Environmental factors controlling vegetation attributes, soil nutrients and hydrolases in South Mediterranean arid grasslands

Abstract

Understanding environmental factors controlling vegetation and soil attributes is crucial for the prediction of future climate change impacts on ecosystem functioning. This study conducted at 10 sites in South Mediterranean grasslands along aridity gradient aims to determine the possible controlling factors (mean annual precipitation MAP, mean annual temperature MAT and soil pH) that may affect vegetation attributes (total plant PC and grass cover GC, plant width density PWD, number of plant patches NPP and plant interdistance PI), soil nutrient contents (organic carbon SOC, total nitrogen TN, and available phosphorus AP) and activities of hydrolase enzymes (β-glucosidase and phosphatase) using multiple linear regressions analysis and structural equation modeling.The current research reported a strong effect of soil pH and climate factors on vegetation attributes. P values obtained from multiple linear regressions are highly significant for PC (P = 0.0001), GC (P = 0.045), PWD (P = 0.0005), PI (P = 0.039) and NPP (P = 0.007). Soil pH exerted negative effects on all the measured variables. MAP and MAT affected positively and negatively PC, GC and PWD, respectively. Decreased MAP and increased MAT resulted in higher interdistance between plant patches (PI) and consequently the dominance of bare ground areas. NPP may increase with decreasing MAP and MAT suggesting higher frequency of facilitative interactions as resource (water, nutrients) limitation increases.SEM analysis revealed a negative total effect of soil pH on SOC (β = -0.49), TN (β = -0.69) and AP (β = -0.76) that can be related to its effect on the leaching of dissolved organic carbon and dissolved organic nitrogen, and on the solubility of soil phosphate. Our results reported also a negative effect of soil pH on the activities of the hydrolases β-glucosidase (β = -0.38) and phosphatase (β = -0.3). Climate factors influence the availability of soil nutrients and the activities of hydrolase enzymes. MAP exerted positive effects on SOC (β = 0.73), TN (β = 0.45) and the measured hydrolases (β = 0.57 for β-glucosidase and β = 0.66 for phosphatase). MAP exerted mainly indirect positive control on TN and on the activity β-glucosidase through its effect on SOC. However, MAT is found not a limiting factor on SOC, TN and activities of hydrolases at the temperature range of 15–20 °C. On the contrary, AP was highly controlled by MAT (β = 0.57) but not MAP (β = -0.06). MAT exerted a positive effect on AP probably by increasing soil secondary mineral phosphorus. Based on our findings, we suggest that when MAP is not limiting factor, MAT increases soil AP content. This positive effect can be attributed to the high sand content of the soils (>50%) in Tunisian arid grasslands.