Spatial Distribution of Integrated Nitrate Reduction across the Unsaturated Zone and the Groundwater Body in Germany

Knoll, Lukas; Häußermann, Uwe; Bach, Martin

doi:10.3390/w12092456

Cited by 14 publications

(5 citation statements)

References 39 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…For example, recent observations and modeling work in Germany found 95% of the N surplus may be denitrified along groundwater pathways before even reaching surface water, while other catchments showed little to any capacity to remove N surplus through subsurface pathways. This database can thus be used for management and further research purposes to explore relationships between surplus and stream export and identify factors that modify this relationship ( Knoll et al 2020 ). Second, ranking subbasins only by mass of N and P would likely lead to larger basins being ranked higher than smaller basins largely due to the difference in area rather than the intensity of fluxes or surpluses.…”

Section: Methodsmentioning

confidence: 99%

Considerations when using nutrient inventories to prioritize water quality improvement efforts across the US

Sabo

Clark

Compton

2021

Environ. Res. Commun.

View full text Add to dashboard Cite

Ongoing water quality degradation tied to nitrogen and phosphorus pollution results in significant economic damages by diminishing the recreational value of surface water and compromising fisheries. Progress in decreasing nitrogen and phosphorus pollution to surface water over the past two decades has been slow. Limited resources need to be leveraged efficiently and effectively to prioritize watersheds for restoration. Leveraging recent nitrogen and phosphorus inventories for the years 2002, 2007, and 2012, we extracted relevant flux and demand terms to help identify US subbasins that are likely contributing a disproportionate amount of point and non-point source nutrient pollution to surface water by exploring the mean spatial distribution of terrestrial anthropogenic surplus, agricultural surplus, agricultural nutrient use efficiency, and point source loads. A small proportion of the landscape, <25% of subbasin area of the United States, contains 50% of anthropogenic and agriculture nitrogen and phosphorus surplus while only 2% of landscape contributes >50% of point source loads into surface water. Point source loads are mainly concentrated in urban areas across the country with point source loading rates often exceeding >10.0 kg N ha−1 yr−1 and >1.0 kg P ha−1 yr−1. However, the ability for future upgrades to wastewater treatment plant infrastructure alone is unlikely to drive further improvement in water quality, outside of local water ways, since point source loads only account for ~4% of anthropogenic N and P surplus. As such, further progress in boosting nutrient use efficiency in agricultural production, usually lowest in areas of intensive livestock production, would likely contribute to the biggest gains to water quality restoration goals. This analysis and the corresponding database integrate multiple streams of information to highlight areas where N and P are being managed inefficiently to give decision makers a succinct platform to identify likely areas and sources of water quality degradation.

show abstract

Section: Methodsmentioning

confidence: 99%

Considerations when using nutrient inventories to prioritize water quality improvement efforts across the US

Sabo

Clark

Compton

2021

Environ. Res. Commun.

View full text Add to dashboard Cite

show abstract

“…Although in certain conditions the groundwater environment enables some decrease of nitrates concentration (e.g. Højberg et al, 2017;Knoll et al, 2020a), the capacity of this mechanism is exhaustible (Knoll, et al, 2020b;Kunkel et al, 2017). Therefore, the dilution of groundwater with less polluted percolation water often remains the most available way to reduce an excessive nitrate concentration in groundwater (Mas-Pla, Menció, 2019).…”

Section: Discussionmentioning

confidence: 99%

Definition of hot-spots to reduce the nitrogen losses from agricultural land to groundwater in Slovakia

Bujnovský¹,

Koco

2022

Ekológia (Bratislava)

