Keywords
swat, remote sensing, evapotranspiration, leaf area index, vit basin
Start Date
1-7-2012 12:00 AM
Abstract
Remote sensing has become an indispensable source of input data and parameters estimation for environmental modelling. In the presented study evapotranspiration (ET) and leaf area index (LAI) were derived from satellite images and used for the evaluation of a hydrological model. Such applications can help to identify simulation inaccuracies and optimize the performance in areas with limited data availability. Moreover daily observations can provide useful information on agricultural management measures, such as harvesting dates. The Soil and Water Assessment Tool (SWAT) was chosen to model the Vit River catchment (area 3200 km²), which is located in the North of Bulgaria. SWAT is an integrated river basin model that performs the calculations on the scale of hydrological response units (HRUs), which are generated by a unique combination of land use, soil type and land slope. This approach allows a high discretization of the entire basin and enables the plot of results in detailed maps. The main objectives were to evaluate the predictions for ET and LAI by visual comparison of the model output with remotely sensed data. The evaluation showed that estimates for ET tend to be slightly underestimated, while those for LAI were visibly overpredicted. Furthermore a very distinct zoning effect of the results into clusters with similar values was observed. Those zones were identified as areas to which SWAT had assigned the same climatic conditions. On the contrary the satellite images were clearly following the land-use pattern of the basin and were showing uniform values for the different types of vegetation. It was concluded that in SWAT, climate showed a considerably greater impact on the ET and LAI simulations than different land-use or vegetation types, which was an unexpected phenomenon that needs further investigation.
The use of satellite images for evaluating a SWAT model: Application on the Vit Basin, Bulgaria
Remote sensing has become an indispensable source of input data and parameters estimation for environmental modelling. In the presented study evapotranspiration (ET) and leaf area index (LAI) were derived from satellite images and used for the evaluation of a hydrological model. Such applications can help to identify simulation inaccuracies and optimize the performance in areas with limited data availability. Moreover daily observations can provide useful information on agricultural management measures, such as harvesting dates. The Soil and Water Assessment Tool (SWAT) was chosen to model the Vit River catchment (area 3200 km²), which is located in the North of Bulgaria. SWAT is an integrated river basin model that performs the calculations on the scale of hydrological response units (HRUs), which are generated by a unique combination of land use, soil type and land slope. This approach allows a high discretization of the entire basin and enables the plot of results in detailed maps. The main objectives were to evaluate the predictions for ET and LAI by visual comparison of the model output with remotely sensed data. The evaluation showed that estimates for ET tend to be slightly underestimated, while those for LAI were visibly overpredicted. Furthermore a very distinct zoning effect of the results into clusters with similar values was observed. Those zones were identified as areas to which SWAT had assigned the same climatic conditions. On the contrary the satellite images were clearly following the land-use pattern of the basin and were showing uniform values for the different types of vegetation. It was concluded that in SWAT, climate showed a considerably greater impact on the ET and LAI simulations than different land-use or vegetation types, which was an unexpected phenomenon that needs further investigation.
