Contrasting Temperature Sensitivity of Boreal Forest Productivity in North America and Eurasia

The seasonal amplitude of atmospheric carbon dioxide (CO₂) has increased by as much as 50% over the last 6 decades, suggesting changes to the Arctic-boreal carbon cycle. Some of this increase is due to increasing seasonality of net ecosystem exchange in boreal and arctic ecosystems, although the mechanisms are still uncertain. For example, it is unknown how much of the increase is due to increases in gross primary productivity (GPP) during the growing season. Such an increase could be attributable either to global-scale CO₂ fertilization or to regional climate impacts, which may vary across the boreal zone. In this study, we use a global, spatially, and temporally contiguous solar-induced chlorophyll fluorescence (CSIF) data product from 2001 to 2019 (0.05°, 4-day resolution) to identify spatial, interannual, and long-term sensitivities (linear slopes) of GPP to temperature across boreal forests in Eurasia and North America. Across all evergreen needleleaf regions, spatial sensitivities of CSIF to temperature are stronger in magnitude when compared to interannual sensitivities, suggesting that there are limitations when performing space for time substitutions. We find that Eurasian forests generally have the strongest spatial sensitivity to temperature. Eurasian deciduous needleleaf forests show the highest growing season mean interannual sensitivity to temperature variations. Although all regions show a positive spatial and interannual relationship between productivity and temperature, in western North America, long-term warming may be curbing productivity gains. Our results suggest that Eurasian and North American boreal forests may show divergent trends as climate continues to warm.