifold described
by the observed estimates and the coupled model and how a mapping between them can
be derived. A promising avenue is the use of fully coupled assimilation systems
(Zhang et al. 2007a).
c. Predictability
Although deterministic atmospheric predictability is limited to approximately two
weeks (e.g., Kleeman 2007), on longer time scales at least two types of predictions may
be possible. The first is a prediction of the internal variability of the climate system based
on an initialized state of the ocean, atmosphere, land and cryosphere system. Coupled
ocean-atmosphere interactions, for instance, are likely important for understanding the
temporal evolution of some extratropical, regional modes of climate variability, such as
the North Atlantic Oscillation (Hurrell et al. 2006) and local modes of coupled variability
in the Atlantic and Indian Ocean basins (e.g., Xie and Carton 2004; Webster 2006).
Moreover, land surface processes, and the influence of the stratosphere on the state of the
troposphere, might also be a significant source of predictability, at least on seasonal time
scales (e.g., Baldwin 2003).
16
First attempts at “decadal prediction” with an AOGCM showed reduced error growth
in large-scale averaged surface temperature over ten year periods as a result of the
initialized climate state (Smith et al. 2007; Keenlyside et al. 2008; Pohlmann et al. 2009).
Decadal scale predictability in the ocean may occur from the thermal inertia of the
initialized anomalies in ocean heat content, but additional predictability may also arise
from fluctuations in gyre and overturning circulations (e.g., Delworth and Mann 2000;
Dong and Sutton 2005), particularly in the Atlantic (Fig. 5). Multi-decadal variations in
Atlantic SSTs have been linked to low-frequency boreal summer changes in rainfall and
drought in the continental United States (e.g., Schubert et al. 2004; Sutton and
Hodson 2005) as well as hemispheric-scale temperature anomalies (Zhang et al. 2007b).
They may also have implications for North Atlantic hurricane forecasts (e.g., Zhang and
Delworth 2006). It is possible that decadal scale predictability exists in the Pacific Ocean
as well (e.g., Meehl and Hu 2006).
In addition to the potential sources of predictability from the initial values of the
system, predictability may also be derived from past and future changes in radiative
forcing (Hansen et al. 2005; IPCC 2007; Smith et al. 2007). Past emissions of greenhouse
gases have committed the climate system to future warming as the ocean comes into
equilibrium with the altered radiative forcing. In addition, the best possible estimates of
future emissions of radiatively important pollutants are needed for making predictions, as
well as modeling capabilities to accurately simulate both how these pollutants affect the
global energy, carbon and sulfur cycles, and how the climate system subsequently
responds to that altered forcing. In this regard, the phase and amplitude of the solar cycle
17
and unpredictable volcanic eruptions can be significant “wild cards” to such predictions
(Ammann and Naveau 2009).
d. Single versus multiple model predictions
The purpose of ensemble prediction is to quantify the uncertainty in the forecast from
errors in the initial conditions, errors in the model (or multiple models), or a fundamental
lack of predictability in the phenomenon itself (e.g., Hawkins and Sutton 2009). This
technique is commonly used for NWP where many ensemble members are generated
from the same model. It is also relevant for seasonal forecasting where more than one
model can be used, since a simulation average across different models is presently more
skillful than a simulation from a single model (e.g., Glecker et al 2008; Kirtman and Min
2009).
The rainfall variability simulated by nine-member ensembles of several state-of-theart AGCMs forced by observed SSTs (Fig. 6) is very different in the rainfall (signal)
variance (first column) despite the common SST forcing. This uncertainty reflects
differences in model formulation, and it is larger than the uncertainty due to initial
conditions (middle column) highlighting the utility of the multi-model approach.
There are a number of different strategies currently employed to combine models for
the purpose of prediction. The simplest and most common approach is to have the various
modeling centers make ensemble predictions and then devise statistical strategies (i.e.,
Bayesian, linear regression) for combining the models (e.g., Palmer et al. 2004). It is also
possible to take a specific model and systematically probe the uncertainty in the model
formulation by varying the parameters in the model (Stainforth et al. 2005). Both
approaches have strengths and weaknesses, but neither strategy is completely satisfactory
18
in terms of adequately resolving the uncertainty. Another recently proposed methodology
is to use stochastic-dynamic parameterization techniques which perturb parameterizations
in such a way as to improve on the benefits of a multi-model ensemble by using a single
model (Palmer et al. 2009).
e. Verification
A quick scan through the Journal of Climate reveals a dizzying array of different
climate metrics both interesting and important. Furthermore, the attraction to use metrics
to select the “best” model for an application is problematic (Gleckler et al. 2008). Metrics
differ in variable, time scale, space scale, or functional representation. The same is not
true in weather prediction, where some estimates of both prediction limits and the impact
of different weather prediction metrics can be determined. The skill of daily weather
forecasts can be verified many times and a quantification of model skill is relatively
straightforward. The problem is more difficult for seasonal prediction since a large
number of seasons and those forecast states must pass in order to build up forecast
verification statistics.
For decadal and longer time scales, the problem of quantifying prediction skill
becomes even more difficult, and the metrics will likely involve how the forecasts are
used in applications. Even if we could test long term climate models with all possible
climate metrics proposed in the last decade of journal papers, we have no current method
to prioritize or weight their impact in measuring uncertainty in predicting future climate
change for temperature, precipitation, soil moisture and other variables of critical interest
to society.
19
There has been some recent progress in this direction using perturbed physics
ensembles (PPEs; Stainforth et al. 2005). PPEs are climate models that perturb uncertain
physical parameterizations instead of initial conditions. The non-dimensional error in
Fig. 7 (from Murphy et al. 2004) is defined as the ratio of climate model rms error versus
observations to the interannual natural variability of the same climate variable metric: in
essence a signal to noise measure. A large range of a given non-dimensional climate
metric indicates sensitivity. The whisker plots in Fig. 7 confirm the intuition that climate
variables associated with energetics (cloud, radiation, sea ice) appear more sensitive than
classical weather dynamical variables (e.g., 500 hPa streamfunction). Further work along
these lines is critically needed to discover methodologies to define rigorous climate
metrics capable of determining climate prediction uncertainty. The essential question is
this: what climate metrics for hindcast climate prediction accuracy can be used to
determine the uncertainty bounds on future climate prediction accuracy? If this question
can be answered, a second benefit will be the ability to more rigorously define climate
observation requirements.
5. Concluding remarks
Strategies for a more unified approach to climate system prediction currently include:
(i) using IPCC class coupled climate models for predictions on time scales of days to
decades; (ii) using NWP class models for seasonal to decadal prediction, after
modification to properly account for changing radiative forcing; and (iii) developing very
high resolution models with mesoscale processes explicitly resolved, either globally or by
nesting high resolution regional models within global climate models. There are other
emerging approaches as well, such as the concept of beginning integrations with higher
20
resolution to satisfy weather forecast requirements, then cascading down to lower
resolution versions of the model with consistent physical parameterization schemes for
longer time scale predictions. All of these approaches attempt to remove the distinction
between weather and climate by taking advantage of the processes and mechanisms that
characterize the climate system at all time and space scales. Questions are being raised as
to whether model development efforts should be focused on improving AOGCMs before
attempting ESMs, with their added complexities of coupled carbon and nitrogen cycles,
chemistry, aerosols, dynamic vegetation and other components. With a unified modeling
approach, the common processes can be addressed in both classes of models and progress
can be made on both fronts.
There are other potential benefits of using similar models for predictions on different
time scales. Among them are skill improvement in both weather and climate forecasts,
stronger collaboration and shared knowledge among those in the weather and climate
“communities” working on physical parameterization schemes, data assimilation schemes
and initialization methods, and shared infrastructure and technical capabilities.
A significant step forward is a planned set of coordinated climate change experiments
called the Coupled Model Intercomparison Project phase 5 (CMIP5; Taylor et al. 2009).
The strategy is to approach the climate change prediction problem in a unified way with
two classes of related climate models to address two time scales: higher resolution
(~50 km) AOGCMs for decadal predications out to about the year 2035 (Meehl
et al. 2009), and lower resolution (~200 km) versions of the same models but with a
coupled carbon cycle and perhaps simple chemistry, dynamic vegetation, and prognostic
aerosols for century and longer climate change integrations. The latter experiments would
21
quantify the magnitude of important feedbacks that will determine the ultimate degree of
climate change in the second half of the 21st century (Meehl and Hibbard 2007;
Hibbard et al. 2007).
Computer resource and other limitations will likely dictate that resolving certain
processes and phenomena could still require alternative strategies for many years into the
future. A case in point is the need to represent hurricanes in a special class of climate
models that could include embedded regional models with resolutions of about 5 km in
order to adequately depict their extreme intensity and their effects on the ocean and the
energy and water cycles.
Additionally, current and future efforts with ESMs will allow for more complete
assessments of the physics of climate change by including additional components and
processes that are not essential to the shorter time scales. The computational burden of
the ESMs will test the feasible limits of explicit resolution of multi-scale interactions and
more regional discrimination of climate change impacts. Moreover, given relatively large
systematic errors, the additional feedbacks from more interactive components of ESMs
clearly increase the uncertainty in the magnitude and nature of the climate changes
projected in future scenario simulations. The time-evolving ingredients required for
future scenario integrations with ESMs also still must be estimated as a range of possible
outcomes based to a large extent on the unpredictable nature of human actions. These,
along with observational data needs, logistical issues related to coupling strategies and
coupled initialization, and the scientific questions related to the myriad of unconstrained
and poorly understood feedbacks, are significant aspects of these emerging ESMs that
will continue to stretch both computational and human resources for the foreseeable
22
future. However, activities that have already begun indicate we are moving into a new
and exciting era of climate system prediction that will, by nature of the converging
interests, modeling tools and methodologies, produce greater interactions among
previously separate communities and thereby provide better predictions of the climate
system at all time and space scales.
Acknowledgements
The authors wish to thank two anonymous referees, as well as Kevin Trenberth, Joe
Tribbia, Greg Holland, Tony Busalacchi and Rick Rosen for their constructive comments
and suggestions on earlier versions of the manuscript. We also wish to thank Adam S.
Phillips and Julie Caron for their assistance with the figures, and James Murphy for the
data in Figure 7.
23
References
Adler, R.F., G.J. Huffman, A. Chang, R. Ferraro, P. Xie, J. Janowiak, B. Rudolf, U.
Schneider, S. Curtis, D. Bolvin, A. Gruber, J. Susskind, and P. Arkin, 2003: The
Version 2 Global Precipitation Climatology Project (GPCP) Monthly Precipitation
Analysis (1979-Present). J. Hydrometeor., 4, 1147-1167.
Ammann C.M., and P. Naveau, 2009: A statistical volcanic forcing-scenario generator for
climate simulations. J. Geophys. Res., doi: 10.1029 (submitted).
Baldwin, M.P., D.B. Stephenson, D.W.J. Thompson, T.J. Dunkerton, A.J. Charlton, A.
O'Neill, 2003: Stratospheric memory and extended-range weather forecasts. Science,
301, 636-640.
Brunet, G., and co-authors, 2009: Toward a seamless process for the Prediction of
weather and climate: The advancement of sub-seasonal to seasonal prediction. Bull.
Amer. Meteor. Soc., submitted.
Collins, W. D., and co-authors, 2006: The Community Climate System Model Version 3
(CCSM3). J. Climate, 19, 2122-2143.
Delworth, T.L., and M.E. Mann, 2000: Observed and simulated multidecadal variability.
Clim. Dyn., 16, 661-676.
Dong, B., and R. T. Sutton, 2005: Mechanism of interdecadal thermohaline circulation
variability in a coupled ocean–atmosphere GCM J. Climate, 18, 1117-1135.
Emanuel, K., 2001: Contribution of tropical cyclones to meridional heat transport by the
oceans. J. Geophys. Res., 106, 14771–14782.
Emanuel, K., 2006: Climate and tropical cyclone activity. A new model downscaling
approach. J. Climate, 19, 4797-4802.
24
Gleckler, P. J., K. E. Taylor, and C. Doutriaux, 2008: Performance metrics for climate
models, J. Geophys. Res., 113, D06104, doi:10.1029/2007JD008972.
Hansen, J. E. and co-authors, 2005: Efficacy of climate forcings. J. Geophys. Res. 110,
D18104, doi:10.1029/2005JD005776.
Hawkins, E. and R. Sutton, 2009. The potential to narrow uncertainty in regional climate
predictions. Bull. Am. Met. Soc. doi:10.1175/2009BAMS2607.1.
Hibbard, K. A., G. A. Meehl, P. Cox, and P. Friedlingstein, 2007: A strategy for climate
change stabilization experiments. EOS, 88, 217, 219, 221.
Hu, A., and G.A. Meehl, 2009: Effect of the Atlantic hurricanes on the oceanic
meridional overturning circulation and heat transport. Geophys. Res. Lett., submitted.
Hurrell, J. W., and co-authors, 2006: Atlantic Climate Variability and Predictability: A
CLIVAR perspective. Journal of Climate, 19, 5100-5121.
IPCC, 2007: Climate Change 2007: The Physical Science Basis. Contribution of Working
Group I to the Fourth Assessment Report of the Intergovernmental Panel on Climate
Change [Solomon, S., D. Qin, M. Manning, Z. Chen, M. Marquis, K.B. Averyt, M.
Tignor and H.L. Miller (eds.)]. Cambridge University Press, Cambridge, United
Kingdom and New York, NY, USA, 996 pp.
Keenlyside, N., M. Latif, J. Junclaus, L. Kornblueh and E. Roeckner, 2008: Advancing
decadal climate scale prediction in the North Atlantic. Nature, 453, 84-88.
Kirtman, B. P., and D. Min, 2009: Multi-model ensemble ENSO prediction with CCSM
and CFS. Mon. Wea. Rev., submitted.
25
Khairoutdinov, M., C. DeMott, and D. Randall, 2008: Evaluation of the simulated
interannual and subseasonal variability in an AMIP-style simulation using the CSU
Multiscale Modeling Framework. J. Climate, 21, 413-431.
Kleeman, R., 2007: Statistical predictability in the atmosphere and other dynamical
systems. Physica D, 230, 65-71.
Korty, R. L., K. A. Emanuel, and J. R. Scott, 2008: Tropical cyclone–induced upperocean mixing and climate: Application to equable climates. J. Climate, 21, 638-654.
Large, W. G., and G. Danabasoglu, 2006: Attribution and impacts of upper-ocean biases
in CCSM3. J. Climate, 19, 2325-2346.
Lorenz, P., and D. Jacob, 2005: Influence of regional scale information on the global
circulation: A two-way nesting climate simulation. Geophys. Res. Lett., 32, L18706,
doi:10.1029/2005GL023351.
Meehl, G. A., and A. Hu, 2006: Megadroughts in the Indian monsoon region and
southwest North America and a mechanism for associated multi-decadal Pacific sea
surface temperature anomalies. J. Climate, 19, 1605–1623.
Meehl, G.A., R. Lukas, G.N. Kiladis, M. Wheeler, A. Matthews, and K.M. Weickmann,
2001: A conceptual framework for time and space scale interactions in the climate
system. Clim. Dyn., 17, 753 - 775.
Meehl, G.A., and K.A. Hibbard, 2007: A strategy for climate change stabilization
experiments with AOGCMs and ESMs. WCRP Informal Report No. 3/2007, ICPO
Publication No. 112, IGBP Report No. 57, World Climate Research Programme:
Geneva, 35 pp.
