Lag in catchment vegetation response to water availability and atmospheric dryness

Catchment water availability for vegetation use (i.e., catchment wetness) and atmospheric water demand (i.e., vapor pressure deficit, VPD) are two of the major abiotic factors that control the intra-annual variability of catchment vegetation carbon uptake (i.e., GPP). This study analyzes 380 catchments distributed across the contagious US to explore the causality and interconnectedness between these two factors and catchment vegetation productivity. We use indices to represent seasonal climatic, hydrologic, and vegetation characteristics: Horton Index (HI), ecological aridity index (EAI), evaporative fraction index (EFI), and carbon uptake efficiency (CUE). Further, we employ statistical methods, including circularity statistics, spearman's correlation, Granger's causality, and PCMCI+, to depict connections between catchment wetness, atmospheric dryness, and vegetation carbon uptake. Our results indicate that catchment water supply-productivity and water demand-productivity cause-effect relations occur within a maximum span of two months (i.e., ±1 month from GPP). The annual scale relationships of these variables are more likely driven by a few dominant months. Moreover, attributed to the lag, hysteresis exists between GPP and catchment wetness and between GPP and VPD. The narrowest hysteresis develops in dry catchments (i.e., HI→1, EFI→1, and CUE have low intra-annual variability), and the wide hysteresis develops in catchments where HI and EFI have strong intra-annual variability, and their seasonal patterns are not in phase. For catchments that are not permanently under water-limited or energy-limited conditions, vegetation is under hydrologic stress (i.e., high HI) during the peak growing period. GPP is at its highest in this period, and CUE is out of phase with HI and in phase with EFI. These findings support the need for developing a direct functional framework between catchment water supply, atmospheric demand, and vegetation productivity. Such a framework can help us track normal and extreme hydrologic and climatic signals' effect on catchment vegetation and vice versa.