Download Unchecked Climate Change Migration V Ramanathan F Forman

Unchecked Climate Change, Mass Migration And Sustainability:
A Probabilistic Case for Urgent Action
Fonna Forman and Veerabhadran Ramanathan*
University of California at San Diego
•
Corresponding Author [email protected]
Paper Submitted to the Pontifical Academy of Sciences, Pontifical Academy of Social Sciences,
Ross Institute and UCLA Workshop on Humanitarianism and Mass Migration
January 18-19, 2017 held at the University of California at Los Angeles
December 31, 2016
1
Summary
Climate migration describes the voluntary and forced movement of people within and
across habitats due to changes in climate. Climate change can act as a causal factor of
migration or as a threat multiplier. With unchecked emissions of pollutants, global warming is
projected to increase to 1.50C within 15 years; to 20C within 35 years and 40C by 2100.1 These
projections are central values with a small (<5%) probability that warming by 2100 can exceed
60C with potentially catastrophic impacts on every child and grandchild.
Climate is already changing in perceptible ways through floods, droughts, fires, heat
waves and sea level rise displacing communities and catalyzing migration. Climate change and
associated migratory shifts have also been statistically linked with civil conflict and political
unrest.
The decades-long drought, agricultural failure, dramatic urbanization and failed
government response in Syria is a powerful case in point. Reliable quantitative estimates of
future climate migration are yet to be made. Reported estimates vary from 25 million to as many
as one billion climate change migrants by 2050.
Quantitative approaches for projecting mass migration face forbidding obstacles due to: 1)
a wide range of projected warming due to uncertainties in climate feedbacks; 2) dependence of
mitigation on non-environmental factors such as culture, economics and politics. It may take
decades to arrive at reliable quantitative estimates. But this creates unacceptable ethical risks.
For this reason, we advocate a probabilistic approach to climate migration that accounts for
both the central and low probability warming projections as the only ethical response to the
unfolding crisis. We conclude that in the absence of drastic mitigation actions, climate change
induced mass migration can become a major threat during latter half of this century. For the
poorest three billion however, who still depend on 1000 year-old technologies for meeting basic
needs such as cooking and obtaining drinking water, forced mass migration will be a reality
much sooner. Climate justice demands an urgent global response for the well being of us all.
1
All warming projections here and elsewhere in the text use pre-industrial temperatures as base
line.
2
I. Climate Change: How Soon And How Large?
Climate change has begun to change our lives already through droughts, mega-floods,
heat waves, intense hurricanes, glacial melting, forest fires and other severe weather phenomena.
There are other dramatic changes such as the melting of the glaciers in Greenland and the west
Antarctic; the retreat of the arctic sea ice; acidification of the oceans; disappearance of coral
reefs and rising sea levels.
Massive amounts of data provide compelling, if not convincing, evidence that much of
the climate changes we are experiencing are caused by the build up of carbon dioxide and other
manmade greenhouse gases from human activities. The major human activity warming the
climate is burning of fossil fuels for energy access. Others include: deforestation; release of the
super warming pollutants, halocarbons (CFCS; HFCs) used for refrigeration and air
conditioning; massive release of the super pollutant methane from use of natural gas, the
growing cattle population and dumping of food and other organic waste into land fills; and the
release of the greenhouse gas nitrous oxide from agriculture and release of black carbon (another
super warming pollutant) from diesel combustion and residential bio-mass burning.
The planet has already warmed by 10C (from pre-industrial temperatures). In about 13
years, the warming would exceed 1.50C, largely from the warming pollutants that are already in
the air. If current emissions of CO2 and the super pollutants continue unabated until 2030, the
warming is likely to exceed 20C in another 33 years (by 2050). Warming of 1.50C to 20C during
the coming decades, is a source for major concern for many reasons including the following:
•
The last time the planet was this warm was about 130000 years ago. But these
earlier warm epochs evolved over thousands of years. Human-induced warming
will happen 30 to 100 times faster in a matter of decades from now, raising
questions about the ability of ecosystems as well as social-systems to adapt.
•
The planet undergoes warm (interglacial) to cold (glacial) epochs and is currently
in a warm epoch called Holocene. The last glacial peaked about 20,000 years ago
when the planet was colder by about 50C. The human-induced warming is
happening on top of the already warm Holocene climate.
If the emissions continue unabated until 2050, it is likely the 21st century will witness warming
of 40C or more by 2100. These projected warming estimates represent at best a 50% probability.
3
Our understanding of the climate system, particularly the feedbacks, are such that, there is a wide
distribution of plausible warming values around the central value of 40C. For example, there is a
5% probability that the warming will be only half as large which would be good for society. But
the problem is, there is an equal probability of 5% that the warming could be more than 60C,
which is likely to be catastrophic for most human beings, rich or poor. In short, there is a 1 in 20
chance our current fossil fuel use can result in catastrophic consequences to our children and
grandchildren. Would we put our child on a plane if the pilot informed us the aircraft has a one in
20 chance of falling down from the sky?
It is within this probabilistic context that we have to assess the impacts of climate change
on migration. It would be misleading to focus just on the central value of about 20C for 2050 and
40C for 2100 since there is a 50% probability such warming estimates will be exceeded. By
focusing just on the central values policy makers would have made an implicit value judgment
that a one in twenty chance of catastrophic events happening to their children and grandchildren
is an acceptable policy choice.
The daunting nature of the tasks ahead for estimating likely migration during this century
are illustrated by the two figures reproduced below from UNEP (2009) report, which is a
summary of IPCC reports until 2009.
The first figure illustrates the pace of climate change
projected for the 21st century compared with the past 1000 years. The main inference to be drawn
is that, studies based on past climate changes to derive the link between climate change and mass
migration, may not be adequate to assess the likely migration during the 21st century because: the
pace of projected changes are drastically larger compared with past changes; second the
magnitude of the changes are also order of magnitude larger. The second figure (Table)
illustrates the dramatic dependence of the impacts on society on the magnitude of the warming.
The impacts on water, food, ecosystems and health reach catastrophic proportions when the
warming exceeds 40C.
4
5
II. Climate Change and Mass Migration: What’s at Stake?
Climate change is projected to cause widespread and serious harm to public health and the
environment upon which life depends, threatening to unravel many of the development and
public health advances of the last century. The burden of harm will fall disproportionately on the
poorest communities, who have the least capacity and resources to adapt to changing
environmental conditions, raising urgent issues of “climate justice” (Forman et al. 2016).
6
According to EM-DAT/CRED, 226 million people each year are impacted by natural disasters.
In the first decade of the 21st century, one million people died in natural disasters, 88% of which
were weather-related. In that same decade, 370,000 people died as a direct cause of extreme
climate conditions, which amounts to a 20% increase over the prior decade, primarily as a result
of the increased incidence of heat waves.