26
Meehl, G.A., and co-authors, 2009: Decadal prediction: Can it be skillful? Bull. Amer.
Meteor. Soc., submitted.
Moncrieff, M. W., Shapiro, M., Slingo, J., Molteni, F., 2007: Collaborative research at
the intersection of weather and climate. WMO Bulletin, 56, 206-211.
Moncrieff, M. W., and co-authors, 2009: A Scientific Framework for Addressing the
Multi-scale Organization of Tropical Convection and its Interaction with the Global
Circulation. Bull. Amer. Meteor. Soc., submitted.
Murphy J. M., David M. H. Sexton, David N. Barnett, Gareth S. Jones, Mark J. Webb,
Matthew Collins, David A. Stainforth, 2004: Quantification of modelling
uncertainties in a large ensemble of climate change simulations. Nature, 430, 768-772.
Navarra, A., and co-authors, 2008: Atmospheric horizontal resolution affects tropical
climate variability in coupled models. J. Climate, 21, 730-750.
Neale, R. B., M. Jochum and J. H. Richter, 2008: The impact of convection on ENSO:
From a delayed oscillator to a series of events. J. Climate, in press, doi:
10.1175/2008JCLI2244.
Palmer, T. N., and co-authors, 2004: Development of a European multimodel ensemble
system for seasonal-to-interannual prediction (DEMETER). Bull. Amer. Meteor. Soc.,
85, 853-872.
Palmer, T. N., F. J. Doblas-Reyes, A. Weisheimer, and M. J. Rodwell, 2008: Toward
seamless prediction. Calibration of climate change projections using seasonal
forecasts. Bull. Amer. Meteor. Soc., 89, 459-470.
Palmer, T.N. and co-authors, 2009: Towards the probabilistic earth system model. J.
Climate, submitted.
27
Phillips, T.J., and co-authors, 2004: Evaluating parameterizations in general circulation
models: Climate simulation meets weather prediction. Bull. Amer. Meteor. Soc., 85,
1903–1915.
Pohlmann H, J. H. Jungclaus, A. Köhl, D. Stammer, J. Marotzke, 2009: Initializing
Decadal Climate Predictions with the GECCO Oceanic Synthesis: Effects on the
North Atlantic. J. Climate, in press.
Randall, D., M. Khairoutdinov, A. Arakawa, and W. Grabowski, 2003: Breaking the
cloud parameterization deadlock. Bull. Amer. Meteor. Soc., 84, 1547-1564.
Randall, D., and co-authors, 2007: Climate Models and Their Evaluation. In: Climate
Change 2007: The Physical Science Basis. Contribution of Working Group I to the
Fourth Assessment Report of the Intergovernmental Panel on Climate Change
[Solomon, S., D. Qin, M. Manning, Z. Chen, M. Marquis, K.B. Averyt, M.Tignor and
H.L. Miller (eds.)]. Cambridge University Press, Cambridge, United Kingdom and
New York, NY, USA.
Rodwell, M. J., and T. N. Palmer, 2007: Using numerical weather prediction to assess
climate models. Quart. J. Roy. Meteor. Soc., 133, 129–146.
Schubert, S. D., M. J. Suarez, P. J. Pegion, R. D. Koster, and J. T. Bacmeister, 2004:
Causes of long-term drought in the United States Great Plains. J. Climate, 17, 485–
503.
Shapiro, M. A., and A. J. Thorpe, 2004: THORPEX International Science Plan.
WMO/TD-No. 1246, WWRP/THORPEX No. 2, 57 pp.
Shapiro, M. A., and co-authors, 2009: A Weather, climate and Earth-system Prediction
Initiative for the 21st Century. Bull. Amer. Meteor. Soc., submitted.
28
Shukla, J., 2009: A Revolution in Climate Prediction is both necessary and possible. Bull.
Amer. Meteor. Soc., submitted.
Smith, D. M., and co-authors, 2007: Improved surface temperature prediction for the
coming decade from a global climate model. Science, 317, 796-799.
Stainforth, D. A., and co-authors, 2005: Uncertainty in predictions of the climate
response to rising levels of greenhouse gases. Nature, 433, 403-406.
Sutton, R. T., and D. L. R. Hodson, 2005: Atlantic Ocean forcing of North American and
European summer climate. Science, 309, 115–118.
Taylor, K., R. J. Stouffer, and G. A. Meehl, 2009: A summary of the CMIP5 experiment
design. WCRP, submitted.
Toniazzo, T., M. Collins and J. Brown, 2008: The variation of ENSO characteristics
associated with atmospheric parameter perturbations in a coupled model. Climate
Dynamics, 30, 643-656.
Trenberth, K. E., 2008a: The impact of climate change and variability on heavy
precipitation, floods and droughts. Encyclopedia of Hydrological Sciences. M. H.
Anderson,
Ed.
J.
Wiley
and
Sons,
Ltd.
Chichester,
UK.
doi:
10.1002/0470848944.hsa211.
Trenberth, K. E., 2008b: Observational needs for climate prediction and adaptation.
WMO Bulletin, 57 (1) 17-21.
Trenberth, K. E., and J. Fasullo, 2007: Water and energy budgets of hurricanes and
implications
for
climate
change.
J.
Geophys.
Res.,
112,
D23107,
doi:10.1029/2006JD008304.
29
Trenberth, K. E., A. Dai, R. M. Rasmussen and D. B. Parsons, 2003: The changing
character of precipitation. Bull. Amer. Meteor. Soc., 84, 1205-1217.
WCRP, 2005: The World Climate Research Programme Strategic Framework 2005-2015:
Coordinated Observation and Prediction of the Earth System (COPES). WCRP-123,
WMO/TD-No. 1291, 65 pp.
WCRP, 2008: World Climate Research Programme position paper on Seasonal
Prediction, B. Kirtman and A. Pirani, Eds. Informal Report No. 3/2008, ICPO
Publication No. 127, 25 pp.
Webster, P. J, 2006: The coupled monsoon system. In The Asian Monsoon. B. Wang, Ed.
Praxis Publishing Ltd, Chichester, UK, 3-66.
Xie, S.-P., and J. A. Carton, 2004: Tropical Atlantic variability: Patterns, mechanisms,
and impacts. Earth’s Climate: The Ocean-Atmosphere Interaction, Geophys. Monogr.,
Vol. 147, Amer. Geophys. Union, 121–142.
Zhang, R., and T. L. Delworth, 2006: Impact of Atlantic multidecadal oscillations on
India/Sahel rainfall and Atlantic hurricanes. Geophysical Research Letters, 33,
L17712, doi:10.1029/2006GL026267.
Zhang, S., M. J. Harrison, A. Rosati, and A. Wittenberg, 2007a: System design and
evaluation of coupled ensemble data assimilation for global oceanic climate studies.
Monthly Weather Review, 135(10), 3541-3564.
Zhang, R., T. L. Delworth, and I. M. Held, 2007b: Can the Atlantic Ocean drive the
observed multidecadal variability in Northern Hemisphere mean temperature?
Geophysical Research Letters, 34, L02709, doi:10.1029/2006GL028683.
30
Figure Captions
Figure 1. Schematic illustrating interactions between various time and space scales in the
climate system. Space scales are indicated on the left, and possible forecasts are shown
on the right. Though “synoptic” is the smallest time scale, these interactions could
continue to infinitely short time scales and small space scales.
Figure 2. Summary statistics of Niño-3.4 (170ºW-120ºW; 5ºN-5ºS) monthly SST
anomalies. Time series and wavelet analysis for 100 simulated years from (a) CCSM3
(Collins et al. 2006); (b) after modifications to the CCSM3 parameterization for deep
convection; c) the most recent 80 years of the observed HadiSST record, in addition to d)
power spectra; e) autocorrelation; and f) average variance for each calendar month, for all
model runs. See Neale et al. (2008) for details.
Figure 3. Space-time spectrum of the 15°S-15°N symmetric component of precipitation,
divided by the background spectrum. Observational estimates from an atmospheric
reanalysis product are in the top panel and the lower panel shows results from a coupled
climate model simulation.
31
Figure 4. (Top) Difference between annual mean precipitation from a multi-century
control simulation with CCSM3 and observational estimates (1979-2007) from the
Global Precipitation Climatology Project (Adler et al. 2003). (Bottom) Changes in
simulated CCSM3 oceanic precipitation in a fully coupled simulation, but with ocean
temperature and salinity restored to observed values in the Subtropical Eastern Boundary
(STEB) regimes off the coasts of Southwest Africa, Peru-Ecuador-Chile, and BajaSouthern California. Note the reduction in rainfall biases not only locally, but across the
tropical oceans. Adapted from Large and Danabasolgu (2006).
Figure 5. One example of decadal scale predictability of the Atlantic MOC as computed
in the GFDL CM2.1 global coupled climate model. A 5-member ensemble of
predictability experiments is shown, in which each ensemble member used identical
initial conditions for the ocean, land, and sea ice. These are taken from 1 January 1101 in
a long control integration. The ensemble members differed in their atmospheric initial
conditions, which come from January 6, 11, 16, 21, and 26 from the same year in the
control integration. The quantity plotted is an index of the MOC, defined as the
maximum stream function value in the North Atlantic each year, indicating the northward
mass flow in the upper layers of the North Atlantic (1 Sverdrup = 106 m3 s-1). The
relatively low spread among ensemble members in the first 10 years suggests substantial
decadal predictability. Additional ensembles were calculated, some if which had similar
predictability, and others of which had very little predictability.
32
Figure 6. Rainfall variability simulated by several AGCMs forced with observed sea
surface temperatures. Each model simulation includes an ensemble of nine initial
conditions, the differences in which are designed to mimic potential observational errors.
The first column shows the rainfall variance of the ensemble mean of each model. This is
the signal variance. The second column shows the variance about the ensemble mean or
the variance due to atmospheric internal dynamics. The last column is the ratio of the
ensemble mean variance divided by the internal dynamics variance, i.e., a signal to noise
ratio. Results from WCRP/Clivar/WGSIP SMIP project and figure courtesy of In Sik
Kang Seoul National University.
Figure 7. Values of the Climate Prediction Index (CPI) of Murphy et al. (2004), and its
32 components (black boxes and bars, representing surface and atmospheric variables)
from the perturbed physics ensemble (PPE). The components are calculated as the r.m.s.
difference between simulated and observed present-day climatological mean patterns
divided by the r.m.s value of the standard deviation of simulated interannual variations.
The plot shows averages of values calculated separately for each season of the year. Bars
show the full range of the ensemble distribution of values, boxes show the range
encompassed by the 5th and 95th percentiles, and the horizontal line within each box
shows the median. The CPI is calculated as the r.m.s. value of the 32 components for a
given ensemble member. Adapted from Murphy et al. (2004: see their article for more
detail).
33
Figure 1. Schematic illustrating interactions between various time and space scales in
the climate system. Space scales are indicated on the left, and possible forecasts are
shown on the right. Though “synoptic” is the smallest time scale, these interactions could
continue to infinitely short time scales and small space scales.
34
Figure 2. Summary statistics of Niño-3.4 (170ºW-120ºW; 5ºN-5ºS) monthly SST
anomalies. Time series and wavelet analysis for 100 simulated years from (a) CCSM3
(Collins et al. 2006); (b) after modifications to the CCSM3 parameterization for deep
convection; c) the most recent 80 years of the observed HadiSST record, in addition to d)
power spectra; e) autocorrelation; and f) average variance for each calendar month, for
all model runs. See Neale et al. (2008) for details.
35
Figure 3. Space-time spectrum of the 15°S-15°N symmetric component of precipitation,
divided by the background spectrum. Observational estimates from an atmospheric
reanalysis product are in the top panel and the lower panel shows results from a coupled
climate model simulation.
36
Figure 4. (Top) Difference between annual mean precipitation from a multi-century
control simulation with CCSM3 and observational estimates (1979-2007) from the
Global Precipitation Climatology Project (Adler et al. 2003). (Bottom) Changes in
simulated CCSM3 oceanic precipitation in a fully coupled simulation, but with ocean
temperature and salinity restored to observed values in the Subtropical Eastern
Boundary (STEB) regimes off the coasts of Southwest Africa, Peru-Ecuador-Chile, and
Baja-Southern California. Note the reduction in rainfall biases not only locally, but
across the tropical oceans. All units are mm day-1. Adapted from Large and Danabasolgu
(2006).
37
Figure 5. One example of decadal scale predictability of the Atlantic MOC as computed
in the GFDL CM2.1 global coupled climate model. A 5-member ensemble of
predictability experiments is shown, in which each ensemble member used identical
initial conditions for the ocean, land, and sea ice. These are taken from 1 January 1101
in a long control integration. The ensemble members differed in their atmospheric initial
conditions, which come from January 6, 11, 16, 21, and 26 from the same year in the
control integration. The quantity plotted is an index of the MOC, defined as the maximum
stream function value in the North Atlantic each year, indicating the northward mass
flow in the upper layers of the North Atlantic (1 Sverdrup = 106 m3 s-1). The relatively
low spread among ensemble members in the first 10 years suggests substantial decadal
predictability. Additional ensembles were calculated, some if which had similar
predictability, and others of which had very little predictability.
38
Figure 6. Rainfall variability simulated by several AGCMs forced with observed sea
surface temperatures. Each model simulation includes an ensemble of nine initial
conditions, the differences in which are designed to mimic potential observational errors.
The first column shows the rainfall variance of the ensemble mean of each model. This is
the signal variance. The second column shows the variance about the ensemble mean or
the variance due to atmospheric internal dynamics. The last column is the ratio of the
ensemble mean variance divided by the internal dynamics variance, i.e., a signal to noise
ratio. Results from WCRP/Clivar/WGSIP SMIP project, and figure courtesy of In Sik
Kang Seoul National University.
39
Figure 7. Values of the Climate Prediction Index (CPI) of Murphy et al. (2004), and its
32 components (black boxes and bars, representing surface and atmospheric variables)
from the perturbed physics ensemble (PPE). The components are calculated as the r.m.s.
difference between simulated and observed present-day climatological mean patterns
divided by the r.m.s value of the standard deviation of simulated interannual variations.
The plot shows averages of values calculated separately for each season of the year. Bars
show the full range of the ensemble distribution of values, boxes show the range
encompassed by the 5th and 95th percentiles, and the horizontal line within each box
shows the median. The CPI is calculated as the r.m.s. value of the 32 components for a
given ensemble member. Adapted from Murphy et al. (2004: see their article for more
detail).
40
Toward A Seamless Process for the Prediction of Weather and Climate: the
advancement of sub-seasonal to seasonal prediction
A weather and climate research community’s collaborative effort for the advancement of the
science of sub-seasonal to seasonal prediction and its socio-economic applications.
Gilbert Brunet, Meteorological Research Division, Environment Canada 1
Melvyn Shapiro, National Oceanic and Atmospheric Administration, USA
Brian Hoskins, Grantham Institute for Climate Change, Imperial College and University of
Reading, UK
Mitch Moncrieff, National Center for Atmospheric Research, University Corporation
for Atmospheric Research, USA
Randal Dole, Earth System Research Laboratory, National Oceanic and Atmospheric
Administration, USA
George N. Kiladis, Earth System Research Laboratory, National Oceanic and Atmospheric
Administration, USA
Ben Kirtman, Center for Ocean-Land-Atmosphere Studies Calverton and Rosenstiel
School for Marine and Atmospheric Science, University of Miami, USA
Andrew Lorenc, Met Office, UK
Brian Mills, Adaptation and Impacts Research Division, Environment Canada
Rebecca Morss, National Center for Atmospheric Research, University Corporation
for Atmospheric Research, USA
Saroja Polavarapu, Meteorological Research Division, Environment Canada
David Rogers, Health and Climate Foundation, USA
1
Corresponding author address: Gilbert Brunet, Meteorological Research Division, Environment Canada,
2121 Trans-Canada Highway, Dorval, Quebec, Canada, H9P 1J3.
E-mail : [email protected]
Page 1
John Schaake, NWS Office of Hydrologic Development, National Oceanic and
Atmospheric Administration
Jagadish Shukla, Earth Sciences and Global Change, George Mason University, USA
Submitted to BAMS, December 29, 2008.
