Mapping of Rill Erosion of the Middle Volga (Russia) Region Using Deep Neural Network
Soil erosion worldwide is an intense, poorly controlled process. In many respects, this is a consequence of the lack of up-to-date high-resolution erosion maps. All over the world, the problem of insufficient information is solved in different ways, mainly on a point-by-point basis, within local are...
Main Author: | |
---|---|
Format: | Article |
Language: | English |
Published: |
MDPI AG
2022-03-01
|
Series: | ISPRS International Journal of Geo-Information |
Subjects: | |
Online Access: | https://www.mdpi.com/2220-9964/11/3/197 |
_version_ | 1797471081849683968 |
---|---|
author | Artur Gafurov |
author_facet | Artur Gafurov |
author_sort | Artur Gafurov |
collection | DOAJ |
description | Soil erosion worldwide is an intense, poorly controlled process. In many respects, this is a consequence of the lack of up-to-date high-resolution erosion maps. All over the world, the problem of insufficient information is solved in different ways, mainly on a point-by-point basis, within local areas. Extrapolation of the results obtained locally to a more extensive territory produces inevitable uncertainties and errors. For the anthropogenic-developed part of Russia, this problem is especially urgent because the assessment of the intensity of erosion processes, even with the use of erosion models, does not reach the necessary scale due to the lack of all the required global large-scale remote sensing data and the complexity of considering regional features of erosion processes over such vast areas. This study aims to propose a new methodology for large-scale automated mapping of rill erosion networks based on Sentinel-2 data. A LinkNet deep neural network with a DenseNet encoder was used to solve the problem of automated rill erosion mapping. The recognition results for the study area of more than 345,000 sq. km were summarized to a grid of 3037 basins and analyzed to assess the relationship with the main natural-anthropogenic factors. Generalized additive models (GAM) were used to model the dependency of rill erosion density to explore complex relationships. A complex nonlinear relationship between erosion processes and topographic, meteorological, geomorphological, and anthropogenic factors was shown. |
first_indexed | 2024-03-09T19:44:33Z |
format | Article |
id | doaj.art-91027a1f2c5145b4bd079d5f74f92100 |
institution | Directory Open Access Journal |
issn | 2220-9964 |
language | English |
last_indexed | 2024-03-09T19:44:33Z |
publishDate | 2022-03-01 |
publisher | MDPI AG |
record_format | Article |
series | ISPRS International Journal of Geo-Information |
spelling | doaj.art-91027a1f2c5145b4bd079d5f74f921002023-11-24T01:28:42ZengMDPI AGISPRS International Journal of Geo-Information2220-99642022-03-0111319710.3390/ijgi11030197Mapping of Rill Erosion of the Middle Volga (Russia) Region Using Deep Neural NetworkArtur Gafurov0Institute of Environmental Sciences, Kazan Federal University, 420008 Kazan, RussiaSoil erosion worldwide is an intense, poorly controlled process. In many respects, this is a consequence of the lack of up-to-date high-resolution erosion maps. All over the world, the problem of insufficient information is solved in different ways, mainly on a point-by-point basis, within local areas. Extrapolation of the results obtained locally to a more extensive territory produces inevitable uncertainties and errors. For the anthropogenic-developed part of Russia, this problem is especially urgent because the assessment of the intensity of erosion processes, even with the use of erosion models, does not reach the necessary scale due to the lack of all the required global large-scale remote sensing data and the complexity of considering regional features of erosion processes over such vast areas. This study aims to propose a new methodology for large-scale automated mapping of rill erosion networks based on Sentinel-2 data. A LinkNet deep neural network with a DenseNet encoder was used to solve the problem of automated rill erosion mapping. The recognition results for the study area of more than 345,000 sq. km were summarized to a grid of 3037 basins and analyzed to assess the relationship with the main natural-anthropogenic factors. Generalized additive models (GAM) were used to model the dependency of rill erosion density to explore complex relationships. A complex nonlinear relationship between erosion processes and topographic, meteorological, geomorphological, and anthropogenic factors was shown.https://www.mdpi.com/2220-9964/11/3/197rillsoil erosionneural networksremote sensingrelationship modeling |
spellingShingle | Artur Gafurov Mapping of Rill Erosion of the Middle Volga (Russia) Region Using Deep Neural Network ISPRS International Journal of Geo-Information rill soil erosion neural networks remote sensing relationship modeling |
title | Mapping of Rill Erosion of the Middle Volga (Russia) Region Using Deep Neural Network |
title_full | Mapping of Rill Erosion of the Middle Volga (Russia) Region Using Deep Neural Network |
title_fullStr | Mapping of Rill Erosion of the Middle Volga (Russia) Region Using Deep Neural Network |
title_full_unstemmed | Mapping of Rill Erosion of the Middle Volga (Russia) Region Using Deep Neural Network |
title_short | Mapping of Rill Erosion of the Middle Volga (Russia) Region Using Deep Neural Network |
title_sort | mapping of rill erosion of the middle volga russia region using deep neural network |
topic | rill soil erosion neural networks remote sensing relationship modeling |
url | https://www.mdpi.com/2220-9964/11/3/197 |
work_keys_str_mv | AT arturgafurov mappingofrillerosionofthemiddlevolgarussiaregionusingdeepneuralnetwork |