@article{fdi:010091908,
  title = {{L}everaging machine learning and remote sensing for water quality analysis in {L}ake {R}anco, {S}outhern {C}hile},
  author = {{R}odríguez-{L}ópez, {L}. and {A}lvarez, {L}. {B}. and {D}uran-{L}lacer, {I}. and {R}uíz-{G}uirola, {D}. {E}. and {M}ontejo-{S}ánchez, {S}. and {M}artínez-{R}etureta, {R}. and {L}ópez-{M}orales, {E}. and {B}ourrel, {L}uc and {F}rappart, {F}. and {U}rrutia, {R}.},
  editor = {},
  language = {{ENG}},
  abstract = {{T}his study examines the dynamics of limnological parameters of a {S}outh {A}merican lake located in southern {C}hile with the objective of predicting chlorophyll-a levels, which are a key indicator of algal biomass and water quality, by integrating combined remote sensing and machine learning techniques. {E}mploying four advanced machine learning models (recurrent neural network ({RNN}s), long short-term memory ({LSTM}), recurrent gate unit ({GRU}), and temporal convolutional network ({TCN}s)), the research focuses on the estimation of chlorophyll-a concentrations at three sampling stations within {L}ake {R}anco. {T}he data span from 1987 to 2020 and are used in three different cases: using only in situ data ({C}ase 1), using in situ and meteorological data ({C}ase 2), using in situ, and meteorological and satellite data from {L}andsat and {S}entinel missions ({C}ase 3). {I}n all cases, each machine learning model shows robust performance, with promising results in predicting chlorophyll-a concentrations. {A}mong these models, {LSTM} stands out as the most effective, with the best metrics in the estimation, the best performance was {C}ase 1, with {R}2 = 0.89, an {RSME} of 0.32 mu g/{L}, an {MAE} 1.25 mu g/{L} and an {MSE} 0.25 (mu g/{L})2, consistently outperforming the others according to the static metrics used for validation. {T}his finding underscores the effectiveness of {LSTM} in capturing the complex temporal relationships inherent in the dataset. {H}owever, increasing the dataset in {C}ase 3 shows a better performance of {TCN}s ({R}2 = 0.96; {MSE} = 0.33 (mu g/{L})2; {RMSE} = 0.13 mu g/{L}; and {MAE} = 0.06 mu g/{L}). {T}he successful application of machine learning algorithms emphasizes their potential to elucidate the dynamics of algal biomass in {L}ake {R}anco, located in the southern region of {C}hile. {T}hese results not only contribute to a deeper understanding of the lake ecosystem but also highlight the utility of advanced computational techniques in environmental research and management.},
  keywords = {water quality ; lake ; remote sensing ; machine learning ; southern {C}hile ; {CHILI}},
  booktitle = {},
  journal = {{R}emote {S}ensing},
  volume = {16},
  numero = {18},
  pages = {3401 [20 p.]},
  year = {2024},
  DOI = {10.3390/rs16183401},
  URL = {https://www.documentation.ird.fr/hor/fdi:010091908},
}