Current estimates of climate migration vary widely, from a low of 25 million to a high of
one billion migrants by 2050. The most commonly cited estimate is 200 million displaced by
2050 (IOM 2008:11-12, Myers 2002). According to recent estimates, between 2008 and 2014,
natural disasters – primarily atmospheric storms and floods – annually displaced an average of
26.4 million individuals (NRC/IDMI 2015:20). The rate of displacement has more than doubled
since 1970, from fewer than 2000 persons per million population to over 4000 persons per
million in 2014 (NRC/IDMI 2015:22). These displacements, even if temporary, have a profound
impact on individuals’ lives, often involving the loss of a home or crops, particularly harming
individuals at the bottom of the socioeconomic ladder who lack the resources to adapt to extreme
weather events and other weather patterns, such as prolonged droughts, resulting from a rapidly
changing climate.
Climate-related displacement also disproportionately impacts those in Asia and Latin
America, and particularly countries just beginning to gain "middle income" status. Annually,
from 2006-2014, per million inhabitants, East Asia had over 6000 displaced, Latin America had
approximately 5700, and South Asia 4500. In absolute terms, Asia accounted for 82.0% of all
climate displaced individuals, or 21.5 million (NRCIDMI 2015:30-31). Consequently, not only
are the most vulnerable individuals impacted heavily, but so too are more vulnerable developing
countries, who often lack the resources to manage large-scale displacements.
In addition to extreme weather events, which often cause sudden, mass displacements and
are growing in frequency, slower, progressive factors, like drought, soil erosion, and forest loss,
seem to have a stronger predictive effect on the likelihood of climate migration (McLeman 2014,
Gutmann and Field 2010). The main geographic impacts are already being felt in Asia, Africa,
Latin America, and the small island states, who also have the largest populations at risk of
becoming climate refugees (Biermann and Boas 2008).
7
Rising sea level around the world will have a disproportionate impact on the already
vulnerable. Since the middle of the 1800s, the rate of sea level rise has been greater than the
average rate during the previous two thousand years and the rate of sea level rise continues to
accelerate (IPCC 2013b). Research suggests that by 2100 average sea levels could rise by 1
meter or more (IPCC 2013c, Nicholls et al. 2014). Globally, the most populous areas that are
vulnerable to increased sea level and coast loss include China, India, Bangladesh, Indonesia, and
Vietnam; although rapid population growth and urbanization in coastal zones in Africa
(especially in Egypt and sub-Saharan countries in Western and Eastern Africa) are also of
concern (Neumann et al. 2015). It is estimated that by 2060 between 729 and 983 million people
will be living in low-elevation coastal zones in Asia, accounting for 70% of all those who live in
such regions globally (Neumann et al. 2015). Moreover, it is also estimated that approximately
280 million of the world’s inhabitants will be underwater if the global temperature increase stays
below 2C; and that approximately 630 million will be underwater at 4C. 74% of the impacted
population is in Asia (which itself comprises 59% of the global population).
At the macro scale, regions with fragile ecosystems and vulnerable geographies, such as
the low-lying mega deltas in Asia and the Sahel Belt in Western Africa are particularly
precarious and susceptible to the impacts of climate change. At meso scale, countries with weak
or underfunded and understaffed bureaucratic and administrative systems are likely to have low
adaptive and responsive capacities, leaving their inhabitants exposed to climactic shifts. At the
micro level, households and individuals in the Global South are most susceptible to the negative
impacts of climate. Those living in rural areas, whose livelihoods tend to be linked with
“climate-sensitive” sectors like agriculture and fishing, are most vulnerable and at highest risk,
and are typically least capable of adapting to changing environmental conditions, including insitu adaptations, as well as out-migration options. The capacity to leave one's home entails
certain financial and social capital, such as education, language skills, or support networks.
III. Historical Overview Of Climate Change And Migration Research
Climate-induced migration was identified as an area of concern for scholarly research in
the 1990s, after the first Intergovernmental Panel on Climate Change report in 1990 contained a
8
section on “Migration and resettlement.”2 For a long time, catastrophic climactic events such as
hurricanes (or cyclones), flash floods, heat waves, wildfires and tornadoes typically received the
most attention, both from the general public as well as from academics While these events are
responsible for significant climate-induced migration, slowly progressing but devastating
changes to the environment are predicted to play a much larger large role in producing
population displacement over the next century
Vulnerability and adaptation
Everything we presently know about vulnerability and adaptation, and the factors people
consider when making decisions about migration, apply to past warming trends, which remained
in the range of 0.5 C warming. While these experiences are instructive, it would be a mistake to
assume we can simply extrapolate from these experiences to anticipate future adaptation
strategies should warming increase in the next decades by a catastrophic 2-4 C, as probabilities
suggest they might. In such a scenario, migration will surely become less an option for many,
and more an imperative of mere survival.
Migration is presently one in a range of adaptation options people consider when responding to
their changing climate and environment (McLeman 2014). For the very poorest, international
migration is unlikely since it requires an upfront injection of cash that many do not have access
to. Thus, those who are impacted hardest by climate change are likelier to move regionally, if
not merely locally. Further, the Intergovernmental Panel on Climate Change, in Working Group
II’s first report noted that “climate change could translate into migration of impoverished people
from rural to urban areas (developing countries), from coastal lowlands (particularly densely
inhabited delta areas) to inland areas, and possibly across national boundaries” (IPCC 1990:511). The IPCC correctly noted that the majority of migration would be within developing
nations, following patterns of urbanization and away from coastal regions.
Additionally, increasing urbanization due to an inability of rural dwellers to survive off
the land may lead to increased “demand on urban services and increasing political pressure on
2
However the relationship between environment and climate and migration had previously been
discussed, at least in the US context (see Cebula and Vedder 1973, Graves 1980, Knapp and
Gravest 1989 and Evans 1990).
9
the state” (Barnett and Adger 2007: 642), which may make macro-level institutions even harder
pressed to provide adaptation support for exploding urban populations, exacerbating civil unrest
and even revolution, as in the case of Syria (Kelly, et al 2015).
For policymakers, the problem is one of identifying local socioeconomic vulnerabilities
likely to be exacerbated by climate change-related events, and especially the impact of extreme
weather events, which are likely to increase as the planet continues to warm. While this requires
a mixed-methods approach, reasonably substantial work mapping the causal linkages exists. The
bigger problem is developing reliable quantitative estimates of these impacts. Some degree of
quantification is necessary to assess the resources needed to address various localized
vulnerabilities.
The Impacts of the Securitization and Militarization of Climate Change and Migration
There has been increased concern about the linkages between climate migration and
security, including attention to immigration, national security, food scarcity and global instability
(Barnett 2003, Brown 2012). Homer-Dixon for example argues that competition over dwindling
natural resources, such as forests, water, and arable land, have the potential to lead to
devastating, prolonged global violent conflict (Homer-Dixon 2010), including ethnic conflict,
urban instability and political insurrection. Barnett warns that the issue of climate change may
be co-opted by “conventional” national security discourses, whereby the nexus of climate change
and security may be used by political leaders to support a conservative approach to borders,
among other issues (Barnett 2003).