View full text Add to dashboard Cite

Agriculture is a significant contributor to nitrate pollution of groundwater which in many cases serves as a source of drinking water. Therefore, targeted reduction of nitrogen leaching losses is fully justified to address this issue. The aim of the study was to define the areas of utilized agricultural land (UAL) in Slovakia, where a nitrogen surplus needs to be reduced. Using the average values of leachable nitrogen in the period 2015-2018 and the long-term amount of percolated water, the nitrate concentration in leachate was calculated. To ensure that agricultural activities will contribute to the gradual reduction of nitrate concentration in groundwater, the nitrate concentration in leachate of 40 mg L-1 was chosen as the target value. This concentration was exceeded at 11.7% of the UAL area. The average indicative amount of nitrogen in industrial fertilizers that needs to be reduced to achieve a stricter nitrate concentration in the leachate in these hot-spots is 16 kg ha-1 with the proviso that in two districts this value exceeds 30 kg ha-1.

show abstract

“…This is only possible if the groundwater managers have decision tools to spatially predict the nitrate concentration. On one hand, there is a lot of research going on in the field of modeling nitrate in groundwater bodies, mainly on the basis of geostatistical methods such as kriging (Wriedt et al, 2019), numerical models (Nguyen and Dietrich, 2018), and tree-based models such as the random forest (Breiman, 2001;Knoll et al, 2020;Mandal et al, 2023;Sarkar et al, 2023) and gradient boost regression trees (Friedman, 2002). Breiman (2001) showed that the random forest model gives an opportunity for support to water managers and authorities in developing strategies for measures to reduce nitrate inputs into the groundwater.…”

Section: Introductionmentioning

confidence: 99%

“…For example, Mendes et al (2016), Ouedraogo et al (2018) and Sarkar et al (2022) applied random forest regression to model groundwater nitrate concentration. Boosted regression trees have been successfully used to fit machine learning models for predicting groundwater nitrate (Ransom et al, 2017(Ransom et al, , 2022Knoll et al, 2020). Although attempts have been made to incorporate information about an observation's surroundings into the predictor variables of a random forest model, this information still lacks a spatial reference.…”

Section: Introductionmentioning

confidence: 99%

Application of machine learning and deep neural networks for spatial prediction of groundwater nitrate concentration to improve land use management practices

et al. 2023

View full text Add to dashboard Cite

The prediction of groundwater nitrate concentration's response to geo-environmental and human-influenced factors is essential to better restore groundwater quality and improve land use management practices. In this paper, we regionalize groundwater nitrate concentration using different machine learning methods (Random forest (RF), unimodal 2D and 3D convolutional neural networks (CNN), and multi-stream early and late fusion 2D-CNNs) so that the nitrate situation in unobserved areas can be predicted. CNNs take into account not only the nitrate values of the grid cells of the observation wells but also the values around them. This has the added benefit of allowing them to learn directly about the influence of the surroundings. The predictive performance of the models was tested on a dataset from a pilot region in Germany, and the results show that, in general, all the machine learning models, after a Bayesian optimization hyperparameter search and training, achieve good spatial predictive performance compared to previous studies based on Kriging and numerical models. Based on the mean absolute error (MAE), the random forest model and the 2DCNN late fusion model performed best with an MAE (STD) of 9.55 (0.367) mg/l, R2 = 0.43 and 10.32 (0.27) mg/l, R2 = 0.27, respectively. The 3DCNN with an MAE (STD) of 11.66 (0.21) mg/l and largest resources consumption is the worst performing model. Feature importance learning from the models was used in conjunction with partial dependency analysis of the most important features to gain greater insight into the major factors explaining the nitrate spatial variability. Large uncertainties in nitrate prediction have been shown in previous studies. Therefore, the models were extended to quantify uncertainty using prediction intervals (PIs) derived from bootstrapping. Knowledge of uncertainty helps the water manager reduce risk and plan more reliably.

show abstract

Spatial Distribution of Integrated Nitrate Reduction across the Unsaturated Zone and the Groundwater Body in Germany

Cited by 14 publications

References 39 publications

Considerations when using nutrient inventories to prioritize water quality improvement efforts across the US

Considerations when using nutrient inventories to prioritize water quality improvement efforts across the US

Definition of hot-spots to reduce the nitrogen losses from agricultural land to groundwater in Slovakia

Application of machine learning and deep neural networks for spatial prediction of groundwater nitrate concentration to improve land use management practices

Contact Info

Product

Resources

About