How much does afforestation’s impact on local land surface temperature vary in space, in time, and during dry and hot extreme events?

Changing the properties of the land surface may be one of the most direct ways to modulate local (and possibly non-local) land-atmosphere interactions, which in turn is of great interest for designing proper land-based climate mitigation and adaptation strategies. When we change the type of vegetation across a landscape, the biophysical properties of that land surface will change, potentially altering both radiative and non-radiative fluxes. Land surface temperature (LST), as measured from remote sensing satellites, provides a useful diagnostic, integrating the effects of these changes in fluxes. When combined with space-for-time substitution approaches, it is possible to derive data-driven estimations of what a given land cover transition could lead to in terms of LST before the actual land cover change occurs. However, the interannual variability of such biophysical effects of land use and land cover change is still understudied, which is an important prerequisite to understand the role these effects may have in alleviating or aggravating the occurrence and impacts of extreme events. 

In this study we present a global analysis of potential afforestation on local afternoon clear-sky LST across the MODIS Aqua record (from 2002 until 2024). This allows us to explore the interannual variability of local increases in forest cover on local LST, which in turns helps us estimate the sensitivity of the effects of afforestation in a changing climate. By combining these results with a dedicated dataset identifying hot and dry extremes from ERA5, we further explore how the effect of afforestation on LST changes under extreme conditions, which the trees would be increasingly more susceptible to encounter once they reach maturity.

Additionally, we take the opportunity to present the processing pipeline that has been developed within the Open-Earth-Monitor cyberinfrastructure (OEMC) project to make such analysis possible and reproducible. This includes improvements to better handle local topographic effects and testing the capacity to run the entire pipeline within a Discrete Global Grid System (DGGS) framework that preserves area and neighbourhood properties within the space-for-time moving window. We expect that these tools will facilitate data integration and model evaluation, thereby assisting research in land-atmosphere interactions and climate extremes.