资源环境科技发展态势分析平台

Climate change has focused attention on the effects of changing temperature, particularly the effect on human health. Thus, robust and accurate spatially and temporally resolved air temperature (T-a) data are of particular importance in the field of epidemiology and public health. However, most health studies to date have matched people to the nearest monitor. In this study, we aimed to develop a robust satellite-based spatio-temporally resolved T-a estimation model across the complex geo-climatic regions of France resulting in daily high-resolution 1 km predicted air temperature (T-ap) estimations. We use a daily calibration approach using a series of processes to generate daily T-ap for every day across the entire study area and period. First, we start by calibrating MODIS (Moderate Resolution Imaging Spectroradiometer) satellite-gridded surface temperature (T-s) data against T-a collected within 1 km of the T-s centroid. The calibration stage adjusted for spatio-temporal predictors, as done in environmental exposure assessment methods such as land use regressions. Second, to estimate T-ap when no T-s data are available we fit a second model which uses the association of predicted grid cells T-ap values (based on satellite T-s) with surrounding T-a monitors and the association with values in neighbouring grid cells. Out-of-sample tenfold cross-validation was used to quantify the accuracy of our predictions. Our model performance was excellent for both days with available T-s and days without T-s observations (overall mean out-of-sample R-2 = 0.95 for both stages). In conclusion, we demonstrate how T-s can be used reliably to predict daily T-ap at high-resolution across France for use in studies looking at the effects of fine resolution Ta exposure on various health outcomes.