IV. The International Response to Climate Migration
A variety of international bodies, organizations, and consortia have considered the case of
climate migration, including the United Nations Framework Convention on Climate Change, the
United Nations High Commissioner for Refugees, and the Intergovernmental Panel on Climate
Change (IPCC). The IPCC integrates vulnerability variables and adaptation strategies into its
approach, recognizing complications in measuring climate-motivated migration. In 2014, the
Fifth Assessment Report of the IPCC was released. Working Group II, which focuses on
impacts, adaptation, and vulnerability, noted that while we are lacking reliable quantitative data,
10
and have “low confidence in quantitative projections of changes in mobility, due to its complex,
multi-causal nature,” (IPCC 2014:21), there is nevertheless high agreement that climate change
“is projected to increase displacement of people,” throughout the 21st century, with highest risks
falling on vulnerable rural and urban demographics whose agency is low, especially in lowincome developing countries. (IPCC 2014:20). The Paris Agreement, decided upon at the 2015
COP, provides a new global platform for emissions reductions, and has legally-binding reduction
targets for all nations, not just developed states. However, climate migration remains outside the
scope of the Paris Agreement.
Biermann and Boas demand the creation of a global, international protocol outlining the
rights of climate refugees and the responsibilities of industrialized nations toward them
(Biermann and Boas 2008). Biermann and Boas urge the international community to draft a
protocol that is framed in planned and organized voluntary resettlement programs” (2008:15), as
voluntary resettlement programs are seen as more successful. The Nansen Initiative is a positive
step in this direction. Launched in 2012 by Norway and Switzerland, it led to the creation of a
Protection Agenda, which addresses “the needs of people displaced across borders in the context
of disasters and climate change” (The Nansen Initiative 2015b). It provides effective state
practices for managing cross-border disaster-displacement and also accounts for measures to
manage disaster displacement risks in origin countries. The Platform on Disaster Displacement
continues the work of the Nansen Initiative and urges regional, bilateral, and international
cooperation, having identified a lack of coordinated effort as a stumbling block to appropriate
adaptation responses. (The Nansen Initiative 2015b).
V. Proposal for a New Mixed-Methods Approach
This section attempts to weave together research on climate change with social scientific
research on migration for a more robust and comprehensive understanding of climate migration,
and the crisis of climate refugees. We need to think about “climate migration” in terms of
adaptation strategies of more-or-less vulnerable populations to climate change, among a range of
options. However, the question remains: to what are we adapting? The range of predictions from
an increase in average global temperatures by 1.5, 2, or 4°C, will each require different
adaptations. If average global temperatures increase by 4°C or more, the only adaptation
pathway for many communities may be migration. Nevertheless, tracing the impact of climate
11
change on migration is not a one-to-one ratio (McLeman 2014:53-4). Moreover, decisions to
migrate usually take place at the individual and/or household level, rather than state-organized
population resettlement, introducing substantial variation across individuals, as well as across
regions and populations. In this section, we identify the primary hurdles to developing reliable
quantitative estimates of climate migration, and discuss two potential approaches to overcoming
these hurdles.
Climate research has been the domain of environmental studies, geography, and earth and
atmospheric sciences, while migration research has come from sociology, political science,
demography studies, and global ethics, drawing on both qualitative and quantitative approaches.
Because climate migration is a convergence of sociopolitical factors and environmental
conditions, a more rigorous approach to assessing and understanding the future impact of climate
on global migratory patterns - which will surely accelerate as the planet continues to warm requires that we develop a more 'integral' methodological approach that merges scientific
research with both quantitative and qualitative methods in the social sciences - charged with an
urgent commitment to climate justice. Nevertheless, while we are committed to developing a
rigorous metric to refine our understanding over time, and encourage advancement in scholarship
and training along these lines, we also insist that an ethical global policy response to the
emerging climate-migration crisis cannot wait for reliable metrics.
Diagnosing the Methodological Challenge
McLeman’s (2014) MESA function, which builds on the vulnerability function presented
in section 3, provides a reasonable starting point for expanding on the challenges of obtaining
accurate estimates of current and future climate-driven migration. Under this function, migration
is a function of a population’s exposure to climate stimulus (E), sensitivity to the stimulus (S),
and adaptation options other than migration (A). However, although it provides a good starting
point, it is simplistic in two respects. As McLeman himself notes, the MESA function “offers no
information about spatial or temporal scale, and gives no more than a simple indication of the
relationship between migration potential and other components of vulnerability” (2014:68).
More generally, there are two principle conceptual hurdles to obtaining reasonable
estimates of climate-driven migration. First, there is no present consensus on what constitutes a
climate migrant. Existing definitions and estimates – such as the much-cited and compelling
12
figure of 200 million migrants by 2050 (Myers 2002, later cited in Adam 2005, Renaud et. al.
2007, Friends of the Earth Australia 2007, Global Humanitarian Forum 2009, and Bierman and
Boas 2010) – fail to differentiate local migration, intra-national vs. international migration, shortterm vs. cyclical/seasonal vs. long-term migration, or even between actual migrants and potential
migrants living in regions at risk (Gemenne 2011:546, Perch-Nielsen et. al. 2008 highlight
similar methodological issues). As Crisp (1999:3) points out: “Any form of enumeration exercise
must be based upon a clearly defined unit of measurement if it is to produce reliable, usable and
comparable data. In the case of refugee statistics, however, such clarity does not always exist.”
(See Appendix B for a brief history of the term “climate refugee”)
Second, causality is not simple in the case of climate change and migration. Migration
itself is a multi-causal phenomenon requiring analysis at a variety of stages and levels of analysis
(Gemenne 2011, McLeman 2014, Perch-Nielsen et. al. 2008, Piguet 2010, Warner 2011).
Climate changes are transmitted along a variety of distinct causal pathways, and their impact
varies depending on bioregion, individual and social vulnerability, and a wide array of other
factors.
Data constraints and insufficient or inaccurate theorization may plague estimation
attempts at each stage of analysis.
Migration researchers typically attempt to tackle these two challenges through the use of
detailed causal maps which highlight various chains of influence at each step in the process from
climate change to migration. Although the specific processes identified vary depending on the
ecosystem effects examined, most process maps outline five interrelated stages: (for a clear
example, see, Perch-Nielsen et. al. 2008:378 and 385):
1. Climate change (Global)
2. Ecosystem effects (Regional)
3. Livelihood effects (Individual)
4.
Adaptive capacity and options (Individual decisions; influenced by local & state-level
social and political resources)
5.
