Modelagem da qualidade da água da sub-bacia do rio Siriri em Sergipe, utilizando o modelo QUAL-UFMG

Abstract

In view of the progress of water quality degradation processes in hydrographic basins due to disorderly growth, economic activities and natural processes, modeling presents itself as a fundamental alternative in the management of water resources. Mathematical models are able to foresee situations, simulate different scenarios, considering diffuse and punctual pollution sources. The present work aimed to model the water quality of the Siriri River Sub-Basin by applying the QUAL-UFMG mathematical model simulating future scenarios and seeking to evidence polluting sources to build a diagnosis of the basin. QUAL-UFMG is a model developed by Von Sperling based on the QUAL2E model, which allows a quick and simple simulation of water quality in rivers based on the following parameters: dissolved oxygen, biochemical oxygen demand, total nitrogen and its fractions, phosphorus and its fractions and thermotolerant coliforms. To this end, water quality variables were used for calibration, from 24 sampling campaigns covering the period from October 2014 to September 2016 and which were analyzed in two dry and rainy campaigns. In addition, flow data from two fluviometric stations provided by the National Water Agency (ANA) for the same period were also used. The modeled parameters were: dissolved oxygen, total phosphorus and thermotolerant coliforms. The simulation was made considering the situation of the basin between 2014 and 2016 with raw effluent being released directly into the watercourse, and three other scenarios: effluent treated with conventional primary treatment, effluent treated with conventional secondary treatment, and reduction of phosphorus concentration by 30% due to the application of agricultural techniques such as direct planting. Among the analyzed variables, values in anthropized areas were observed that indicate a greater degradation of water quality, especially in point 5, which receives a greater load of untreated effluent. Calibration performed manually showed satisfactory adjustments between measured and estimated data for the dry and rainy periods at all points. The kinetic coefficients used showed values that corroborate those found in the literature. As for the simulated scenarios, there was a increase for dissolved oxygen about 10% for the rainy season and 20% for the dry season, reaching values close to 7mg/l in one of the scenarios; for phosphorus, an attenuation of 27% of the charge was achieved, with values below 0.040 mg/l in almost all stretches; coliforms showed better results in the dry period, with decreases in organic load close to 30%. Results showed the need for the application of effluent treatment in the Siriri River sub-basin, both primary and secondary, which would cause significant reductions in portions of the polluting load currently present in the basin, in addition to influencing the conceptions of the river classification classes. As for the model, it was easy to handle and understand, with demonstrably satisfactory results.