Learning from humanitarian needs for better anticipatory action in a changing climate

Abstract:

The forced displacement of individuals and communities as a result of flood, drought and other extreme weather events, and the impact of human-induced climate change, has been described as one of the greatest humanitarian challenges of the 21st Century. Within the discussions on the humanitarian dimension of climate-related displacement, there has been increasing recognition of the specific importance of anticipatory initiatives. Practitioners like the Red Cross Red Crescent Climate Centre, for instance, have been starting to apply tests of forecast-based financing to inform short-term humanitarian assistance based on disaster warnings from scientific forecasts. We take this credo, anticipation instead of reaction, as a motivation to explore forecasting possibilities of humanitarian needs such as food, water and shelter for internally displaced people in Somalia.

Applying panel econometric methods to monthly within-country priority need observations together with weather data, we will quantify the impacts of extreme weather on priority needs in recent years. We will quantify which types of priority needs are prevalent during the past and ongoing drought, flood or compound extreme weather conditions, and predict near-future needs to provide urgent response before compound extreme and climate-related displacement cascade. We propose a logistic regression model, specifically designed for analyzing categorical response variables such as needs. We will train the model using observed data and the results can then be interpreted to understand the effect of climate on the probability of different needs across regions. More importantly, these log-linear models will be used to make early predictions to draw early action on climate-related forced displacement.

The research will provide quantitative results on the strength of weather impacts on priority needs for internally displaced people in Somalia. Our research findings will inform short-term, early action humanitarian assistance based on risk-informed forecasts and thus, help to reduce or even prevent additional displacement in this part of the world.