Methodological sensitivities when identifying dryland degradation: case studies from two Australian regions

Quantifying dryland degradation is essential for effective land management. Remote sensing data provides continuous records of dryland vegetation dynamics, enabling the identification of degraded areas. However, uncertainties remain regarding the sensitivities introduced by different satellite vegetation products, varying spatial resolutions, and the potential spatial variability overlooked when using gridded climate data. This study evaluates these sensitivities using the Time Series Segmented Residual Trend (TSS-RESTREND) V0.3.2 method and offers recommendations for more robust dryland degradation assessment. Our results indicate that the low spatial variability in gridded rainfall products may introduce errors in the precipitation–vegetation relationship. Although GIMMS and MODIS NDVI generally align, differences in the timing of peak NDVI can affect degradation assessments, particularly those driven by climate. We recommend using MODIS NDVI when long-term data are not essential, as its higher spatial resolution effectively identified degradation areas covering 1.16% of Fowlers Gap and 4.26% of Boolcoomatta—areas that were undetected with coarser data. This highlights the importance of higher spatial resolution for detecting dryland degradation. In contrast, higher temporal resolution provided limited additional value, suggesting that monthly data are sufficient. Additionally, results from different vegetation indices were broadly consistent in degradation detection.