Item

Abstract

A methodology to assess future development in patterns of vulnerability is presented which can support the assessment of global policies with regard to their impacts on specific vulnerabilities on the regional or local scale. Patterns of vulnerability, formalized by vulnerability profiles (e.g. for the livelihoods of dryland smallholder farmers) were investigated under different consistent indicator scenarios reflecting different global policies. After unfolding several principal possibilities to do such an analysis of temporal change in vulnerability patterns we could conclude that the concept of “Clusters of Change” (CoCs) is the most straight forward and promising approach. The main arguments are that each interpretation has necessarily to consider both, the starting situation and it’s change over time (”poor and heavily improving”, ”rich and stagnating” etc.). This implies that we are looking for patterns which represent typical combinations of present states AND expected future changes. An application of the CoC-concept to the drylands vulnerability patterns considering the indicator set for the present situation and the same indicator set for 2050 under a baseline scenario was performed as a test. Comparison of the present vulnerability cluster partition with the spatial distribution of the CoCs revealed that most of these clusters are separated into an improving and a deteriorating part which shows where winners and losers of the baseline scenario are – an interesting result which illustrates the appropriateness of the CoC – method. To explore the potential of CoCs for the dryland vulnerability we applied the method to two different sets of scenarios until 2050: a baseline vs. Climate policy scenario (OECD, 2012) and a ”policy first” scenario vs. ”security first” scenario (UNEP, 2007). The first one serves as an example for a policy assessment while the second compares the vulnerability consequences of two scenarios based on different story-lines of further global development. The main conclusion to be drawn from these calculations is that the CoCs are rather insensitive with regard to the small differences between the scenarios. Regarding the first set of scenarios the relatively short time horizon of relevant influences of climate policies on climate change impacts and several indicators which are not influenced at all generate only a very small difference. The only significant change in the resulting vulnerability profiles was in the values of change in water scarcity: it was lower for all profiles in the climate policy case. The second set of scenarios is not directly related to policy decisions but to different global story-lines which deviate stronger. This resulted in an increasing cluster number from 4 (policy first) to 5 (security first) clusters, about 20% of the pixels changing cluster membership, 3 clusters showing the same spatial extent for both scenarios but the 4th cluster (“policy first”) “losing” India which generates a separate cluster in the “security first” scenario. This allows for the interpretation that a further development according to the “security first”-storyline compared to the “policy first”-storyline would make a difference particularly for India. Closer inspection of the respective profile shows a qualitatively different situation indicating increased vulnerability compared to the “policy first” scenario where India shares one cluster with e.g., Northern Africa.