Theme 1-1: Rangeland/Grassland Ecology--Oral Sessions
Description
Being able to confidently predict net primary production within intensively used rangeland systems has become the goal of many rangeland scientists. The communally-managed rangelands of Africa, with their very high livestock numbers, represent such a challenge, as above-ground biomass is most often depleted before it has had chance to accumulate. In order to provide more accurate estimates of NPP for these rangelands, we explored the use of direct measurements of canopy cover, standing biomass and leaf area index, against standard vegetation indices from the Sentinel 2 sensors. Following intensive field surveys, we developed regression models of the relationships between Sentinel 2 NDVI and these three bio-physical attributes. Repeated measurements along an NDVI gradient revealed a good relationship between NDVI and LAI, which was used to predict the ET across landscapes under communal tenure arrangements. Although un-improved grasslands and abandoned cultivated lands are the dominant land cover classes in the study area, woody invasive alien plants (IAPs) along seep-lines have increased in spatial extent since their introduction in the middle of the last century. These IAPs are now responsible for most of the catchment ET in these highly modified ecosystems. Grass biomass production for the upper reaches of the Umzimvubu catchment was calculated using the relationship between Sentinel 2 NDVI and grass canopy cover.
Included in
Modelling Catchment-Scale Evapotranspiration and Net Primary Production in Sub-Humid African Grasslands: Towards Understanding Carbon and Water Trade-Offs in Communal Rangelands
Being able to confidently predict net primary production within intensively used rangeland systems has become the goal of many rangeland scientists. The communally-managed rangelands of Africa, with their very high livestock numbers, represent such a challenge, as above-ground biomass is most often depleted before it has had chance to accumulate. In order to provide more accurate estimates of NPP for these rangelands, we explored the use of direct measurements of canopy cover, standing biomass and leaf area index, against standard vegetation indices from the Sentinel 2 sensors. Following intensive field surveys, we developed regression models of the relationships between Sentinel 2 NDVI and these three bio-physical attributes. Repeated measurements along an NDVI gradient revealed a good relationship between NDVI and LAI, which was used to predict the ET across landscapes under communal tenure arrangements. Although un-improved grasslands and abandoned cultivated lands are the dominant land cover classes in the study area, woody invasive alien plants (IAPs) along seep-lines have increased in spatial extent since their introduction in the middle of the last century. These IAPs are now responsible for most of the catchment ET in these highly modified ecosystems. Grass biomass production for the upper reaches of the Umzimvubu catchment was calculated using the relationship between Sentinel 2 NDVI and grass canopy cover.
