Download ICT research for Climate Change

Climate models, interpreting results,
and impacts
Why model?
Modelling
Carib Effort
Some results
Impacts
Capacity-Building Workshop:
Climate Change Adaptation and Water Resources
in the Caribbean Region
March 22, 2011
Michael A. Taylor and Tannecia S. Stephenson
Climate Studies Group, Mona
Department of Physics
University of the West Indies, Mona
1
3 Reasons for Modelling…
Why model?
Modelling
Carib Effort
Some results
Impacts
1. The Caribbean is climate sensitive…
Our lives and livelihood revolve around or are closely
linked to climate.
• Economic:
E.g. Tourism, agriculture, mining, fishing
• Infrastructure:
E.g. Location of major cities, water
quality and storage
•Recreation
• Wellbeing/health
Size and topography enhances sensitivity: E.g. Hilly
backbone, limited landspace, infrastructure few miles from
coast.
3 Reasons for Modelling…
Why model?
Modelling
Carib Effort
2. The Caribbean is therefore vulnerable…
Changes in climate (short term or long term) can and do
alter Caribbean existence.
Some results
Impacts
Droughts and floods;
Hurricanes;
Hot days, nights,
Long term climate change etc.
Impact felt throughout many areas of Caribbean life
directly or indirectly:
Agriculture, Health, Water, Tourism, Disaster
Management/Infrastructure, Sport, Finance
3 Reasons for Modelling…
Why model?
Modelling
Carib Effort
2. The Caribbean is however not helpless…
If we could anticipate the change in climate then we could
better plan for it.
Some results
Impacts
Modelling gives us a clue into the long term changes
in climate.
So why model?
To build climate resilience by:
- offering insight into likely ways the climate might
change in our region
- enabling the use of this knowledge for planning
purposes.
Premise…
Why Model?
Modelling
Carib Effort
1. Climate has been changing due
primarily to human activity –
primarily through the addition of
greenhouse gases.
Some results
2. The result is the earth has
warmed over the last
century - Earth is 0.75
Impacts
degrees warmer than in 1860.
3. Rising temperatures
result in other effects
e.g. sea level rise
and changing rainfall
patterns
Premise…
Why Model?
Modelling
Carib Effort
1. Suppose we could guess how
the concentrations of
greenhouse gases will
change going into the future
scenarios
Some results
So what?
models
3. Generate pictures of how
the climate of the earth, or
region on the earth will look
in the future.
2. Put these concentrations
in computer models that
simulate all the physical
processes of the earth.
projections
Modelling…
Why Model?
Modelling
Carib Effort
Some results
Impacts
Models
+
Scenarios
=
Future Climate
Computer simulations of future climates given conjectures of how the
greenhouse gases will change going to the end of the century.
Caribbean Modelling
Initiative
Why Model?
Modelling
In 2003 a group of modellers got together in Havana
Cuba.
Carib Effort
Some results
Impacts
• 4 Countries: Jamaica, Cuba, Babados, Belize
• 4 Institutions: UWI (Cave Hill), UWI (Mona), INSMET,
CCCCC
• Disheartened that no model projections existed for
Caribbean and at scale of Caribbean.
• Deliberate collaborative effort to produce Caribbean
climate projections at scale of Caribbean.
• Premised on shared workload to get results out quickly.
• Premised on building of capacity in the region.
Caribbean Modelling Initiative
Why Model?
Chose a DOMAIN
Modelling
•All of Caribbean
•Run at 50 km
•Smaller domains at 25 km
