Please use this identifier to cite or link to this item: http://hdl.handle.net/10609/147355
Title: Evaluación de algoritmos de aprendizaje automático para la determinación de parámetros de calidad del agua mediante teledetección
Author: Navarro Oronoz, Vanesa
Tutor: Muñoz Bollas, Anna
Others: Solé-Ribalta, Albert  
Abstract: Monitoring water quality is a priority for supply and consumption needs, and also for the proper functioning of natural ecosystems. This paper studies the potential of monitoring indicative parameters of water quality using advanced remote sensing techniques and machine learning algorithms in the El Val reservoir (Arag´on), since it is classified as a -Sensitive Zoneand it is in a vulnerable and environmental risk situation. This paper presents the results of the validation of thermal images of the TIRS sensor on board the Landsat missions to obtain the surface temperature of the water, and the validation of optical images captured by the MSI sensor on board the Sentinel mission to obtain the turbidity and chlorophyll-a concentration between January 2018 and December 2022. In estimating temperature, the images have been calibrated with field data, obtaining a linear regression model (R2=0.98) with an error of 1 °C. For the estimation of chlorophyll-a and turbidity, the Decision Tree, Random Forest and SVM algorithms have been evaluated with automatic search for the best combination of hyperparameters and cross validation with 5 stratified partitions. In the estimation of chlorophyll-a, the performance of Decision Tree and Random Forest has been similar, explaining 88 % of the variance of the model and with errors between 10-15 mg/l. The models evaluated to estimate turbidity have not been able to adequately represent the behavior of the variable.
Keywords: water quality
remote sensing
machine learning
Document type: info:eu-repo/semantics/masterThesis
Issue Date: Jan-2023
Publication license: http://creativecommons.org/licenses/by-nc-nd/3.0/es/  
Appears in Collections:Treballs finals de carrera, treballs de recerca, etc.

Files in This Item:
File Description SizeFormat 
vnoronozTFM0123memoria.pdfMemoria del TFM6,3 MBAdobe PDFThumbnail
View/Open
Share:
Export:
View statistics

This item is licensed under aCreative Commons License Creative Commons