Remote Monitoring of NH<sub>3</sub>-N Content in Small-Sized Inland Waterbody Based on Low and Medium Resolution Multi-Source Remote Sensing Image Fusion
In applying quantitative remote sensing in water quality monitoring for small inland rivers, the time-frequency of monitoring dramatically impacts the accuracy of time-spatial changes estimates of the water quality parameters. Due to the limitation of satellite sensor design and the influence of atm...
Main Authors: | , , , |
---|---|
Format: | Article |
Language: | English |
Published: |
MDPI AG
2022-10-01
|
Series: | Water |
Subjects: | |
Online Access: | https://www.mdpi.com/2073-4441/14/20/3287 |
_version_ | 1797469556542799872 |
---|---|
author | Jian Li Meiru Ke Yurong Ma Jian Cui |
author_facet | Jian Li Meiru Ke Yurong Ma Jian Cui |
author_sort | Jian Li |
collection | DOAJ |
description | In applying quantitative remote sensing in water quality monitoring for small inland rivers, the time-frequency of monitoring dramatically impacts the accuracy of time-spatial changes estimates of the water quality parameters. Due to the limitation of satellite sensor design and the influence of atmospheric conditions, the number of spatiotemporal dynamic monitoring images of water quality parameters is insufficient. Meanwhile, MODIS and other high temporal resolution images’ spatial resolution is too low to effectively extract small inland river boundaries. To solve the problem, many researchers used Spatio-temporal fusion models in multisource data remote sensing monitoring of ground features. The wildly used Spatio-temporal fusion models, such as FSDAF (flexible spatial-temporal data fusion), have poor performance in heterogeneous changes of ground objects. We proposed a spatiotemporal fusion algorithm SR-FSDAF (Super-resolution based flexible spatiotemporal data fusion) to solve the problem. Based on the FSDAF, it added ESPCN to reconstruct the spatial change prediction image, so as to obtain better prediction results for heterogeneous changes. Both qualitative and quantitative evaluation results showed that our fusion algorithm obtained better results. We compared the band sensitivity of the images before and after fusion to find out that the sensitive band combination of NH<sub>3</sub>-N has not changed, which proved that the fusion method can be used to improve the time-frequency of NH<sub>3</sub>-N inversion. After the fusion, we compared the accuracy of linear regression and random forest inversion models and selected the random forest model with better accuracy to predict the NH<sub>3</sub>-N concentration. The inversion accuracy of NH<sub>3</sub>-N was as follows: the R<sup>2</sup> was 0.75, the MAPE was 23.7% and the RMSE was 0.15. The overall concentration change trend of NH<sub>3</sub>-N in the study area was high-water period < water-stable period < low water period. NH<sub>3</sub>-N pollution was serious in some reaches. |
first_indexed | 2024-03-09T19:22:59Z |
format | Article |
id | doaj.art-8186954c31584b10aef02d3f442e28ba |
institution | Directory Open Access Journal |
issn | 2073-4441 |
language | English |
last_indexed | 2024-03-09T19:22:59Z |
publishDate | 2022-10-01 |
publisher | MDPI AG |
record_format | Article |
series | Water |
spelling | doaj.art-8186954c31584b10aef02d3f442e28ba2023-11-24T03:12:57ZengMDPI AGWater2073-44412022-10-011420328710.3390/w14203287Remote Monitoring of NH<sub>3</sub>-N Content in Small-Sized Inland Waterbody Based on Low and Medium Resolution Multi-Source Remote Sensing Image FusionJian Li0Meiru Ke1Yurong Ma2Jian Cui3School of the Geo-Science & Technology, Zhengzhou University, Zhengzhou 450001, ChinaSchool of Water Conservancy Science & Engineering, Zhengzhou University, Zhengzhou 450001, ChinaLibrary of Zhengzhou University, Zhengzhou University, Zhengzhou 450001, ChinaHenan Institute of Regional Geological Survey, Zhengzhou 450001, ChinaIn applying quantitative remote sensing in water quality monitoring for small inland rivers, the time-frequency of monitoring dramatically impacts the accuracy of time-spatial changes