DeSherbinin 2013 CCHotspots ClimaticChange
DeSherbinin 2013 CCHotspots ClimaticChange
DeSherbinin 2013 CCHotspots ClimaticChange
Abstract
In the past five years there has been a proliferation of efforts to map climate change hotspots
regions that are particularly vulnerable to current or future climate impacts, and where human security
may be at risk. While some are academic exercises, many are produced with the goal of drawing policy
maker attention to regions that are particularly susceptible to climate impacts, either to mitigate the risk
of humanitarian crises or conflicts or to target adaptation assistance. Hotspots mapping efforts address
a range of issues and sectors such as vulnerable populations, humanitarian crises, conflict, agriculture
and food security, and water resources. This paper offers a timely assessment of the strengths and
weaknesses of current hotspots mapping approaches with the goal of improving future efforts. It also
highlights regions that are anticipated, based on combinations of high exposure, high sensitivity and low
adaptive capacity, to suffer significant impacts from climate change.
Keywords: hotspots, spatial vulnerability assessment, climate impacts
Note: This is an unformatted post-print version of the article. The final published version of this paper in
Climatic Change is available at http://dx.doi.org/10.1007/s10584-013-0900-7 or via
http://link.springer.com/.
1
Revised Accepted Manuscript submitted to Climatic Change (8 August 2013).
Contents
1. Introduction .................................................................................................................................. 3
2. Common Conceptual, Data and Methodological Issues ............................................................... 3
3. Assessment of Hotspots Mapping Efforts ..................................................................................... 5
3.1. Climate change exposure hotspots ........................................................................................ 5
3.2. Population dynamics and migration hotspots ....................................................................... 6
3.3. Disasters and humanitarian crises ......................................................................................... 7
3.4. Agriculture and food security hotspots.................................................................................. 8
3.5. Water resources hotspots ...................................................................................................... 9
4. Discussion and Conclusions .......................................................................................................... 9
4.1 Data and methodologies ............................................................................................................. 9
4.2 Where are the hotspots? .......................................................................................................... 10
4.3 Design and communication of results ....................................................................................... 11
4.4 Hotspots maps as tools for decision making ............................................................................. 11
4.5. Suggestions for future research ............................................................................................... 12
5. Conclusion ....................................................................................................................................... 13
Acknowledgements ............................................................................................................................. 13
References........................................................................................................................................... 13
Supplementary Online Material Climate Change Hotspots Mapping .............................................. 17
SOM 1. Regional hotspots mapping efforts ........................................................................................ 17
SOM 1.1. Hotspots Maps for Africa................................................................................................. 17
SOM 1.2. Hotspots Map for Southeast Asia .................................................................................... 19
SOM 1.3. Hotspots Maps for Europe .............................................................................................. 19
SOM 2. Hotspots mapping based on expert opinion ......................................................................... 20
SOM 2.1. Population dynamics and migration hotspots ................................................................ 20
SOM 2.2. Security and conflict hotspots ......................................................................................... 20
Annex: Tables and Figures .................................................................................................................. 22
2
Revised Accepted Manuscript submitted to Climatic Change (8 August 2013).
1. Introduction
Maps depicting climate change hotspots have been issued with increasing regularity in recent
years by researchers, advocacy groups, and NGOs. By identifying likely climate change impacts and
conveying them in a map format with strong visual elements, hotspots maps can help to communicate
issues in a manner that may be easier to interpret than text. Hotspots maps are developed with a
number of goals in mind. Academic researchers are generally seeking to vet data and methodologies,
applied researchers may be interested in guiding institutional strategies, and NGOs are often
communicating climate impacts. In addition, building on early roots in biodiversity hotspots mapping
(Myers 1990) where hotspots were developed to target conservation efforts, hotspots maps are often
explicitly developed to help aid organizations in priority setting and strategic planning with regards to
climate adaptation projects (Kok et al. 2011; Midgley et al. 2011; Yusuf and Francisco 2009). At a time of
increasing pressure on donors and development organizations to show that scarce public resources are
being used in a responsible manner, spatial indicators and hotspots maps hold the promise of
transparent, scientific, and defensible priority setting (Barnett et al. 2008). Although hotspots mapping
holds great promise for informing policy, there are a number of risks as well, which are reviewed in the
discussion section. This paper offers a timely assessment of the strengths and weaknesses of current
hotspots mapping approaches with the goal of improving future efforts. It also highlights regions that
are anticipated, based on combinations of high exposure, high sensitivity and low adaptive capacity, to
suffer significant impacts from climate change.
This review focuses on global data-driven GIS or modeling approaches to hotspots identification.
Unlike national level hotspots mapping, these efforts capture subnational variation in vulnerability by
combining spatial data layers, generally by converting each layer to a unitless scale and aggregating the
layers to reveal vulnerability levels. In this approach, hotspots emerge from the spatial analysis, being
revealed through the integration of spatial layers. In the supplementary online material (SOM) I also
review several regional GIS-based and global expert-based hotspots mapping efforts.
