Soil depth and vegetation type influence ecosystem functions in urban greenspaces

Abstract

The contributions of soil and plant diversity to supporting ecosystem functions in urban greenspaces remains little known. Soil compaction and nutrient enrichment constitute major threats, and can give rise to the loss of ecosystem services in urban greenspaces. Our study focused on spontaneous vegetation in urban greenspaces in a Mediterranean city. We aimed to study the network correlations between the physical-chemical and biological properties of the soil at different depths along gradients of soil disturbance; and to analyse how vegetation and soil depth influence soil functioning. We focused our study on the following four spontaneous ruderal vegetation types: open vegetation on trampled soils, roadside vegetation, annual grasslands, and perennial forbs. They are widely distributed in Mediterranean urban greenspaces along soil compaction and nutrient enrichment gradients. We determined soil physical-chemical properties (organic matter, available nutrient content) and soil microbial activity relating to the main macro-nutrient cycles in the soil under each vegetation type at different soil depth (0–5, 5–20 cm). We used Spearman's bivariate correlations to study the relationships between soil variables at different soil depths by means of network analysis. We performed two-way ANOVAs to determine the influence of the plant-community type and soil depth on soil physical-chemical parameters and enzyme activity. We found that physical-chemical variables such as total organic carbon and bulk density represented the main drivers of soil functionality in the surface horizon (0–5 cm). Enzyme activity, however, associated with cycles of macronutrients such as arylamidase, arylsulfatase and phosphatase, had a greater influence on ecosystem functions in the subsurface horizon (5–20 cm). Regardless of the soil depth (0–5 cm, 5–20 cm), ANOVAs revealed a significant increase in bulk density between trampled soils (lowest values) and roadside vegetation (highest values); and in available phosphorus between annual grasslands (lowest scores) and perennial forbs (highest scores). Soils under annual grasslands and perennial forbs showed the greatest number of significant differences in soil between horizons. The highest number of significant differences were found between vegetation types in the surface soil horizon (0–5 cm) for soil organic matter and enzyme activities. This research therefore suggests that in urban Mediterranean greenspaces soil horizons are dissociated in terms of their soil parameter drivers, and that the surface horizon (0–5 cm) is more closely related to the response of the vegetation to soil disturbance gradients than the subsurface horizon (5–20 cm).