Modeling freshwater inflow to a tropical estuary using stochastic and neural-network approaches under contrasting hydroclimatic conditions in Brazil.
4 affiliations
- 1Programa de Pós-Graduação Em Ecologia E Recursos Naturais, Laboratório de Ciências Ambientais, Centro de Biociências E Biotecnologia, Universidade Estadual Do Norte Fluminense Darcy Ribeiro, Av. Alberto Lamego, 2000 - Parque Califórnia, CEP: 28013-602, Campos Dos Goytacazes, Rio de Janeiro, Brazil. [email protected].
- 2Laboratório de Ciências Químicas, Centro de Ciência E Tecnologia, Universidade Estadual Do Norte Fluminense Darcy Ribeiro, Av. Alberto Lamego, 2000 - Parque Califórnia -, CEP: 28013-602, Campos Dos Goytacazes, Rio de Janeiro, Brazil.
- 3Programa de Pós-Graduação Em Ecologia E Recursos Naturais, Laboratório de Ciências Ambientais, Centro de Biociências E Biotecnologia, Universidade Estadual Do Norte Fluminense Darcy Ribeiro, Av. Alberto Lamego, 2000 - Parque Califórnia, CEP: 28013-602, Campos Dos Goytacazes, Rio de Janeiro, Brazil.
- 4Departamento de Ciências Exatas, Universidade Estadual de Feira de Santana, Feira de Santana, Bahia, Brazil.
Abstract
Freshwater inflow regulates estuarine circulation, salinity gradients, sediment transport, and biogeochemical fluxes, yet long-term discharge records remain scarce in many tropical coastal basins. We compared linear regression, stochastic time-series models (ARMA, ARIMA, and SARIMA), and multilayer perceptron (MLP) neural networks for reconstructing monthly discharge at the Ituberá gauging station, upstream of the Serinhaém estuary (southern Bahia, Brazil), using records from 1966 to 2019. No model family was uniformly superior across rolling-origin selection and independent validation. Predictor-matched comparisons showed no general advantage of MLPs over linear formulations, and extending calibration through 2011 did not consistently improve post-2012 performance. Mean monthly discharge was approximately 15% lower after 2012, while changes in variance and the general seasonal pattern were not detected and evidence for altered temporal dependence remained inconclusive. One-step reconstructions transferred more consistently to held-out data than recursive and longer-horizon predictions, which showed error accumulation. The reconstructed series provides an upstream freshwater boundary indicator for future estuarine modeling, while ecological consequences require explicit hydrodynamic and biogeochemical simulation.
Related
- Integration of multi-source hydroclimatic variables for enhanced drought monitoring
- Hydroclimatic controls and interaction-driven sensitivity of reference evapotranspiration across Indian climate regimes
- Diurnally asymmetric warming and hydroclimatic stress structure alpine ecosystem vulnerability in the Three-River headwaters region
- A 256-year daily rainfall record reveals multiscale rainy-season variability and emerging hydroclimatic hazard over northern East Asia
- Climate risk assessment in tropical wetlands through integrated hydroclimatic, land cover, and community knowledge
BibTeX
@article{viana2026modeling,
title = {{Modeling freshwater inflow to a tropical estuary using stochastic and neural-network approaches under contrasting hydroclimatic conditions in Brazil.}},
author = {Luísa Maria de Souza Viana and Eloá Corrêa Lessa Tostes and Wendel Dias Constantino and Felipe Henrique Rossi Luze and Marcos Sarmet Moreira de Barros Salomão and Taíse Bonfim de Jesus and Carlos Eduardo Veiga de Carvalho},
journal = {Environmental monitoring and assessment},
year = {2026},
month = oct,
doi = {10.1007/s10661-026-16034-1},
url = {https://europepmc.org/article/MED/42891723},
note = {CC BY},
}
RIS
TY - JOUR
TI - Modeling freshwater inflow to a tropical estuary using stochastic and neural-network approaches under contrasting hydroclimatic conditions in Brazil.
AU - Luísa Maria de Souza Viana
AU - Eloá Corrêa Lessa Tostes
AU - Wendel Dias Constantino
AU - Felipe Henrique Rossi Luze
AU - Marcos Sarmet Moreira de Barros Salomão
AU - Taíse Bonfim de Jesus
AU - Carlos Eduardo Veiga de Carvalho
PY - 2026
DA - 2026/10/17
JO - Environmental monitoring and assessment
DO - 10.1007/s10661-026-16034-1
UR - https://europepmc.org/article/MED/42891723
AB - Freshwater inflow regulates estuarine circulation, salinity gradients, sediment transport, and biogeochemical fluxes, yet long-term discharge records remain scarce in many tropical coastal basins. We compared linear regression, stochastic time-series models (ARMA, ARIMA, and SARIMA), and multilayer perceptron (MLP) neural networks for reconstructing monthly discharge at the Ituberá gauging station, upstream of the Serinhaém estuary (southern Bahia, Brazil), using records from 1966 to 2019. No model family was uniformly superior across rolling-origin selection and independent validation. Predictor-matched comparisons showed no general advantage of MLPs over linear formulations, and extending calibration through 2011 did not consistently improve post-2012 performance. Mean monthly discharge was approximately 15% lower after 2012, while changes in variance and the general seasonal pattern were not detected and evidence for altered temporal dependence remained inconclusive. One-step reconstructions transferred more consistently to held-out data than recursive and longer-horizon predictions, which showed error accumulation. The reconstructed series provides an upstream freshwater boundary indicator for future estuarine modeling, while ecological consequences require explicit hydrodynamic and biogeochemical simulation.
SN - 0167-6369
SN - 1573-2959
ER -