I exclude from this review hotpots mapping efforts that use countries as the units of analysis, since
these are essentially repackaging of country level indicators (e.g., Birkmann et al. 2011; Yohe et al.
2006), with all the limitations inherent in those approaches (Barnett et al. 2008). I also limit this review
to mapping efforts whose primary goal is explicitly to identify hotspots or geographic areas where
impacts will be greatest (even if not labeled per se as hotspots), rather than maps describing impacts
that are incidental to a publication or report.
Papers meeting these criteria were identified through Google Scholar searches on climate change
hotspots and hot spots, announcements, and bibliographies of other spatial vulnerability assessment
reviews. This paper identifies some common conceptual, data and methodological issues (Section 2);
and then moves to a review of hotspots mapping efforts in the two broad classes (Section 3). It then
proceeds to discussion (Section 4) and conclusions (Section 5).
3
Revised Accepted Manuscript submitted to Climatic Change (8 August 2013).
vulnerability mapping is the most common, and also the most conceptually challenging owing to the
proliferation of frameworks and definitions (Eakin and Luers 2006; Adger 2006). OBrien et al. (2007)
describe two broad categories of vulnerability definitions, one that identifies contextual vulnerability or
the intrinsic characteristics of a system, which is rooted in political economy, and another, outcome
vulnerability, which combines information on potential climate impacts and on the capacity of society to
cope and adapt. Vulnerability hotspots mapping efforts generally adopt the latter, which is closer to the
IPCC definition of vulnerability as a function of exposure, sensitivity, and coping capacity (Parry et al.
2007). While there is no one correct way to characterize vulnerability, there are certainly wrong ways to
do so. Fssel (2007) argues that quantitative vulnerability assessment requires definition of the system
of analysis (what is vulnerable?), the valued attributes of concern (why is it important?), the external
hazard (to what is the system vulnerable?), and a temporal reference (when?).
Beyond conceptual definitions and frameworks, there are significant measurement challenges
(Birkmann and Wisner 2006). The exposure aspect of vulnerability generally presents fewer problems,
since biophysical data sets are reasonably well advanced, and the uncertainties in the data are, for the
most part, quantifiable. However, owing to data gaps, the socioeconomic aspects are often measured
through the use of proxies. Thus for sensitivity it is common to use close surrogates such as poverty
levels and income, and for coping or adaptive capacity, measures might include education, institutional
capacity, funding levels for disaster risk reduction, or insurance coverage. These are often less-than-
adequate proxies for intrinsic vulnerability, and many of them are difficult to measure or data may be
difficult to obtain. As Kasperson et al. (2005: 149) write, political and social marginalization, gendered
relationships, and physiological differences are commonly identified variables influencing vulnerability,
but incorporating this conceptual understanding in global mapping remains a challenge.
Not all hotspots mapping efforts actually incorporate future climate change and variability. Some
use past variability or extreme events as a proxy for future changes. However, those that do use general
circulation model (GCM) outputs run into a number of issues. A fundamental challenge for vulnerability
mapping that relies on accurate prediction of extremes, such as that for hazards or human vulnerability,
is the limited ability of GCMs to capture historical variance or future extremes (IPCC 2012; Brown &
Wilby 2012). The use of multi-model ensembles only tends to further reduce variance. The spatial
resolution of the model outputs ranging in resolution from 1 to 2 degree grid cells is also a concern,
and some efforts do not follow the best practice of using multiple models for a given SRES scenario.
Specific climate parameters that are required will differ based on the kind of hotspots assessment.
For agricultural systems, water management, or natural hazard prediction, the most important variables
would be anticipated change in rainy season onset, gaps in rainfall during growing seasons, changes in
drought periodicity, changes in rainfall duration and intensity, and temperature increases beyond
certain crop thresholds. These parameters are not easy to calculate, so hotspots efforts require a certain
amount of expertise in climate data analysis. Although GCM outputs have uncertainties, it should be
noted that hotspots mapping efforts that rely on long-term precipitation reanalysis data are also
inaccurate in some regions, especially in developing countries where there is sparse rain gauge data.
Finally, Preston et al. (2011) highlight a number of temporal and scale issues that tend to plague
vulnerability mapping. Often data layers are from inconsistent dates, scale mismatches in underlying
data sets create spatial artifacts in the maps, and for mapping efforts that do use GCM outputs, the
4
Revised Accepted Manuscript submitted to Climatic Change (8 August 2013).
sensitivity and adaptive capacity variables represent current rather than future states. These issues are
addressed again below.
5
Revised Accepted Manuscript submitted to Climatic Change (8 August 2013).
tropics and high latitudes. The CCI has the advantage that results are presented on a pixel basis, and
they are also summarized by country for the benefit of policy audiences, and the scores are also
relatively easy to interpret. The CCI depicts similar hotspots to the RCCI, except for northern southern
Africa and the Amazon, which are found to have much greater exposure to climate risks when compared
to the present (Figure 2).
