Multi-temporal remote sensing of ground cover reveals beneficial effects of soil carbon farming in a semi-arid landscape

Sustainable land management practices are a strategic tool for addressing land degradation processes that threaten agroecosystem services supply. Currently, carbon credit schemes are important promoters of the adoption of such practices, yet their effectiveness on providing services other than carbon sequestration is not frequently assessed. Particularly, vegetation cover may not only be a mediator of CO₂ fixation, but also act as protection against soil erosion and prevent water quality deterioration. The overarching aim was to develop generalizable methods to assess the effectiveness of sustainable land management practices for maintaining agroecosystem integrity. To achieve this, we assessed the effect of soil carbon sequestration practices on remotely sensed groundcover levels and its stability, and on its response to short-term antecedent accumulated rainfall. These methods were tested in the Cowra Trough, an agricultural region of semi-arid New South Wales, Australia. Time series statistics (mean and standard deviation) and non-parametric tests were used to analyse temporal change in remotely sensed groundcover on paddocks undergoing different land management change intensities. This was complemented with a regional scale analysis of the effect of land use to contextualize paddock-scale results. Moreover, sequential linear regressions of remotely sensed vegetation cover response to antecedent rainfall through a moving temporal window were employed to assess trends in this relationship. A significant effect of land management change was demonstrated: over 90% of the sites implementing sustainable practices had increased and more stable ground cover levels, and the same number (though not the same sites) decreased their ground cover dependence on rainfall. The size of the effect was not related to the intensity of management change implemented for soil carbon sequestration. Land use type proved to be an important spatiotemporal predictor of ground cover and its stability at the Cowra Trough scale with cropping performing worse than grazing systems. Notably, the implementation of carbon farming practices was found to have a more prevalent positive impact on ground cover than on soil carbon contents, suggesting that such practices may provide co-benefits even when no carbon sequestration occurred. This study advances the possibility of monitoring agroecosystem multifunctionality and the development of integrative ‘payment for ecosystem services’ schemes.