Deep Learning Prediction of Streamflow in Portugal

The transformative potential of deep learning models is felt in many research fields, including hydrology and water resources. This study investigates the effectiveness of the Temporal Fusion Transformer (TFT), a deep neural network architecture for predicting daily streamflow in Portugal, and bench...

Full description

Saved in:
Bibliographic Details
Main Authors: Rafael Francisco, José Pedro Matos
Format: Article
Language:English
Published: MDPI AG 2024-12-01
Series:Hydrology
Subjects:
Online Access:https://www.mdpi.com/2306-5338/11/12/217
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1846104393739403264
author Rafael Francisco
José Pedro Matos
author_facet Rafael Francisco
José Pedro Matos
author_sort Rafael Francisco
collection DOAJ
description The transformative potential of deep learning models is felt in many research fields, including hydrology and water resources. This study investigates the effectiveness of the Temporal Fusion Transformer (TFT), a deep neural network architecture for predicting daily streamflow in Portugal, and benchmarks it against the popular Hydrologiska Byråns Vattenbalansavdelning (HBV) hydrological model. Additionally, it evaluates the performance of TFTs through selected forecasting examples. Information is provided about key input variables, including precipitation, temperature, and geomorphological characteristics. The study involved extensive hyperparameter tuning, with over 600 simulations conducted to fine–tune performances and ensure reliable predictions across diverse hydrological conditions. The results showed that TFTs outperformed the HBV model, successfully predicting streamflow in several catchments of distinct characteristics throughout the country. TFTs not only provide trustworthy predictions with associated probabilities of occurrence but also offer considerable advantages over classical forecasting frameworks, i.e., the ability to model complex temporal dependencies and interactions across different inputs or weight features based on their relevance to the target variable. Multiple practical applications can rely on streamflow predictions made with TFT models, such as flood risk management, water resources allocation, and support climate change adaptation measures.
format Article
id doaj-art-df7febfea1b449cd8243ea4c0c760611
institution Kabale University
issn 2306-5338
language English
publishDate 2024-12-01
publisher MDPI AG
record_format Article
series Hydrology
spelling doaj-art-df7febfea1b449cd8243ea4c0c7606112024-12-27T14:29:49ZengMDPI AGHydrology2306-53382024-12-01111221710.3390/hydrology11120217Deep Learning Prediction of Streamflow in PortugalRafael Francisco0José Pedro Matos1Instituto Superior Técnico, Universidade de Lisboa, Av. Rovisco Pais 1, 1049–001 Lisboa, PortugalCivil Engineering Research and Innovation for Sustainability (CERIS), Instituto Superior Técnico, Universidade de Lisboa, Av. Rovisco Pais 1, 1049–001 Lisboa, PortugalThe transformative potential of deep learning models is felt in many research fields, including hydrology and water resources. This study investigates the effectiveness of the Temporal Fusion Transformer (TFT), a deep neural network architecture for predicting daily streamflow in Portugal, and benchmarks it against the popular Hydrologiska Byråns Vattenbalansavdelning (HBV) hydrological model. Additionally, it evaluates the performance of TFTs through selected forecasting examples. Information is provided about key input variables, including precipitation, temperature, and geomorphological characteristics. The study involved extensive hyperparameter tuning, with over 600 simulations conducted to fine–tune performances and ensure reliable predictions across diverse hydrological conditions. The results showed that TFTs outperformed the HBV model, successfully predicting streamflow in several catchments of distinct characteristics throughout the country. TFTs not only provide trustworthy predictions with associated probabilities of occurrence but also offer considerable advantages over classical forecasting frameworks, i.e., the ability to model complex temporal dependencies and interactions across different inputs or weight features based on their relevance to the target variable. Multiple practical applications can rely on streamflow predictions made with TFT models, such as flood risk management, water resources allocation, and support climate change adaptation measures.https://www.mdpi.com/2306-5338/11/12/217streamflowhydrological predictionhydrological modeldeep learningtemporal fusion transformerprobabilistic prediction
spellingShingle Rafael Francisco
José Pedro Matos
Deep Learning Prediction of Streamflow in Portugal
Hydrology
streamflow
hydrological prediction
hydrological model
deep learning
temporal fusion transformer
probabilistic prediction
title Deep Learning Prediction of Streamflow in Portugal
title_full Deep Learning Prediction of Streamflow in Portugal
title_fullStr Deep Learning Prediction of Streamflow in Portugal
title_full_unstemmed Deep Learning Prediction of Streamflow in Portugal
title_short Deep Learning Prediction of Streamflow in Portugal
title_sort deep learning prediction of streamflow in portugal
topic streamflow
hydrological prediction
hydrological model
deep learning
temporal fusion transformer
probabilistic prediction
url https://www.mdpi.com/2306-5338/11/12/217
work_keys_str_mv AT rafaelfrancisco deeplearningpredictionofstreamflowinportugal
AT josepedromatos deeplearningpredictionofstreamflowinportugal