Estimativa da Descarga Líquida e da Concentração de Sedimentos em Suspensão por meio de Perfiladores Acústicos de Corrente por Doppler no Canal São Gonçalo - RS

Abstract

The São Gonçalo Channel (76 km long) is responsible for linking Mirim Lake to Patos Lake which compound the largest coastal lagoon system in Latin America and the largest shallow coastal environment in the world. However, this channel is unassisted by continuous monitoring systems about its water and suspended-sediment discharges, much of this because this channel has complex flow characteristics (inverse, stationary and bidirectional flows). In this way, this work aimed to develop models capable to estimate continuous data of discharge and Suspendes-sediment Concentration (CSS), by Acoustic Doppler Current Profilers (ADCPs). Two sections (GS1 and GS2) were monitored in the channel, one located near the mouth and the other near the outlet, 60 km apart. To estimate the water discharges, linear and multilinear regression models were developed, based on the Indexed Velocity Method - it consists of the relationship between height and area of the cross-section, and average velocity and indexed velocity (velocity at one or more specific points in the cross-section), considering the Integrated Velocity Cell (IVC) and Multi-cell profile (MCV) approaches. The IVC and IVC+MCV models presented results such as coefficient of determination r2 of 0.986, and 0.998, at GS1 respectively, and r2 of 0.986, and 0.995 at GS2. The results expressed the accuracy of the models for the different approaches. From the models, time series of discharge were developed for the sections, these capable of identifying moments of inversion and stagnation of the water flow, which matches the in situ discharge data used for calibration process. Also, models for estimating the SSC were developed from the intensity of the corrected return pulse (backscatter – resulting from the interaction of the pulse with suspended particles present in the water). The corrections considered the attenuations caused by the acoustic wave, by the properties of the water, by the properties and presence of the sediments. The results pointed to models with Good and Very Good performance in GS1 and GS2, respectively; models considering the mean corrected backscatter, the sediment absorption coefficient and the water temperature were the ones that presented the best performance in both monitored sections. The models allowed the construction of SSC time series for the sites, which matches the in situ samples used during calibration process. The joint use of these models provides to the São Gonçalo Channel accurate water discharge and suspended-sediment discharge data in a short timescale (15-minute intervals).