High-resolution Downscaling of Disposable Income in Europe using Open-source Data

Poverty maps have been extensively used for identifying populations vulnerable to global changes. The frequency and intensity of extreme events are likely to increase in coming years as a result of climate change. In this context, several studies have hypothesized that the economic and social impact of extreme events depends on income. However, to rigorously test this hypothesis, it is necessary to have income data on a fine spatial scale, compatible with the analysis of extreme climatic events. In order to produce reliable high-resolution income data, we have developed an innovative machine learning framework, based on random forests, that we applied to produce a 1 km-gridded dataset of disposable income for 2015 in Europe. This dataset was generated by downscaling disposable income data available for more than 120,000 administrative units. Our learning framework showed high accuracy levels, and outperformed other existing approaches used in the literature for downscaling income. Using SHAP values, we explored the contribution of the model input factors to income predictions and found that, in addition to geographic inputs (country, latitude, longitude), distance to public transport or nighttime light intensity were key drivers of income predictions. Finally, we illustrated how this new dataset can help identifying poverty areas in Europe. More broadly, this dataset offers an opportunity to explore the relationships between economic inequality and environmental degradation in health, adaptation or urban planning sectors. It can also facilitate the development of future income maps that align with the Shared Socioeconomic Pathways, and ultimately enable the assessment of future climate risks.