Carib Effort
Chose a MODEL
PRECIS - Providing REgional
Climates for Impact Studies
• Hadley Centre, UK
Some results
•
Impacts
•
•
•
•
Dynamical Downscaling
Model (RCM)
Driven by the HADAM3P
GCM and ECHAM but can be
forced at its boundaries by
other GCM’s
Has a resolution of up to
25km
Built by UK Hadley Centre
but run locally
Can be used for any part of
the Globe
Caribbean Modelling Initiative
Why Model?
Chose SCENARIOS
Divided up the runs
Modelling
•IPCC SRES Scenarios
•A2 (high emissions) and B2
(low emissions)
•Considered capacity
•Considered available
computing power
•Considered interest
Carib Effort
Some results
Impacts
Cuba
Carib basin
B1 (30 yrs) & A2 (30 yrs)
(INSMET)
50 x 50 km
Baseline (30 yrs)
Reanalysis (15 yrs)
Jamaica –
UWI (Mona)
Carib Basin
A2 (30 yrs) & B2 (30 yrs)
50 x 50 km
Baseline (30 yrs)
Barbados –
UWI (Cave
Hill)
Eastern Caribbean
A2 (30 yrs) & B2 (30 yrs)
25 x 25 km
Baseline (30 yrs)
Belize - 5C’s
Caribbean and
Eastern Caribbean
Multiple runs
Caribbean Modelling Initiative
Why Model?
Modelling
Carib Effort
Some results
Impacts
Chose Methodology
•Simulate historical conditions (e.g. 1970-present)
•Simulate future conditions under scenarios (end of century)
•Determine absolute or percentage change between future and
present.
Caribbean Modelling Initiative
Why Model?
PRECIS Project Timeline
Modelling
Carib Effort
Some results
Impacts
2003
2004
Sept: Initial
Cuba meeting
2005
2006
2007
Apr: Runs restarted
Dec: Space
limitations noted
Sept: Computers purchased.
Oct: Runs initiated
Nov: Start of computer woes
2008
2009
Analysis and
publications
New runs
begun
Aug: All initial runs completed.
Analysis begun
Warmer Temperatures
Why Model?
Modelling
A2
Carib Effort
Irrespective of
scenario the
Caribbean expected
to warm.
Some results
Warming between 1
and 5oC
Impacts
Warming greater
under A2 scenario.
B2
Warming consistent
with projections for
other parts of globe.
Warming far exceeds
natural variability
Mean changes in the annual mean surface temperature for 2071-2099
with respect to 1961-1989, as simulated by PRECIS (ECHAM) and PRECIS
(HADCM3) for SRES A2 (high emissions) and SRES B2 (low emissions).
Idea of scales at which
information can be
provided
Why Model?
Modelling
Carib Effort
Warmer Temperatures
Irrespective of
scenario the
Caribbean expected
to warm.
Some results
Warming between 1
and 5oC
Impacts
Warming greater
under A2 scenario.
Warming consistent
with projections for
other parts of globe.
Warming far exceeds
natural variability
Mean changes in the annual mean surface temperature for 2071-2099
with respect to 1961-1989, as simulated by PRECIS_Had for SRESA2.
Warmer Temperatures –
Country level
Absolute Temperature
Change (oC)
Why Model?
Country
Modelling
Cayman Islands
2.1 - 2.7
Anguilla
2.3 - 3.2
Barbados
2.0 - 2.6
St. Vincent and
the Grenadines
2.1 - 2.7
Guadeloupe
2.1 - 2.8
Carib Effort
Some results
Impacts
Antigua and
Barbuda
1.9 - 2.4
Turks and Caicos
2.1 - 2.7
St. Kitts and
Nevis
1.7 - 2.2
Irrespective of
scenario the
Caribbean expected
to warm.
Warming between 1
and 5oC
Warming greater
under A2 scenario.
Warming consistent
with projections for
other parts of globe.
Warming far exceeds
natural variability
Examples of future projections of average absolute temperature change
(degrees C) for the period 2071-2099 using the PRECIS (Hadley) model.
Each column contains a range of values across the A2 and B2 scenarios.
Temperature Conclusion
Why Model?
Modelling

