Evaluating a process‐guided deep learning approach for predicting dissolved oxygen in streams
Fuente:
Wiley Open Access
Salvato in:
| Autori principali: | , , , , |
|---|---|
| Natura: | Artículo Open Access |
| Pubblicazione: |
Wiley
2024
|
| Soggetti: | |
| Accesso online: | |
| Tags: |
Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
|
| _version_ | 1867002546827034624 |
|---|---|
| author | Jeffrey M. Sadler Lauren E. Koenig Galen Gorski Alice M. Carter Robert O. Hall |
| author_facet | Jeffrey M. Sadler Lauren E. Koenig Galen Gorski Alice M. Carter Robert O. Hall Jeffrey M. Sadler Lauren E. Koenig Galen Gorski Alice M. Carter Robert O. Hall |
| collection | Wiley Open Access |
| contents | Evaluating a process‐guided deep learning approach for predicting dissolved oxygen in streams Jeffrey M. Sadler Lauren E. Koenig Galen Gorski Alice M. Carter Robert O. Hall Hydrological Processes AbstractDissolved oxygen (DO) is a critical water quality constituent that governs habitat suitability for aquatic biota, biogeochemical reactions and solubility of metals in streams. Recently introduced high‐frequency sensors have increased our ability to measure DO, but we still lack the capacity to understand and predict DO concentrations at high spatial resolutions or in unmonitored locations. Machine learning (ML) has been a commonly used approach for modelling DO, however, conventional ML models have no representation of the limnological processes governing DO dynamics. Here we implement and evaluate two process‐guided deep learning (PGDL) approaches for predicting daily minimum, mean and maximum DO concentrations in rivers from the Delaware River Basin, USA. In both cases, a multi‐task approach was taken in which the PGDL models predicted stream metabolism and gas exchange rates in addition to the DO concentrations themselves. Our results showed that for these sites, the PGDL approaches did not improve upon baseline predictions in temporal and spatially similar holdout experiments. One of the approaches did, however, improve predictions when applied to spatially dissimilar sites. Although this particular PGDL approach did not improve predictive accuracy in most cases, our results suggest that process guidance, perhaps a more constrained approach, could benefit a data‐driven DO model. 10.1002/hyp.15270 http://onlinelibrary.wiley.com/termsAndConditions#vor |
| doi_str_mv | 10.1002/hyp.15270 |
| format | Artículo Open Access |
| id | wiley_oa_10_1002_hyp_15270 |
| institution | Wiley Open Access |
| license_str_mv | http://onlinelibrary.wiley.com/termsAndConditions#vor |
| publishDate | 2024 |
| publisher | Wiley |
| record_format | wiley_oa |
| spellingShingle | Evaluating a process‐guided deep learning approach for predicting dissolved oxygen in streams Jeffrey M. Sadler Lauren E. Koenig Galen Gorski Alice M. Carter Robert O. Hall Hydrological Processes Evaluating a process‐guided deep learning approach for predicting dissolved oxygen in streams Jeffrey M. Sadler Lauren E. Koenig Galen Gorski Alice M. Carter Robert O. Hall Hydrological Processes AbstractDissolved oxygen (DO) is a critical water quality constituent that governs habitat suitability for aquatic biota, biogeochemical reactions and solubility of metals in streams. Recently introduced high‐frequency sensors have increased our ability to measure DO, but we still lack the capacity to understand and predict DO concentrations at high spatial resolutions or in unmonitored locations. Machine learning (ML) has been a commonly used approach for modelling DO, however, conventional ML models have no representation of the limnological processes governing DO dynamics. Here we implement and evaluate two process‐guided deep learning (PGDL) approaches for predicting daily minimum, mean and maximum DO concentrations in rivers from the Delaware River Basin, USA. In both cases, a multi‐task approach was taken in which the PGDL models predicted stream metabolism and gas exchange rates in addition to the DO concentrations themselves. Our results showed that for these sites, the PGDL approaches did not improve upon baseline predictions in temporal and spatially similar holdout experiments. One of the approaches did, however, improve predictions when applied to spatially dissimilar sites. Although this particular PGDL approach did not improve predictive accuracy in most cases, our results suggest that process guidance, perhaps a more constrained approach, could benefit a data‐driven DO model. 10.1002/hyp.15270 http://onlinelibrary.wiley.com/termsAndConditions#vor |
| title | Evaluating a process‐guided deep learning approach for predicting dissolved oxygen in streams |
| topic | Hydrological Processes |
| url | https://onlinelibrary.wiley.com/doi/10.1002/hyp.15270 |