Page 2
ABSTRACT
The World Weather Research Programme (WWRP) and World Climate Research
Programme (WCRP) have identified collaborations and scientific priorities to accelerate
advances in analysis and prediction at sub-seasonal to seasonal time scales in order to: i)
advance knowledge of mesoscale to planetary-scale interactions and their prediction; ii)
develop high-resolution global-regional climate simulations with advanced physical
processes representation to improve the predictive skill of the sub-seasonal and seasonal
variability of high-impact events, such as seasonal drought and floods, blocking, tropical
and extratropical cyclones; iii) contribute to the improvement of data-assimilation methods
for monitoring and prediction of the coupled ocean-atmosphere-land and Earth system; iv)
develop and transfer related diagnostic and prognostic information tailored to user needs
for accurate weather and climate forecasts and their socio-economic decision making. The
document puts forward specific underpinning research, linkage and requirements
necessary to achieve the goals of the proposed collaborative effort.
Page 3
1. Introduction
This document was prepared by scientists associated with the WWRP and WCRP. The
leading component within the WMO/WWRP is The Observing-system Research and
Predictability EXperiment (THORPEX) that aspires to accelerate improvements in the
accuracy of 1-day to 2-week high-impact weather forecasts and the use of this information
for the benefit of society. The WCRP is sponsored by the International Council for Science
(ICSU), the World Meteorological Organization (WMO) and the Intergovernmental
Oceanographic Commission (IOC) of UNESCO, with overarching objectives to determine
the predictability of climate and the effect of human activities on climate. WWRP and
WCRP share the responsibility to advance scientific knowledge and the science
infrastructure to provide society and decision-makers with: i) accurate predictions of
high-impact weather, climate and environmental events; ii) information for the reduction
of socioeconomic and environmental losses related to high-impact weather, climate
variability and change.
The collaboration is timely because of recent advances in observing technologies, field and
laboratory process studies, data-assimilation techniques, and coupled numerical models of
weather and climate prediction aided by high-performance computing advances. The
challenge is to leverage these advances to develop and apply new forecast and diagnostic
products and increase their applications to the economy and society. The next generation of
numerical climate and weather prediction operational systems, based on coupled
ocean-land-atmosphere models, will greatly benefit from this broad effort.
Page 4
The
collaborative research is described within the framework of background, underpinning
research, linkage and requirements. The research objectives are:
•
Seamless weather/climate prediction including Ensemble Prediction Systems
(EPSs)
•
The multi-scale organisation of tropical convection and its two-way interaction
with the global circulation
•
Data assimilation for coupled models as a prediction and validation tool for
weather and climate research
•
Utilization of sub-seasonal and seasonal predictions for social and economic
benefits
2. Seamless Weather/Climate Ensemble Prediction Systems (EPSs)
Background: A fundamental principle of seamless prediction is that the global coupled
atmosphere-ocean-land-cryosphere system exhibits a wide range of dynamical, physical,
biological and chemical interactions involving a scale of spatial and temporal variability of
the overall system. The traditional boundaries between weather and climate are
non-physical. As explained in Hurrell et al. (2009) for example, the slowly varying
planetary-scale circulation preconditions the environment for the “fast-acting” micro-scale
Page 5
and mesoscale processes of daily high-impact weather and regional climate. As an example,
there is evidence that natural climate variations, such as ENSO and the North Atlantic
Oscillation/Northern Annular Mode, significantly alter the intensity, track and frequency
of hazardous weather, such as extratropical and tropical cyclones and associated
high-impact weather.
Longer-term regional variations have been observed, such as
decadal variability in tropical cyclones and the multi-decadal drought in the Sahel region.
Conversely, small-scale processes have significant up-scale effects on the evolution of the
large-scale circulation and the interactions among the components of the global climate
system. The challenge facing the weather and climate science communities is to improve
the prediction of the spatial-temporal continuum of the interactions between
weather/climate and Earth-system 2 . Bridging the gap between forecasting high-impact
events at daily-to-seasonal timescales would benefit from coordinating the activities of the
WCRP CLIVAR Climate-system Historical Forecast Project (CHFP) and the THORPEX
Interactive Grand Global Ensemble (TIGGE). There is evidence that weather and climate
ensemble prediction has greater use and value from a Multi-model Ensemble Prediction
System (MEPS) approach. Compelling evidence is that different physics, numerics and
initial conditions of the contributing EPSs provide more useful probability density
functions (PDFs) than those obtained from a single EPS. Moreover, the MEPS approach
identifies which outcomes are EPS independent, and hence likely to be robust.
2
In this context the Earth-system is understood to mean the atmosphere and its chemical composition, the
oceans, sea-ice and other cryosphere components, the land-surface, including surface hydrology and
wetlands, lakes, and short-time scale phenomena that result from the interaction between one or more
components, such as ocean waves and storm surge. On longer (e.g. climate) time scales, the terrestrial and
ocean ecosystems, including the carbon and nitrogen cycles, and slowly varying cryosphere components
such as the large continental ice sheets and permafrost, are also considered to be part of the Earth-system.
Page 6
While a goal of weather and climate EPS is to produce model outputs that are unbiased,
together with ensemble forecasts that properly account for uncertainty, models have biases
and atmospheric ensemble predictions only partially account for the true uncertainty.
Because of the complexity of the weather and climate systems, it is a non-trivial task to
post-process model output for their applications. This is especially true for seamless
forecast applications where a wide range of future time horizons must be addressed.
Efforts have largely focused on Model Output Systems (MOS) approaches that are very
useful but only partially meet the needs of users. This is because the skill and uncertainty
of weather and climate forecasts are highly space and time-scale dependent. Accounting
for this dependency is critical for many EPS applications that are sensitive to the
space-time variability of weather and climate events. The necessity of evaluating MEPs
biases and forecasting skill in the sub-seasonal time scale will require hindcast
experiments.
Ensemble prediction systems are widely used for weather and environmental (e.g.
hydrological) prediction by operational services throughout the world. Ensemble forecasts
not only offer an estimate of the most probable future state of a system, but also provide an
estimate of the range of possible outcomes. From the perspective of decision-making,
assessing how climate sub-seasonal to seasonal variations may alter the frequencies,
intensities, and locations of high-impact events is an issue of the highest priority. Many
users are risk averse, in that they are often more concerned with quantitative estimates of
the probability of occurrence of high-impact events than with the most probable future
state.
Page 7
Interaction among spatial and temporal scales are fundamental to weather and climate.
Atmospheric, hydrologic, biospheric, cryospheric and oceanic models, and their
interactions, are essential to both weather and climate predictions.
Modelling and
predicting seasonal climate anomalies requires a realistic treatment of the effects of
sea-surface temperature, sea ice, snow, soil wetness, vegetation, stratospheric processes,
and chemical composition. It is well recognized that sub-seasonal and seasonal prediction
systems must realistically represent day-to-day weather fluctuations and their statistical
characteristics. The collaboration between the CHFP and TIGGE activities will provide the
unique opportunity to bring the weather and climate communities together to address
MEPS on time scale from weeks-to-season. As an example, there is indication (Palmer et al.
2008) that the statistical characteristics of dynamical and physical processes, like the
atmospheric response to sea surface temperature, in seasonal forecast based on a MEPS
can be utilize to assess the reliability of climate projection done with the same MEPS.
Operational weather forecast systems provide our best representation of synoptic-scale and
mesoscale weather events. However, these short-range to medium-range (~10 days)
forecast models traditionally have not addressed key interactions, e.g. at the air-sea-ice
interface. We know that this is problematic on time scales beyond two weeks. It may well
be an impediment to improving forecasts on shorter time scales, particularly for
high-impact weather events. These events include extratropical and tropical cyclones that,
through interaction with the ocean mixed layer, can precondition the atmosphere-ocean
interface for subsequent storms. Seasonal prediction systems, on the other hand, typically
Page 8
include such coupled interactions, yet they fail to adequately resolve mesoscale weather
systems.
Consideration of current operational ENSO forecasting indicates that the coupled models
give good guidance as to the future evolution of SST up to 6 months in advance. However,
there are profound gaps in our prediction capabilities in part due to large systematic errors
in the coupled models. These errors are perhaps, most noticeable in the equatorial Atlantic
where the coupled models fail to capture the east-west SST contrast. These mean state
errors and errors in the evolution of climate anomalies have been addressed
semi-empirically, such as to improve the individual physical parameterizations in the
models and allow for imperfect models in EPSs and MEPSs (Hurrell et al, 2009). The
seamless prediction approach raises a third problem, that current climate models poorly
represent the statistics of weather events for which there is predictive skill. Moreover, the
specification of accurate initial conditions has an impact on the skill of daily-to-seasonal
prediction (Section 4). The issue then becomes: What are the important missing elements
of the statistics of the physical processes and data-assimilation aspects?
The typical assumption for sub-grid scale parameterization is to assume that the statistics
of sub-grid scale processes can be parameterized in terms of the grid-scale variables. An
alternative strategy is to increase the grid-resolution of the model and explicitly represent
key dynamical/thermodynamical processes (Section 3). While this approach has yielded
some improvements, it is limited by the available computing capacity and incomplete
explicit representations.
Page 9
Underpinning Research: A central science issue is the development and use of
ensemble-based modelling in order to improve probabilistic estimates of the likelihood of
high-impact events. Because of the relatively small scale of many events, there will be an
ongoing need to improve model resolution and develop alternative downscaling techniques,
especially for specific user-applications such as hydrology. The requirements for both
ensemble prediction methods and greatly increased spatial resolution imply a need for
substantial future improvements in computational power and data storage.
There are a wide range of scale interactions to be considered within the context of
improving EPSs. One of the most critical is the manner in which moist convection and its
associated mesoscale organisation drive regional and global circulations. Such interactions
directly affect the capability to simulate important climatic and weather features such as
ENSO and tropical cyclones, respectively. They also affect simulations at higher latitudes,
as the export of wave energy and momentum from the tropics is an important driver of
mid-latitude circulations. These research issues are discussed in Section 3.
There is also the problem of forecasting initial conditions in the stratosphere. The
stratosphere has more intrinsic memory than the troposphere, with significant
intra-seasonal memory (during winter) in the polar regions, and inter-annual memory in the
tropical winds. Thus, correct stratospheric initial conditions, and realistic retention of those
conditions in the model, are required for accurate sub-seasonal and seasonal prediction
(Section 4).
Page 10
Linkages: Both the TIGGE and CHFP communities are planning multi-model
multi-institutional numerical experiments using state-of-the-art models and computing
systems, which generate large data sets that need to be shared. The sharing of sub-seasonal
to seasonal prediction datasets, from both retrospective forecasts and near real-time
forecasts, requires a common framework for comparison and diagnostic activities to bridge
the TIGGE and CHFP communities.
Requirements: Terms of reference for collaboration between TIGGE and CHFP must be
established. In particular, a limited number of data archive centres need to be identified and
support the scientific and user communities. These databases will require unprecedented
storage capacity. Note that TIGGE is already providing such support to the participating
centres for MEPS, but limited to two-week forecasts.
3. Tropical convection and its two-way interaction with the global circulation
Background: Tropical convection exhibits a remarkable variability and organization
across space- and time-scales, ranging from individual cumulus clouds, to mesoscale cloud
clusters to super clusters (families of mesoscale clusters) organized within synoptic-scale
disturbances. Tropical synoptic activity is often associated with equatorially-trapped wave
modes of the atmospheric circulation (Wheeler et al. 2000; Yang et al. 2007), which in turn
organize tropical convection: a highly nonlinear scale-interaction problem. Forecast skill
in the tropics is therefore dependent upon representing both equatorial waves and
convective organization. The limitation of contemporary weather and climate-prediction
Page 11
models to realistically represent the life-cycle of equatorial waves and organized
convection is usually attributed to inadequacies in parameterizations of moist physical
processes. Such basic inadequacies
compromise the skill of forecasts on timescales of
days to weeks and beyond, including projections of climate change. It follows that the
parameterization of organized tropical convection is a critical issue.
In the Madden Julian Oscillation (MJO) and other convectively coupled equatorial waves
and in monsoons, precipitating convection organizes into coherent structures (convective
clusters) on spatial scales up to 1000 times that of an individual cumulonimbus. The MJO
excites Rossby wave trains that propagate into extratopical Pacific and North American
and North Atlantic regions causing episodes of high-impact weather. In Table 1, a
time-lagged likelihood study (Lin et al. 2008) shows that the NAO index for different
phases of the MJO indicates a two-way connection between the NAO and the tropical
convection of the MJO. A possible triggering and/or amplifying mechanism for MJO is
linked to an important change of upper zonal wind in the tropical Atlantic associated with
the NAO.
Advances in the representation of the tropical-extratropical interactions would lead to more
skilful prediction of regional-to-global weather and climate. This success will translate into
socio-economic applications for improving early-warning systems for weather-climate
induced hazards, e.g. agriculture, water management, and health. The MJO is considered to
be a significant aspect of ENSO through its forcing of the equatorial ocean (e.g. Kutsuwada
and McPhaden 2002). We are poised to advance the representation of convective
Page 12
organization in global weather, climate and Earth-prediction models given our knowledge
of the systemic properties of organized tropical convection.
There is substantial interaction between the tropical and extratropical atmosphere from
synoptic to decadal time scales and beyond. At the synoptic scale, energy originating at
high latitudes propagates into the tropics through Rossby wave dispersion, initiating in
regions of upper-level westerly flow in the Pacific and Atlantic storm tracks. Such wave
trains frequently excite convection within the Inter Tropical Convergence Zone (ITCZ),
transporting moisture from the tropical boundary layer into the upper-troposphere and
transporting it poleward ahead of troughs extending into extratropics of both hemisphere
(Knippertz 2007). Moisture transport between the tropics and extratropics is enhanced
during such synoptic-scale events. Australia, Europe and North and South America are
impacted by the tropical moist intrusions, sometimes leading to sustained episodes of
heavy precipitation and flooding. Other aspects of sub-seasonal variability within the
extratropical storm tracks that pose problems for both weather forecast and climate models
are the initiation and maintenance of blocking and wave-mean flow interactions. These
contribute to variability in teleconnection patterns, such as the North Atlantic Oscillation
(NAO) and Pacific-North American (PNA).
Underpinning Research: Research with cloud-system resolving models (CSRMs) should
be undertaken using a horizontal grid-spacing of 1 km or finer, as well as CSRMS nested in
coarser global simulations. Present computer capacity precludes cloud-resolving
representations of moist convection in global sub-seasonal-to-seasonal deterministic
Page 13
prediction models and EPSs. It is therefore essential to accelerate efforts to improve
traditional convective parameterizations.
The following research foci were identified at the March 2006, THORPEX and WCRP
international workshop at the ICTP, Trieste, Italy (Moncrieff et al., 2007):
•
Develop metrics/description of the daily, sub-seasonal, and seasonal characteristics
of the MJO and organized convection that encapsulate our knowledge of these
interactions, and assess model/forecast validation with the aim of guiding future
research.