In the future, the Caribbean is very likely to be:
Carib Effort
Some results
Impacts
HOTTER!!
Warmer Temperatures –
Results from other
approaches

Substantial increases in the frequency of days and nights
that are considered ‘hot’ in current climate by the 2090s.
(McSweeney et al. 2008)

Decreases in the frequency of days and nights that are
considered ‘cold’ in current climate. (McSweeney et al.
2008).

Temperatures are projected to increase at 3 stations
located in Trinidad, Jamaica and Barbados by the end of
century from statistical downscaling (Chen et al. 2006).
Less Rainfall
Why Model?
Modelling
A2
Carib Effort
Some results
Impacts
General tendency
for drying (main
Caribbean basin)
by end of the
century.
Drying between 25%
and 30%
B2
Possibly wetter far
north Caribbean
NDJ and FMA.
Drying exceeds
natural variability
June-October – wet
season dryer!
Mean changes in the annual rainfall for 2071-2099 with respect to
1961-1989, as simulated by PRECIS (ECHAM) and PRECIS (HADCM3) for
SRES A2 (high emissions) and SRES B2 (low emissions).
Less Rainfall
Why Model?
Modelling
Carib Effort
General tendency
for drying (main
Caribbean basin)
by end of the
century.
Some results
Drying between 25%
and 30%
Impacts
Possibly wetter far
north Caribbean
NDJ and FMA.
Drying exceeds
natural variability
June-October – wet
season dryer!
Mean changes in the annual rainfall for 2071-2099 with respect to
1961-1989, as simulated by PRECIS (HADCM3) for SRES A2 and SRES B2.
How certain?
Why Model?
Modelling
Carib Effort
Some results
Impacts
Multiple
uncertainties in
models
Consensus diagrams
useful
In some regions, all
scenarios predict
drier.
In some regions all
simulations predict
wetter.
Number of simulations projecting precipitation increase for 2080s.
Rainfall – Conclusion
Why Model?
Modelling

In the future, Caribbean is very likely to be:
Carib Effort
Some results
Impacts
DRIER !!
Rainfall – Results from Other
Approaches
Why Model?
Modelling

Statistically downscaled results for stations in Jamaica
show a summer rainfall decrease through the 2080s
but an increase for the analyzed stations in Trinidad
and Barbados (Chen et al. 2006).

The proportion of total rainfall that falls in heavy
events for the Caribbean islands decreases in most
GCM model projections (McSweeney et al. 2008).
Carib Effort
Some results
Impacts
Other parameters…
Why Model?
Modelling
Vertical Shear
Carib Effort
Some results
Impacts
Hurricanes?
Cloud Cover
All results…
Why Model?
Modelling
Carib Effort
Some results
Impacts
User friendly website
All results
User specify desired variables and
future period.
http://precis.insmet.cu/Precis-Caribe.htm
Multiple plot types. predict wetter.
Rising Sea Levels
Why Model?
Modelling

By the end of the century under an A1B scenario
(emissions are somewhere between A2 and B2
values), global sea levels are expected to rise by 0.21
to 0.48 meters (IPCC 2007).

If ice flow dynamics considered IPCC estimates may at
least double (e.g. Vermeer and Rahmstorf 2009)

Recent studies also suggest that sea level rise in the
Caribbean may be more pronounced than in other
regions because of its proximity to the equator (e.g.
Bamber et al. 2009).
Carib Effort
Some results
Impacts
Rising Sea Levels
Why Model?
Modelling

Simpson et al. (2010) examine the consequences
of 1 and 2 m sea level rise in the Caribbean,
which they consider to not be unreasonable
estimates by the end of the century.

