{"title":"Bridging temporal gaps: AI-based temporal downscaling of biweekly NH3 to daily scale with spatial transferability","authors":"Saman Malik, Eunjin Kang, Yoojin Kang, Jungho Im","doi":"10.1016/j.jhazmat.2025.139166","DOIUrl":null,"url":null,"abstract":"Ammonia (NH<sub>3</sub>) is a gaseous pollutant with significant environmental and health implications. Over recent decades, its increasing concentrations, driven by industrialization and agriculture, have necessitated high-resolution monitoring. However, limited daily ground-based observations hinder comprehensive analysis. This study developed machine learning-based frameworks—deep neural network (DNN), random forest, and light gradient boosting machine—to predict biweekly NH<sub>3</sub> concentrations and downscale them to daily estimates across the United States during 2017–2022. The models integrate NH<sub>3</sub> column concentrations, meteorological variables, land cover data, livestock information, and ground-based measurements. Among the models, DNN showed superior performance in both spatial cross-validation and independent testing, achieving a correlation coefficient of 0.79, a root mean square error of 0.98<!-- --> <!-- -->µg/m³, and an index of agreement of 0.83— effectively capturing fine-scale spatial variations at a 9<!-- --> <!-- -->km resolution. Shapley additive explanations analysis identified temporal dynamic factors—such as day of year and meteorological variables—as key predictors, along with land cover and cattle density, highlighting the model’s ability to support the temporal downscaling of NH<sub>3</sub> from biweekly to daily scale. When applied to the UK, the model demonstrated its potential for spatial transferability via the leave-one station-out and leave-one year-out cross validations. These findings highlight the ability of machine learning in bridging temporal gaps and generating high-resolution daily NH<sub>3</sub> estimates.","PeriodicalId":361,"journal":{"name":"Journal of Hazardous Materials","volume":"687 1","pages":""},"PeriodicalIF":12.2000,"publicationDate":"2025-07-07","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"Journal of Hazardous Materials","FirstCategoryId":"93","ListUrlMain":"https://doi.org/10.1016/j.jhazmat.2025.139166","RegionNum":1,"RegionCategory":"环境科学与生态学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"Q1","JCRName":"ENGINEERING, ENVIRONMENTAL","Score":null,"Total":0}
引用次数: 0
Abstract
Ammonia (NH3) is a gaseous pollutant with significant environmental and health implications. Over recent decades, its increasing concentrations, driven by industrialization and agriculture, have necessitated high-resolution monitoring. However, limited daily ground-based observations hinder comprehensive analysis. This study developed machine learning-based frameworks—deep neural network (DNN), random forest, and light gradient boosting machine—to predict biweekly NH3 concentrations and downscale them to daily estimates across the United States during 2017–2022. The models integrate NH3 column concentrations, meteorological variables, land cover data, livestock information, and ground-based measurements. Among the models, DNN showed superior performance in both spatial cross-validation and independent testing, achieving a correlation coefficient of 0.79, a root mean square error of 0.98 µg/m³, and an index of agreement of 0.83— effectively capturing fine-scale spatial variations at a 9 km resolution. Shapley additive explanations analysis identified temporal dynamic factors—such as day of year and meteorological variables—as key predictors, along with land cover and cattle density, highlighting the model’s ability to support the temporal downscaling of NH3 from biweekly to daily scale. When applied to the UK, the model demonstrated its potential for spatial transferability via the leave-one station-out and leave-one year-out cross validations. These findings highlight the ability of machine learning in bridging temporal gaps and generating high-resolution daily NH3 estimates.
期刊介绍:
The Journal of Hazardous Materials serves as a global platform for promoting cutting-edge research in the field of Environmental Science and Engineering. Our publication features a wide range of articles, including full-length research papers, review articles, and perspectives, with the aim of enhancing our understanding of the dangers and risks associated with various materials concerning public health and the environment. It is important to note that the term "environmental contaminants" refers specifically to substances that pose hazardous effects through contamination, while excluding those that do not have such impacts on the environment or human health. Moreover, we emphasize the distinction between wastes and hazardous materials in order to provide further clarity on the scope of the journal. We have a keen interest in exploring specific compounds and microbial agents that have adverse effects on the environment.