Probabilistic Uncertainty Consideration in Regionalization and Prediction of Groundwater Nitrate Concentration

In this study, we extend our previous work on a two-dimensional convolutional neural network (2DCNN) for spatial prediction of groundwater nitrate, focusing on improving uncertainty quantification. Our enhanced model incorporates a fully probabilistic Bayesian framework and a structure aimed at opti...

Full description

Saved in:
Bibliographic Details
Main Author: Divas Karimanzira
Format: Article
Language:English
Published: MDPI AG 2024-09-01
Series:Knowledge
Subjects:
Online Access:https://www.mdpi.com/2673-9585/4/4/25
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1846104031890505728
author Divas Karimanzira
author_facet Divas Karimanzira
author_sort Divas Karimanzira
collection DOAJ
description In this study, we extend our previous work on a two-dimensional convolutional neural network (2DCNN) for spatial prediction of groundwater nitrate, focusing on improving uncertainty quantification. Our enhanced model incorporates a fully probabilistic Bayesian framework and a structure aimed at optimizing both specific value predictions and predictive intervals (PIs). We implemented the Prediction Interval Validation and Estimation Network based on Quality Definition (2DCNN-QD) to refine the accuracy of probabilistic predictions and reduce the width of the prediction intervals. Applied to a model region in Germany, our results demonstrate an 18% improvement in the prediction interval width. While traditional Bayesian CNN models may yield broader prediction intervals to adequately capture uncertainties, the 2DCNN-QD method prioritizes quality-driven interval optimization, resulting in narrower prediction intervals without sacrificing coverage probability. Notably, this approach is nonparametric, allowing it to be effectively utilized across a range of real-world scenarios.
format Article
id doaj-art-c35c2d5137e640dfb1c8a0bbaf1b5a02
institution Kabale University
issn 2673-9585
language English
publishDate 2024-09-01
publisher MDPI AG
record_format Article
series Knowledge
spelling doaj-art-c35c2d5137e640dfb1c8a0bbaf1b5a022024-12-27T14:34:42ZengMDPI AGKnowledge2673-95852024-09-014446248010.3390/knowledge4040025Probabilistic Uncertainty Consideration in Regionalization and Prediction of Groundwater Nitrate ConcentrationDivas Karimanzira0Fraunhofer Institute for Optronics, System Technique and Image Exploitation IOSB, Am Vogelherd 90, 98693 Ilmenau, GermanyIn this study, we extend our previous work on a two-dimensional convolutional neural network (2DCNN) for spatial prediction of groundwater nitrate, focusing on improving uncertainty quantification. Our enhanced model incorporates a fully probabilistic Bayesian framework and a structure aimed at optimizing both specific value predictions and predictive intervals (PIs). We implemented the Prediction Interval Validation and Estimation Network based on Quality Definition (2DCNN-QD) to refine the accuracy of probabilistic predictions and reduce the width of the prediction intervals. Applied to a model region in Germany, our results demonstrate an 18% improvement in the prediction interval width. While traditional Bayesian CNN models may yield broader prediction intervals to adequately capture uncertainties, the 2DCNN-QD method prioritizes quality-driven interval optimization, resulting in narrower prediction intervals without sacrificing coverage probability. Notably, this approach is nonparametric, allowing it to be effectively utilized across a range of real-world scenarios.https://www.mdpi.com/2673-9585/4/4/25uncertainty analysisspatial predictionregionalizationconvolutional neural networkshigh-quality definition
spellingShingle Divas Karimanzira
Probabilistic Uncertainty Consideration in Regionalization and Prediction of Groundwater Nitrate Concentration
Knowledge
uncertainty analysis
spatial prediction
regionalization
convolutional neural networks
high-quality definition
title Probabilistic Uncertainty Consideration in Regionalization and Prediction of Groundwater Nitrate Concentration
title_full Probabilistic Uncertainty Consideration in Regionalization and Prediction of Groundwater Nitrate Concentration
title_fullStr Probabilistic Uncertainty Consideration in Regionalization and Prediction of Groundwater Nitrate Concentration
title_full_unstemmed Probabilistic Uncertainty Consideration in Regionalization and Prediction of Groundwater Nitrate Concentration
title_short Probabilistic Uncertainty Consideration in Regionalization and Prediction of Groundwater Nitrate Concentration
title_sort probabilistic uncertainty consideration in regionalization and prediction of groundwater nitrate concentration
topic uncertainty analysis
spatial prediction
regionalization
convolutional neural networks
high-quality definition
url https://www.mdpi.com/2673-9585/4/4/25
work_keys_str_mv AT divaskarimanzira probabilisticuncertaintyconsiderationinregionalizationandpredictionofgroundwaternitrateconcentration