Migration options (Individual decisions; influenced by local & state-level social and
political resources)
Note that the temporal sequencing of various adaptations, migration decisions, and livelihood
effects can interact in complex ways. Thus, while adaptations may often follow on observed
climate changes, pre-existing adaptations to potential problems (e.g. substantial levees in regions
13
prone to riparian floods) may raise the threshold at which ecosystem changes impact individual
livelihoods or mitigate livelihood effects in advance (e.g. pre-existing crop insurance or flood
insurance).
Broadly speaking, it is difficult to clearly identify the type and magnitude of effect that
changes in one stage will have on the immediate subsequent stages. Consider the following
examples of variation at each stage:
Climate Change à Ecosystem Effects:
Variation occurs both depending on the type of ecosystem and location.
1. Types: Desert (or near-desert), River Valley, Lakeshore, Coastal, Small Island, Arctic.
For instance, flooding and erosion may be issues in areas near water, while desertification
may be an issue in semi-arid regions.
2. Location specific: For example, two river valleys may face divergent outcomes, as one
river valley faces extreme flooding due to increased ice melt, while a different river
valley encounters declining water levels as drought occurs near the river’s headwaters.
Ecosystem Effects à Livelihood Effects:
Variations in livelihood mean that different individuals will face different impacts from
the same regional change. Consider two farmers faced with drought, one growing a high-yielding
dwarf wheat and the other growing sorghum. Since the dwarf wheat requires substantially more
water than the sorghum, the drought will impact the first farmer more severely. Similarly, if both
farmers grow dwarf wheat, but one has access to mechanical groundwater pumps while the other
does not, the first farmer will be able to mitigate drought conditions for a longer period than the
second farmer, though at the expense of the groundwater table.
Livelihood Effects à Adaptation:
The ability to adapt one’s livelihood to changing climactic conditions depends on a
variety of factors, including the individual’s material and social capital, as well as governmental
social policies and safety nets. Examples include access to crop insurance, flood insurance,
government construction of levees and dams, local charity and religious networks, and subsidies
for changing crops or installing new forms of irrigation, among others.
Livelihood Effects + Adaptation à Migration decisions:
An individual’s decision to migrate depends on a number of factors. The first is the relative
success of the individual’s ability to adapt and the severity of impact on their livelihood by
14
climate changes. Second, even if an individual has decided to migrate, the migration decision
can take many forms, which are in turn influenced by the individual’s material and social capital.
Although there are many ways of disaggregating migration patterns (see McLeman 2014: chapter
2 for an excellent discussion), for the purpose of environmental displacement and adaptation,
two distinct axes of categorization are important: spatial and temporal. We address each in turn.
Spatial
Migration can be categorized along spatial lines, in general terms, according to the distance an
individual moves from their original residence.
1. Short distance migration (intra-urban, intra-regional rural, or short distance rural-urban)
2. Long distance intrastate migration (rural-rural, urban-urban, rural-urban, or urban-rural)
3. International migration
Generally speaking, most individuals move short distances before moving longer distances, even
when such moves are permanent. For example, a family that has lost property due to river
flooding may first move to a new location, at a higher elevation, in the same locality. However,
the exact pattern of migration depends largely on material and social capital. Typically,
individuals moving longer distances tend to be more affluent, or at least above an affluence
threshold. (although intense political and civil conflict, as in the case of Syria, can alter this)
Furthermore, choice of destination is often influenced by an individual's social networks/social
capital. To use a simple example, long-distance rural-urban migration is more likely if an
individual has relatives or friends in the destination city, and often is influenced by ethnic
enclaves at the destination as well (McLeman 2014:34).
Temporal
Migration may also be categorized on temporal lines, depending on the relative permanence of
the migration.
1. Temporary migration
2. Cyclical migration (seasonal or otherwise)
3. Permanent migration
These temporal categories, too, are heavily influenced by material and social capital. In general,
temporary migration precedes permanent migration, and a majority of temporary migrants return
to their original residence (Perch-Nielsen et al. 2008). However, even this regularity is modified
15
by material and social capital. For example, following Hurricane Katrina, while a majority of
white migrants had returned to their homes within a year, a majority of black migrants,
especially residents of the Ninth Ward, had yet to return (McLeman 2014:98-99). Individuals
may also undertake forms of cyclical migration to diversify income streams and mitigate risk of
crop loss, or, in the case of American "snow birds," affluent individuals may adopt seasonal
migration patterns to avoid undesirable seasonal weather conditions. Contrary to recent
"securitization" narratives, international migration is particularly unlikely for the poorest
individuals, absent countervailing social networks, due to the upfront costs required (McLeman
2014:47, IPCC 1990:5-11).
Potential Solutions – Modeling Complex Causality
Due to the complex nature of causation and vulnerability in determining migration
patterns, two approaches show promise, and are increasingly studied in the quantitative literature
(for a good summary article, see Piguet 2010).
1. Multi-level Modelling / Multivariate event history analysis:
a. Description: Analyzes individual survey data on various forms of vulnerability –
human
capital
(e.g.
education),
social
networks/capital
(e.g.
family
migrants/migrant communities), physical capital (e.g. money), and natural capital
(e.g. landholdings) – and links to local ecological changes (rainfall, etc.) with
fixed geolocations (e.g. wet vs. dry states).
b. Strength: Incorporates data at both the individual and community levels. Allows
for stratified analysis, enabling policymakers to better target potential migration
“hotspots” (both sources and destinations) and identify vulnerable groups. Merges
individual survey methods with national statistical collection.
c. Challenges: Fine-grained spatial breakdowns of weather patterns often lacking or
limited to state-level data – so the spatial breakdown of the phenomenon may not
map onto available spatial hierarchies. Could better account for return migration
patterns – often uses survival analysis methods, which have difficulty accounting
for return migration.
16
d. Good example studies: Gray 2009; Gray and Mueller 2012a and 2012b; Gray and
Bilsborrow 2013; Henry, Schoumaker, and Beauchemin 2004; Nawrotzki,
Riosmena, and Hunter 2013.
2. Agent-Based Modeling:
a. Description: Computer simulation technique allowing simulated responses of
individuals and households to climate signals, accounting for heterogeneity in
agents’ vulnerability (including the various forms of capital discussed above), and
social factors such as collective action problems, social learning, and network
effects (Patt and Siebenhüner 2005, Piguet 2010).
b. Strength: Strong ability to account for heterogeneity among agents, as well as
social factors. Relies on both individual and community level data. Higher
predictive power vs. other methods, since it can account for some future
adaptations in the system. Can account for return migration.
c. Challenges: Depends on the ability to identify rules of behavior in response to
stimuli, at the subgroup level. Requires excellent data on pre-existing responses of
heterogeneous subgroups to climate shocks.
d. Good example studies: Beine and Parsons 2015; Entwisle et. al. 2016; HassaniMahmooei and Parris 2012.
VI. Case Studies
To substantiate the human impact of climate change while climate migration metrics
continue to develop in the coming years, we advocate a mixed-methods approach that combines
quantitative prediction with compelling qualitative case studies that capture the human
experience of climate migration, particularly within vulnerable demographics.
