Velocity of climate change revisited: MATCHing spatial shifts into a continuous field

Describing the spatial velocity of climate change is essential to assess the ability of natural and human systems to follow its pace by adapting or migrating sufficiently fast. We propose a fully-determined approach, "MATCH", to calculate a realistic and continuous velocity field of any climate parameter, without the need for \emph{ad hoc} assumptions. We apply this approach to the displacement of isotherms predicted by regional and global climate models between 1950 and 2100 under the RCP 8.5 emission scenario, and show that it favors comparisons between models as well as the analysis of local or regional features. Furthermore, the local properties of the velocity field obtained using the MATCH method define a trajectory regularity index which offers a quantitative perspective on the discussion of climate sinks and sources.

/!\ WARNING: Please delete file fast_cubic_spline.c before running "python setup.py build_ext" in the fast-cubic-spline package, otherwise you will most certainly get errors during compilation. /!\