Retrieving Total Suspended Matter in Tropical Reservoirs within a Cascade System with Widely Differing Optical Properties

• Grant number: 17/17841-9
• Support Opportunities: Regular Research Grants - Publications - Scientific article
• Start date: October 01, 2017
• End date: March 31, 2018
• Field of knowledge: Physical Sciences and Mathematics - Geosciences
• Principal Investigator: Enner Herenio de Alcântara
• Grantee: Enner Herenio de Alcântara
• Host Institution: Instituto de Ciência e Tecnologia (ICT). Universidade Estadual Paulista (UNESP). Campus de São José dos Campos. São José dos Campos , SP, Brazil

Abstract

Total suspended matter (TSM) is an important indicator of water quality and has an adverse effect on light propagation in the water affecting the primary production. Moderate Resolution Imaging Spectroradiometer (MODIS) data have been widely used to develop numerous semi-empirical models all over the world. In this study, many of these models perform satisfactorily either in a single water body or in many similar turbid water environments, however, their performance drops when data from oligo-to-eutrophic environments are used to monitor all reservoirs simultaneously. Most of them tend to overestimate TSM at low concentration. Two separate model calibrations were used to monitor the spatio-temporal variability of TSM in two oligo-to-eutrophic reservoirs, Nova Avanhandava (Nav) and Barra Bonita (BB), situated in the cascade system of São Paulo State, Brazil. The exponential model using MODIS band 1 yielded the best accuracy for BB (RMSE = 3.66 mg/l) and MODIS band 1 based linear model showed the best performance for Nav (RMSE = 0.37 mg/l). The distribution of TSM in Nav showed comparatively low and homogenous concentrations for the entire reservoir but the magnitude doubled for a period between May-October. The same seasonal variability occurred in BB showing increased TSM trend in October compared to May, but the spatial variability was not homogeneous because of the runoff from Piracicaba River, which is the main contributor of TSM. Rainfall associated runoff and land use and land cover were the principal factors for TSM variability. (AU)