River reach-level machine learning estimation of nutrient concentrations in Great Britain

Frontiers in Water

Water

Matthew Fry¹  David Huxley²  Chak-Hau Michael Tso³  Eugene Magee⁴  Michael Eastman⁵ 

[1] Centre of Excellence for Environmental Data Science, Lancaster, United Kingdom;UK Centre for Ecology and Hydrology, Wallingford, United Kingdom;Formerly Data Science MSc Programme, School of Computing and Communications, Lancaster University, Lancaster, United Kingdom;UK Centre for Ecology and Hydrology, Lancaster, United Kingdom;Centre of Excellence for Environmental Data Science, Lancaster, United Kingdom;UK Centre for Ecology and Hydrology, Wallingford, United Kingdom;Formerly Data Science MSc Programme, School of Computing and Communications, Lancaster University, Lancaster, United Kingdom;UK Centre for Ecology and Hydrology, Wallingford, United Kingdom;Met Office, Exeter, United Kingdom;

关键词: river network;    machine learning;    nutrients;    water quality;    random forest;   

DOI  :  10.3389/frwa.2023.1244024

 received in 2023-06-21, accepted in 2023-08-17,  发布年份 2023

来源: Frontiers

【 摘 要 】

Nitrogen (N) and phosphorus (P) are essential nutrients necessary for plant growth and support life in aquatic ecosystems. However, excessive N and P can lead to algal blooms that deplete oxygen and lead to fish death and the release of toxins that are harmful to humans. Estimates of N and P levels in rivers are typically calculated at station or grid (>1 km) scale; therefore, it is difficult to visualise the evolution of water quality as water travels downstream. Using a high-resolution reach-scale river network and associating each reach with land cover fractions and catchment descriptors, we trained random forest models on aggregated data (2010–2020) from the Environmental Agency Open Water Quality Data Archive for 2,343 stations to predict long-term nitrate and orthophosphate concentrations at each river reach in Great Britain (GB). We separated the model training and predictions for different seasons to investigate the potential difference in feature importance. Our model predicted concentrations with an average testing coefficient of determination (R2) of 0.71 for nitrate and 0.58 for orthophosphate using 5-fold cross-validation. Our model showed slightly better performance for higher Strahler stream orders, highlighting the challenges of making predictions in small streams. Our results revealed that arable and horticultural land use is the strongest and most reliable predictor for nitrate, while floodplain extents and standard percentage runoff are stronger predictors for orthophosphate. Nationally, higher orthophosphate concentrations were observed in urbanised areas. This study shows how combining a river network model with machine learning can easily provide a river network understanding of the spatial distribution of water quality levels.

【 授权许可】

Unknown   
Copyright © 2023 Tso, Magee, Huxley, Eastman and Fry.