•
Promote collaboration on the use of Numerical Weather Prediction (NWP)-type
experiments for exploring error growth in simulations of the MJO and other modes
of organized convection and of two-way interactions between tropical and
extratropical weather and climate systems.
•
Integrate process studies of observed organized convection based on satellite and
ground-based remote sensing (including 3D Doppler radar), with in situ
measurements to provide improvements and validation of high-resolution models
as it is proposed in the WWRP/WCRP Year of Tropical Convection (Moncrieff and
Shapiro, 2009).
•
Consider the feasibility and strategy for the design and implementation of field
campaigns on organized convection (e.g., over the Indian Ocean) guided by
high-resolution modelling studies.
Page 14
•
Develop a strategy for demonstration and assessment of socio-economic benefits
and applications arising from advanced knowledge and predictive skill of
multi-scale tropical weather/climate events on timescales of days to seasons.
Linkages: In order to advance the aforementioned research objectives it will be necessary
to: i) promote collaboration on forecast demonstration experiments coupling prediction
systems based on statistical techniques with those based on dynamical models to assess the
value of improved MJO/organized convection simulations for deterministic and ensemble
prediction on timescales up to 4 weeks; ii) promote the transfer of advanced knowledge
and predictive skill of organized convection into improvements for operational NWP and
climate models through links with key groups within GEWEX/CLIVAR/THORPEX, as
well as operational prediction centers; iii) promote international collaboration on CSRM
studies for exploring the upscale energy cascade associated with organized convection in
order to optimize use of computing resources and to share the development of data analysis
tools.
Requirements: Achieving the goals of the above research objectives will require: i)
operational high-resolution global analysis and 10-day forecasts; ii) access to
High-Performance Computing (HPC) centers for high-resolution regional and global
numerical weather, climate and environmental science activities (Shapiro et al., 2009,
Shukla et al., 2009). This will lead to efficient numerical models, advanced experimental
design, and improved data processing, distribution and analysis; iii) maintaining existing
and implementing planned satellite missions that observe tropical cloud and precipitation
Page 15
to provide long-term capability for process studies, data assimilation and prediction in
collaboration with WMO Global Climate Observing System (GCOS). This includes access
to experimental in situ and satellite observations to establish data bases for diagnostic
studies.
4. Data assimilation for coupled models as a prediction and validation tool for weather
and climate research
Background: Fundamental issues, related to data-assimilation at different scales, must be
addressed before we can design “seamless” Earth system prediction systems. Historically,
data assimilation research and its applications have focussed mostly on the requirements of
operational forecast deliverables for short to medium-range applications. As operational
forecasting has extended into sub-seasonal prediction, improved data-assimilation in the
tropics, ocean, upper atmosphere and Earth-system information have become necessary. A
unified system will accelerate improvement of weather/climate models and applications as
advocated by Palmer and Webster (1995). The return on the investment by society in
existing and new observations will be significantly increased by advances in data
assimilation systems.
Data assimilation is the process of fitting a numerical prediction model to observations,
allowing for the error characteristics of both. The model carries information in time,
allowing more observations to influence the estimated state at any time, so using a good
model is crucial. The models used in existing NWP systems are therefore designed to
utilise available computer power: future Earth system models are anticipated to do
Page 16
likewise. Practical assimilation methods have to allow for the error characteristics of model
forecasts, without the enormous computer costs required to calculate them fully.
Data assimilation innovations (observation minus prediction) allow a diagnosis of errors
while they are still small, before they interact significantly with other fields.
This
established NWP approach is also proving beneficial for climate models, see Rodwell and
Palmer (2007), through application in centres where climate and NWP modelling is unified
and by the WGNE Transpose AMIP programme. The method permits direct comparison of
parameterised variables such as clouds and precipitation with synoptic observations,
satellite and field campaign measurements.
Underpinning Research: It will not be possible, with foreseeable computers, to conceive
of a single data-assimilation method for an Earth system model with the complexity
required for seamless prediction. What is possible is a composite system, applying
different assimilation steps to different scales and components of the total Earth system
model. These can be based on the methods currently used in specialised systems, such as
NWP. Recent attempts to build such a composite system use a two-way interaction model
for the forecast step, but apply assimilation to each component separately. Ideally, the
assimilation should be coupled, so that observed information in one component is used to
correct fields in the other coupled component. One of the few attempts to do this is coupled
land-atmosphere assimilation, where soil moisture is corrected based on errors in
atmospheric forecasts of near-surface temperature and humidity. Yet many land surface
modellers distrust such soil moisture analyses; because of compensating errors they can
Page 17
give soil moistures which reduce atmospheric forecast errors but do not correspond to
actual soil moistures nor do they conserve the water budget. Coupled data assimilation
must be accompanied by a much better characterization of the errors and biases in
components of a coupled model (e.g. atmosphere and the upper ocean). Only then can we
successfully correct them as part of the data assimilation process.
In addition the following topics need addressing for a seamless coupled system:
One challenge that needs to be addressed in the seamless prediction problem is the fact that
assimilation methods attempt to estimate only a certain range of scales (temporal and
spatial). For example, all current operational implementations of 4D variational
assimilation compare measurements to model forecasts over a fixed assimilation window
and assume that the flow evolution is weakly nonlinear during this window. This implies
that time and spatial scales for which this is true can be well resolved if those same scales
are observed. Thus, for the global NWP assimilation problem, with an assimilation
window of 6 or 12 hours, synoptic scale flow can be estimated since it is well
observed. Similarly, a high resolution cloud-scale model with a very short assimilation
window can resolve fine scales if they are observed, for example by Doppler
radar. However, when a model can simulate a very wide range of scales of motion, this
method of assimilation can be limiting. Thus, for a global model with extremely high
resolution, synoptic scale flow has a nonlinear time scale commensurate with the
assimilation window, but convective scale motions do not. In addition, convective scale
motion is not completely observed over the whole globe. Thus 4D variational assimilation
methods (as implemented currently) rely on the fact that larger scales can generate smaller
Page 18
scales through nonlinear interactions. The final analysis of the global model with
extremely high resolution will contain fine scale features which are developed during the
nonlinear forecast. However, these features may not match the observed flow on these fine
scales. An alternate approach to the global prediction problem uses ensembles of
forecasts. Then an ensemble of possible fine scale structures will be obtained. In this case,
it would be useful to know not only which ensemble members are most accurate, but also
whether observations can help constrain the range of the ensemble in terms of the power at
scales which are not completely observed. In other words, how can assimilation methods
make use of information about power on these scales? Such information might come
directly from satellite images or indirectly from measurements of eddy-fluxes for instance
from observations of large scales of the MJO, plus an understanding of the convection
necessary to drive them, or from estimates of an eddy-flux needed to give the observed
ocean state, or from measurements of the age of stratospheric air and the Brewer-Dobson
circulation plus an understanding of the vertical eddy fluxes which drive this
circulation. Part of the problem with these types of indirect measurements is that they
contain information about time scales which are much longer than those resolved by
current assimilation schemes. Thus, a major challenge will be to build a composite
assimilation system for the Earth-system capable of dealing with a wide range of time
scales, from atmospheric to oceanic time scales.
Besides obtaining an initial condition for launching a weather forecast, data assimilation
methodology can also be used for parameter estimation. Since the largest uncertainties
with climate and weather models are associated with their physical parameterizations,
Page 19
improvements in these schemes may reap great benefits. This process has been used in the
NWP context, but is relatively new for climate models. It is simple enough to determine
uncertain parameters for a given parameterization scheme (such as gravity wave drag or
convective schemes). However, if the scheme is not of the correct form (i.e. it has too few
or too many parameters, or is missing processes) then the results of assimilation may not
lead to useful parameter estimates. However, the failure of the assimilation process could
provide an indication of the inappropriateness of a given scheme without directly
indicating how it should be improved. Thus, close collaboration with model developers is
needed to interpret assimilation results and address flaws in specific schemes. As an
example, Figure 1 shows that a gravity wave drag (GWD) scheme can provide the
dominant forcing in the mesosphere for some phenomena (here a stratospheric sudden
warming). The GWD parameterization is responsible for the mesospheric cooling
(upwelling) simulated at 80 km over the polar cap in an ensemble of forecasts that captures
the stratospheric warming (thick solid). For the forecasts that miss the stratospheric
warming (thin solid curve), the upwelling and associated cooling due to parameterized
waves is much smaller. In these experiments, observations were only assimilated below 50
km so the mesospheric response occurs entirely through the model dynamics. This
example shows that the zonal-mean mesosphere is slaved to the stratosphere through
GWD, raising the prospect of using data assimilation of mesospheric observations to
constrain GWD parameters.
Linkages: Data assimilation, fitting models to observations, should be a key component in
model development. This already occurs for NWP, but is less common for climate models.
Page 20
The trend toward unified weather/climate models, and activities such as the WGNE
Transpose-AMIP programme should help this. The coupled seamless prediction system
requires unified data assimilation and model development. There is need to test “climate
modelling in a deterministic prediction mode” as advocated by Morel (2007). The
utilization of deterministic forecasts to validate climate models has already been forged for
the middle atmosphere through the WCRP SPARC programme. Ocean assimilation is now
part of seasonal prediction programmes. Environmental monitoring initiatives have linked
assimilation with models of atmospheric composition.
Requirements: New resources would accelerate the development of seamless coupled
model and data assimilation systems discussed above. One mechanism to achieve this is
through the various re-analysis projects which are designed to provide a historical record
for climate studies as promoted also in Trenberth (2008). In the past these have been based
on operational NWP systems; most of the resources were directed towards gathering and
quality controlling the observations and performing the assimilation. Next-generation
developments can no longer rely on operational forecast systems but require an
interdisciplinary research programme on data-assimilation methodologies.
5. Socio-Economic Applications of Sub-Seasonal and Seasonal Predictions
Background: The primary rationale for pursuing a seamless prediction process is that the
resulting information will influence decisions that contribute to the achievement of societal
objectives, including: i) protection of life and property; ii) enhancement of socioeconomic
well-being; iii) improvement of the quality of life; and iv) sustainability of the environment
Page 21
(Shapiro et al., 2009). Such objectives are embodied in the broad mandates, missions, and
visions of national governments or international organizations (e.g., Rogers et al. 2007).
The rationale has merit in that weather and climate predictions to date have made
significant albeit variable contributions to a wide range of economic sectors and public
policy issues, including: i) agriculture, water resources and the natural environment; ii)
human health; iii) tourism and human welfare; iv) energy, transport and communications;
v) urban settlement and sustainable development; vi) economics and financial services
(WMO 2008). Weather forecasts have been proven useful for many forms of tactical
decision-making in these sectors and the number of applications to longer-term operational
and planning decisions—including those related to climate change—is growing. However,
there is considerable evidence of underutilization of weather and climate information that
may be rooted as much in a lack of understanding of the decision-making context and
requirements of users as in the precision or accuracy of atmospheric predictions. A variety
of constraints make it difficult for decision-makers to benefit fully from scientific
information and for the science to satisfy users’ needs (Jasanoff and Wynne 1998, Morss et
al. 2005, Rayner et al. 2005). A seamless prediction process may help resolve this problem.
Take public health for example. Figure 2 depicts a sample of public health decisions
covering a wide range of temporal scales. Each may be influenced by weather, climate and
even climate change predictions, but only in combination with other pieces of information
(e.g., expected disease outbreak patterns, available medical supplies and resources, etc.)
that more directly relate to important health outcomes of interest to decision-makers. In
Page 22
this sense the term seamless extends beyond the realm of atmospheric predictions to
include consideration of biophysical and socio-economic factors that are pertinent to
successful decision-making and outcomes from the user’s perspective. Potential benefits
are greatest in developing nations, especially in Africa where at least 30 climate-sensitive
diseases pose a major threat to the lives and livelihoods of millions of people. More than
500 million Africans live in regions endemic to malaria, which is highly correlated with the
seasonal climate, and a further 125 million live in regions prone to epidemic malaria,
which is correlated with climate anomalies (Connor and Thomson 2005). The response
time to a particular outbreak or epidemic is greater than one week and often much longer,
depending on the time it takes to identify cases and integrate the information from different
clinics—this makes this issue ideal for atmospheric prediction applications at seasonal and
sub-seasonal timescales, the key integrating period for seamless weather and climate
prediction.
Developing an enriched concept of seamless prediction with WWRP/WCRP will require
active involvement of researchers from physical and social science disciplines along with
service providers and users/decision-makers (Morss et al. 2008). Previous experience at
the science-society interface has demonstrated that much scientific research provides little,
if any, direct benefit to society without focused interdisciplinary efforts dedicated to
providing relevant, useable information that helps key decision-makers address specific
societal issues (Fothergill 2000, White et al. 2001).
Page 23
Underpinning Research: The specific type of research required will depend on the
weather- or climate-sensitive issue, geographic area, and decision context, as will the most
appropriate disciplinary expertise and methodologies. Priority projects may be selected
based on their potential contribution to the societal objectives listed previously (e.g.,
locations with greatest weather- and climate-related mortality or property loss; vulnerable
economic sectors) or where existing programmes, activities and interdisciplinary
collaborations can be leveraged (e.g., MERIT project 3 ). Regardless, efforts should focus
on: i) understanding the relevance of weather and climate information to the issue, the
decision context, and decision-maker’s information needs; ii) identifying new or improved
weather- and climate-related information that is likely to help decision-makers address the
socioeconomic issue; iii) exploring the most effective mechanisms for generating and
communicating the decision-relevant weather and climate information; iv) assessing the
use and value in decision-making, making refinements as needed; v) implementing
strategies for sustainable, effective provision of the most valuable new weather- and
climate-related information and vi) transferring knowledge and experiences to other
regions.
Linkages:
Linkages are needed with: i) policy- and decision-makers from industry, government, and
non-governmental organizations involved in managing the weather- or climate-sensitive
issue or activity of concern; ii) weather, climate and hydrometeorological service
3
Meningitis Environmental Risk Information Technologies (MERIT) project
http://merit.hc-foundation.org/aboutMERIT.html
Page 24
providers; and iii) social and interdisciplinary scientists with expertise in the issue or
activity of concern. Examples of specific existing programs to link with include: i) the
Hydrological Ensemble Prediction Experiment (HEPEX), an international project to
advance technologies for hydrological forecasting comprised primarily of researchers,
forecasters, water managers, and users; ii) Climate for Development in Africa Programme
(ClimDev Africa), a major effort designed to increase the availability of climate
information to communities and economic sectors throughout Africa; iii) the World Bank’s
Disaster Risk Reduction program, which plans to help modernize service providers so that
the operational services can take advantage of scientific advances in a timely manner; and
iv) the Global Environmental Change and Human Health initiative (Confalonieri and
McMichael 2007), one of four joint projects of the Earth System Partnerhsip (ESSP) and
geared to quantifying and modeling health impacts and vulnerability and evaluating
adaptation measures. Efforts should also contribute to the WMO Madrid Conference
Action Plan (WMO 2007).
Requirements: The specific capacity needed depends on the socioeconomic application
area, the country or region, and the project. Within each country or region, it is important
to determine where the greatest potential for use of sub-seasonal to seasonal forecasts
exists, and where largest social benefit and biggest buy-in can be realized. These areas can
then be used as a focus for articulating requirements. Another need is for closer ties among
weather and climate research centres that can provide state-of-the-art weather and climate
forecast information in a form that can be easily accessed by non-atmospheric scientists,
Page 25
user groups interested in applying the information, and intermediaries who understand
both the scientific and socioeconomic issues (Shapiro et al., 2009). Liaison with
programmes such as WAS*IS (Demuth et al. 2007), START (http://www.start.org) and
DISCCRS (http://www.disccrs.org/) will be of particular assistance in developing the pool
of interdisciplinary scientists.