The RCCI and CCI contribute to the literature by describing changes to climate parameters that
could serve as inputs to broader vulnerability assessments. However, despite claims to the contrary,
both are clearly for academic audiences. By contrast, the UK Met Office has produced a map depicting
the regional temperature changes associated with a 4oC rise in global mean temperature (Figure 3). The
map, produced in poster and online interactive forms, is intended for policy audiences, utilizing circles of
various colors to highlight likely impacts. Areas with the greatest temperature changes include the Arctic
and high northern latitudes, the western US, the Amazon, West Africa, southern Africa, and Central Asia.
A hybrid approach, based on climate parameters but tied to thresholds in four important sectors
(water, agriculture, ecosystems and health), was recently developed by Piontek et al. (2013). The
authors use the outputs of three GCMs simulating the highest representative concentration pathway
(RCP8.5) to feed multiple Global Impact Models (GIMs), and then identify temperature thresholds in
each sector where impacts could be considered to be severe. For example, the thresholds for the water
and agriculture sectors are defined as the 10th percentile of the reference period distribution (1980-
2010) of river discharge and crop yields, respectively. For each GIM-GCM combination and at each grid
cell they define a crossing temperature that is the global mean temperature change (GMT) at which
the sectoral metric crosses the respective impact threshold. Hotspot regions where thresholds are
crossed for two or three sectors for a 4.5oC GMT are found in Figure 4, with high impacts found in the
Amazon, the Andes, southern Mexico and Central America, southern and eastern Europe, the African
highlands and parts of West Africa, and the Ganges basin. These results should be seen as conservative,
given the stringent criteria for inclusion of severe impacts (>50% of GIM-GCM combinations agreeing)
used in the study.
6
Revised Accepted Manuscript submitted to Climatic Change (8 August 2013).
A second climate-demography vulnerability index (CDVI) adds a third dimension, rapid population
growth, to shows areas in which conditions that already support high population densities and where
there is rapid population growth will see a decline in climate conditions. The authors ran their model
using several climate projections. Figures 5 and 6 show the resulting maps. For the CVI, hotspots of high
vulnerability are found in the Amazon basin, North Africa, Sudan, southern Africa, Central China,
Mongolia, and eastern Australia. For the CDVI, the same regions become hotter, while new areas are
added in Central America, the U.S. Southwest, most of Africa, the Arabian Peninsula, Afghanistan, and
Indonesia. Many of these are areas where climate change will amplify the conditions currently
supporting low population densities, e.g., hot and arid regions that will become drier. Yet the CDVI
clearly identifies a number of tropical humid regions (Amazon, Central Africa, and Indonesia) as hotspots
as well. One limitation is the treatment of populations as homogenous, and therefore having similar
sensitivity and adaptive capacity to climate change impacts.
A number of efforts have sought to identify hotspots of population vulnerability to sea level rise
(SLR). Here I review a representative global assessment by McGranahan et al. (2007), which utilizes a
Low Elevation Coastal Zone (LECZ) mask, representing coastal elevations from 0-10m, to identify the
regions that will be most affected by climate change impacts. Results are provided in spatial and tabular
formats, providing estimates of population exposure within the LECZ for urban and rural areas by
country. The method constitutes a simple overlay of the LECZ grid on a year 2000 population grid. The
maps identify highly populated areas at high risk of coastal flooding and SLR, especially the Asian mega-
deltas (Figure 7). The strength of this effort is that the methodology is simple and easy to understand,
and the impacts of SLR are relatively certain, though the timing of specific sea level increments is a
matter of some debate, and local impacts are hard to predict with global scale data sets.
7
Revised Accepted Manuscript submitted to Climatic Change (8 August 2013).
that the authors do not combine the climate risks (past hazards and future scenarios) with the human
vulnerability index in such a way as to draw out hotspots at the intersection of climate and societal
vulnerability.
8
Revised Accepted Manuscript submitted to Climatic Change (8 August 2013).
there are swaths or resource poor areas with severe to moderate poverty that are more vulnerable to
global change, including the Sahel, the Horn of Africa, Afghanistan, and small areas of western China.
The Andes, southern Maghreb, Arabian peninsula, Iran and Pakistan and the rest of western China are
deemed to be resource poor with only moderate poverty.
9
Revised Accepted Manuscript submitted to Climatic Change (8 August 2013).
availability and mapping scales for a number of socioeconomic variables relevant to the mapping of
sensitivity and adaptive/coping capacity. Often, proxies such as poverty and malnutrition are used to
measure social vulnerability, yet it is acknowledged in the literature that this is a multidimensional, time
dependent and often complex concept that is difficult to capture in static maps (Kasperson et al. 2005).
More fundamentally, the SREX framework (IPCC 2012) which separates exposure and vulnerability -
may yield better results for policy audiences since it translates more easily into a risk management
approach.
