Extended Data Fig. 8: Structural equation model for the direct and indirect effects of climate variations in preceding periods on the subsequent drought recovery.
From: Widespread spring phenology effects on drought recovery of Northern Hemisphere ecosystems
The direct effects of climate anomalies during drought year (including mean temperature, Tempdrought_year, total precipitation, Precipdrought_year, and mean vapour pressure deficit, VPDdrought_year), during the dormant period (including the mean temperature, Tempdormant, and snow water equivalent, SWEdormant), and during the spring of the subsequent year (including mean temperature, Tempsubsequent_year_spring, and total precipitation, Precipsubsequent_year_spring), on drought recovery within the subsequent growing season were considered. The indirect effects of Tempdrought_year, Precipdrought_year, VPDdrought_year, Tempdormant, and SWEdormant on the spring phenology of subsequent year and then the drought recovery was considered. The arrows indicate the hypothesized direction of causation, with positive and negative relationships in blue and red, respectively. Coloured lines represent significant (p < 0.05) relationships by the two-sided t-test. Line thickness is proportional to the strength of the relationship and to the standardized path coefficients adjacent to each arrow. We used the χ2 test (and associated p values), goodness-of-fit index (GFI), an adjusted goodness-of-fit index (AGFI), and the root mean square error of approximation (RMSEA), to evaluate the fit of the established structural equation model. The SEM model is with a χ2 = 0.002, RMSEA = 0.000, GFI=1.000, and AGFI=1.000.
