Fig. 4: Structural equation models (SEMs) for the direct and indirect effects on the peak of the growing season (POS). | Communications Earth & Environment

Fig. 4: Structural equation models (SEMs) for the direct and indirect effects on the peak of the growing season (POS).

From: Increased early-season productivity drives earlier peak of vegetation photosynthesis across the Northern Hemisphere

Fig. 4

A–D The results for FluxSat data (A), VPM data (B), CSIF data (C), and Flux data (D). Statistical parameters of the SEMs are shown for each SEM. The solid arrows in the SEMs indicate significant correlations with P < 0.05, while dotted arrows represent nonsignificant correlations with P > 0.05. The error bars in the histograms represent 95% confidence intervals. EGS and LPS represent the early growing season and late peak season, respectively. The indirect effects represent the effects of the SOS, EGS radiation (EGS-Rad), EGS temperature (EGS-Tem), EGS precipitation (EGS-Pre), and EGS soil moisture (EGS-SM) on POS via EGS gross primary productivity (GPP), and the effects of EGS factors, and LPS radiation (LPS-Rad), LPS temperature (LPS-Tem), and LPS precipitation (LPS-Pre) on POS via LPS soil moisture (LPS-SM), respectively. Time-series data for each variable from each pixel/site of each dataset were standardized using Z-scores.

Back to article page