VÁZQUEZ AMABILE, GABRIEL G., BOSCH, NICOLAS, RICCA, ALEJANDRA P., ROJAS, DANTE E., ORTIZ DE ZÁRATE, MARÍA L., LASCOMBES, JOAQUÍN, FEIGUÍN, MARÍA F., and CRISTOS, DIEGO
Groundwater table depth and nitrate (NO3-N) concentration were studied in agricultural fields, in western Buenos Aires Province, Argentina. DRAINMOD, a process-based and field-scale hydrological model was calibrated and validated for studying groundwater table depth control strategies, at 12 farms located within a radius of 100 km, using records from 2004 to 2012. Pedotransfer functions were evaluated and applied to setup model inputs. A second set of observation wells were installed in 2011 to take monthly groundwater samples for studying NO3-N concentration under different crops, at three landscape positions (upper, middle and lower hill) at seven farms. Model performance was evaluated in order to be used as a decision support system to estimate the impact of potential management, land use and climate scenarios, on groundwater table dynamics. Likewise, groundwater control strategies were analyzed for agricultural sandy soils of Western Pampa, combining tile drains, crop rotations and pastures, on the landscape. All observations considered, DRAINMOD predicted daily groundwater table depth with an average Nash-Sutcliffe efficiency (NSE) of 0.88, and an average RMSE of 25 cm. At validation, the model presented an average RMSE of 39 cm and NSE of 0.73. Drainmod was also used to estimate groundwater table changes under different land use changes, As for NO3-N concentration, a significant effect of the landscape position and water table depth was observed. Considering 10 mg L-1 NO3-N as a standard limit, 52% of the observations exceeded this value, regardless N fertilization rates and mostly related to unusual precipitations events during winter 2012. Results showed that a significant proportion of leached nitrate would came from the natural process of mineralization of soil organic matter. [ABSTRACT FROM AUTHOR]