Predicting climate change impacts on ecosystems is necessary for informing and assessing policy decisions. This is particularly true in much of Africa, a continent with a fast-growing human population that will drive rapid environmental changes in the coming decades. However, ecosystem change can be challenging to forecast because of a lack of available ecological data, and because individual processes will respond differently to different pressures. We solve these problems by using a mechanistic general ecosystem model, the Madingley model, which allows us to predict ecosystem change through multiple, individual-level processes. Mechanistic models are ideal for predicting ecosystem change as they require less data than statistical models, and they allow emergent large-scale properties to be forecast from interactions between and among individuals and their environment. The Madingley model has previously been shown to capture observed properties of individual organisms and broad structures of ecosystems reasonably well under environmental conditions with no human impacts. We have updated the model to explicitly include human impacts in order to predict ecosystem structure under different climate scenarios. We develop a modelling framework to investigate the individual and combined contributions of three climate-dependent ecological processes to overall ecosystem change: net primary productivity (NPP), metabolism, and predator-prey interactions.
Results/Conclusions
We show that mechanistic models are a valuable tool for predicting future ecosystem structure and function in data-poor regions, such as Africa. Using the updated Madingley model we can identify likely mechanisms for drivers of change under different climate scenarios at the ecosystem level. Preliminary modelling results indicate that altered predator-prey interactions are likely to be the largest driver of shifts in ecosystem structure and function due to climate change. In addition, our modelling framework allows for testing the resilience of different African regions to future climate scenarios, and identifying regions where policy decisions should be focused.