Oklahoma & Texas Agriculture (2024 Spring)
Team: Ashley Bañuelos (Project Lead), Alyssa Spencer, Joseph Dale, Yezzen Khazindar
Summary: Remote sensing offers valuable insights into the impacts of rangeland management practices and the enhancements of soil biogeochemical models. This collaborative project partners with the Noble Research Institute, US Department of Agriculture Agricultural Research Service, and Colorado State University to understand the intricate relationship between pasture and rangeland management decisions and ecosystem health. We conducted a comprehensive evaluation of the MOD17, Rangeland Analysis Platform (RAP), Robinson Landsat, and Robinson MODIS net primary production (NPP) models derived from NASA Earth observations across 2001 to 2019. This evaluation aimed to assess the variability of the NPP products at numerous spatial and temporal scales for select ranches in southern Oklahoma and northern Texas. We found that MOD17 and Robinson Landsat values were most similar across time and ranches. When we evaluated the NPP models by vegetation types, we found that some vegetations resulted in considerably different values between the NPP models. Furthermore, our team validated the RAP biomass data against field biomass data from Noble Research Institute from 2022 and 2023 to determine its accuracy. We found the temporal variability in R2 values between 2022 and 2023 underscores the importance of considering temporal dynamics when assessing the accuracy and reliability of NPP models in rangeland ecosystems. This study advances our understanding of rangeland ecosystem productivity applications within Oklahoma and Texas and informs the public about the potential of using remote sensing for evaluating rangeland health.