They note that even if GHGs emissions were
stabilized now SLR would continue to rise beyond
the end of the century, and suggest that ‘the
question is not if the Caribbean will face SLR of
1m or 2m under either a 2.0°C or 2.5°C global
warming scenario, but rather when’.
Carib Effort
Some results
Impacts
Climate Will Change...
...Sea Level Rise
• The Caribbean is projected to
experience greater SLR than
most areas of the world due
to its location closer to the
equator and related
gravitational and geophysical
factors.
•SLR will continue for centuries after 2100, even if
global temperatures are stabilised at 2°C or 2.5°C and
therefore represents a chronic and unidirectional,
negative threat to coastal areas in the Caribbean and
globally.
UNDP/CASRIBSAVE (2010)
• Even in the absence of
increased intensity or
frequency of tropical storms
and hurricanes, SLR will
intensify their impact on
coastlines in the Caribbean.
We Are At Risk
Impacts from a 1m SLR on CARICOM nations - UNDP/CARIBSAVE (2010)
1. Nearly 1,300 km2 land area lost (e.g., 5% of The Bahamas, 2% Antigua and
Barbuda, 1% T&T).
2. Over 100,000 people displaced (e.g., 5% of population in The Bahamas, 3%
Antigua and Barbuda, 1% T&T).
3. At least 149 multi-million dollar tourism resorts damaged or lost, with beach
assets lost or greatly degraded at many more tourism resorts (33% T&T).
4. Damage or loss of 5 power plants (0% T&T)
5. Over 1% agricultural land lost, with implications for food supply and rural
livelihoods (e.g., 5% in Dominica, 6% in The Bahamas, 3% T&T).
6. Inundation of known sea turtle nesting beaches (e.g., 35% in The Bahamas and
St. Kitts and Nevis, 44% in Belize and Haiti, 50% in Guyana, 15% T&T).
7. Transportation networks severely disrupted.
8. Loss or damage of 21 (28%) CARICOM airports (50% T&T).
9. Lands surrounding 35 ports inundated (out of 44) (100% T&T)..
10. Loss of 567 km of roads (e.g., 14% of road network in The Bahamas, 12%
Guyana, 14% in Dominica, 1% T&T).
More Intense Hurricanes
Why Model?
Modelling

It is likely that with increased sea surface
temperatures ‘future tropical cyclones (typhoons
and hurricanes) will become more intense, with
larger peak wind speeds and more heavy
precipitation’ (IPCC 2007).

There is less consensus on frequency.

Much more research needs to be done to arrive
at a consensus on hurricane trends. This is work
we are currently undertaking.
Carib Effort
Some results
Impacts
Next steps
Why Model?
Modelling
Carib Effort

Adaptation hinges on more specific knowledge of threat
within each sector, more than what is currently available.

Coordinated data gathering and research, particularly
research which defines the climate-sectoral linkages are
needed.

A research agenda must be a feature of adaptation
response.

There is sufficient information available to begin addressing
questions on the impacts of climate change in individual
sectors.

Some of this work is already underway . . . .
Some results
Impacts
Next steps
Why Model?
Modelling

How do we use the climate change information to
help us identify adaptation strategies?

Four steps:
Carib Effort
Some results
Impacts
1.
Identify the link (between climate and sector).
2.
Look at how climate will change
3.
Identify likely change in sector
4.
Propose adaptation strategy
Next steps –
Health Example: Dengue Case
Study
Why Model?
Modelling
Carib Effort

Increased dengue fever associated with warmer
temperatures (Chen et al., 2006).

Study included data gathering and analysis to make the
climate sectoral linkage, assessment of vulnerability and
likely impact of the projected change.

Impacts of projected change used to craft a raft of possible
adaptation strategies ranging from public education alone
to scientific, technical and behavioural solutions dependent
on the availability of resources.