Climate projections tend to be more common in hotspots studies of biophysical systems, especially
as inputs to other models, and are rarely used for social or general vulnerability assessments
(exceptions include Midgley et al. 2011, CARE and Maplecroft 2008). Apart from Busby et al. (2011),
climate and security mapping efforts appear to be less sophisticated, relying to a greater extent on
expert opinion (e.g., Schubert et al. 2007). Overall, there are strong disciplinary influences reflected in
each of the approaches. The RCCI and CCI are entirely grounded in climate science, the CVI in ecology,
the livelihood systems mapping in integrated assessment (Kok et al. 2010) and development practice
(Warner et al. 2009, Thornton et al. 2008), and most agriculture and water hotspots maps are generated
by modelers.
Many hotspots mapping efforts are affected by the spatial scale and uncertainties in the available
global data sets. Kok et al. point out that there is a gap between local vulnerability assessment, which
address context-specific situations with more detailed data, and the kinds of analyses possible for global
VAs, which are based on aggregated data and rather crude assumptions about the underlying
mechanisms being assessed. Bridging this gap will prove to be difficult.
10
Revised Accepted Manuscript submitted to Climatic Change (8 August 2013).
11
Revised Accepted Manuscript submitted to Climatic Change (8 August 2013).
investments (see Table 2 for funders), it seems unlikely that national or local policy makers will take up
the maps as planning tools. Many country level decision makers distrust global/regional mapping efforts
because they dont understand/accept the methods, or question data inputs if they are not from their
own national agencies. There is little direct evidence that the maps actually influence investments or
adaptation activities, but Preston et al. (2011) suggest that maps may serve as boundary objects that
facilitate discourse.
More broadly, there is a risk, should the maps actually influence decisions, that quantification gives
decision-makers the false impression that the information is more objective (Preston et al. 2011). Yet
the framing of issues and selection of indicators cannot be presented as purely the result of objective
scientific criteria. By reifying vulnerability and resilience, and relying on proxies, other qualitative
aspects such as culture, power relations, and local ecological knowledge can be overlooked or
downplayed (Adger 2006; Kasperson et al. 2005). Seemingly innocent and value neutral, maps could play
an important role in framing societal responses to climate change and its impacts in ways that are surely
not neutral.
With increased attention and funding being devoted to adaptation, there is now a financial
incentive for countries to portray themselves as highly vulnerable to climate impacts. Yet, should donors
use hotspots maps to guide investments, there is a potential to reward countries with poor governance
should they be identified as the most vulnerable. Conversely, there is a moral hazard that countries
could suffer funding triage if they are deemed overly vulnerable to climate impacts. The role of
hotspots maps in political discourse and guiding decisionmaking deserves more attention.
12
Revised Accepted Manuscript submitted to Climatic Change (8 August 2013).
5. Conclusion
This paper reviews a number of global and regional hotspots mapping efforts, assessing data and
methods, the hotspots identified, and their efficacy as tools for risk communication and decisionmaking.
Efforts to date can largely be characterized as supply-driven academic exercises rather than responding
to demands from the policy community. Yet in a world where human security is potentially imperiled by
temperature increases of >4oC, and where loss and damage has become part of the UN Framework
Convention on Climate Change lexicon, demand for hotspots maps will likely increase as decisionmakers
seek to identify where impacts will be greatest and what adaptation measures, if any, are possible.
Acknowledgements
The author would like to acknowledge comments on an earlier version of this paper by Richard Sliuzas of
ITC/University of Twente and by three anonymous reviewers. The author also presented earlier versions
of this paper and benefited from exchanges with researchers at the ICARUS II and Adaptation Futures
conferences in May 2012.
References
Abson, D. J., A. J. Dougill, and L. C. Stringer. 2012. Using Principal Component Analysis for information-
rich socio-ecological vulnerability mapping in Southern Africa. Applied Geography, 35, 515-524.
Adger, N. 2006. Vulnerability. Global Environmental Change, 16: 268-281.
Baettig, M.B., M. Wild, and D.M. Imboden. 2007. A climate change index. Geophysical Research Letters,
34, L01705
Barnett, J., S. Lambert, and I. Fry (2008). "The hazards of indicators: insights from the Environmental
Vulnerability Index." Annals of the Association of American Geographers 98(1): 102-119.
Birkmann, J., D. Krause, N. Stiadi, D. Suarez, T. Welle, J. Wolfertz. 2011. World Risk Report. Bonn,
Germany: UNU and IEHS.
Birkmann, J., and B. Wisner. 2006. Measuring the Un-Measurable. UNU-EHS SOURCE No. 5, Bonn,
Germany: UNU-EHS.
Black, R., S.R.G. Bennett, S.M. Thomas, and J.R. Beddington. 2011. Migration as Adaptation. Nature 478:
447-449.
Brown, C. and R. L. Wilby. 2012. An Alternate Approach to Assessing Climate Risks. Eos, 93(41): 401-
402.
Busby, J.W., T.G. Smith, and K.L. White. 2011. Locating Climate Insecurity: Where are the most
vulnerable places in Africa? Climate Change and African Political Stability Program Policy Brief No.
3.
CARE and Maplecroft. 2008. Humanitarian implications of climate change: Mapping emerging trends
and risk hotspots. Geneva, Switzerland: CARE International.
