Banca de DEFESA: Iara Resende da Fonseca

Uma banca de DEFESA de MESTRADO foi cadastrada pelo programa.
STUDENT : Iara Resende da Fonseca
DATE: 15/12/2023
TIME: 09:00
LOCAL: Sala de Informática do PTARH com Transmissão pela Plataforma MS Teams
TITLE:

Spatio-temporal behavior of chlorophyll-a concentration in water bodies of Federal District through remote sensing

KEY WORDS:

Chlorophyll-a, Landsat, Urban water supply reservoirs, Google Earth Engine.

PAGES: 105
BIG AREA: Engenharias
AREA: Engenharia Sanitária
SUBÁREA: Recursos Hídricos
SUMMARY:

Urban water supply reservoirs are complex lentic systems that require continuous monitoring to detect diffuse pollution interferences that could compromise water quality and pose risks to public health. Remote sensing stands out as a tool that can enable quick action by authorities in response to quality-related adverse situations due to its high temporal and spatial potential. However, continuous empirical studies are necessary to understand the environments and their relationships with the data obtained from orbital, terrestrial, or airborne sensors. In the case of the Federal District in Brazil, the proximity of urban centers to public water supply sources presents prominent risks of sudden changes in water quality. Despite this characteristic, such aquatic environments are challenging for remote sensing due to the low concentrations of limnological variables that have been observed in recent years. In this sense, the present study aims to propose a precise and accurate spatiotemporal estimation method for chlorophyll-a in the reservoirs of the Federal District, using a semi-empirical model based on the historical series of Landsat images. To achieve this, chlorophyll-a data collected in the field by CAESB (Water and Sanitation Company of the Federal District) and surface reflectance data from Landsat satellites from 1984 to 2023 were obtained through cloud processing in Google Earth Engine. The study sought to find a semi-empirical model for predicting chlorophyll-a concentrations for all reservoirs in the Federal District: Paranoá Lake, Descoberto, and Santa Maria. As a result, a model was found for the Landsat 5 data series from 1984 to 2011, using the ratio of blue and green bands in a satisfactory polynomial fit (R² = 0.69 and RMSE = 12.88 µg/L). Another model was developed for the OLI sensor data, corresponding to Landsat 8 and 9 satellites, using the ratio of red and green spectral bands and an exponential fit of the data (R² = 0.69 and RMSE = 3.36 µg/L). The models allowed the production of spatial distribution maps of chlorophyll-a throughout the historical series from 1985 to 2023, revealing the evolution of water quality in Paranoá Lake. Therefore, remote sensing facilitated the study of water bodies in the Federal District and showed promising prospects for the region.

COMMITTEE MEMBERS:
Externo à Instituição - DHALTON LUIZ TOSETTO VENTURA - ANA
Interna - 1509161 - RAQUEL MORAES SOARES
Presidente - 1732111 - REJANE ENNES CICERELLI
Interno - 2189333 - RICARDO TEZINI MINOTI