Cascading environmental, phenological, and physiological influences on spatial variability of annual gross primary productivity in the Northern Hemisphere

Accurately assessing spatiotemporal variations in terrestrial gross primary productivity (GPP) is crucial for understanding the interactions between the terrestrial carbon cycle and climate change. Environmental factors influence annual GPP (AGPP) either directly or indirectly through plant phenology and physiology. However, it remains unclear how environment, plant phenology, and physiology interact to influence the spatial patterns of global AGPP. In this study, we analyzed the geographic patterns and primary controls of the phenological and physiological properties of GPP using 827 site-years of eddy covariance data from 101 sites in the Northern Hemisphere. Specifically, the cascading relationships among environmental, phenological, and physiological factors that contribute to the spatial patterns of AGPP were tested. While the majority of ecosystems across different biomes displayed unimodal GPP seasonal patterns, significant geographical variations were observed in their phenological and physiological properties. The growing season length (GSL) decreased with increasing latitude (P < 0.001), while the maximum photosynthetic capacity (GPP_max) increased from 20°N to 70°N (P < 0.001). The foremost drivers of the spatial variation in the start date of the growing season (SGS), the end date of the growing season (EGS), and GPP_max were winter air temperature, summer precipitation, and spring solar radiation, respectively, which in turn influenced the spatial variation in AGPP. The cascade effects (0.95) of environmental factors on AGPP were larger than the direct effects (0.28). The cascading relationships among environmental factors, SGS, EGS, and GPP_max explained 93 % of the spatial pattern in AGPP. GPP_max exerted the strongest direct influence (0.60) on AGPP, followed by SGS (0.33). Environmental factors influenced the spatial variability of AGPP through cascading effects mediated by plant phenology and physiology. These findings not only provide fundamental parameters for model validation but also enhance our understanding of the intricate environmental and biotic controls governing the spatial pattern of AGPP.