Aknowledgement: We would like to thank the comments and/or support of Antonio J.
Busalacchi, Philippe Bougeault, Eric Brun, David Burridge, John Church, Pascale
Decluse, Michel Déqué, Russel Elsberry, Tom Hamill, Ann Henderson-Sellers,
Gudrun Magnusdottir, Martin Miller, L. Aogallo, Steven Pawson, Kamal Puri,
Adrian Simmons, Julia Slingo, Soroosh Sorooshian, Istvan Szunyogh, Theodore
Shepherd, Ilana Wainer, Duane Waliser and Laurie Wilson.
2.4 References
Confalonieri, U. and A. McMichael (eds.), 2007: Global Environmental Change and
Human Health Science Plan and Implementation Strategy. Earth System Science
Partnership (DIVERSITAS, IGBP, IHDP, and WCRP) Report No.4; Global
Environmental Change and Human Health Report No.1 (Confalonieri, U. and A.
McMichael, co-chairs) (http://www.igbp.net/documents/GECHH-SP.pdf)
Connor, S.J. and M.C. Thomson, 2005: Epidemic malaria: preparing for the unexpected. A
policy brief, in SciDevNet dossier on malaria, SciDevNet.
Demuth, J. L., E. Gruntfest, R. E. Morss, S. Drobot, and J. K. Lazo, 2007: Weather and
Society * Integrated Studies (WAS*IS): Building a community for integrating
meteorology and social science. Bull. Amer. Meteor. Soc., 88, 1729-1737.
Page 26
Fothergill, A., 2000: Knowledge transfer between researchers and practitioners. Nat.
Hazards Rev., 1, 91–98.
Hurrell, James, G.A. Meehl, D. Bader, T. Delworth, B. Kirtman and B. Wielicki, 2009:
Climate system prediction. To appear in BAMS.
Jasanoff, S., and B. Wynne, 1998: Science and decisionmaking. Human Choice and
Climate Change, S. Rayner and E. L. Malone, Eds., Battelle Press, 1–87.
Kutsuwada, K., and M. McPhaden, 2002: Intraseasonal variations in the upper equatorial
Pacific Ocean prior to and during the 1997-1998 El Niño. J. Phys. Oceanogr., 32,
1133-1149.
Knippertz, P., 2007: Tropical-extratropical interactions related to upper-level troughs at
low latitudes. Dyn. Atmos. Oceans, 43, 36-62.
Lin, H., G. Brunet, J. Derome 2008 An observed connection between the North Atlantic
Oscillation and the Madden-Julian Oscillation. To appear in Journal of Climate.
Moncrieff, M.W., M.A. Shapiro, J. Slingo and F. Molteni, 2007: Collaborative research at
the intersection of weather and climate. WMO Bulletin, Vol.56 (3), July 2007,
p.204-211.
Moncrieff, M.W., M.A. Shapiro, 2009: A scientific basis for addressing the multi-scale
organization of tropical convection and its interaction with the global circulation.
To appear in BAMS.
Morel, P. 2007: Can GEWEX become the cutting edge of WCRP? GEWEX/WCRP News,
Vol.17, No. 4,7-11)
Page 27
Morss, R. E., O. V. Wilhelmi, M. W. Downton, and E. Gruntfest, 2005: Flood risk,
uncertainty, and scientific information for decision-making: Lessons from an
interdisciplinary project. Bull. Amer. Meteor. Soc., 86, 1593–1601.
Morss, R. E., J. K. Lazo, B. G. Brown, H. E. Brooks, P. T. Ganderton, and B. N. Mills,
2008: Societal and economic research and applications for weather forecasts:
Priorities for the North American THORPEX program. Bull. Amer. Meteor. Soc.,
89, 335-346.
Palmer. T.N. and P.J.Webster, 1995: Towards a unified approach to climate and weather
prediction. Global Change. EUR 15158, 427pp, European Commission. ISBN
92-826-7757-5
Palmer, T.N., F. J. Doblas-Reyes, A. Weisheimer, and M. J. Rodwell 2008: Toward
Seamless Prediction: Calibration of Climate Change Projections Using Seasonal
Forecasts. Bull. Amer. Meteor. Soc., 89, 459–470.
Rayner, S., D. Lach, and H. Ingram, 2005: Weather forecasts are for wimps. Climatic
Change, 69, 197-227.
Ren, S., S. M. Polavarapu, and T. G. Shepherd (2008), Vertical propagation of information
in a middle atmosphere data assimilation system by gravity-wave drag feedbacks,
Geophys. Res. Lett., 35, L06804, doi:10.1029/2007GL032699.
Rodwell, M and T.N.Palmer, 2007: Using numerical weather prediction to assess climate
models. Q.J.R.Meteorol.Soc., 133, 129-146.
Rogers, D.P., C. Clarke, S.J. Connor, P. Dexter, J. Guddal, A.I. Korshunov, J. K. Laso,
M.I. Smetanina, B. Stewart, Tang Xu, V.V. Tsirkunov, S.I Ulatov, Pai-Yei Whung,
Page 28
and D.A. Wilhite, 2007: Deriving societal and economic benefits from
meteorological and hydrological services. WMO Bulletin, 56, 15-22.
Shapiro, M., J. Shukla, M. Béland, J. Church, K. Trenberth, B. Hoskins, G. Brasseur, M.
Wallace, G. McBean, A. Busalacchi, G. Asrar, D. Rogers, G. Brunet, L. Barrie, D.
Parsons, D. Burridge, T. Nakazawa, M. Miller, P. Bougeault, R. Anthes, Z. Toth, J.
Meehl, R. Dole, M. Moncrieff, H. Le Treut, A. Troccoli, T. Palmer, J. Marotzke, J.
Mitchell, A. Simmons, B. Mills, Ø. Hov, H. Olafsson and J. Caughey, 2009: A
Weather, Climate and Earth-system Prediction Initiative for the 21st Century.
Submitted to BAMS.
Trenberth, K. E., 2008: Observational needs for climate prediction and adaptation. WMO
Bulletin, 57 (1), 17-21.
Wheeler, M., G.N. Kiladis, and P.J. Webster (2000), Large-scale dynamical fields
associated with convectively-coupled equatorial waves, J. Atmos. Sci., 57,
613-640.
White, G. F., R. W. Kates, and I. Burton, 2001: Knowing better and losing even more: The
use of knowledge in hazards management. Environ. Hazards, 3, 81–92.
WMO (World Meteorological Organization) 2007: Madrid Conference Statement and
Action Plan. Adopted by the International Conference on Secure and Sustainable
Living: Social and economic benefits of weather, climate and water services.
Madrid, Spain 19-22 March 2007. WMO JN 13753.
Page 29
WMO (World Meteorological Organization) 2008: Secure and Sustainable Living: The
findings of the Madrid Conference on Social and Economic Benefits of Weather,
Climate and Water Services. To be published.
Yang, G. –Y., B. Hoskins, and J. Slingo (2007), Convectively coupled equatorial waves.
Part I: Horizontal and vertical structures, J. Atmos. Sci., 64, 3406-3423.
Page 30
Phase
Lag −5
1
2
3
−35
−40
4
Lag −4
Lag −3
5
6
7
+49
+49
+52
+46
−40
8
+46
Lag −2
+50
Lag −1
Lag 0
Lag 1
Lag 2
+47
Lag 3
+48
+45
−42
+47
+45
−46
+50
+42
Lag 4
Lag 5
−41
−41
−41
−48
−39
−48
−42
−41
Table 1: Lagged probability composites of the NAO index with respect to each MJO phase.
Lag n means that the NAO lags the MJO of the specific phase by n pentads, while Lag –n
Page 31
represents that the NAO leads the MJO by n pentads. Positive values are for upper tercile,
while negative, while negative for low tercile. Values shown are only for those pass a 0.05
significance level according to a Monte Carlo test. (From Lin et al. 2008)
Page 32
Figure 1: Residual vertical velocity (w*) induced by the resolved waves (dashed) and
non-orographic gravity wave drag (GWD) parameterization (solid) in the steady,
“downward control” limit. The calculation is area weighted and averaged over the polar
cap (60°-90°S) and over the period Sept. 25 – Oct. 1, 2002 for an ensemble of 3 forecasts
that capture (thick curves) or miss (thin curves) the 2002 stratospheric sudden warming in
the southern hemisphere. Positive w* is associated with cooling, and negative w* with
warming. The GWD parameterization is responsible for the mesospheric cooling
(upwelling) around 80 km in the forecast “hits” (thick solid). (From Ren et al. 2008)
Page 33
Figure 2. Simplified set of public health- related decisions and supporting information
Page 34
The Multi-scale Organization of Tropical Convection and its Interaction with the
Global Circulation: Year of Tropical Convection (YOTC)
Capsule: International research at the weather-climate intersection utilizing observations, highresolution prediction systems, cloud-system models and theory with emphasis on representing
convective organization in global models.
Mitchell W. Moncrieff1, National Center for Atmospheric Research, USA
Duane E. Waliser, Jet Propulsion Laboratory, USA
Melvyn . A. Shapiro, CIRES/NOAA/Univ. Colorado, USA; Geophysics Institute, University of
Bergen, Norway
Ghassem R. Asrar, World Meteorological Organization, Geneva, Switzerland
Leonard A. Barrie, World Meteorological Organization, Geneva, Switzerland
Submitted to Bull. Amer. Meteor. Soc.
1
Corresponding author address: Mitchell W. Moncrieff, Mesoscale and Microscale Meteorology Division, National Center
for Atmospheric Research, 1850 Table Mesa Drive, Boulder, CO 80305, USA.
E-mail: [email protected]
1
ABSTRACT
Improving the representation of precipitating convection and its multi-scale organization in regionalto-global weather prediction and climate models is a critical challenge addressed by a new
international research project, the Year of Tropical Convection (YOTC). This project provides the
framework, infrastructure and research objectives for an unsurpassed integrated observationalnumerical research resource with emphasis on time scales up to seasonal, thereby addressing
issues at the intersection weather and climate. The research employs high-resolution operational
global prediction models, operational/research satellite-borne, air-borne and surface-based
observations, ultra-high-resolution regional-to-global cloud-system resolving models, and idealized
models. The YOTC is a key element of “seamless prediction” described in Brunet et al.; Shapiro et
al; Shukla et al., this issue.
1. Introduction
The prediction of weather and climate and the global hydrological cycle involves moist physical
processes and how they are coupled to the large-scale circulation of the atmosphere and the
oceanic boundary layer. Precipitating convection organized into cloud systems on mesoscales
(~10-1000 km) represents a continuum of interaction whose thermodynamic and dynamical
properties affect the regional-to-global atmospheric circulation. Improving the representation of
moist convection in models used for regional-to-global weather prediction and climate prediction
requires a more realistic treatment of organized convection. New approaches are available to
comprehensively address convective organization, an issue that has confronted prediction models
since their inception over four decades ago.
In response to this challenge, a team of scientists inaugurated an integrative research project, the
Year of Tropical Convection (YOTC). This project has been encouraged and endorsed as a priority
activity by the World Climate Research Programme (WCRP) and the World Weather Research
2
Programme (WWRP), as summarized in Waliser and Moncrieff, 2008. The emphasis on the tropics
recognizes its key role in the Earth’s weather-climate system, as well as the uncertainties weather
and climate models encounter in association with tropical convection and convection-related
processes. The emphasis on convective organization recognizes that contemporary convective
parameterizations do not adequately represent key processes, including the mesoscale dynamics
that are central to organized convective systems and their scale interactions.
There are important distinctions between the initiation, organization and maintenance of moist
convection in the extratropical compared to the tropical atmosphere. Figure 1 illustrates the basis
of this distinction in terms of the global distribution of vertically integrated water vapor: the moist
tropics and the relatively dry extratropics. These two zones are clearly demarcated apart from the
atmospheric “rivers” of moisture flowing episodically from the tropics to the extratropics.
•
Extratropical moist convection is primarily initiated through a down-scale cascade of energy
from planetary-scale to the synoptic-scale (extratropical cyclones), to the mesoscale (fronts and
jet streams), and to the small scale (cumulus and turbulence). This cascade is associated with
a quasi-balanced baroclinic redistribution of energy and involves interaction between
convection, the
planetary
boundary-layer
and
radiation.
Parameterized
and
explicit
representations of convection in contemporary prediction models have reasonable fidelity in the
extratropics owing, in part, to the successful reproduction in these models of the synopticbaroclinic life-cycle that controls interaction between synoptic and mesoscale motions and the
embedded convection. The subsequent organization of convection has an upscale component
especially in sheared environments, such as squall lines and cloud clusters. This dynamical
coherence modulates the mid-latitude baroclinic lifecycles, subsequent down-stream Rossbywave dispersion, and the maintenance of the mean zonal flow.
3
Tropical moist convection is more complex than extratropical convection because in the tropics
the direct baroclinic control of convection is a secondary element.
Tropical convection is
modulated by mean-state circulations, such as the Intertropical Convergence Zone (ITCZ), El NinoSouthern Oscillation (ENSO), the monsoon systems, tropical waves, and the variability of the
ocean mixed layer. The large-scale tropical circulation is responsive to, if not primarily forced by,
the upscale effects of multi-scale convective organization up to thousands of times the scale of
individual cumulus. Individual cumulus convection is a comparatively disorganized process that
does not effectively couple synoptic-to-planetary scale motion. Hence the spatial-temporal
organization of convection in the cloud-to-synoptic spectral range is requisite to the accurate
representations of the large-scale tropical circulation (Moncrieff 2004). Conversely, extratropical-totropical Rossby-wave dispersion and winter-time cold surges from the Asian continent propagating
into the tropics exert a downscale control that on occasion activates organized tropical convection
and the MJO, e.g., Kiladis, 1998.
There are important connections between the tropics and extratropics in which convection plays a
central role that can involve hemispheric and global scales. The weather systems of the midlatitude
storm tracks (extratropical cyclones) are driven by the potential energy released in global-scale
almost horizontal slantwise convective motion -- conveyor belts of warm, moist tropical/sub-tropical
air and cold, dry polar air that act to equilibrate the pole-equator gradient of diabatic heating. The
slantwise convection conditions the environment in the extratropics (thermodynamic stratification
and vertical shear), thereby affecting the organization and intensity of the extratropical moist
convection as indicates above.
An association of large-scale organized tropical heating with Rossby wave global teleconnections
and the modulation of the midlatitude storm tracks is illustrated in Fig. 2, adapted from Shapiro and
Thorpe 2004. This aspect was addressed by Hoskins and Karoly, 1981; Trenberth et al., 1988.
4
Additional global teleconnections have been proposed by Branstator (2002) and Weickmann and
Berry (2007).
The effects of physiography (e.g., coasts, land-use, and orography) modulate the convective
lifecycles, notably over Africa, the Americas, China, India, and the Indonesian Maritime Continent.
Propagating organized systems in the lee of mountain chains affect the diurnal cycle of
precipitation and radiative energy in the tropics and extratropics, e.g., Laing and Fritsch, 1997;
Carbone et al., 2002; Knievel et al., 2004. Neither the physiography nor the convective
organization are resolved in present-day climate models.
