Use of Remote Sensing (MODIS) Data and Rainfall to Estimate Forage Production in Arid Rangeland

Abstract

This study investigated the use of remote sensing data from the Moderate Resolution Imaging Spectroradiometer (MODIS) and rainfall data to estimate forage production in the arid rangelands of the Richtersveld National Park, South Africa. The research aimed to assess the spatial and temporal variability in forage production across five vegetation types and three landscape units, and to examine the relationships between the fraction of photosynthetically active radiation (fPAR), rainfall, and biomass production. Field measurements of above-ground biomass were collected during the peak growing season in 2007 and compared with corresponding MODIS fPAR data. The results revealed significant spatial heterogeneity in forage production, with higher biomass observed in the Central Richtersveld Mountain and Northern Richtersveld Scorpionstailveld vegetation types, and in the mountain landscape unit. Leaf and stem succulents contributed the most to the available forage, while grasses and forbs dominated in the desert vegetation types. A strong positive linear relationship was found between MODIS fPAR and field biomass measurements, indicating the potential of using remote sensing data as a reliable proxy for forage production. Rainfall emerged as a key driver of vegetation dynamics, with both fPAR and biomass showing strong positive correlations with precipitation. The study highlights the importance of understanding the spatial and temporal variability in forage resources for effective rangeland management and conservation planning in arid environments. The findings suggest that the integration of remote sensing data and rainfall records can provide valuable insights into the dynamics of arid rangelands and support the development of adaptive management strategies in the face of increasing climate variability and land use pressures.