Effects of aboveground biomass and soil moisture drought on VOD-based global isohydricity estimates

Ecosystem scale isohydricity derived from satellite vegetation optical depth (VOD) has been increasingly used to investigate the ecosystem responses to drought, yet the robustness of existing VOD-based metrics remains poorly constrained. In particular, seasonal variations in vegetation biomass and episodic drought events can dominate the seasonal canopy water dynamics represented by VOD, potentially masking the diurnal VOD variations that are used to derive ecosystem isohydricity. However, these effects have not been systematically evaluated by previous studies. In this study, we evaluate the robustness of two VOD-derived isohydricity metrics, σ and Rslope, using multiple passive microwave VOD datasets. Our results reveal that estimates of σ are strongly influenced by biomass dynamics, with higher seasonal variations in biomass consistently yielding higher σ values. σ also exhibits substantial differences in spatial patterns and global mean values across VOD datasets. In contrast, Rslope is largely independent of biomass and shows greater spatial consistency across VOD datasets and microwave frequencies. Explicitly accounting for uneven sampling across drought severity through binning alters both metrics. In particular, Rslope shows improved relationships with ecosystem isohydricity derived from eddy covariance measurements across both VOD datasets, whereas the improvements on σ are limited. Lastly, we derived a reliability map for VOD-based isohydricity estimates considering the agreements across dataset, isohydricity metric, and sampling sufficiency. Higher reliability was concentrated in seasonally dry regions with moderate biomass. Overall, these results demonstrate fundamental differences among VOD-based isohydricity metrics and highlight the importance of explicitly accounting for biomass dynamics, drought sampling, and dataset differences in large-scale assessments of ecosystem water use strategies.