2. Convective organization in the context of weather-climate models
While researchers have addressed the parameterization of organized convection, progress has
been limited because organized convection is distinct from cumulus convection especially in terms
of its dynamical morphology and transport properties (Moncrieff and Liu 2006, Fig. 1).
Contemporary parameterizations do not represent the dynamics and scale-interactions associated
with multi-scale convective systems, (e.g., the super-cluster in Fig. 3) with the result that global
models have difficulty in reproducing multi-scale convective organization associated with the
Madden-Julian Oscillation (MJO) illustrated by Nakazawa, 1988; c.f., Miura et al., 2008, and
Sperber et al., 2008.
Organized convection interacts with the synoptic-to-large scale atmospheric circulation through
thermodynamic (e.g., diabatic heating) and dynamical processes (e.g., vertical transport of
horizontal momentum), exemplified as the stratiform precipitation and mesoscale downdraft
regions of mesoscale convective systems (MCS). The prevalence for stratiform precipitation in the
tropics was a finding of the Tropical Rainfall Measurement Mission (TRMM), e.g., Kummerow et al.,
(1998), Houze (1989), Schumacher and Houze (2003), Schumacher et al. (2004), and Tao et al.
5
(1998). Properties of MCS have been quantified by dynamical models (e.g., Moncrieff 1992),
numerical simulations (e.g., Wu and Moncrieff 1996), observational analysis, e.g., LeMone 1983;
Nesbitt et al., 2006; Tung and Yanai, 2002. Houze (2004) reviewed the current knowledge and the
substantial advances made in recent decades. Nevertheless, many important large-scale effects
of organized convection remain to be quantified, such as the transfer of energy and momentum
across scales and their role in atmospheric variability,
scale-selection mechanisms; and the
parameterization of organized convection for climate models. Further aspects are described in
ECMWF, 2003; ECMWF 2005; Moncrieff, 2007.
Researchers are addressing organized convection represented explicitly by non-hydrostatic cloudsystem resolving models (CSRM; grid-spacing ~ 1 km) as a more realistic approach than
parameterization. This fidelity was demonstrated by the WCRP GEWEX Cloud System Study,
(GCSS; Moncrieff et al. 1997, and Randall et al. 2003). The explicit representation of mesoscale
convective organization by CSRMs provides insight into the interaction between convective
organization and the large-scale circulation. The Satoh et al. (2008) global simulation at 3.5 km
grid-spacing, corresponding to a physical resolution of about 20 km (Bryan et al., 2003), heralds
the arrival of global CSRMs.
That uncertainties generated at the unresolved scales propagate rapidly to the larger scales raises
a paradox: we do not yet have the computational capacity to fully resolve convection in global
models, and convective parameterizations have not been improved to the level where they reliably
replicate convection.
In the interim, regional models are a test-bed for future explicit global
models. Regional and short-term operational weather prediction models having non-hydrostatic
dynamical cores already utilize explicit convection, e.g., the prediction systems of the UK MetOffice
and
the
Environment
Canada.
Operational
global
models
will
require
parameterized
representations of convection for at least another decade, particularly in relation to the use of
ensembles of forecasts. It follows that improved parameterizations of convection and the physical
6
processes related to convection (e.g., planetary boundary layer, turbulence, and cloud-radiation
interaction) are an imperative. Interaction between the atmospheric boundary layer and the ocean
boundary layer is a key element for both shallow and deep convection. Shallow cumulus
convection plays an important part in the diurnal cycle and the transition from stratocumulus to
deep convection in the Hadley circulation associated with the ITCZ. Despite notable advances, the
representation of coupled atmosphere-ocean processes in models remains less than optimal.
Inertial-gravity waves,
Kelvin waves, Rossby-gravity waves, and the MJO (e.g., Wheeler and
Kiladis 1999) coupled to moist convection modulate weather and climate on regional-to-global
scales and time scales of days and beyond, e.g., i) convectively generated inertial-gravity waves
affect the life-cycle of cloud clusters, e.g., Straub and Kiladis 2002, Haertel and Kiladis 2004; ii)
Rossby-wave packets generated by organized convection interconnect the tropics and
extratropics; iii) Rossby waves propagating into the tropics may initiate the MJO, e.g., Kiladis,
1998; Wheeler and Kiladis, 1999; Kiladis and Weickmann, 1992; Ray et al., 2009.
Climate models run in the initial-value or NWP mode (e.g., 3-day integrations) is a resourceefficient way to assess and quantify shortcomings associated with the representation of moist
physics and its interaction with the global circulation. Errors in numerical weather prediction (NWP)
models accorded to moist convection typically emerge within days, as shown by the spatialtemporal of errors in the precipitation distribution. This issue is being examined as part of the
Climate Change Prediction Program – ARM Parameterization Testbed (CAPT) of the Program for
Climate Model Diagnostics and Intercomparison (PCMDI) conducted at the U.S. Department of
Energy (DOE) Livermore National Laboratory, see Boyle et al., 2008. Figure 4 compares the
precipitation errors in the composited 3-day integrations (weather bias) with the climate bias. It is
clear that climate bias arises in a matter of days; namely, on the weather time scale.
7
Within a few years deterministic global numerical weather prediction (NWP) models will have ~10km grid-spacing, and so should standard climate models within a decade. At ~10-km grid-spacing,
however,
convective
parameterization
may
assume
a
hybrid
character:
convective
parameterization and explicit ‘grid-scale’ circulations are simultaneously in action. The grid-scale
circulations,
a primitive form
of convective organization are, nevetheless, superior to the
contempory parameterizations which have difficulty in representing the dynamics of convective
organization. Moncrieff and Klinker (1997) showed that grid-scale circulations in the ECMWF T213
(grid-spacing about 80 km) represent tropical super-clusters. The organized grid-scale fluxes
dominated the parameterized ones. It follows that ~10-km grid-spacing presents both issues and
opportunities for convective representation at a level intermediate between the explicit CSRM
approach and contemporary parameterization.
The organization of precipitating convection is associated with the following important
meteorological phenomena:
•
Madden-Julian Oscillation (MJO) and convectively-coupled equatorial waves are
intra-seasonal phenomena (Fig. 5) incompletely represented in global and regional
prediction models. Difficulties are attributed to deficiencies in convective parameterizations,
e.g., Lin et al., 2006. Advances in our knowledge of morphology, scale-selection and scaleinteraction are anticipated to improve the skill of forecasting tropical weather, monsoon
onsets and breaks, and bridge weather and climate aspects.
Further information are
available in review papers by Lau and Waliser, 2005; Zhang, 2005, and Kiladis et al., 2008.
•
Tropical cyclones and easterly waves. Tropical cyclones are catastrophic events
associated with easterly waves. Progress in this area is crucial for achieving: i) skillful
forecasting of high-impact weather; ii) advanced simulation of
land-atmosphere-ocean
interaction and their impact on mean-state, such as the ITCZ; iii) the role of organized
8
convection in initiating tropical depressions, storms and cyclones, e.g., Thorncroft and
Hodges 2001, Hopsch et al. 2007.
•
Monsoon systems are integrative phenomena that embody a cumulative response to the
diurnal cycle, tropical waves, the MJO, and land-atmosphere-ocean interaction. Intraseasonal and sub-seasonal variability strongly affect the monsoon circulations and the
distribution and frequency of precipitation on continental/ocean-basin scales. This implies a
potential for improving predictive skill on regional-to-global scales. Studies of monsoon
systems are focal areas of the Asian Monsoon Years (AMY 2007-2012), which is an
observational and modeling effort coordinated by the WCRP. Wang (2008) reviewed the
properties of monsoon systems.
•
Tropical-extratropical interaction. Tropical convection is major element of the two-way
exchange of energy between the tropics and extratropics. It has significant effects on
forecast skill at sub-seasonal-to-interannual timescales, such as the excitation of Rossby
wave trains that emanate from the tropics and, conversely, the excitation of convective
flare-ups associated with
extratropical Rossby wave trains propagating from the
extratropics into the tropics, e.g., Ferranti et al., 1990; Kiladis, 1998. Tropical-extratropical
interaction is part of the predictability and dynamical processes component of The
Observing-System Research and Predictability Experiment (THORPEX) described in
Shapiro and Thorpe, 2004.
•
Diurnal cycle. Shortcomings in representing the diurnal cycle, a fundamental forced mode
of atmospheric variability, demands attention in all classes of prediction models. Both local
and long-range effects are involved. Observations and models indicate that processes
associated with the diurnal cycle may impact upon longer time scales, e.g., monsoonal
9
precipitation, and the effects of the Indonesian Maritime Continent. The diurnal cycle of
convection in the tropics is particularly noticeable in the presence of coasts and elevated
terrain, e.g., Yang and Slingo 2001, Tian et al. 2006.
3.
The Way Forward: Year of Tropical Convection (YOTC)
The Year of Tropical Convection (YOTC) Science Plan was published as a WMO Technical
Document compiled by Waliser and Moncrieff (2008) on behalf of the YOTC Science Planning
Group. It is available at http://www.wmo.int/pages/about/sec/documents/YOTC-Science-Plan.pdf.
The “year” is a nominal timeframe for archiving and integrating the observational and numerical
data sets that extends from May 2008 through October 2009.
3.1 Historical background
Early efforts to integrate observations and models to address the interaction between convective
cloud systems and the synoptic-to-planetary scales were coordinated by the Global Atmospheric
Research Programme (GARP) and the Tropical Ocean-Global Atmosphere (TOGA) program.
These efforts were implemented as the First GARP Global Experiment (FGGE, 1979); the GARP
Atlantic Tropical Experiment (GATE, 1974); the Winter and Summer Monsoon Experiments
(MONEX, 1978-79); and the TOGA Coupled Ocean-Atmosphere Response Experiment (TOGA
COARE, 1992-93). These regional field campaigns set the stage for advances in understanding
tropical convection variability, and its prediction in global models, e.g., Betts, 1974; Greenfield and
Krishnamurti, 1979; Johnson and Houze, 1987; Moncrieff and Klinker; 1997; Mapes and Houze
1993; Webster and Lucas 1992; Mapes and Houze 1995; Houze et al., 2000; Chen et al. 1996;
Chen and Houze, 1997; Godfrey et al., 1998; Johnson et al.,1999.
The regional field-campaigns also addressed the formulation and fidelity of convective
parameterizations. A crucial assumption of convective parameterization is that the cumulus and
10
synoptic scales interact across a spectral gap, and that the role of the intermediate scales is a
secondary consideration. More than three decades ago, the analysis of the GATE observations
revealed that organized convection (e.g., squall systems and cloud clusters) populates the
intermediate (meso) scale between the cumulus and the synoptic scales, Houze and Betts, 1981.
This conflicts with the scale-separation assumption. The Zipser (1969) analysis of the Line Islands
Experiment data showed that mesoscale downdrafts led to the rapid demise of a synoptic-scale
tropical disturbance. These added complexities undoubtedly retarded the development of
convective parameterization.
Progress in multi-scale modeling, observational analysis and theoretical insights motivated the
WCRP and WWRP to convene the 2006 Workshop on the Organization and Maintenance of
Tropical Convection and the Madden-Julian Oscillation (MJO), with co-sponsorship by the
International Centre for Theoretical Physics (ICTP) in Trieste, Italy. The scientific context and
proceedings of the Workshop are summarized in Moncrieff et al., 2007. The Workshop reviewed
the state-of-knowledge of organized tropical convection, such as: i) the interaction with the
regional-to global circulation; ii) the degree and manner convection is influenced by, and feedbacks
onto, the large-scale atmospheric circulation; iii) the mechanisms that link the cloud, mesoscale,
synoptic and planetary scales, including influences on and from microphysical processes; iv) the
two-way interaction between organized tropical convection and the extratropics.
In recognition of an emerging new era, the Workshop made the following recommendations: i)
develop an internationally coordinated virtual computational-observational laboratory [research
resource]; ii) prepare a coordinated observing, modeling, and forecasting activity with emphasis on
organized tropical convection and its influence on predictive skill in the western Pacific and Indian
Ocean regions. This led to the framework of the YOTC project: the virtual computational-
11
observational research resource illustrated in Fig. 6, which embraces the elements described in
sections 4.1 - 4.4 below.
4. The Elements of YOTC
The enormous range of scales associated with organized tropical convection, especially the MJO
(~ 1 km- to-planetary), poses a formidable observational challenge whose complete measurement
requires integrated satellite, in-situ and field-campaign datasets. Note that high-resolution global
operational model analyses and forecasts, as well as CSRMs with global or ocean-basin sized
computational domains, are ready to be utilized. In other words, the elements of a global-scale
virtual field campaign are in place. The YOTC project provides the necessary integration.
4.1 Observational analysis
Satellite-based measurements: The deployment of the Earth Observing Systems (EOS) satellites
that began in the 1990s provided a new capability to profile the moist atmosphere and cloud
systems. The pioneering TRMM (1997) radar and radiometer measurements give rain-rate and
vertical structure, and SST measurements in the presence of clouds. The constellation of EOS
satellites, the A-train, makes nearly simultaneous measurements of a range of quantities from
Terra
(launched in 1999), Aqua (in 2002),
Aura (in 2004), and CloudSat/Calipso (in 2006).
Geostationary satellite measurements of organized tropical convection range from planetary-tosynoptic scales, down to the diurnal cycle in the temporal dimension. Extending back to 1983, the
International Satellite Cloud Climatology Project (ISCCP) provides global cloud characterization at
3-hour intervals, e.g., Rossow and Duenas, 2004; Rossow and Schiffer, 1991, 1999; Rossow et al.,
2004.
Land-based in-situ measurements: The U.S. Department of Energy (DOE) Atmospheric
Radiation Measurement (ARM) program implemented surface-based monitoring sites at three sites
12
in the tropical western Pacific, Ackernan and Stokes, 1993. These facilities began collecting data
in August 1996 at Manus, Papua New Guinea; in November of 1998 on the island of Nauru; and in
April 2002 at Darwin, Australia.
The continental-scale radar network of the U.S. National Weather Service in the form of the NextGeneration Weather Radar (NEXRAD), along with the rain-gage network, provides is a
comprehensive measure of the spatial-temporal distribution of precipitation and, in particular,
propagating organized convection, Carbone et al., 2002. Continental-scale radar networks are in
the process of being deployed in China and India. Country-scale networks are already deployed in
Taiwan and the U.K.
The Coordinated Enhanced Observing Period (CEOP), established in 2001 by WCRP's Global
Energy and Water Cycle Experiment (GEWEX), is an international effort focused on measuring,
understanding and modeling the water and energy cycles within the climate system.
Ocean-based measurements: The TOGA program and the WCRP World Ocean Circulation
Experiment (WOCE) program led to extensive ocean-based observing systems. In the Pacific,
the Tropical Atmosphere Ocean/Triangle Trans-Ocean Buoy Network (TAO/TRITON) moored
array provide subsurface thermal information and surface meteorology information, in some cases
including radiation measurements.
The Pilot Research Moored Array in the Tropical Atlantic
(PIRATA) provides similar capabilities. The Indian Ocean Observing System, presently being
deployed, will quantify air-sea coupling and intra-seasonal variability, including monsoons and the
MJO. The drifter/float program includes information for the initialization of ocean models and data
assimilation. The global drifter program provides surface-state information for prediction models,
ground-truth for satellite SST products, and estimates of ocean surface velocity.
13
Field-campaigns play a lead role as previously summarized. In regard to YOTC, the summer
and winter THORPEX Pacific Area Regional Campaign (TPARC), provide measurements of
opportunity for quantifying and predicting tropical convection and its global interactions.
