Global climate change is having marked influences on species distributions, phenology and ecosystem composition and raises questions as to the effectiveness of current conservation strategies. Conservation planning has only recently begun to adequately account for dynamic threats such as climate change. We propose a method to incorporate climate-dynamic environmental domains, identified using specific environmental correlates of floristic composition, into conservation strategies, using the province of KwaZulu-Natal, South Africa as a case study. The environmental domains offer an approach to conservation that conserves diversity under current and future climates, recognising that the species constituting diversity may change through time. We mapped current locations of domains by identifying their positions in a multi-dimensional environmental space using a non-hierarchical iterative k-means clustering algorithm. Their future locations were explored using an ensemble of future climate scenarios. The HadCM2 and GFDL2.1 models represented the extreme ranges of the models. The magnitude of change in each environmental domain was calculated using Euclidean distances to determine areas of greatest and least stability for each future climate projection. Domains occurring in the savanna biome increase at the expense of domains occurring in the grassland biome, which has significant negative consequences for the species rich grasslands. The magnitude of change maps represents areas of changed climatic conditions or edaphic disjunctions. The HadCM2 model predicted the greatest overall magnitude of change across the province. Species with specific soil requirements may not be able to track changing climatic conditions. A vulnerability framework was developed that incorporated climatic stability and habitat intactness indices. The mean magnitude of change informed the potential speed of transition of domains between the vulnerability quadrants. The framework informs appropriate conservation actions to mitigate climate change impacts on biodiversity. The study explicitly links floristic pattern and climate variability and provides useful insights to facilitate conservation planning for climate change.
Jewitt, Debbie; Erasmus, Barend F.N.; Goodman, Peter S.; O'Connor, Timothy G.; Hargrove, William W.; Maddalena, Damian M.; Witkowski, Ed. T.F. 2015. Climate-induced change of environmentally defined floristic domains: A conservation based vulnerability framework. Applied Geography. 63: 33-42. 10 p. http://dx.doi.org/10.1016/j.apgeog.2015.06.004