Assessment of the soil water content in the Pampas region using SWAT

- SWAT model was used to estimate soil water content in an ungauged Argentinian basin.
- The study area is situated in the main agricultural activity zone of the Pampas region.
- The reliability of modelled soil water content was tested using NDVI and SPI indices.
- The methodology could be replicated in other similar watershed of the Pampas region.
- The results could have important implications for drought management at basin level.

The Pampas region has recently experienced an expansion of agriculture towards more fragile environments, which has been associated with an increase in the frequency of droughts affecting the whole region. In the present study, we obtained a long-term record of soil water content (SWC) using a Soil and Water Assessment Tool (SWAT) model, for the first time in Argentina. The reliability of this model was contrasted with the temporal variation of the Standard Precipitation Index (SPI) and the Normalized Difference Vegetation Index (NDVI) to characterise episodes of drought. We also estimated the correlation between SWC anomaly (aSWC) and SPI, as well as the correlation between NDVI anomaly (aNDVI) and SPI. The model performance was reasonably satisfactory. The model calibration showed determination coefficient (R2) and Nash-Sutcliffe coefficient (NS) values of 0.70 and 0.59, respectively, and the model validation showed R2 and NS values of 0.77 and 0.75, respectively. The aNDVI showed a relatively low correlation with aSWC (0.26 ≤ r ≤ 0.45). In contrast, the SPI presented significantly positive correlations with aSWC (0.67 ≤ r ≤ 0.83). This work showed that SWAT is a suitable tool to measure SWC in poorly gauged geographical areas such as the Pampas region. Additionally, our approach could be applied to other systems resembling that studied here, without any significant reduction in performance.