Fig. 1: Relative importance of daily maximum (T_max) and minimum temperatures (T_min) in interannual vapor pressure deficit (VPD) variations from 1980 to 2023.

a Spatial distribution of the trend in diurnal temperature range (DTR) over land areas. Inset shows boxplots of trends across observation stations (solid boxes) and ERA5-Land grid points (hollow boxes). Pie chart shows the percentage of land area with significantly (p < 0.05) positive (red), weak positive (light red), weak negative (light blue), and significantly negative (blue) trends, based on ERA5-Land data. b Spatial distribution of ridge regression (RR) coefficients of interannual VPD with respect to DTR, derived from the RR model defined in Eq. (2). c, e Spatial distribution of RR coefficients of interannual VPD with respect to T_max (c) and T_min (e), derived from the RR model defined in Eq. (3). Insets in b, c, and e show boxplots of RR coefficients across observation stations (solid boxes) and grid points (hollow boxes). Pie charts show the percentage of land area with positive (red), negative (blue), and non-significant (light grey) RR coefficients based on ERA5-Land data. “Non-significant” refers to cases where none of the coefficients in the RR model are statistically significant. Stations (1.07% in b; 1.29% in c, e) and areas (1.10% in b; 1.11% in c, e) with non-significant RR coefficients are masked in light grey and excluded from RR model-based analysis. d, f Relative importance of interannual T_max, T_min, and soil moisture (SM) in driving interannual VPD variability based on the Random Forest regression model defined in Eq. (3), evaluated across stations (d) and ERA5-Land grid points (f). All variables in the regressions are detrended and standardized annual averages. In all boxplots, the height of each box represents the interquartile, with the thick black line indicating the median, and the edges denoting the first and third quartiles. Whiskers extend to the 2.5th and 97.5th percentiles.