Improving the BACI method to account for asymmetrical changes in weather station solar masks using panoramic photography: a case-study of two urban forests in Paris.

To adapt to climate change, cities are implementing various urban cooling techniques to mitigate heat stress during heat waves. In Paris (France), several such projects have been carried out, including the use of cool pavements vegetation, or the renovation of schoolyards. The cooling efficiency of these solutions can be assessed following the BACI method. This method uses fixed onsite microclimatic measurements performed before and after site transformation at case and control locations. For heat mitigation studies, the analysis focuses on radiative days (clear sky and low wind).

When case and control stations are not placed in different sections of the same street, their solar masks can exhibit significant differences. In principle, the BACI method is designed to capture the effect of these pre-existing differences. However, a number of sites analyzed in Paris present statistically significant results corresponding to the creation (or destruction) of shade where no trees have been planted (uprooted) nor shading devices installed (removed). Indeed, non-uniform data sampling can bias the before or after characterization phases and lead to the detection of false-positives.

This communication proposes a method to account for such natural asymmetries in solar mask changes between paired stations by characterizing black globe solar masks precisely using panoramic photography. These are used to analyse the sun's trajectory at different dates of the year and identify potential false positives. Two newly-planted urban forests in Paris are used as a case study. In addition, the use of local time for field data analysis rather than solar time can also cause the detection of false positives as the position of the sun can vary by up to 6° for two symmetrical dates in spring and summer. The results and methodological limits are discussed.