13
Revised Accepted Manuscript submitted to Climatic Change (8 August 2013).
Davies, R.A.G., and S.J.E. Midgley. 2010. Risk and Vulnerability Mapping in Southern Africa: A Hotspots
Analysis. Cape Town, South Africa: OneWorld Sustainable Investments (Pty) Ltd.
de Sherbinin, A., K. Warner, and C. Ehrhart. 2011. Casualties of Climate Change, Scientific American,
January 2011: 64-71.
De Stefano, L., J. Duncan, S. Dinar, K. Stahl, K. Strzepek, and A.T. Wolf. 2010. Mapping the Resilience of
International River Basins to Future Climate Change-Induced Water Variability. Water Sector Board
Discussion Paper Series No. 15. Washington DC: The World Bank.
Diffenbaugh, N. S., F. Giorgi, and J. S. Pal. 2008. Climate change hotspots in the United States,
Geophysical Research Letters, 35, L16709.
Dilley, M., R. Chen, U. Deichmann, A. Lerner-Lam, and M. Arnold. 2005. Natural Disaster Hotspots.
Washington, DC: World. Bank.
Dll, P. 2009. Vulnerability to the impact of climate change on renewable groundwater resources.
Environmental Research Letters, 4:035006. doi:10.1088/1748-9326/4/3/035006
Dransch, D., H. Rotzoll, and K. Poser. 2010. The contribution of maps to the challenges of risk
communication to the public. Intl Journal of Digital Earth, 3(3):292-311.
Eakin, H., and A. Luers. 2006. Assessing the vulnerability of socio-environmental systems. Annual Review
of Environment and Resources. 31:36594.
Ericksen, P., P. Thornton, A. Notenbaert, L. Cramer, P. Jones, M. Herrero. 2011. Mapping hotspots of
climate change and food insecurity in the global tropics. CCAFS Report no. 5. Copenhagen,
Denmark.
ESPON Climate. 2011. Climate Change and Territorial Effects on Regions and Local Economies. Applied
Research 2013/1/4. Final Report, Version 31/5/2011.
Fraser, E.D.G., E. Simelton, M. Termansen, S.N. Gosling, and A. South. 2013. Vulnerability hotspots:
Integrating socio-economic and hydrological models to identify where cereal production may
decline in the future due to climate change induced drought. Agricultural and Forest Meteorology,
170: 195-205.
Fssel, H.-M. 2007. Vulnerability: A generally applicable conceptual framework for climate change
research. Global Environmental Change, 17:155-167.
Giorgi, F. 2006. Climate change hot-spots, Geophysical Research Letters, 33, L08707,
doi:10.1029/2006GL025734.
Hagenlocher, M., S. Lang, D. Holbling, et al. 2013. Modeling Hotspots of Climate Change in the Sahel
Using Object-Based Regionalization of Multidimensional Gridded Datasets. IEEE Journal of Selected
Topics in Applied Earth Observations and Remote Sensing. doi: 10.1109/JSTARS.2013.2259579
IPCC (Intergovernmental Panel on Climate Change). 2012. Managing the Risks of Extreme Events and
Disasters to Advance Climate Change Adaptation. Field, C.B., et al. (eds.). Cambridge University
Press, Cambridge, UK, and New York, NY, USA, 582 pp.
Kasperson R.E., K. Dow, E. Archer, D. Caceres, T. Downing, et. al. 2005. Vulnerable People and Places.
Chapter 6 in: Millennium Ecosystem Assessment. Washington, DC: Island Press.
Kok, M.T.J., M.K.B. Ldeke, T. Sterzel, P.L. Lucas, C. Walter, P. Janssen and I. de Soysa. 2010.
Quantitative analysis of patterns of vulnerability to global environmental change. Den
Haag/Bilthoven, Netherlands: PBL.
14
Revised Accepted Manuscript submitted to Climatic Change (8 August 2013).
Liu, J., S. Fritz, C.F.A. van Wesenbeeck, M. Fuchs, L. You, M. Obersteiner, H. Yang. 2008. A spatially
explicit assessment of current and future hotspots of hunger in Sub-Saharan Africa in the context of
global change. Global and Planetary Change, 64 (2008): 222235.
Ldeke, M.K.B., G. Petschel-Held, H.J. Schellnhuber. 2004. Syndromes of Global Change. GAIA -
Ecological Perspectives for Science and Society, 13(1):42-49.
McGranahan, G., D. Balk, and B. Anderson. 2007. Low Elevation Coastal Zone (LECZ) Urban-Rural
Population Estimates. Palisades, NY: NASA SEDAC.
Midgley, S.J.E., R.A.G. Davies, and S. Chesterman. 2011. Climate Risk and Vulnerability Mapping: Status
quo (2008) and future (2050). Report produced for UK Department for International Development
(DFID).
Montello, D.R., and S. Freundschuh. 2005. "Cognition of Geographic Information". In: R.B. Mc.Master
and E.L. Usery (eds.), A Research Agenda for Geographic Information Science. Boca Raton FL: CRC
Press.
