Get reliable rangeland science

Inadequate environmental management in drylands can have serious consequences shifting ecosystems into alternate stable states where key ecosystem services are compromised. Open access Earth observation data can provide continuous and consistent measurements globally, thus providing valuable information on remote dryland locations. The most widely employed remote sensing indicators are spectral indices of vegetation greenness, such as the Normalized Difference Vegetation Index (NDVI) and the Enhanced Vegeta tion Index (EVI) and are not specifically fit for drylands where background soil has a strong influence and senescent vegetation often dominates. Methodological advances in the past decade now allow the quantification of vegetation fractional cover (VFC) into three categories: photosynthetic vegetation (PV), non-photosynthetic vegetation (NPV) and bare soil (BS). Although monitoring applications have been derived from this product, significant knowledge gaps remain regarding the relationships between vegetation cover fractions and descriptors of ecosystem condition. Our work addressed these gaps at regional and continental scale. At the regional scale we explored links between VFC time series statistics and field measurements of woodland ecosystem quality (soil health and vegetation structure complexity) and elastic net regression to assess combinations of fractional cover statistics for predicting ecosystem quality. We found that time series statistics were robust predictors of soil health and vegetation structure complexity. At the continental scale we demonstrated how vegetation structure components (i.e. woody and herbaceous cover) can be accurately predicted by VFC time series statistics (RMSE<14.75%). Moreover, we found links between field-measured vegetation structure complexity metrics (i.e., growth form and height class diversity) and different combinations of cover-fraction time-series statistics depending on vegetation type and climate.