4.2 Global analysis and forecasts
High-resolution operational global prediction systems utilizing data-assimilation provide complete
estimates of atmospheric variability. A mainstay of the YOTC project, high-resolution analyses and
forecast data will be used extensively, e.g., i) basic weather analysis and diagnostic studies; ii)
high-resolution short-range (< 48 hr) prediction experiments utilizing explicit convection; iii) lateral
boundary and initial conditions for ultra-high resolution CSRM simulations; iv) meridional boundary
conditions for tropical channel models used to quantify the effects of extra-tropical forcing on
tropical convection; v) forecast experiments (hind-casts) of selected events occurring during
summer and winter phases of the THORPEX Pacific Area Regional Campaign (TPARC). A number
of organized field campaigns associated with the Asian Monsoon Years, as well as individual
campaigns, provide an unprecedented examination of the monsoon and organized convection
aspects.
Presently, ECMWF T799 (25 km grid) data are being archived (i.e., complete analyses, 10-day
forecasts, and special diagnostics that include diabatic fluxes) and are available via the internet.
There are plans to obtain NCEP Global Forecasting System (GFS) forecasts and analysis data, as
well as NASA GEOS5 products. The meteorological phenomena summarized at the end of section
2 will be targeted by YOTC researchers.
4.3
Cloud-system resolving models (CSRM) are a key element of the YOTC project,
especially models with large computational domains capable of representing meso-to-large scale
interactions. The review paper by Tao and Moncrieff (2009) demonstrates the extensive range
14
of application of CSRMs. By coupling mesoscale dynamics to the parameterized small-scale
processes (e.g., cloud-microphysics, turbulence, and surface exchange), CSRMs bridge the gap
between
the
cumulus-scale
and
the
synoptic-scale
motion
assumed
by
convective
parameterization. Developments of multi-scale modeling occurred in conjunction with TOGA
COARE (e.g., Wu et al. 1998), TRMM (Tao et al. 1993), and GCSS, Randall et al. 2003.
The advancing computer capacity in the late 1990s, two decades after the field campaign, enabled
regimes of convective organization in African easterly waves observed in GATE to be simulated
using CSRMs.
A three-dimensional simulation (Grabowski et al. 1998) confirmed dynamically
based relationships between convective organization, vertical shear and convective available
potential energy. A two-dimensional global-scale CSRM showed that multi-scale convective
organization can evolve from uniform initial conditions, principally as a positive feedback involving
convective momentum transport, Grabowski and Moncrieff, 2001. Mesoscale downdrafts and
convectively generated inertial-gravity waves are key to the onset, life-cycle and scale of
convective organization, e.g., Nicholls et al. 1991; Mapes 1993; Liu and Moncrieff 2004; Tulich et
al. 2007.
Super-parameterization sidesteps issues of conventional parameterization by placing twodimensional CSRMs in each grid-column of a global circulation model (GCM), an approach
referred to as the Multiscale Modeling Framework (MMF). The GCM provides global coverage
while the CSRMs simulate scale interaction, convection dynamics, cloud-overlap, cloud-radiation
interaction, interaction between convective downdrafts and surface fluxes. The time-dependent
two-way interaction between the CSRMs and the parent GCM bridges the aforementioned scalegap. Super-parameterization per se was developed by Grabowski (2001) utilizing an idealized
global model, and has since been applied in a full climate model, Khairoutdinov et al., 2005. Figure
7 shows that MMF generates MJO-like large-scale convective organization, whereas contemporary
15
convective parameterization has comparatively little success. The evaluation of MMFs against
observations, the treatment of orography in CSRMs, the implications of two-dimensionality, and
improved MMFs models are active areas of research, e.g., Tao et al. 2009, Tao and Moncrieff
2009.
4.4
Idealized models reduce complex processes and scale-interactions to first principles.
Examples of reductionism include the role of momentum transport in convective organization and
multiscale interaction, (Moncrieff 1992; Biello et al., 2007); multi-scale convective organization in
the MJO and convectively-coupled tropical waves (Yano et al., 1996; Moncrieff 2004; Khouider and
Majda 2007; Majda and Stechmann 2009); stochastic-dynamical models (Shutts, 2004); stochastic
convective parameterization (Craig et al. 2005
and Tompkins 2005); and parameterization of
convective organization utilizing cellular automata, Berner et al., 2005.
5. Conclusions
The representation of tropical convection, its multiscale organization, and scale-interactions are
leading impediments to skillful regional-to-global prediction on hourly-to-seasonal time scales and
beyond. In response to this challenge, the World Climate Research Programme (WCRP) and the
World Weather Research Programme (WWRP) are coordinating the Year of Tropical Convection
(YOTC), an integrated computational-observational research resource and a key element of
“seamless prediction” described in Brunet et al.; Shapiro et al; Shukla et al., in this issue of BAMS.
The emphasis is on multi-scale convective organization, thereby addressing a long-standing
uncertainty in global models. The YOTC Implementation Plan, currently being prepared, targets
meteorological events at the weather-climate interface that occur during the Year, with emphasis
on the
MJO and convectively coupled waves; easterly waves and tropical cyclones; tropical-
extratropical interaction; the monsoons; and the diurnal cycle.
The project will engage with
coordinated activities already involved with tropical convection research at the international level,
e.g.,
NCAR’s Prediction Across Scales initiative; NSF’s Center for Multiscale Modeling of
16
Atmospheric Processes (CMMAP);
NERC’s
Cascade project; Asian Monsoon Years (AMY);
Working Group on Numerical Experimentation (WGNE); THORPEX, CLIVAR, and GCSS. It will
also engage with focused efforts in academia and research institutions.
Acknowledgements: We acknowledge the lasting encouragement for the YOTC project by the
WCRP and WWRP and, in particular, for their support of the YOTC Science Planning Group: J.
Caughey, R. Elsberry, R. Houze, C. Jakob, R. Johnson, T. Koike, J. Matsumoto, M. Miller, M.
Moncrieff (co-chair), J. Petch, W. Rossow, M. Shapiro, I. Szunyogh, C. Thorncroft, Z. Toth, D.
Waliser (co-chair), B. Wang, M. Wheeler, and S. Woolnough. Duane Waliser's contribution to this
project was carried out on behalf of the Jet Propulsion Laboratory, California Institute of
Technology, under a contract with the National Aeronautics and Space Administration (NASA). The
National Center for Atmospheric Research is sponsored by the National Science Foundation.
17
REFERENCES
Ackerman, T.P., and G. M. Stokes, 2003: The Atmospheric Radiation Measurement Program,
Physics Today 56, 38-44.
Berner, J., G. Shutts, and T. Palmer, 2005: Parameterizing the multiscale structure of organized
convection using cellular automaton, Proceedings of the ECMWF Workshop on
Representation of Sub-grid Processes using Stochastic-Dynamic Models, 6-8 June, 2005,
129-139.
Betts, A.K., 1974: The scientific basis and objectives of the U.S. convective subprogram for the
GATE. Bull. Amer. Meteor. Soc., 55, 304-313.
Branstator, G.W., 2002: Circumglobal teleconnections, the jet stream waveguide, and the North
Atlantic Oscillation. J. Climate, 15, 1893-1910.
Bryan, G.H., J. C. Wyngaard, and J. M. Fritsch. 2003: Resolution requirements for the
simulation of deep moist convection. Mon. Wea. Rev, 131, 2394–2416.
Biello, J., A. Majda, M.W. Moncrieff, 2007: Meridional momentum flux and superrotation in the
multiscale IPESD MJO model. J. Atmos. Sci., 64, 1636–1651.
Boyle, S. Klein, G. Zhang, S. Xie, and X. Wei, 2008: Climate model forecast experiments for
TOGA COARE. Mon. Wea.Rev.,136, 808–832
Carbone, R.E., J. D. Tuttle, D. Ahijevych, and S. B. Trier, 2002: Inferences of predictability
associated with warm season precipitation episodes. J. Atmos. Sci., 59, 2033-2056.
Chen, S.S., R. A. Houze, B. E. Mapes, 1996: Multiscale variability of deep convection in relation
to large-scale circulation in TOGA COARE. J. Atmos. Sci., 53, 1380-1409.
Chen, S.S., and R. A. Houze Jr., 1997: Interannual variability of deep convection over the
tropical warm pool. J. Geophys. Res., 102 (D22), 25783-25795.
18
Craig, G.C., B.G. Cohen, and R.S. Plant, 2005: Statistical mechanics and stochastic convective
parameterization. Proceedings of the ECMWF Workshop on Representation of Sub-grid
Processes using Stochastic-Dynamic Models, 6-8 June, 2005, 107-114.
ECMWF, 2003: Proceedings of the ECMWF/CLIVAR Workshop on Simulation and Prediction of
Intra-Seasonal Variability with Emphasis on the MJO, 3-6 November, 2003.
ECMWF, 2005: Proceedings of the ECMWF Workshop on Representation of Sub-grid
Processes using Stochastic-Dynamic Models, 6-8 June, 2005.
Ferranti, L., T. N. Palmer, F. Molteni, and E. Klinker, 1990: Tropical-extratropical interaction
associated with the 30–60 Day oscillation and its impact on medium and extended range
prediction. J. Atmos. Sci., 47, 2177–2199.
GEWEX Cloud System Science Team (A. Betts, K.A. Browning, P.R. Jonas, R. Kershaw, M.
Manton, P.J. Mason, M. Miller, M.W. Moncrieff, H. Sundqvist, W.K. Tao, and M. Tiedtke),
1993: The GEWEX Cloud System Study (GCSS). Bull. Amer. Meteor. Soc., 73, 387-399.
Godfrey, J., R. Houze Jr., R. Johnson, R. Lukas, J.-L. Redelsperger, A. Sumi, and R. Weller,
1998: Coupled Ocean-Atmosphere Response Experiment (COARE): An interim report. J.
Geophys. Res., 103(C7), 14395-14450.
Grabowski, W.W., 2001: Coupling cloud processes with the large-scale dynamics using the
Cloud-resolving Convection parameterization (CRCP). J. Atmos. Sci., 58, 978-997.
Grabowski, W.W., and M.W. Moncrieff 2001: Large-scale organization of tropical convection in
two-dimensional explicit numerical simulations. Quart. J. Roy. Meterol. Soc., 127, 445-468.
Grabowski, W. W., X. Wu, M. W. Moncrieff, and W. D. Hall, 1998: Cloud-resolving Modeling of
Cloud Systems during Phase III of GATE. Part II: Effects of resolution and the third spatial
dimension. J. Atmos. Sci., 55, 3264-3282.
Greenfield, R.S., and T. N. Krishnamurti, 1979: The Winter Monsoon Experiment: Report on the
December 1978 field phase. Bull. Amer. Meteor. Soc., 60, 439-444.
19
Haertel, P. T., and G.N. Kiladis, 2004: Dynamics of 2-Day equatorial waves. J. Atmos. Sci., 61,
2707-2721.
Hopsch, S.B., C.D. Thorncroft, K. Hodges, and A. Aiyyer, 2007: West African storm tracks and
their relationship to Atlantic tropical cyclones. J. Climate, 20, 2468–2483.
Hoskins, B.J., and D.J. Karoly, 1981: The steady linear response of a spherical atmosphere to
thermal and orographic forcing. Quart. J. Roy. Meteor. Soc., 38, 1179-1196.
Houze, R. A., and A. Betts, 1981: Convection in GATE, Rev. Geophys., 19 , 541-576.
Houze, R. A., 2004: Mesoscale convective systems. Rev. Geophys., 42, RG4003,
doi:101029/2004RG000150.
Houze, R.A., 1989: Observed structure of mesoscale convective systems and implications for
large-scale heating. Quart. J. Roy. Meteor. Soc., 425-461)
Houze, R.A., 1997: Stratiform precipitation in regions of convection: A meteorological paradox?
Bull. Amer. Meteor. Soc., 78, 2179–2196.
Houze, R.A., S. S. Chen, D. E. Kingsmill, Y. Serra, S. E. Yuter, 2000: Convection over the
Pacific warm oool in relation to the atmospheric Kelvin-Rossby waves. J. Atmos.Sci., 57,
3058–3089.
Johnson, R.H., T. M. Rickenbach, S.A. Rutledge, P. E. Ciesielski, and W. H. Schubert 1999:
Trimodal characteristics of tropical convection. J. Climate. 12, 2397–2418.
Johnson, R.H., R.A. Houze, 1987: Precipitating Cloud Systems of the Asian Monsoon. Monsoon
Meteorology C.-P. Chang, T. N. Krishnamurti, Eds. (Clarendon Press, Oxford, 1987), 298353.
Khairoutdinov, M., D. Randall, and C. DeMott, 2005: Simulations of the atmospheric general
circulation using a cloud-resolving model as a superparameterization of physical processes.
J. Atmos. Sci., 62, 2136–2154
Khouider, B., and A. J. Majda, 2007: A simple multicloud parameterization for convectively
coupled tropical waves. Part II: Nonlinear simulations. J. Atmos. Sci., 64, 381–400.
20
Kiladis, G.N., 1998: Observations of Rossby waves linked to convection over the Eastern
Tropical Pacific. J. Atmos. Sci., 55, 321-339.
Kiladis, G. N., and K. M. Weickmann, 1992: Extratropical forcing of tropical Pacific convection
during northern winter. Mon. Wea. Rev., 120, 1924–1938.
Kiladis, G.N., M.C. Wheeler, P.T. Haertel, K.N. Straub, P.E. Roundy, 2008: Convectively
coupled equatorial waves. Rev. Geophys, in press.
Knievel, J.C., D.A. Ahijevych, and K. W. Manning, 2004: Using temporal modes of rainfall to
evaluate the performance of a numerical prediction model. Mon Wea. Rev., 132, 2995-3009.
Kuell, V., A. Gassmann, and A. Bott, 2007: Towards a new hybrid cumulus parametrization
scheme for use in non-hydrostatic weather prediction models. Quart. J. Roy. Meteor. Soc.,
133, 479-490.
Kummerow, C., W. Barnes, T. Kozu, J. Shiue, J. Simpson, 1998: The Tropical Rainfall
Measuring Mission (TRMM) Sensor Package. J. Atmos. Ocean Tech., 15, 809–817.
Laing, A.G., and J. M. Fritsch, 1997: The global population of mesoscale convective complexes.
Quart. J. Roy. Meteor. Soc., 123, 1756-2776..
Lau, W. K. M., and D. E. Waliser (Eds.) 2005: Intraseasonal Variability of the Atmosphere-Ocean
Climate System, Springer, Heidelberg, Germany, 474 pp.
LeMone M. A., 1983: Momentum transport by a line of cumulonimbus. J. Atmos. Sci., 40, 1815–
1834.
Lin, J.-L., G. N. Kiladis, B. E. Mapes, K. M. Weickmann, K. R. Sperber, W. Lin, M.C.
Wheeler, S. D. Schubert, A Del Genio, L. J. Donner, S. Emori, J-F. Gueremy, F.
Hourdin, P.J. Rasch, E. Roeckner, and J. F. Scinocca, 2006: Tropical intraseasonal
variability in 14 IPCC AR4 climate models. Part I: Convective signals. J. Climate, 19, 2665–
2690.
Liu, C. and M.W. Moncrieff, 2004: Effects of convectively generated gravity waves and rotation
on the organization of convection. J. Atmos. Sci., 61, 2218-2227.