Myers, N. 1990. Threatened Biotas: Expanded Hot-Spots Analysis, The Environmentalist 10:243-256.
OBrien, K.L., S. Eriksen, L. Nygaard, and A. Schjolden. 2007. Why Different Interpretations of
Vulnerability Matter in Climate Change Discourses. Climate Policy 7: 73-88
Parry, M.L., O.F. Canziani, J.P. Palutikof, P.J. van der Linden and C.E. Hanson (eds). 2007. Climate Change
2007: Impacts, Adaptation and Vulnerability. Cambridge, UK: Cambridge University Press.
Parish, E.S., E. Kodra, K. Steinhauser, and A.R. Ganguly. 2012. Estimating future global per capita water
availability based on changes in climate and population. Computers & Geosciences, 42: 79-86.
Patt, A., and S. Dessai. 2005. Communicating uncertainty: lessons learned and suggestions for climate
change assessment. C.R. Geoscience, 337:425441.
Piontek, F., C. Mller, T.A.M. Pugh, et al. 2013. Multisectoral climate impacts in a warming world.
Proceedings of the National Academy of Sciences (early online edition),
doi:10.1073/pnas.1222471110.
Preston, B., E.J. Yuen, R.M. Westaway. 2011. Putting vulnerability to climate change on the map: a
review of approaches, benefits, and risks. Sustainability Science, 6:177-202.
Samson, J., D. Berteaux, B.J. McGill and M.M. Humphries. 2011. Geographic disparities and moral
hazards in the predicted impacts of climate change on human populations. Global Ecology and
Biogeography doi:10.1111/j.1466-8238.2010.00632.x
Scheffran, J., and A. Battaglini. 2011. Climate and conflicts: the security risks of global warming. Regional
Environmental Change 11(Suppl 1):S27-S39. doi:10.1007/s10113-010-0175-8
Schrter, D., W. Cramer, R. Leemans, et al.. 2005. Ecosystem Service Supply and Vulnerability to Global
Change in Europe. Science 310:1333-1337.
Schubert, R., H.J. Schellnhuber, N. Buchmann, A. Epiney, R. Griehammer, M. Kulessa, D. Messner, S.
Rahmstorf, and J. Schmid. 2007. Climate Change as a Security Risk. London: Earthscan and German
Advisory Council on Global Change (WBGU).
Shen, X., T.E. Downing and M. Hamza (eds). 2010. Tipping Points in Humanitarian Crisis: From Hot Spots
to Hot Systems. Studies of the University: Research, Counsel, Education (SOURCE) Publication Series
of UNU-EHS, No. 13/2010.
15
Revised Accepted Manuscript submitted to Climatic Change (8 August 2013).
16
Revised Accepted Manuscript submitted to Climatic Change (8 August 2013).
17
Revised Accepted Manuscript submitted to Climatic Change (8 August 2013).
In the second vulnerability hotspots mapping effort for southern Africa, Abson et al. (2012) created
vulnerability maps based on principal components analysis (PCA). They argue that the standard practice
of averaging or summing indicator scores hides important information regarding the relations between
the original variables. Because the principal components (PCs) are uncorrelated, the scores associated
with each PC encapsulate a unique aspect of the overall socio-ecological vulnerability represented by
the original set of vulnerability indicators. However, since the components are statistically derived, it can
be difficult to attribute meaning to a specific component. For example, their first PC, which they term
poverty and health vulnerability, includes infant mortality, poverty, agricultural constraints, and
malnutrition, which is straightforward enough. But their third PC, termed infrastructure poverty and
population pressure vulnerability, combines population per net primary productivity, infrastructure
poverty (a measure of population divided by night time lights), and travel time to major cities. It is hard
to make sense of this except perhaps as a proxy for population density.
Figure 16 provides a comparison of the results by these two efforts, revealing broadly similar
patterns but also some notable differences. For example, Midgley et al. find Zimbabwe and southern
Zambia to be highly vulnerable but Abson et al. find them to be less so. Conversely, Abson et al. find
most of the Congo and Angola to be highly vulnerable, but Midgley et al. find them to be less so. While
the results are not directly comparable owing to the use of different indicator sets, it does serve to
illustrate the fact that depictions of vulnerability patterns in spatial index approaches depend heavily on
data and methods.
Liu et al. (2008) focus on hunger hotspots using multiple crop modeling outputs. They identify
areas of high population density and current undernutrtion problems that are likely to see decreases in
per capita calorie availability of 0-30% and >30% (Figure 17). A major area of current and future
vulnerability is the highlands of Ethiopia; Areas stretching from western Tanzania to Mozambique are
projected to see >30% declines in calorie intake, and the lakes region, northern Nigeria, and parts of
southern Nigeria are considered currently vulnerable but without significant changes in future calorie
intake.
