Random forest model to predict long-term seasonal nitrate and orthophosphate concentrations in British river reaches
https://doi.org/10.5285/ba208b6c-6f1a-43b1-867d-bc1adaff6445
Download/Access
This model is available under the terms of the Open Government Licence
By accessing or using this model, you agree to the terms of the relevant licence agreement(s). You will ensure that this model is cited in any publication that describes research in which the data have been used.
Providing access to this model does not signify any endorsement or approval. By accessing and using the resource, you acknowledge that it is entirely at your own risk and you are solely responsible for any loss or liability that may arise
Uses Environment Agency water quality data from the Water Quality Archive
This resource comprises two Jupyter notebooks that includes the model code in python to train a random forest model to predict long-term seasonal nitrate and orthophosphate concentrations at each river reach in Great Britain. The input features considered are catchment descriptors and land cover matched to the reaches. The training data is obtained from the Environmental Agency Water Quality Archive, 2010-2020. This method provides an effective way to map water quality data from stations to the river network.
A live demo of a web application to visualize the dataset can be viewed at https://moisture-wqmlviewer.datalabs.ceh.ac.uk/wqml_viewer
A live demo of a web application to visualize the dataset can be viewed at https://moisture-wqmlviewer.datalabs.ceh.ac.uk/wqml_viewer
Publication date: 2023-09-05
284 downloads
2,957 views
Formats
Jupyter notebook, Comma-separated values (CSV)
Spatial information
Study area
Spatial representation type
Tabular (text)
Spatial reference system
OSGB 1936 / British National Grid
Spatial resolution
50000 metres
Temporal information
Temporal extent
2010-01-01 to 2020-12-31
Provenance & quality
The machine learning model is trained on the following datasets:
* Land cover map
* NRFA Catchment descriptors
* UKCEH digital river network of Great Britain (1:50,000)
* Environmental Agency water quality archive
More information is provided in the supporting documentation accompanying this resource.
Please note that only a subset of the data (HA 72 Wyre and Lune) is provided for demonstration. This is because the underlying river network is separately licensed (https://catalogue.ceh.ac.uk/documents/7d5e42b6-7729-46c8-99e9-f9e4efddde1d).
* Land cover map
* NRFA Catchment descriptors
* UKCEH digital river network of Great Britain (1:50,000)
* Environmental Agency water quality archive
More information is provided in the supporting documentation accompanying this resource.
Please note that only a subset of the data (HA 72 Wyre and Lune) is provided for demonstration. This is because the underlying river network is separately licensed (https://catalogue.ceh.ac.uk/documents/7d5e42b6-7729-46c8-99e9-f9e4efddde1d).
Licensing and constraints
This model is available under the terms of the Open Government Licence 
Cite this model as:
Magee, E.; Huxley, D.; Tso, C.-.M. (2023). Random forest model to predict long-term seasonal nitrate and orthophosphate concentrations in British river reaches. NERC EDS Environmental Information Data Centre. https://doi.org/10.5285/ba208b6c-6f1a-43b1-867d-bc1adaff6445
Uses Environment Agency water quality data from the Water Quality Archive
Related
Supplemental information
Correspondence/contact details
Tso, M.
UK Centre for Ecology & Hydrology
Lancaster Environment Centre, Library Avenue, Bailrigg
Lancaster
Lancashire
LA1 4AP
UNITED KINGDOM
enquiries@ceh.ac.uk
Lancaster
Lancashire
LA1 4AP
UNITED KINGDOM
Other contacts
Publisher
NERC EDS Environmental Information Data Centre
info@eidc.ac.uk
Rights holder
UK Centre for Ecology & Hydrology
enquiries@ceh.ac.uk