Linking land–atmosphere processes and hydrological response under prolonged drought in a semi-arid basin

Abstract

Semi-arid river basins are highly sensitive to hydroclimatic variability, where rainfall seasonality and soil moisture conditions strongly influence vegetation dynamics, surface energy balance, and hydrological response. This study investigates seasonal eco-hydroclimatic variability in the Apodi–Mossoró River Basin (northeastern Brazil), focusing on rainfall–runoff coupling and land–atmosphere interactions under prolonged drought and regulated flow conditions. An integrated framework combined Moderate Resolution Imaging Spectroradiometer (MODIS)-derived Normalized Difference Vegetation Index (NDVI), Land Surface Water Index (LSWI), Land Surface Temperature Anomaly (LSTA), and albedo products, potential evapotranspiration (PET) from the TerraClimate reanalysis dataset, observed precipitation and streamflow records, and large-scale climate indices for the 2002–2023 period, together with hydrological modeling and wavelet analyses. Results revealed a strongly seasonal hydroclimatic regime, with precipitation concentrated between February and May (Seasonality Index = 0.908), while water-deficit conditions predominated throughout the year. Higher NDVI and LSWI values occurred during wetter periods, whereas spatial changes in LSTA and albedo were more evident during dry conditions suggesting altered land–atmosphere interactions under prolonged drought conditions. Trend analyses showed generally weak and heterogeneous long-term changes across the basin, while PET exhibited relatively low temporal variability throughout the study period. The Pettitt test identified a statistically significant shift in streamflow around 2013 despite the absence of a significant precipitation change point. Wavelet coherence analyses revealed non-stationary precipitation–streamflow coupling modulated by El Niño–Southern Oscillation and tropical Atlantic variability. The Soil Moisture Accounting Procedure (SMAP) model satisfactorily reproduced streamflow variability during calibration and validation, despite limitations under low-flow conditions. Streamflow dynamics were associated with prolonged drought and hydrological regulation effects. Land–atmosphere interactions revealed strong seasonal coupling among vegetation, moisture, and surface energy dynamics. The SMAP model reproduced streamflow variability under contrasting hydroclimatic conditions. Wavelet coherence revealed significant climate controls on rainfall–runoff coupling across multiple timescales. A significant post-2013 streamflow regime shift occurred without corresponding precipitation changes. Streamflow dynamics were associated with prolonged drought and hydrological regulation effects. Land–atmosphere interactions revealed strong seasonal coupling among vegetation, moisture, and surface energy dynamics. The SMAP model reproduced streamflow variability under contrasting hydroclimatic conditions. Wavelet coherence revealed significant climate controls on rainfall–runoff coupling across multiple timescales. A significant post-2013 streamflow regime shift occurred without corresponding precipitation changes.