It may take
decades for the numbers to be validated and causality to be clarified; but we can advance
narratives now that illustrate cause and effect with compelling human stories of loss and
displacement. Narratives that drill deep into the human experience play an essential role in
advancing public awareness and motivating collective response.
Here we offer two such
narratives, one detailing the disproportionate impact of a catastrophic weather event on a city’s
most vulnerable demographics (Hurricane Katrina); the second an incremental climate-change
impact in which millions have perished: the 30 year drought still ravaging sub-Saharan Africa.
17
Case Study I: Hurricane Katrina, USA
In late August of 2005, Hurricane Katrina formed from interaction between a tropical
wave and a cyclone over the Bahamas. One of the most devastating natural disasters in the
history of the United States, the storm moved westward towards the continental United States,
first striking two cities in Florida (Hallandale Beach and Aventura) on August 25. Then, moving
through the Gulf of Mexico, the storm intensified and was classified as a Category 5 hurricane,
and subsided slightly to a Category 3 hurricane before making a second landfall on August 29 in
southeast Louisiana (Greater New Orleans) (Beven et al. 2008:1131). The coasts of Mississippi
and Alabama were also impacted by the storm.
The devastation was widespread, including tragic loss of life, irreparable damage to
buildings, and significant financial loss. Hurricane Katrina took 1833 lives, from both direct and
indirect causes, with the majority (1577) occurring in Louisiana (Beven et al. 2008:1140).
Hurricane Katrina is the costliest natural disaster in US history, with an estimated $81.2 billion
in damages (Johnson 2006:1). The majority of the major flooding in Greater New Orleans was
caused by levee failure and floodwall breaching (American Society of Civil Engineers 2007:2532).
Estimates suggest that over one million people were displaced by Hurricane Katrina in
the Greater New Orleans area, including pre- and post-hurricane evacuations, accounting for
approximately 80-90% of the population (American Society of Civil Engineers 2007:43). Studies
show that of those evacuees in temporary accommodations in Houston, Texas more than 90 per
cent were African American and had low household incomes (Brodie et al. 2006). Of those in
temporary accommodations in Houston, approximately 60% had not evacuated before the storm
hit (Brodie et al. 2006). The impacts of the disaster were amplified for those who were already
poor, disabled, elderly, and sick. These people also tended to be those who were unable to
evacuate when the disaster struck.
Relative vulnerability also affected who was able to return to Louisiana and neighboring
states after the fact. For instance, the population of New Orleans in 2015, approximately
390,000, remained well below pre-Katrina levels, approximately 462,000 in 2004 (U.S.
Department of Commerce 2016). It should be remembered that New Orleans was already
18
experiencing out-migration even before Hurricane Katrina hit, due to various factors, including
economic decline, as evidenced in decreased employment in manufacturing (Vigdor 2008). Thus,
the impact of the storm aggravated pre-existing conditions in the region.
One's ability to return to New Orleans after evacuation was racially and socioeconomically determined: white residents who evacuated to areas adjacent to New Orleans
(Frey, Singer, and Park 2007), were likelier to return, especially those were educated and who
were employed and/or retired (Sastry 2009). Black residents tended to evacuate to further
destinations, including Baton Rouge, Houston, Dallas, and Atlanta, (Frey, Singer, and Park
2007), due either to preference or to what was provided by government assistance; and fewer
black residents were able to return than white residents within three months of Katrina (Fussell,
Sastry, and VanLandingham 2010).
Despite the lower socio-economic status of the Vietnamese community in New Orleans,
and its location in the damaged east end of the city, the return rates for Vietnamese residents
were above average, about 67%, after one year (Vu et al. 2009). This high return rate has been
primarily attributed to high levels of social capital within the Vietnamese community (Airriess et
al. 2008). Additionally, the city of New Orleans experienced a striking increase in its Latino
population after Hurricane Katrina, due to an increased need for manual laborers to rebuild the
city. The Latino population rapidly doubled from around 60,000 to over a 100,000 (Fussell,
Sastry, and VanLandingham 2010).
Moreover, return rates to New Orleans were strongest four months after Hurricane
Katrina hit and declined slowly. Evacuees who remained away from the city for longer periods
were less likely to return; this was especially true for those who had been away for nine months
(or longer) (Fussell, Sastry, and VanLandingham 2010). Further, home-ownership status also
determined return-migration as around only 25% of those who rented returned to New Orleans,
while around 75% of those who owned a still-habitable home returned (Elliott and Pais 2006).
Overall, damage to housing played a key determinant role in whether or not residents
returned to New Orleans after Hurricane Katrina, with those whose houses had minimal and/or
reparable damage most likely to return (McLeman 2014:103). Further, the slow and uneven
institutional and government responses (at all levels) to relief and recovery efforts after the storm
further stalled the return and reintegration of residents (Green, Bates, and Smyth 2007). Again,
African American communities, who were already marginalized in New Orleans, were
19
disproportionately impacted by Hurricane Katrina, and also suffered excessively in the lagging
response to restoration and recovery projects (Hartman 2006). Strong institutional support is
necessary to help vulnerable populations adapt in times of displacement, especially when there
are strong pressures against return migration. Nevertheless, as the example of the Vietnamese
community in New Orleans demonstrates, strong social capital and networks may mitigate some
of the effects of weak institutional support.
Case study 2: Multi-decadal drought in the Sahel Belt
The Sahel region extends across Africa just south of the Saharan desert. It is a transition
region with extreme desert to the north and the tropical rain belt to the south. As a result its
precipitation is very variable with frequent drought years. However, it experienced a pronounced
drought period that began in the 1970s (Figure that follows), and has not let up in its entirety to
this day. The summer rainfall decreased by as much as 30% (Figure).
More than 100,000
Sahelians died due to the drought and close to a million sought food aid in just one year in 1974
alone (UNEP; African Environment Outlook). Impact on migration will be described in the final
version.
Regional Trends in Summer Rainfall from 1950 to 2002
(Chung and Ramanathan 2012)
Red shaded areas show decrease in rainfall and blue areas denote increase in rainfall.
20
VI. Conclusion
Climate change is real and happening now. Data driven science has provided compelling
evidence that most of the observed changes are due to human activities, largely from combustion
of coal, oil and gas. Models that simulate many of the observed changes are projecting a 2C
warming by mid century and unprecedented warming in the range of 2C to 4 C towards the latter
half of this century. Because of non-linear feedbacks, there is a small probability (5% to 10%)
the warming can reach catastrophic levels of much larger than 4C by 2100. At such levels of
warming, mass migration may emerge as a dominant issue of human suffering induced by
climate change. Migration research is at its infancy and reliable estimates may take much too
long to be societally relevant, unless massive infusion of funds are made to develop quantitative
models of the sort proposed in this study.
