Nation-wide estimation of groundwater redox conditions and nitrate concentrations through machine learning
The protection of water resources and development of mitigation strategies require large-scale information on water pollution such as nitrate. Machine learning techniques like random forest (RF) have proven their worth for estimating groundwater quality based on spatial environmental predictors. We...
| Main Authors: | , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
IOP Publishing
2020-01-01
|
| Series: | Environmental Research Letters |
| Subjects: | |
| Online Access: | https://doi.org/10.1088/1748-9326/ab7d5c |
| _version_ | 1797747528234434560 |
|---|---|
| author | Lukas Knoll Lutz Breuer Martin Bach |
| author_facet | Lukas Knoll Lutz Breuer Martin Bach |
| author_sort | Lukas Knoll |
| collection | DOAJ |
| description | The protection of water resources and development of mitigation strategies require large-scale information on water pollution such as nitrate. Machine learning techniques like random forest (RF) have proven their worth for estimating groundwater quality based on spatial environmental predictors. We investigate the potential of RF and quantile random forest (QRF) to estimate redox conditions and nitrate concentration in groundwater (1 km × 1 km resolution) using the European Water Framework Directive groundwater monitoring network as well as spatial environmental information available throughout Germany. The RF model for nitrate achieves a good predictive performance with an R ^2 of 0.52. Dominant predictors are the redox conditions in the groundwater body, hydrogeological units and the percentage of arable land. An uncertainty assessment using QRF shows rather large uncertainties with a mean prediction interval (MPI) of 53.0 mg l ^−1 . This study represents the first nation-wide data-driven assessment of the spatial distribution of groundwater nitrate concentrations for Germany. |
| first_indexed | 2024-03-12T15:51:52Z |
| format | Article |
| id | doaj.art-6e010542b3e6460ead4b87e37a56f451 |
| institution | Directory Open Access Journal |
| issn | 1748-9326 |
| language | English |
| last_indexed | 2024-03-12T15:51:52Z |
| publishDate | 2020-01-01 |
| publisher | IOP Publishing |
| record_format | Article |
| series | Environmental Research Letters |
| spelling | doaj.art-6e010542b3e6460ead4b87e37a56f4512023-08-09T15:06:46ZengIOP PublishingEnvironmental Research Letters1748-93262020-01-0115606400410.1088/1748-9326/ab7d5cNation-wide estimation of groundwater redox conditions and nitrate concentrations through machine learningLukas Knoll0https://orcid.org/0000-0001-7928-0778Lutz Breuer1https://orcid.org/0000-0001-9720-1076Martin Bach2https://orcid.org/0000-0002-8695-7529Institute for Landscape Ecology and Resources Management (ILR), Research Centre for BioSystems, Land Use and Nutrition (iFZ), Justus Liebig University Giessen , Giessen, GermanyInstitute for Landscape Ecology and Resources Management (ILR), Research Centre for BioSystems, Land Use and Nutrition (iFZ), Justus Liebig University Giessen , Giessen, GermanyInstitute for Landscape Ecology and Resources Management (ILR), Research Centre for BioSystems, Land Use and Nutrition (iFZ), Justus Liebig University Giessen , Giessen, GermanyThe protection of water resources and development of mitigation strategies require large-scale information on water pollution such as nitrate. Machine learning techniques like random forest (RF) have proven their worth for estimating groundwater quality based on spatial environmental predictors. We investigate the potential of RF and quantile random forest (QRF) to estimate redox conditions and nitrate concentration in groundwater (1 km × 1 km resolution) using the European Water Framework Directive groundwater monitoring network as well as spatial environmental information available throughout Germany. The RF model for nitrate achieves a good predictive performance with an R ^2 of 0.52. Dominant predictors are the redox conditions in the groundwater body, hydrogeological units and the percentage of arable land. An uncertainty assessment using QRF shows rather large uncertainties with a mean prediction interval (MPI) of 53.0 mg l ^−1 . This study represents the first nation-wide data-driven assessment of the spatial distribution of groundwater nitrate concentrations for Germany.https://doi.org/10.1088/1748-9326/ab7d5cgroundwater qualitynitrate pollutionredox conditionsrandom forestuncertaintylarge-scale |
| spellingShingle | Lukas Knoll Lutz Breuer Martin Bach Nation-wide estimation of groundwater redox conditions and nitrate concentrations through machine learning Environmental Research Letters groundwater quality nitrate pollution redox conditions random forest uncertainty large-scale |
| title | Nation-wide estimation of groundwater redox conditions and nitrate concentrations through machine learning |
| title_full | Nation-wide estimation of groundwater redox conditions and nitrate concentrations through machine learning |
| title_fullStr | Nation-wide estimation of groundwater redox conditions and nitrate concentrations through machine learning |
| title_full_unstemmed | Nation-wide estimation of groundwater redox conditions and nitrate concentrations through machine learning |
| title_short | Nation-wide estimation of groundwater redox conditions and nitrate concentrations through machine learning |
| title_sort | nation wide estimation of groundwater redox conditions and nitrate concentrations through machine learning |
| topic | groundwater quality nitrate pollution redox conditions random forest uncertainty large-scale |
| url | https://doi.org/10.1088/1748-9326/ab7d5c |
| work_keys_str_mv | AT lukasknoll nationwideestimationofgroundwaterredoxconditionsandnitrateconcentrationsthroughmachinelearning AT lutzbreuer nationwideestimationofgroundwaterredoxconditionsandnitrateconcentrationsthroughmachinelearning AT martinbach nationwideestimationofgroundwaterredoxconditionsandnitrateconcentrationsthroughmachinelearning |