Plant biomass and stress responses
Water stress and soil sterilisation treatments significantly reduced the total T. triandra biomass recorded compared to control (no-stress) and live soil conditions (both p<0.001; Figure 1b; Table S2). The water stress-sterilisation interaction was significant, with the most notable difference being increased biomass in the live control (no-stress) soil treatment (p<0.001; Table S2; Figure S6). Plants in the live high aridity soils had greater biomass than those in live low aridity soils (p<0.001; Figure 1b; Table S2), and the soil aridity-sterilisation interaction was also significant, and showed plant biomass was higher under high aridity soil conditions (p=0.022; Table S2; Figure S6).
The aboveground biomass of T. triandra plants was lower in the water stress (p<0.001; Figure 2c; Table S2) and sterilisation treatments (p<0.001; Figure 2c; Table S2). A water stress-sterilisation interaction was also present, where aboveground plant biomass was greater under live soil, with control water conditions, while unaffected by sterile soil conditions under both water availabilities (p<0.001; Table S2; Figure S7). Like total biomass, we found higher T. triandra aboveground biomass for plants grown under high compared to low aridity soil conditions (p<0.001; Figure 2c; Table S2). We also found significant interactions between soil aridity-sterilisation (p=0.032; Figure S7; Table S2), soil aridity-water stress (p<0.001; Table S2; Figure S7), and soil aridity-sterilisation-water stress (p=0.046; Table S2; Figure S7). Here, there was a stronger increase in aboveground biomass in the live high aridity soils compared to the sterile high aridity soils. We also found that the aboveground biomass increase was greater between the water stress and the control treatments in high aridity soils than in the low aridity soils (Figure S7).
We found that belowground biomass decreased when under water stress (p<0.001; Figure 2d; Table S2) and sterilisation treatments (p<0.001; Figure 2c; Figure 2d; Table S2). High aridity soils also led to increased belowground biomass than low aridity soils (p<0.001; Figure 2d; Table S2). Belowground biomass was also affected by a water stress-sterilisation interaction (p= 0.014; Figure 2d; Table S2; Figure S8). Here, sterilisation reduced belowground biomass more under low aridity soil conditions than under high aridity soil conditions.
There was no effect of water stress on root-mass fraction. However, sterilisation of low aridity soils increased the root-mass fractions (p=0.003; Figure 2e; Table S2) and sterilisation of high aridity soils reduced the root-mass fraction (p=0.002; Figure 2e; Table S2).
All plant soil feedback ratios were positive in each treatment, though we found significantly higher plant soil feedback ratios in the low aridity soils compared to high aridity soils for total, aboveground, and belowground biomass and root mass fractions (see Table S3; Figure S10). The elevated plant soil feedback ratios in low aridity soils appear to be driven by the very low biomass outcomes when these soils were sterilised (Figure 1b-d). In the low aridity soils, the plant soil feedback ratios were higher in the control treatments compared to water stress treatments for total, aboveground, and belowground biomass (Table S3; Figure S10a-c).