The richest 1 billion people on the planet are responsible for about 50% of greenhouse
gas emissions; while the poorest 3 billion, without access to affordable fossil fuels, are
responsible for about 5% (Ramanathan, 2014; Dasgupta and Ramanathan, 2015). In contrast, the
bottom 3 billion suffer the greatest harms associated with climate change. Susceptibility of the
poor to forced climate migration is perhaps the most vivid example of disproportionate harm.
Climate justice, as a species of social justice, places an ethical imperative on the most
advantaged populations to mitigate harms and improve conditions for the least advantaged.
These demands become acute when the advantage of a few is responsible for producing or
aggravating the deprivation of many, living or yet to be born. Climate justice demands urgent
global intervention to reduce planetary warming, reasserting our global commitment to human
rights and public goods, and a broader conception of self-interest that includes the sustainability
of human settlements on our planet.
In this paper, we have summarized significant methodological barriers to obtaining
verifiable estimates on climate migration rates. Climate-induced migration is in its infancy. It is
ill-defined, causally complex, and it could take decades to produce verifiable estimates. In this
paper we advance two key recommendations in face of these limitations.
First, on ethical
grounds, we argue for a probabilistic approach to climate migration, and a corresponding call for
urgent mitigation to slow global warming trends. When there is insufficient time to quantify an
impeding crisis, ethics demands a urgent probabilistic response. When even seemingly small
21
probabilities are combined with the solidity of climate science, with changes in observed
environmental phenomena, and with the documented displacement of millions of vulnerable
people in recent decades by dramatic catastrophic weather events and slower-moving climate
effects, often exacerbated by civil unrest, we have a sufficient ethical basis for urgent action to
mitigate global warming.
For this reason, secondly, we advocate a mixed-methods approach to estimating climate
migration that merges science and evolving quantitative metrics with powerful qualitative case
studies that document the human impacts of climate change. Climate justice is as much a battle
of public opinion as it is about redistributing responsibilities; and stories can humanize the
impact of climate change better than cold numbers can. We believe that human stories will play
an increasingly important role in communicating the harms that climate change is already
inflicting on the world’s most vulnerable people.
Acknowledgments: Our sincere thanks to University of California, San Diego PhD candidates,
Sean Morgan (Political Science) and Vanessa Loedermeier (Anthropology) for their support and
assistance during the research and writing of this paper. VR was supported by the Alderson
Foundation; FF was supported by the UCSD Center on Global Justice.
22
References
Adam, D. 2005. "50m Environmental Refugees by End of Decade, UN Warns." The Guardian.
London.
Adger, W Neil, Saleemul Huq, Katrina Brown, Declan Conway, and Mike Hulme. 2003.
“Adaptation to Climate Change in the Developing World.” Progress in Development
Studies 3 (3): 179–95.
American Society of Civil Engineers. 2007. “The New Orleans Hurricane Protection System:
What Went Wrong and Why.” Reston, Virginia.
Bang, Guri, Jon Hovi, and Tora Skodvin. 2016. “The Paris Agreement: Short-Term and LongTerm Effectiveness.” Politics and Governance 4 (3): 209–18.
Barnett, Jon. 2003. “Security and Climate Change.” Global Environmental Change 13 (1): 7–17.
doi:10.1016/S0959-3780(02)00080-8.
Barnett, Jon, and W. Neil Adger. 2007. “Climate Change, Human Security and Violent
Conflict.” Climate Change and Conflict 26 (6): 639–55.
doi:10.1016/j.polgeo.2007.03.003.
Bettini, Giovanni. 2013. “Climate Barbarians at the Gate? A Critique of Apocalyptic Narratives
on ‘climate Refugees.’” Risky Natures, Natures of Risk 45 (March): 63–72.
doi:10.1016/j.geoforum.2012.09.009.
Beven, John L, Lixion A Avila, Eric S Blake, Daniel P Brown, James L Franklin, Richard D
Knabb, Richard J Pasch, Jamie R Rhome, and Stacy R Stewart. 2008. “Atlantic Hurricane
Season of 2005.” Monthly Weather Review 136 (3): 1109–73.
Biermann, Frank, and Ingrid Boas. 2008. “Protecting Climate Refugees: The Case for a Global
Protocol.” Environment: Science and Policy for Sustainable Development 50 (6): 8–17.
doi:10.3200/ENVT.50.6.8-17.
Boas, Ingrid. 2015. Climate Migration and Security: Securitisation as a Strategy in Climate
Change Politics. Vol. 24. Routledge.
Brodie, Mollyann, Erin Weltzien, Drew Altman, Robert J Blendon, and John M Benson. 2006.
“Experiences of Hurricane Katrina Evacuees in Houston Shelters: Implications for Future
Planning.” American Journal of Public Health 96 (8): 1402–8.
Brown, Lester R. 2012. Full Planet, Empty Plates: The New Geopolitics of Food Scarcity. WW
Norton & Company.
Cebula, Richard J., and Richard K. Vedder. 1973. “A Note on Migration, Economic Opporuntity,
and the Quality of Life.” Journal of Regional Science 13 (2): 205–11.
doi:10.1111/j.1467-9787.1973.tb00395.x.
Dasgupta, P., and Ramanathan, V. 2014, 19 Sep, "Pursuit of the common good: Religious
institutions may mobilize public opinion and action. Science, 345: 6203. DOI:
http://dx.doi.org/10.1126/science.1259406
Forman, Fonna, Ginal Solomon, Racel Morello-Forsch and Keith Pezzoli, 2016 "Bending the
Curve and Closing the Gap: Climate Justice and Public Health, Collabra 2 (1): 1-17.
DOI: http://dx.doi.org/10.1525/collabra.67.
Graves, Philip E. 1980. “Migration and Climate.” Journal of Regional Science 20 (2): 227–37.
doi:10.1111/j.1467-9787.1980.tb00641.x.
Hartmann, Betsy. 2010. “Rethinking Climate Refugees and Climate Conflict: Rhetoric, Reality
and the Politics of Policy Discourse.” Journal of International Development 22 (2): 233–
46. doi:10.1002/jid.1676.
23
Homer-Dixon, T.F. 2010. Environment, Scarcity, and Violence. Princeton University Press.
https://books.google.com/books?id=B6B-3CugWG0C.
IPCC. 1990. “Climate Change: The IPCC Impacts Assessment.” Camberra, Australia:
Intergovernmental Panel on Climate Change.
———. 2013a. “IPCC Factsheet: What Is the IPCC?” World Meteorological Organization.
———. 2013b. “Summary for Policymakers.” Climate Change 2013: The Physical Science
Basis. Contribution of Working Group I to the Fifth Assessment Report of the
Intergovernmental Panel on Climate Change.
———. 2013c. “The Physical Science Basis. Contribution of Working Group I to the Fifth
Assessment Report of the Intergovernmental Panel on Climate Change.” K., Tignor, M.,
Allen, SK, Boschung, J., Nauels, A., Xia, Y., Bex, V., Midgley, PM, Eds, 1535.
