Natural or anthropogenic chemical compounds of different origins often accumulate in estuarine regions. These compounds may alter the water quality. Therefore, It is important to constantly monitor the quality of estuarine regions. A combination of remote sensing and traditional sampling can lead to a better monitoring program for water quality parameters. The objective of this work is to assess the spatiotemporal variability of the physicochemical properties of water in the lower region of the Mearim River and estimate water quality parameters via remote sensing. Samples were collected at 16 points, from Baixo Arari to the mouth of the watershed, using a multiparameter meter and Landsat 8 satellite images. The physicochemical parameters of the water had high salinity levels, between 2.30 and 20.10 parts per trillion; a high total dissolved solids content, between 2.77 and 19.70 g/L; and minimum dissolved oxygen values. Estimating the physicochemical properties of the water via remote sensing proved feasible, particularly in the dry season when there is less cloud cover.
31 July 2022 in Revista Ibero-Americana de Ciências Ambientais
Urban noise affects human health and impacts fauna and flora. Mapping noise levels may contribute to the development of noise mitigation strategies. This paper proposes a noise estimation method for the construction of an acoustic map in urban spaces using satellite images. The method was tested using data from the city of São Luís (MA), Brazil. For one year, noise was measured monthly at 25 points, with different characteristics of noise sources related to types of urban landscapes and rural areas. The month was randomly chosen for data acquisition from the noise monitoring period to estimate an acoustic map using remote sensing data from Landsat 8 satellite and vegetation indices, such as the Normalized Difference Vegetation Index (NDVI). A linear regression model and a second-order polynomial regression function were used to estimate the noise from spectral remote sensing data. The NDVI was the best predictor variable (r2 = 0.8554, p < 0.05) considered to be satisfactory, as the mean square error was 3.8 dB. The tested method and proceedings have been shown to be appropriate for monitoring noise in the city and constructing the acoustic map. Therefore, the results demonstrate a good relation between noise and NDVI allowing to use the regression model to construct acoustic maps using free and easily accessible data, and it has the potential to be used as a tool for planning noise control strategies.