21
Mapes, B.E., and R. A. Houze, 1995: Diabatic Divergence Profiles in Western Pacific
Mesoscale Convective Systems. J. Atmos. Sci., 52, 1807–1828.
Moncrieff, M.W., 1992: Organized convective systems: Archetypal models, mass and
momentum flux theory, and parameterization. Quart. J. Roy. Meteor. Soc., 118, 819-850.
Moncrieff, M.W.,1997: Momentum transport by organized convection. NATO Advanced Study
Series: C. The Physics and Parameterization of Moist Atmospheric Convection (R. Smith,
Ed.), Klumer Academic Publishers, Netherlands, 231-253.
Moncrieff, M.W., 2004: Analytic representation of the large-scale organization of tropical
convection. J. Atmos. Sci., 61,1521-1538.
Moncrieff, M.W., and E. Klinker, 1997: Mesoscale cloud systems in the Tropical Western Pacific
as a process in general circulation models. Quart. J. Roy. Met. Soc., 123, 805-827.
Moncrieff, M. W., and C. Liu, 2006: Representing convective organization in prediction model by
a hybrid strategy. J. Atmos, Sci., 63, 3404-3420.
Moncrieff, M.W., M. Shapiro, J. Slingo, and F. Molteni, 2007: Collaborative research at the
intersection of weather and climate. WMO Bulletin, 56, 204-211.
Neale R.B., and B.J. Hoskins, 2000: A standard test for AGCMs including their physical
parameterizations. I: The proposal. Atmos.Sci. Lett, 1, 101-107.
Nesbitt, S. W., R. Cifelli, and S. A. Rutledge: 2006: Storm morphology and rainfall
characteristics of TRMM precipitation features. Mon. Wea. Rev., 134, 2702-2721.
Nicholls, M.E., R.A. Pielke, and W. R. Cotton, 1991: Thermally forced gravity waves in an
atmosphere at rest. J. Atmos. Sci., 48, 1869–1884.
Randall, D., M. Khairoutdinov, A. Arakawa, and W. Grabowski, 2003: Breaking the cloud
parameterization deadlock Bull. Amer. Meteor. Soc., 84, 1547–1564.
Randall, D., S. Krueger, C. Bretherton, J. Curry, P. Duynkerke, M. Moncrieff, B. Ryan, D.
Starr, M. Miller, W. Rossow, G. Tselioudis, and B. Wielicki, 2003: Confronting models with
data: The GEWEX Cloud Systems Study. Bull. Amer. Meteor. Soc., 84, 455–469.
22
Ray, P., C. Zhang, J. Dudhia, and S.S. Chen, 2009: A numerical case study on the initiation of
the Madden-Julian Oscillation. J. Atmos. Sci., 66, in press.
Rossow, W. B., and E. Duenas 2004: The International Satellite Cloud Climatology Project
(ISCCP) Web site: An online resource for research, Bull. Am. Meteorol. Soc., 85, 167–172.
Rossow, W. B., and R. A. Schiffer 1991: ISCCP cloud data products, Bull. Amer. Meteorol. Soc.,
72, 2–20.
Rossow, W. B., and R. A. Schiffer 1999: Advances in understanding clouds from ISCCP, Bull.
Amer. Meteorol. Soc., 80, 2261–2287.
Rossow, W.B., G. Tselioudis, A. Polak and C. Jakob, 2005: Tropical climate described as a
distribution of weather states indicated by distinct mesoscale cloud property mixtures.
Geophys. Res. Lett., 32, doi 10.1029/2005GL024584.
Satoh, M., T. Matsuno, H. Tomita, H. Miura, T. Nasuno, and S. Iga, 2008: Nonhydrostatic
icosahedral atmospheric model (NICAM) for global cloud resolving simulations. J. Comp.
Phys., 227, 3486-3514.
Schumacher, C., and R. A. Houze, 2003: Stratiform Rain in the Tropics as Seen by the TRMM
Precipitation Radar. J. Climate, 16, 1739-1756.
Schumacher, C., R. A. Houze Jr., and I. Kraucunas, 2004: The Tropical Dynamical Response to
Latent Heating Estimates Derived from the TRMM precipitation radar. J. Atmos. Sci., 61,
1341–1358.
Shapiro, M. E., and A.J. Thorpe. 2004: THORPEX International Science Plan, WMO/TD-No.
1245, WWRP/THORPEX No. 2, 51 pp.
Slingo, J. M., K. R. Sperber, J. S. Boyle, J. P. Ceron, M. Dix, B. Dugas, W. Ebisuzaki, J. Fyfe, D.
Gregory, J. F. Gueremy, J. Hack, A. Harzallah, P. Inness, A. Kitoh, W. K. M. Lau, B.
McAvaney, R. Madden, A. Matthews, T. N. Palmer, C. K. Park, D. Randall, and N. Renno
1996: Intraseasonal oscillations in 15 atmospheric general circulation models: Results from
an AMIP diagnostic subproject. Clim. Dyn., 12, 325-357.
23
Slingo, J. M., P. M. Inness, and K. R. Sperber 2005: Modeling, in Intraseasonal Variability of the
Atmosphere-Ocean Climate System, edited by W. K. M. Lau and D. E. Waliser, pp. 361-388,
Springer, Heidelberg, Germany.
Sperber, K. R., J. M. Slingo, D. E. Waliser, and P. M. Inness (2008), Coarse-resolution models
only partly cloudy (Comment on “A Madden-Julian oscillation event simulated by a global
cloud-resolving model” by Miura et al.), Science, 320, 612.
Sperber, K.R., and D. E. Waliser, 2008: New Approaches to Understanding, Simulating, and
Forecasting the Madden-Julian Oscillation, Bull. Amer. Meteor. Soc., 89, 1917-1920.
Straub, K.H., and G. N. Kiladis, 2002: Observations of a convectively coupled Kelvin Wave in the
eastern Pacific ITCZ. J. Atmos. Sci., 59, 30-53.
Tao, W.-K., S. Lang, W. S. Olson, R. Meneghini, S. Yang, J. Simpson, C. Kummerow, E.
Smith, and J. Halverson, 2001: Retrieved Vertical Profiles of Latent Heat Release Using
TRMM Rainfall Products for February 1998. J. Appl., Meteor., 40, 957–982.
Tao, W-K, and M.W. Moncrieff, 2009: Multiscale cloud-system modeling. Rev. Geophys.,
accepted subject to minor revision.
Tao, W.-K., J.-D, Chern, R. Atlas, D. Randall, M. Khairoutdinov, J-L. Li, D. E. Waliser, A.Hou, X.
Lin, C. Peters-Lidard, W. Lau, J. Jiang, and J. Simpson, 2009: A Multi-scale Modeling
System: Developments, Applications and Critical Issues. Bull. Amer. Meteor. Soc., in press.
Thorncroft, C. D., and K. I. Hodges, 2001: African easterly wave variability and its relationship to
Atlantic tropical cyclone activity. J. Climate, 14, 1166–1179.
Tian, B., D. E. Waliser, E. J. Fetzer, 2006: Modulation of the diurnal cycle of tropical deep
convective clouds by the MJO, Geophys. Res. Lett., 33, L20704,
doi:10.1029/2006GL027752.
24
Tompkins, A., 2005: Stochastic input to convection based on subgrid humidity distributions.
Proceedings of the ECMWF Workshop on Representation of Sub-grid Processes using
Stochastic-Dynamic Models, 6-8 June, 2005, 115-127.
Trenberth, K. E., G. W. Branstator, D. Karoly, A. Kumar, N-C. Lau, and C. Ropelewski, 1998:
Progress during TOGA in understanding and modeling global teleconnections associated
with tropical sea surface temperatures. J. Geophys. Res., 103 (special TOGA issue), 1429114324.
Tulich, S.N., D. A. Randall, and B. E. Mapes, 2007: Vertical-mode and cloud decomposition of
large-scale convectively coupled gravity waves in a two-dimensional cloud-resolving model.
J. Atmos. Sci., 64, 1210–1229.
Waliser, D. E., and M. Moncrieff, 2007: Year of Tropical Convection - Joint WCRP-THORPEX
activity to address the challenge of tropical convection. WCRP GEWEX News, 17, No. 2,
page 8.
Waliser, D.E., M.W. Moncrieff, 2008: Year of Tropical Convection (YOTC) Science
Plan,WMO/TD-No. 1452, WCRP -130, WWRP/THORPEX- No 9, 26 pp.
Wang, B. (Ed.) 2006: The Asian Monsoon, 844 pp., Springer, Heidelberg, Germany.
Webster, P.J., and R. Lukas, 1992: The Coupled Ocean-Atmosphere Response Experiment.
Bull. Amer. Meteor. Soc., 73, 1377-1416.
Weickmann, K.M., and E. Berry, 2007: A synoptic-dynamical model of subseasonal atmospheric
variability. Mon. Wea. Rev., 135, 449-474.
Wu, X, and M. W. Moncrieff, 1996: Collective effects of organized convection and their
approximation in general circulation models. J. Atmos.Sci., 53, 1477-1495.
Wu, X., W.W. Grabowski, and M.W. Moncrieff, 1998: Long-term behavior of cloud systems in
TOGA COARE and their interactions with radiative and surface processes. Part I: Twodimensional Modeling Study. J. Atmos. Sci., 55, 2693-2714.
Yang, G.Y., J. Slingo, 2001: The Diurnal Cycle in the Tropics, Mon. Wea. Rev., 129, 784-801.
25
Yano, J. I., J. C. McWilliams, M. W. Moncrieff, and K. A. Emanuel, 1996: Hierarchical tropical
cloud systems in an analog shallow-water model. J. Atmos. Sci, 52, 1724-1742.
Zhang, C., 2005: Madden-Julian Oscillation, Rev.,Geophys., 43, RG2003, doi:
10.1029/2004RG000158.
Zhang, C., M. Dong, S. Gualdi, H. H. Hendon, E. D. Maloney, A. Marshall, K.R. Sperber and W.
Wang , 2006: Simulations of the Madden–Julian oscillation in four pairs of coupled and
uncoupled global models. Climate Dynamics, 27, 573-592.
Zipser, E.J., 1969: The role of organized unsaturated convective downdrafts in the structure and
rapid decay of an equatorial disturbance. J. Appl. Meteor., 8, 799-814.
26
FIGURE 1: Snapshot of the global distribution of total precipitable water (TPW) derived
from SSMI and AMSRE satellite data using the Morphed Integrated Microwave Imagery at
CIMSS (MIMIC) product. High water content in the tropics (red hues) drops-off rapidly in the
extratropics. TPW maxima indicate cloud systems organized on large-scales, e.g., MJO in
the Indian Ocean. Rivers of moisture flow poleward from the tropics due to the incursion of
midlatitude fronts, planetary waves, and the extratropical transition of tropical disturbances.
[Courtesy: Tony Wimmers/Chris Velden, CIMSS, University of Wisconsin at Madison.]
NOTE: In the Digital On-Line BAMS Edition, Fig. 1 will be a movie.
27
W
LO
TF
U
O
Convection
JET
MJO
FIGURE 2: Left, Schematic atmospheric response to large-scale divergent outflow in the northern
hemisphere from organized tropical convection at the equator. An anticyclonic pair of vortices (H)
straddle the equator, and the downstream Rossby-wave train shifts the storm track southward, leading
to diminished storminess to the north. (Adapted by Trenberth et al. 1988, from Hoskins and Karoly,
1981). Right, 7-day-mean (18-24 December 1999) of a MJO tropical convective flare-up and downstream Rossby wave train for the period preceding the 24-26 December 1999 development of the ~100
ms-1 North Atlantic upper-level jet stream associated with the European extratropical cyclone “Lothar”
and its destructive wind storm. Upper panel: satellite-derived outgoing long-wave radiation, where blue
into purple colour shading denotes radiation from cold cloud tops. The MJO flare-up is centered over
Indonesia (red arrow), with an outflow plume extending north-eastward into mid latitudes (white dashed
arrow). Lower panel: 250-mb wind velocity; direction, black arrows and speed, colour shaded; light
blue, 20 ms-1; red > 90 ms-1. Centroid of the MJO convection (blue circle); ray path of the Rossby wave
train (red arrows). [From Shapiro and Thorpe 2004].
28
Multiscale convective organization: super-cluster
Multi-scale dynamics of super-cluster
FIGURE 3: Top, a super cluster in an MJO during the TOGA COARE field campaign. A supercluster
consists of a family of mesoscale convective systems embedded within tropical waves and displaying
strong scale interaction. Bottom, dynamical model of a supercluster consisting of organized convection
interlocked with Rossby-gyre dynamics (Moncrieff 2004).
29
PPT Day 3 CAM - CMAP DJF
o
30 N
Forecast Bias
0o
30o S
0o
NCAR
60oE
120oE
180oE
120oW
60oW
CAM - CMAP _ DJF 1992 - 3
30o N
Climate Bias
0o
30o S
0o
60oE
120oE
180oE
120oW
60oW
PPT Day 3 AM 2 - CMAP DJF
30o N
Forecast Bias
0o
30o S
0o
GFDL
60oE
120oE
180oE
120oW
60oW
AM 2 - CMAP_ DJF_1992 - 3
o
30 N
0o
Climate Bias
30o S
0o
-10.0
60oE
-6.0
120oE
-3.0
-1.0
180oE
0.0
120oW
1.0
60oW
3.0
6.0
10.0
mm/day
FIGURE 4: Comparisions of weather and climate bias in terms of precipitation rate (mm/day)
for two modeling systems, GFDL and NCAR. The weather bias is the difference between
predicted and observed precipitation rate for composited 3-day forecasts for December, January
and February (DJF), 1992-1993. The climate bias is from a single integration of these modeling
systems with prescribed SSTs (AMIP mode) for DJF 1992-1993. [Courtesy: Boyle et al., 2008.]
30
Month-Day
FIGURE 5: Left, 3-day running mean anomalies of observed outgoing long-wave radiation (OLR)
for the latitudinal belt 7.5N-7.5 S from 17 November 2003 through 23 March 2004. Right, weektwo (days 8-14) forecasts of precipitation anomalies using a 1998 version of NCEP's Medium
Range Forecast model. The colour bars show units and shading levels. The line contours are
time-space filtered observed OLR anomalies and define three different convectively coupled
tropical modes (Wheeler and Kiladis, 2000). The blue contours slanting downward from left to
right represent the Madden Julian Oscillation, the green contours represent the Kelvin wave and
the black contours slanting from right to left represent the equatorial Rossby wave [Courtesy:
Klaus Weickmann, NOAA/CDC].
31
Global Prediction
Integrated Observations
High‐resolution operational deterministic‐model data sets
Satellite, field‐campaign, in‐situ
data sets
YOTC
Year of Tropical
Convection
Research
Attribution studies of global data sets; parameterized, superparameterized, and explicit convection in regional‐to‐global models; theoretical studies
FIGURE 6: Framework of the YOTC integrated computational-observational research resource
for investigating tropical convection and its interaction with the global circulation, with emphasis
on multi-scale convective organization. The YOTC project consists of four integrated elements: i)
Global analysis and prediction involving high-resolution operational systems; ii) observational
analyses of satellite, field-campaign and in-situ data; iii) multiscale convective organization
simulated by cloud-system resolving models; iv) idealized numerical and dynamical models.
32
FIGURE 7: Left, weak MJO activity in the standard version of the Community Atmospheric
Model (CAM). Right, strong MJO activity occurs in the version of CAM (SP-CAM) which applies
superparamerization. [Courtesy: Khairoutdinov et al. 2005]
33