Thornton et al. (2008) map hotspots of climate change and poverty in Africa using principle
components analysis on 14 indicators measuring five livelihood capitals (Carney 1998): natural capital
(e.g. soil degradation), physical capital (e.g., accessibility to markets), social capital (e.g., governance),
human capital (e.g., malaria and HIV prevalence), and financial capital (e.g., agricultural GDP). Regions
identified as most vulnerable include the Highlands of Ethiopia, southern Chad, southern Niger, and
Rwanda and Burundi, followed by most of the rest of Africa, with only Guinea, southern Ghana,
Namibia, and Zimbabwe and portions of South Africa near Johannesburg showing up as less vulnerable.
The selection of Guinea and Zimbabwe as less vulnerable is puzzling, and may have to do with data
limitations.
Finally, Hagenlocher et al. (2013), in a climate-focused approach similar to that of Baettig et al.s
CCI, develop an innovative modeling approach using historical climatological and vegetation index data
sets to delineate areas with relatively high climate change impacts in West Africa. Hotspots are
identified as areas where temperature and precipitation trends are pronounced and drought and flood
events over the past 24-36 years have been severe, with a focus on the rainy season from May to
October. The map (Figure 18) reveals both the areas of high impacts, and the proportion of the impact
that can be attributable to given impacts. For example, flood impacts dominate in the hotspots of
18
Revised Accepted Manuscript submitted to Climatic Change (8 August 2013).
Burkina Faso, Ghana, and southern Niger, whereas precipitation trends dominate in western Mauritania.
The maps lack any reference to population vulnerability, but some hotspots do coincide with larger
population centers, such as the flood hotspots in northern Nigeria (around Kano) and in southern
Burkina Faso. The approach also does not differentiate between increasing and decreasing trends in
precipitation, such that the rebound in precipitation following the great Sahelian droughts of the early
1970s and 1980s would be considered as contributing to climate hotspots in some regions.
19
Revised Accepted Manuscript submitted to Climatic Change (8 August 2013).
specific ecosystem services, such as changes to cropland areas and water stress. Here again, southern
Europe appears to be most impacted across multiple scenarios.
1
Climate Refugees, Not Found: Discredited by reality, the U.N.'s prophecies go missing. Wall Street Journal,
21 April 2011.
20
Revised Accepted Manuscript submitted to Climatic Change (8 August 2013).
risks (Figure 24). Examples include droughts and general water scarcity (e.g., in the Mediterranean basin
and western and central Asia), recurrent flooding (e.g., in coastal East Asia and parts of the Caribbean),
loss of ecosystems and ecosystem services (e.g., across the arctic), extreme events (e.g., in Central
America and Indonesia), and loss of coastal areas owing to SLR (e.g. in Oceania). Apart from illegibility
and poor cartography, the map fulfills the purpose of distilling major issues. A weakness of both
mapping efforts is the lack of underlying data and over-reliance on the authors subjective assessments.
21
Revised Accepted Manuscript submitted to Climatic Change (8 August 2013).
22
Revised Accepted Manuscript submitted to Climatic Change (8 August 2013).
23
Revised Accepted Manuscript submitted to Climatic Change (8 August 2013).
24
Revised Accepted Manuscript submitted to Climatic Change (8 August 2013).
Table 2. Summary table of global climate change hotspots mapping efforts (continued)
McGranahan et al. 2007 CARE & Maplecroft and Thow & Ericksen et al. 2011 Fraser et al. 2012
de Blois 2008
Primary focus/foci Sea level rise Natural hazard vulnerability Food production Soil moisture and agriculture as it
affects wheat and maize
productivity
Target audience Researchers (urban) Policymakers and humanitarian Agricultural research community Agricultural research community
actors
Geographic scope Coastal Areas The developing world Tropical regions Global
Framework Focus on exposure and sensitivity IPCC vulnerability framework IPCC vulnerability framework n/a
Methods Overlay of coastal low elevation Combination of climate scenario Looks at change in growing season
band up to 10m in elevation on a data, important climate thresholds soil moisture in relation to
population grid with urban and for agriculture, data on natural adaptive capacity. Adaptive
rural identifiers resource degradation, and capacity was modeled based on
indicators of food availability, socioeconomic variables that have
access, and utilization. a high correlation with the crop
yield impacts of past drought
events.
Index None Maps are produced based on n/a
combinations of high-low
exposure, sensitivity, and capacity
Regions Africa Alexandria Sahel, Horn, Central Africa, Portions of the Sahel (rainfall Southern Africa (wheat and maize)
Identified Southern Africa variability) and moist tropical West
Africa and Rwanda/Burundi
(temperature thresholds)
Asia Coastal cities including Shanghai, Central Asia, Afghanistan and Most of India (especially for Western China (wheat)
Ho Chi Minh City Pakistan, Myanmar, Mongolia, rainfall variability but also
Borneo temperature thresholds)
Europe Amsterdam, Hamburg, London Balkans (wheat and maize)
Latin Am. & Buenos Aires, Rio De Janeiro Andes, Northern Mexico, None Southern cone (wheat and maize)
Caribbean Argentina
North Miami, New Orleans U.S. great plains (wheat)
America
Oceania Sydney, Melbourne None
Funder NASA Socioeconomic Data and UN Office for the Coordination of Consultative Group for UK National Environment Research
Applications Center Humanitarian Affairs and CARE International Agricultural Research Council (NERC)
(CGIAR) with funding from aid
agencies
25
Revised Accepted Manuscript submitted to Climatic Change (8 August 2013).
