Choose an application

• Reference Manager
• EndNote
• RefWorks (Direct export to RefWorks)

The Katari basin is a catchment in the La Paz department in the West of Bolivia. Due to previous mining activity and rapid urban growth during the past decades, there are severe contamination problems. Also, the basin discharges in Cohana bay of Lake Titicaca, which is heavily impacted and even underwent a severe anoxic event in 2015. There are four main zones of anthropogenic influence in the Katari basin, namely a mining area, the urban area centered around El Alto, the agricultural zone and finally Cohana bay. A SWAT+ model can assist in understanding the contamination flows and occurring mechanisms. In addition, the effect and efficiency of multiple management practices can be evaluated with this tool. This study aims at inspecting the applicability of SWAT+ for representing the Katari basin and erosion processes in the catchment. Although all necessary inputs were available, the analysis has constrained detail due to the few observations. A monthly streamflow calibration and validation was carried out from respectively 2004-2010 and 2011-2015. Sensitivity analysis and automatic calibration were carried out with SWAT+ Toolbox. Goodness-of-fit between simulated and observed streamflow was appraised with NSE and PBIAS statistics. Calibration NSE of 0.872 and PBIAS of 15.576 were obtained, while validation showed a NSE of 0.595 and PBIAS of 0.586. Overall, these statistics indicate a good model and the inspected water balance was satisfactory. Sediment transport and erosion were assessed without executing any further calibration. Firstly, total suspended sediments (TSS) were analysed. A monthly analysis illustrates that the seasonal hydrological regime of the catchment is clearly reflected in the simulated sediment flows. However, an obvious disparity in observed and simulated values, where the model strongly overestimates observations, demonstrates the model is inadequate for obtaining quantitative sediment-related results. Furthermore, a visualisation of the sediment yield throughout the study area illustrated that the regions at the (north)eastern side of the catchment are prone to severe erosion, and mitigation practices would thus be most worthwhile in these areas. Although some interesting results were obtained, the acquisition of more experimental data such as point source contamination and discharge series is central for attaining a model which can be used in water resources management and mitigation of contamination. Further elaboration of the model such as the incorporation of irrigation and agricultural practices, calibration of other parameters and management practices could surely provide an added value to the model.