Study also makes the important point that the responsibility
for adaptation lies at all levels - individual, community and
national levels.
Some results
Impacts
Next steps –
Health Example
Why Model?
Recent Impacts
Likely Impacts
Adaptation Examples
Modelling
Reported cases of
dengue
are
correlated
with
both temperature
and rainfall, with
warming of early
months of the year
bringing
earlier
onset of reported
dengue cases and
epidemics
e.g.
Trinidad
and
Tobago 1997-1998
(Amarakoon et al.,
2006)
A 2°C increase of
temperature by 2099 is
expected to increase
the transmission of
dengue fever threefold. (Chen et al.,
2006)
The development and
implementation of early
warning systems for
some diseases e.g.
dengue.
Carib Effort
Some results
Impacts
Improving the
capabilities of national
and regional disaster
units to warn of, and
(react) respond to
disasters.
Extreme weather
events could lead to
increases in heat stress,
respiratory
complications, and
diarrhoeal distress
Ensuring efficient water
(Taylor et al., 2009, Ebi monitoring and
et al., 2006)
management.
Next steps –
Water Resources
Example
Why Model?
Modelling
Carib Effort
Some results
Impacts
Recent Impacts
Likely Impacts
Adaptation Examples
Potworks Reservoir,
an important surface
water source for the
island of Antigua, was
left dry after a 2003
drought (Farrell et al.,
2007)
Most small Caribbean
islands will experience
extreme water stress
regardless of SRES
scenario (Cashman et al.,
2010)
Improved management of
watersheds and catchment
areas to optimise
groundwater recharge.
During the 2004–
2005 Cuban drought
2.6 million people
were forced to rely
solely on truck-borne
water (Pulwarty et al.,
2010).
Sea-level rise will increase
the risk of saltwater
intrusion into coastal
aquifers, particularly those
that are already at risk from
over abstraction. (Simpson
et al., 2009)
Improved water resource
monitoring for conservation
and improved water
distribution to reduce loss.
Increase water storage
capacity to mitigate the
effects of drought.
Desalination for periods of
water shortages.
Next stepsWater Resources Example
Jamaica
Why Model?
Modelling

A National Adaptation Strategy for water was
examined for Jamaica – MACC initiative
1.
Climate Change

It is very likely that Jamaica will become warmer by the 2050s and
2080s.

It is likely that Jamaica will become drier in the mean and
particularly in June-August by the 2050s and 2080s.

Based on multiple approaches – global model output, PRECIS model
output and statistical downscalling (rainfall and streamflow)
Carib Effort
Some results
Impacts
Next stepsWater Resources Example
Jamaica
Why Model?
Modelling
Carib Effort
2.
Main threats to sector as a consequence of the climate
change scenarios

Water Supply:
Some results
Increasing length of dry season will particularly
impact communities affected by low drought yields.
Increasing frequency of intense rains will impact
sediment loads and treatment capability
Impacts

Flooding Regimes:
In highland areas of slope instability increased
landslides and flooding can be expected. High
turbidity and damaged distribution infrastructure
Next stepsWater Resources Example
Jamaica
Why Model?
Modelling
3. Propose Adaptation Strategy
Carib Effort

Some results
Impacts
Although the impacts of climate change on the water
sector are wide-ranging and significant, there are a
number of positive measures that can be taken to
increase the adaptive capacity of the sector.
(i) Investment in hydrological and water quality
monitoring and dissemination of data to the
stakeholder community
(ii) Development of appropriate hydrological and water
resources modeling tools in parallel with capacity
building within key stakeholder organizations
Next steps?
Why Model?
Modelling
Carib Effort
Some results
Impacts
More Analysis
Lots of data generated over the years. Get it out there.
Analysis left to be done:
Validation Extreme Analysis Dynamics Hurricanes
Haven’t begun to look at impacts on other sectors.
More Models and Modeling
Must offer multiple realisations of the future.
One way is to use different regional models. PRECIS but one
option.
More Partnerships
Too difficult, time consuming for any one country or institution.
Collaboration heightens efficiency for producing usable results.
Builds synergies/support groups across institutions.
Collaborate with non-English speaking Caribbean
Discernible impacts
Why Model?
Modelling
Carib Effort
Some results
Impacts
Transforming our talk:
Modelling in the Caribbean is moving us from general,
hypothetical conversations on climate change premised on
studies not done in or for the Caribbean to more contextually
relevant science language.
Access Answers
Modelling in the Caribbean is enabling us to answer our own
impact and vulnerability questions (e.g impact of climate
change on dengue fever, or sugar cane crops, or river
streamflows in eastern Jamaica, etc.)
Or…
Modelling enabling region to address our sustainable
development issues.
END
Why Model?
Modelling
Carib Effort
Some results
Impacts
Thank You