estimates of the water quality parameters. Due to the limitation of satellite sensor design and the influence of atmospheric conditions, the number of spatiotemporal dynamic monitoring images of water quality parameters is insufficient. Meanwhile, MODIS and other high temporal resolution images’ spatial resolution is too low to effectively extract small inland river boundaries. To solve the problem, many researchers used Spatio-temporal fusion models in multisource data remote sensing monitoring of ground features. The wildly used Spatio-temporal fusion models, such as FSDAF (flexible spatial-temporal data fusion), have poor performance in heterogeneous changes of ground objects. We proposed a spatiotemporal fusion algorithm SR-FSDAF (Super-resolution based flexible spatiotemporal data fusion) to solve the problem. Based on the FSDAF, it added ESPCN to reconstruct the spatial change prediction image, so as to obtain better prediction results for heterogeneous changes. Both qualitative and quantitative evaluation results showed that our fusion algorithm obtained better results. We compared the band sensitivity of the images before and after fusion to find out that the sensitive band combination of NH<sub>3</sub>-N has not changed, which proved that the fusion method can be used to improve the time-frequency of NH<sub>3</sub>-N inversion. After the fusion, we compared the accuracy of linear regression and random forest inversion models and selected the random forest model with better accuracy to predict the NH<sub>3</sub>-N concentration. The inversion accuracy of NH<sub>3</sub>-N was as follows: the R<sup>2</sup> was 0.75, the MAPE was 23.7% and the RMSE was 0.15. The overall concentration change trend of NH<sub>3</sub>-N in the study area was high-water period < water-stable period < low water period. NH<sub>3</sub>-N pollution was serious in some reaches.https://www.mdpi.com/2073-4441/14/20/3287NH<sub>3</sub>-Nwater quality monitoringspatiotemporal fusion modelLandsat-8MODISremote sensing |
spellingShingle | Jian Li Meiru Ke Yurong Ma Jian Cui Remote Monitoring of NH<sub>3</sub>-N Content in Small-Sized Inland Waterbody Based on Low and Medium Resolution Multi-Source Remote Sensing Image Fusion Water NH<sub>3</sub>-N water quality monitoring spatiotemporal fusion model Landsat-8 MODIS remote sensing |
title | Remote Monitoring of NH<sub>3</sub>-N Content in Small-Sized Inland Waterbody Based on Low and Medium Resolution Multi-Source Remote Sensing Image Fusion |
title_full | Remote Monitoring of NH<sub>3</sub>-N Content in Small-Sized Inland Waterbody Based on Low and Medium Resolution Multi-Source Remote Sensing Image Fusion |
title_fullStr | Remote Monitoring of NH<sub>3</sub>-N Content in Small-Sized Inland Waterbody Based on Low and Medium Resolution Multi-Source Remote Sensing Image Fusion |
title_full_unstemmed | Remote Monitoring of NH<sub>3</sub>-N Content in Small-Sized Inland Waterbody Based on Low and Medium Resolution Multi-Source Remote Sensing Image Fusion |
title_short | Remote Monitoring of NH<sub>3</sub>-N Content in Small-Sized Inland Waterbody Based on Low and Medium Resolution Multi-Source Remote Sensing Image Fusion |
title_sort | remote monitoring of nh sub 3 sub n content in small sized inland waterbody based on low and medium resolution multi source remote sensing image fusion |
topic | NH<sub>3</sub>-N water quality monitoring spatiotemporal fusion model Landsat-8 MODIS remote sensing |
url | https://www.mdpi.com/2073-4441/14/20/3287 |
work_keys_str_mv | AT jianli remotemonitoringofnhsub3subncontentinsmallsizedinlandwaterbodybasedonlowandmediumresolutionmultisourceremotesensingimagefusion AT meiruke remotemonitoringofnhsub3subncontentinsmallsizedinlandwaterbodybasedonlowandmediumresolutionmultisourceremotesensingimagefusion AT yurongma remotemonitoringofnhsub3subncontentinsmallsizedinlandwaterbodybasedonlowandmediumresolutionmultisourceremotesensingimagefusion AT jiancui remotemonitoringofnhsub3subncontentinsmallsizedinlandwaterbodybasedonlowandmediumresolutionmultisourceremotesensingimagefusion |