Jump to Main Content
Potential grazing indicator forbs for two mesic grasslands in South Africa
- Morris, Craig D., Scott-Shaw, Rob
- Ecological indicators 2019 v.107
- correspondence analysis, cost effectiveness, ecosystem services, environmental indicators, forage, forbs, grasses, grasslands, indicator species, indigenous species, livestock, monitoring, overgrazing, perennials, sandstone, species richness, South Africa
- Mesic grasslands in South Africa harbour a diverse community of herbaceous perennial forb species that outnumber grass species by up to 5–6:1, provide various ecosystem services as well as forage for livestock, and are sensitive to overgrazing. However, despite their prevalence and ecological importance, the potential of forb species as grazing indicators has not been evaluated nor are they routinely included in grassland condition assessments. We aimed, therefore, to identify a subset of potential grazing indicator forb species that had a consistent response to a range of grazing intensities historically applied at sites in two mesic grasslands, Midlands Mistbelt grassland (n = 123) and KwaZulu-Natal Sandstone Sourveld grassland (n = 55). Canonical correspondence analysis was used to assess forb composition changes in response to an ordinal index of grazing intensity to identify the most responsive abundant forbs in each grassland. Most (88–92%) forbs had a neutral or inconsistent response to grazing but 24 and 32 species in Mistbelt and Sandstone grassland, respectively, were sufficiently abundant and responsive to be considered potential key grazing indicators. Alternative, somewhat less reliable indicator species were also identified. Grazing-sensitive indicator forbs that declined under grazing (Decreaser species) mostly were erect with elevated growing points whereas grazing-resistant species (Increasers) were predominantly prostrate species. A weighted sum of the abundance of indicators and their grazing sensitivity weights based on the relative position of their centroids along the grazing gradient provides an index (1–100), the forb condition score (FCS), of the overall impact of grazing on the forbs at a site, with high FCS indicating minimal impact. The FCS adequately predicts overall indigenous forb species richness (square-root transformed) in both Mistbelt (r2 = 0.657) and Sandstone grassland (r2 = 0.795). The key grazing indicator forbs identified in our study could be used together with other assessment methods of grass condition and cover for rapid and cost-effective assessment and monitoring of the impact of grazing on the integrity of mesic grassland. Further testing and refinement of the proposed grazing indicator forbs and the FCS index is, however, required.