Table 2. Summary table of global climate change hotspots mapping efforts (continued)
Kok et al. 2011 Parish et al. 2011 Dll 2009
Primary focus/foci Vulnerability profiles based on resource Water resources Renewable groundwater resources
endowments, water, soil, and
development status
Target audience Researchers (integrated assessment) Researchers Researchers
Geographic scope Global Global Global
Framework SUST vulnerability framework Unclear Unclear
Methods Integrated assessment models, geospatial GCMs coupled to hydrological models GCMs coupled to the WaterGap
data, and cluster analysis to answer hydrological model
questions: (1) What are main exposures,
vulnerable groups and their sensitivities?
(2) What are V creating mechanisms? (3)
Where do they manifest? (4) How will
future changes affect wellbeing? (5) What
coping/ adaptation responses are
possible?
Index N/A Water stress Vulnerability index (human vulnerability
to climate change induced changes in
freshwater supply)
Africa Extreme poverty in Sahel and Horn of North Africa (most scenarios), with North Africa, extreme western Africa
Africa isolated spots in SS Africa (Mauritania/Senegal), southwestern
Africa (Angola, Namibia, western South
Africa)
Asia Extreme poverty in Afghanistan; Eastern China (A2 scenario) Portions of Central Asia (the stans) and
Moderate to Extreme poverty in NW western China
China; Extreme overuse in Pakistan and
Regions
Western India, and in NE China
Identified
Europe None None Parts of southern Europe
Latin Am. & Caribbean Moderate poverty Andes Central America, Northeastern Brazil Northeastern Brazil, coastal Peru and
(under A2 scenario) Chile
North America Marginal lands in West and Southwest West of the Great Lakes Ogalala aquifer in western Texas (in two
U.S. scenarios)
Oceania Marginal lands in West and Southwest None Western Australia (in most scenarios)
U.S., Marginal lands in Australia
Funder Netherlands Environmental Assessment Oak Ridge National Laboratory, U.S. Dept Not listed
Agency (PBL) of Energy
26
Revised Accepted Manuscript submitted to Climatic Change (8 August 2013).
27
Revised Accepted Manuscript submitted to Climatic Change (8 August 2013).
Figure 4. Multisectoral hotspots of impacts for two (orange) and three (red) overlapping sectors
Notes: Multisectoral hotspots of impacts for two (orange) and three (red) overlapping sectors in the strict
assessment, with 50% of GIM-GCM combinations agreeing on the threshold crossing in each sector, for a GMT
change of up to 4.5 C. An overlap of all four sectors does not occur in the strict assessment. Regions in light gray
are regions where no multisectoral overlap is possible. The dark gray shows the additional regions affected by
multisectoral pressures under the worst-case assessment, where a minimum of 10% of all sectoral GIM-GCM
combinations have to agree on the threshold crossing.
29
Revised Accepted Manuscript submitted to Climatic Change (8 August 2013).
30
Revised Accepted Manuscript submitted to Climatic Change (8 August 2013).
31
Revised Accepted Manuscript submitted to Climatic Change (8 August 2013).
32
Revised Accepted Manuscript submitted to Climatic Change (8 August 2013).
33
Revised Accepted Manuscript submitted to Climatic Change (8 August 2013).
Figure 15. Vulnerability hotspots (a. 2008 and b. 2050). (Red values indicate hotspots where
people are most likely to be most in need of help adapting to climate stressors, while the blue areas
indicate areas of resilience.)
a. b.
Figure 16. Comparison of vulnerability maps produced by Midgley et al. and Abson et al. (data
and methods are discussed in Section 1.1 of the SOM)
Sources: Davies & MIdgley 2010 and Midgley et al. 2011 (left) and Abson et al. 2012a (right).
34
Revised Accepted Manuscript submitted to Climatic Change (8 August 2013).
Figure 17. Number of people with current undernutrition problems in relation to future potential
hotspots of food insecurity in the 2030s
35
Revised Accepted Manuscript submitted to Climatic Change (8 August 2013).
36
Revised Accepted Manuscript submitted to Climatic Change (8 August 2013).
Figure 21. Projected percent change in rainfall runoff by 2080 overlaid on population distribution
Notes: Map insets are for (clockwise from top): rainfall runoff (1960-1980 baseline), cyclone frequency (1980-
2000), and rainfed agricultural areas.
37
Revised Accepted Manuscript submitted to Climatic Change (8 August 2013).
Source: Emmanuelle Bournay, UNEP-GRID Arendal, Fifty million climate refugees by 2010
http://www.grida.no/graphicslib/detail/fifty-million-climate-refugees-by-2010_71db
38
Revised Accepted Manuscript submitted to Climatic Change (8 August 2013).
Figure 23. Security risks associated with climate change: Selected hotspots
39