———. 2014. “Summary for Policymakers.” In Climate Change 2014: Impacts, Adaptation, and
Vulnerability. Part A: Global and Sectoral Aspects. Contribution of Working Group II to
the Fifth Assessment Report of the Intergovernmental Panel on Climate Change, edited
by C. B. Field, V. R. Barros, D. J. Dokken, K. J. Mach, M. D. Mastrandrea, T. E. Bilir,
M. Chatterjee, et al., 1–32. Cambridge, United Kingdom, and New York, NY, USA:
Cambridge University Press.
Johnson, David L. 2006. “Service Assessment: Hurricane Katrina, August 23–31, 2005.” US
Department of Commerce, National Oceanic and Atmospheric Administration, National
Weather Service: Silver Spring, MD.
Kelley, Colin P., et al. Climate change in the Fertile Crescent and implications of the recent
Syrian drought, PNAS , vol. 112, No. 11 (March 17, 2015): 3241-3246
www.pnas.org/cgi/doi/10.1073/pnas.1421533112
McLeman, Robert A. 2014. Climate and Human Migration: Past Experiences, Future
Challenges. Cambridge: Cambridge University Press.
https://www.cambridge.org/core/books/climate-and-humanmigration/0D0D207E8BA3D169452A84FEB7D779CB.
Neumann, Barbara, Athanasios T Vafeidis, Juliane Zimmermann, and Robert J Nicholls. 2015.
“Future Coastal Population Growth and Exposure to Sea-Level Rise and Coastal
Flooding-a Global Assessment.” PloS One 10 (3): e0118571.
OCHA, UN. 1998. “Guiding Principles on Internal Displacement.” Geneva: United Nations
Office for the Coordination of Humanitarian Affairs.
Reuveny, Rafael. 2007. “Climate Change-Induced Migration and Violent Conflict.” Climate
Change and Conflict 26 (6): 656–73. doi:10.1016/j.polgeo.2007.05.001.
Sen, A. 1981. Poverty and Famines: An Essay on Entitlement and Deprivation. Oxford
Scholarship Online. OUP Oxford. https://books.google.com/books?id=FVC9eqGkMr8C.
Seto, Karen C, Michail Fragkias, Burak Güneralp, and Michael K Reilly. 2011. “A MetaAnalysis of Global Urban Land Expansion.” PloS One 6 (8): e23777.
Solomon, Susan, Dahe Qin, Martin Manning, Z Chen, M Marquis, KB Averyt, M Tignor, and
HL Miller. 2007. “Contribution of Working Group I to the Fourth Assessment Report of
the Intergovernmental Panel on Climate Change, 2007.”
The Nansen Initiative. 2015a. “AGENDA FOR THE PROTECTION OF CROSS-BORDER
DISPLACED PERSONS IN THE CONTEXT OF DISASTERS AND CLIMATE
CHANGE.”
———. 2015b. “The Nansen Initiative: Towards a Protection Agenda for People Displaced
across Borders in the Context of Disasters and the Effects of Climate Change.”
24
UNFCCC. 2011. “The Cancun Agreements: Outcome of the Work of the Ad Hoc Working
Group on Long-Term Cooperative Action under the Convention.” United Nations
Framework Convention on Climate Change.
———. 2014. “Background on the UNFCCC: The International Response to Climate Change.”
United Nations Framework Convention on Climate Change.
http://unfccc.int/essential_background/items/6031.php.
UN General Assembly. 1951. “Convention Relating to the Status of Refugees.” United Nations,
Treaty Series 189: 137.
U.S. Department of Commerce. 2016. “United States Census Bureau.”
25
APPENDIX A.
The International Response to Climate Migration
Intergovernmental Panel on Climate Change
The Intergovernmental Panel on Climate Change (IPCC) was established in 1988 by the
World Meteorological Organization (WMO) and the United Nations Environment Program
(UNEP). It “provide[s] policymakers with regular assessments of the scientific basis of climate
change, its impacts and future risks, and options for adaptation and mitigation” (IPCC 2013a).
The IPCC, through its assessments, keeps state governments and the international community
informed of climate research, which can aid in the development of evidence-based climaterelated policies.
Explicit in the IPCC’s approach to climate migration is the inclusion of vulnerability
variables and adaptation strategies. The first IPCC report in 1990 contained a section on
“Migration and resettlement,” noting that “climate change could translate into migration of
impoverished people from rural to urban areas (developing countries), from coastal lowlands
(particularly densely inhabited delta areas) to inland areas, and possibly across national
boundaries” (IPCC 1990:5-11). Pointedly using the mitigating term of possible migration across
national boundaries, early on, the IPCC recognized the complications in measuring climatemotivated migration, and understood that most movement would be occurring within developing
countries. The Fourth IPCC Assessment Report from Working Group I (The Physical Science
Basis) released in 2007, was one of the most forceful and direct messages that had, as of that
time, been sent to the international community about climate change. It asserted, “[w]arming of
the climate system is unequivocal, as is now evident from observations of increases in global
average air and ocean temperatures, widespread melting of snow and ice, and rising global
average sea level” (Solomon et al. 2007:5). Moreover, it made clear that the observed increased
in global average temperatures since the middle of the 20th century are “very likely due to the
observed increase in anthropogenic greenhouse gas concentrations” (Solomon et al. 2007:10).
Citing human-originated causes for climate change is a necessary, but insufficient, condition for
committing states to reducing greenhouse gas emissions. In 2014, the Fifth Assessment Report of
the IPCC was released. Working Group II, which focuses on impacts, adaptation, and
vulnerability, notes that while there is high agreement that, throughout the 21st century, climate
change “is projected to increase displacement of people,” there is medium evidence in scholarly
literature on which to base this agreement (IPCC 2014:20). Moreover, it amplifies the previously
reviewed scholarly literature, which argues that displacement risk is highest for those whose
vulnerability is high and whose agency is low, especially in low-income developing countries,
and for those in both rural and urban areas (IPCC 2014:21). Moreover, Working Group II of the
IPCC urges that “[t]here is low confidence in quantitative projections of changes in mobility, due
to its complex, multi-causal nature,” (IPCC 2014:21), which highlights the need for more reliable
quantitative data in this regard.
United Nations Framework Convention on Climate Change
The United Nations Framework Convention on Climate Change (UNFCCC) was signed
in 1992 at the Earth Summit in Río de Janerio, and has been in effect as of 1994. Ratified by 197
countries, it is the framework for international cooperation on combatting climate change. The
26
Parties to the Convention are divided into three groups, based on their different commitments:.
