Análise remota das concentrações de clorofila- a e carbono orgânico total e estimativas das emissões naturais e antrópicas de nutrientes em reservatórios e bacias hidrográficas do semiárido brasileiro

Abstract

Among aquaculture activities, fish farming is the most frequent in reservoirs in northeastern Brazil, and also the one that most contributes to the deterioration of water quality of these reservoirs. Therefore, it is of utmost importance to develop techniques that can facilitate the monitoring of water quality conservation, not only with an emphasis on aquaculture activities, but also in maintaining water quality for human consumption and animal drink. In this context, the present work had as objectives: to characterize the contribution basins of the studied reservoirs, both from the morphometric point of view, as to the use and occupation; estimate nutrient inputs from natural and anthropogenic emissions; and remotely analyze the distribution of optically active limnological variables aimed at analyzing the water quality of these reservoirs. To this end, six collections of limnological data were carried out every two months, between October 2017 and August 2018, in two reservoirs in the semi-arid region of Rio Grande do Sul, Umari and Mendubim, and their respective contribution basins were characterized. The reservoirs have exoreic contribution basins, with a predominantly elongated shape, low tendency to flooding, drainage networks with low density, slope and flow speed, formed by permeable substrates; having as main activities agriculture and livestock. It is estimated that these reservoirs receive annual loads of nitrogen and phosphorus around 579 tons per year and 136 tons per year for Umari, respectively, and 329 tons per year and 106 tons per year for Umari. for Mendubim, the main natural sources of nitrogen and phosphorus being the physical-chemical denudation of the soil and atmospheric deposition, and anthropic to livestock and agriculture. The distribution of optically active limnological variables, Chlorophyll a (Chla) and Total organic carbon (TOC) were analyzed using images from the OLI / Landsat 8 sensor, for this purpose linear regression models were developed between combinations of bands B2 to B6 of the images Landsat and the chlorophyll and carbon concentrations estimated in the laboratory from the field collection, in order to recover concentration values of these variables from the spectral data of these images. According to the trophic state index, the reservoirs were Umari and Mendubim were classified, respectively, as mesotrophic (EIT = 49.74) and Eutrophic (EIT = 54.14). There was a high correlation between the spectral data of the LS8 / OLI sensor and the concentrations of Chl-a (R² = 0.74) and TOC (r² = 0.70). The recovery of the Chl-a and TOC concentrations estimated from spectral data proved to be efficient