کد مقاله | کد نشریه | سال انتشار | مقاله انگلیسی | نسخه تمام متن |
---|---|---|---|---|
4577237 | 1630004 | 2012 | 14 صفحه PDF | دانلود رایگان |
![عکس صفحه اول مقاله: Predicting shallow water table depth at regional scale from rainfall and soil data Predicting shallow water table depth at regional scale from rainfall and soil data](/preview/png/4577237.png)
SummaryShallow water table levels can be predicted using several approaches, based either on climatic records, on field evidences based on soil morphology, or on the outputs of physically based soil–water balance models. In this study, data from a water table monitoring network were used to develop an empirical method for predicting water table depth in space and time from a soil map, cumulative rainfall data and long term water table characteristics, in order to optimize irrigation management at regional scale.Records of 160 piezometer sites available from 1997 to 2008 were analyzed to detect the overall temporal trend in water table depth in a relevant agricultural area of Northern Italy (extent of about 12,000 Km2). A clear trend in water table lowering is observed over the 12 years of available observations, with an average rate of 4.5 cm per year. Precipitation data for the years 2004–2008 and the records from 25 selected sites (N = 2299) for the same period were used to calibrate a predictive tool based on the assumption of a sinusoidal behavior of the water table depth with a bimodal yearly cycle. The proposed model uses as inputs the cumulative rainfall in the previous 365 days to that of prediction and the geostatistical estimates of the long-term (2005–2008) yearly average water table depth and oscillation. The model, calibrated for the whole data set and for subsets of sites grouped in terms of functional soil properties, was validated against the WT depths observed in 7 independent sites during 2005–2008 (N = 654). Validation results showed a mean absolute error of 37.1 cm for the general model and 33.1 cm for the model calibrated on the soil groups’ subsets. The accuracy of prediction was higher in years with precipitation amounts closer to the observed average. The general model was eventually extended to the whole study area, providing spatio-temporal maps of the water table depth. The results were further validated against the water table depths observed in all the piezometers at three dates in 2009: error statistics computed for the whole set of available readings at the three selected dates (N = 322), resulted in a mean absolute error of 37.7 cm. Although based on a limited amount of inputs, the accuracy of the proposed models was comparable to the results reported in the literature for more complex models.
► We model the spatio-temporal changes of shallow water table depth at regional scale.
► The model assumes a sinusoidal behavior of the water table with a bimodal yearly cycle.
► The predictive tool is based on cumulative rainfall data and long term water table characteristics.
► Integrating soil information will increase the accuracy of the model.
► Spatio-temporal maps of the water table depth will be used to optimize irrigation management.
Journal: Journal of Hydrology - Volumes 414–415, 11 January 2012, Pages 374–387