The ultimate aim of the UNFCCC is “Preventing ‘dangerous’ human interference with the
climate [earth’s] system” (UNFCCC 2014). The Framework Convention focuses on limiting the
increase in global average temperatures, mitigating the effects of climate change that are already
being experienced, and it expects developed nations to take the lead to decrease carbon
emissions (UNFCCC 2014). Further, beyond existing financial assistance targets, industrialized
nations that are party to the Convention agree to financially support actions to counter climate
change in developing countries. The UNFCCC document of 1992 itself contains no mention of
climate migration. The Conference of the Parties (COP) to the Convention regularly convenes
every year, with the first meeting held in 1995. When COP convenes, the Parties to the
Convention review implementation of the Convention and other agreements and protocols of the
UNFCCC and decide on institutional and administration actions that impact the implementation
of the Convention (UNFCCC 2014). The most recent COP was held in Morocco in November
2016.
The Kyoto Protocol
The Kyoto Protocol is considered the enforcement side of the UNFCCC, as it commits
developed countries to meet the expectations outlined in the Convention, with compliance
requirements. While the Kyoto Protocol was adopted in 1997, it did not enter into force until
2005, due to a complicated and beleaguered ratification process. Key industrialized nations have
not ratified the Protocol, including the United States of America; Canada, which had formerly
ratified the Protocol, withdrew from it in 2012. The Kyoto Protocol, “structured on the principles
of the Convention… only binds developed countries [to emission reduction targets] because it
recognizes that they are largely responsible for the current high levels of GHG emissions in the
atmosphere, which are the result of more than 150 years of industrial activity. KP [Kyoto
Protocol] places a heavier burden on developed nations under its central principle: that of
‘common but differentiated responsibility’” (UNFCCC 2014). Like the UNFCCC, the Kyoto
Protocol does not contain a consideration of climate migration. The first commitment period,
with binding emission reduction targets was from the five-year period 2008-2012. The Doha
Amendment to the Kyoto Protocol was adopted in 2012 in Qatar and launched the second
commitment period, from 2013 until 2020; due to not having sufficient ratifications, the Doha
Amendment has not entered into effect. While the majority of carbon emissions must come from
on-shore developed nations, the Kyoto Protocol has mechanisms that enable market-based
emissions trade. These market-based mechanisms allow for reducing greenhouse gas emissions
in developing nations, while also, ideally, generating the creation of green energy capabilities in
developing states, “leap-frogging” older energy sources that cause more pollution (UNFCCC
2014). One flexible mechanism, known as the Clean Development Mechanism (CDM), which
functions as a cap and trade system, enables developing countries to earn emission certified
emission reduction (CER) credits, each equivalent to one tonne of CO2 (UNFCCC 2014). The
CDMs function as an adaption option for developing nations, which simultaneously help
developed nations meet their emissions reductions targets. Moreover, the Kyoto Protocol
identifies which states are most vulnerable to negative repercussions of climate change and those
states most able to reduce emissions, based on economic power and stability.
The Paris Agreement
When the Kyoto Protocol Doha Amendment expires in 2020, the Paris Agreement,
decided upon at the 2015 COP, provides the new global platform for emissions reductions.
Unlike the Kyoto Protocol, the Paris Agreements has legally-binding reduction targets for all
27
nations, not just developed states. Some aspects of the Agreement are legally-binding, while
others are not. Moreover, the Agreement primarily functions on international goodwill, not
having compliance or enforcement mechanisms. The legally-binding status of the Agreement can
function as a double-edged sword, as it can prevent some countries from ratifying it in the first
place if in place, and on the other hand, if not in place, leaves an “easy-out” option available to
states (Bang, Hovi, and Skodvin 2016). A consideration of climate migration remains outside the
scope of the Paris Agreement.
Cancun Adaptation Framework
Unlike the UNFCCC, Kyoto Protocol, and Paris Agreement, the Cancun Adaptation
Framework, given its focus on responses to climate change, considers the role of states in the
coordination of climate change induced migration and displacement (UNFCCC 2011). Following
the language of the UNFCCC, the Cancun Adaptation Framework recognizes the differentiated
abilities and responsibilities of states in responding to adaptation concerns, such as coordinating
planned relocation when it comes to forced displacement due to climate change (UNFCCC
2011).
Appendix B: The United Nations High Commission for Refugees and the term
“Climate Refugee”
The United Nations Convention Relating to the Status of Refugees provides a definition for the
term refugee for use consistent in international law. It specifically refers to those who have been
displaced from their home countries due to political or social persecution, based on their identity
characteristics in their home country, including race, ethnicity, political affiliation, and religion,
among others. In this definition, act of persecution against a stateless person implies the
responsibility or culpability of a human actor (UN General Assembly 1951). When considering
the term “climate refugees,” there is no similar responsible human actor (McLeman 2014:31).
Thus, the term is inconsistent with international law. Those who have been forcibly displaced
due to progressive climate change or environmental disasters are thus not considered “refugees”
according to this definition of a political refugee. In order to remain consistent with international
law, the term “forced climate migrant” may be applicable in cases where displacement, whether
it is regional, national or international occurs. Following existing terminology enables
consistency within the international arena and also for domestic policy decisions. Moreover, the
Guiding Principals on Internal Displacement, which is a non-legally binding document, outlines
the rights and of displaced persons, including those displaced by “natural or human-made
disasters,” although in the context of intra-state displacement (OCHA 1998).
In “Protecting Climate Refugees: The Case for a Global Protocol,” Biermann and Boas
(nevertheless still using the term “refugee”) argue for the creation of a global, international
protocol outlining the rights of climate refugees and the responsibilities of industrialized nations
toward them (Biermann and Boas 2008). Biermann and Boas urge the international community
to draft a protocol that is not “framed in terms of emergency response and disaster relief[,] but in
planned and organized voluntary resettlement programs” (2008:15), as voluntary resettlement
programs are seen as more successful.
28
The Nansen Initiative, the Protection Agenda, and the Platform on Disaster Displacement
While not a Protocol, the Nansen Initiative does meet some of what Biermann and Boas call for.
Launched in 2012 by Norway and Switzerland, the Nansen Initiative was a state-led consultative
process, which led to the creation of a Protection Agenda, which addresses “the needs of people
displaced across borders in the context of disasters and climate change” (The Nansen Initiative
2015b). The Protection Agenda “[c]onceptualizes a comprehensive approach to disaster
displacement that primarily focuses on protecting cross-border disaster-displaced persons” (The
Nansen Initiative 2015a). It provides effective state practices for managing cross-border disasterdisplacement and also accounts for measures to manage disaster displacement risks in origin
countries. The Platform on Disaster Displacement continues the work of the Nansen Initiative, in
implementing the recommendations of the Protection Agenda. The Platform urges regional,
bilateral, and international cooperation and coordination in responding to displacement and
migration border resulting from disasters and climate change, having identified a lack of
coordinated effort as a stumbling block to appropriate adaptation responses. The work of the
Platform also deals with cross-cutting issues, including “humanitarian assistance and protection,
human rights, migration management, refugee protection, [and] disaster risk reduction” (The
Nansen Initiative 2015b).
29
30