FIGURE LEGENDS
Figure 1 . Location of the five study sites.
Figure 2 .Box plots of twig water potential in Haloxylon ammodendron across five sites differing in depth-to-groundwater (DGW) at early (EGS) and late (LGS) growing season. The global assessment of a statistical difference among the sites with different DGW was conducted using ANOVA. Different letters represent significant differences among different DGWs as assessed using Tukey’s HSD test (P < 0.05).
Figure 3 .Twig water potential in Haloxylon ammodendronbetween the early (EGS) and late growing season (LGS) with the same depth-to-groundwater (DGW). To compare the difference between twig water potential in the EGS and LGS at every site with contrasting DGW, independent-samples t -tests were used.
Figure 4. Relationships between twig water potential and depth-to-groundwater (DGW) of Haloxylon ammodendron in early growing season (EGS; a, c and e) and late growing season (LGS; b, d and f). Blue lines indicate linear trends of DGW changes of turgor pressure, and shallow blue bands represent 95% confidence intervals.
Figure 5. Twig water relation traits in Haloxylon ammodendron across five sites differing in depth-to-groundwater (DGW) at early (EGS; a and c) and late (LGS; b and d) growing season. The global assessment of a statistical difference among the sites with different DGW was conducted by ANOVA. Different letters represent significant differences among the different DGWs as assessed using Tukey’s HSD test (P < 0.05).
Figure 6. Relationships between twig P-V parameters and depth-to-groundwater (DGW) of Haloxylon ammodendron in the early growing season (EGS; a and c) and late growing season (LGS; b and d). Values are means ± SEs (n = 5).
Figure 7. Seasonal changes in leaf area per branch (a) and growth rate of branch area (b) of Haloxylon ammodendron at five sites with different depth-to-groundwater (DGW). Error bars represent standard errors of means. Asterisks (*) indicate significant differences among the five sites with different DGWs on the same trial day or during the same growth period at P = 0.05.
Figure 8. Relationship between growth rates, Huber values, and depth-to-groundwater (DGW) of Haloxylon ammodendron during the early growing season (EGS; a and c) and the late growing season (LGS; b and d). Blue lines indicate linear trends of DGW changes of growth rate and HV, and shallow blue bands represent 95% confidence intervals.
Figure 9 . Relationship between soluble sugars and full turgor (π 100; a and b), betweenπ 100 and growth rate (c and d), and between growth rate and Huber value (e and f) of Haloxylon ammodendronduring the early growing season (EGS; a, c and e) and the late growing season (LGS; b, d and f). Values are means ± SEs (n = 5–10). Blue lines indicate linear trends between soluble sugar and full turgor, between full turgor and growth rate, and between growth rate and Huber Value. Dashed blue lines represent 95% confidence intervals of linear trendlines.