Modeling soil erosion susceptibility using GIS-based different machine learning algorithms in monsoon dominated diversified landscape in India

Severe rainfall causes soil to erode, which has far-reaching effects on the natural environment and anthropogenic activity. It is regulated by the interplay of “lithology, orography, hydrography, land use, and vegetation” and is dependent on rainfall distribution (e.g., “intensity, duration, the cum...

Celý popis

Uloženo v:
Podrobná bibliografie
Vydáno v:Modeling earth systems and environment Ročník 9; číslo 2; s. 2927 - 2942
Hlavní autoři: Chakrabortty, Rabin, Pal, Subodh Chandra
Médium: Journal Article
Jazyk:angličtina
Vydáno: Cham Springer International Publishing 01.06.2023
Springer Nature B.V
Témata:
Algorithms
Anthropogenic factors
Chemistry and Earth Sciences
Computer Science
Decision making
Earth and Environmental Science
Earth Sciences
Earth System Sciences
Ecosystems
Environment
Geographical information systems
Hydrography
Land use
Learning algorithms
Lithology
Machine learning
Math. Appl. in Environmental Science
Mathematical Applications in the Physical Sciences
Modelling
Monsoons
Natural environment
Original Article
Orography
Physics
Precipitation
Rainfall
Rainfall distribution
Soil erosion
Statistics for Engineering
Wind
India
GIS
Machine learning method
Erosion susceptibility
MaxEnt
Monsoonal climate
ISSN:2363-6203, 2363-6211
On-line přístup:Získat plný text
Tagy: Přidat tag
Žádné tagy, Buďte první, kdo vytvoří štítek k tomuto záznamu!
Popis
Shrnutí:Severe rainfall causes soil to erode, which has far-reaching effects on the natural environment and anthropogenic activity. It is regulated by the interplay of “lithology, orography, hydrography, land use, and vegetation” and is dependent on rainfall distribution (e.g., “intensity, duration, the cumulative per event”). More than 75 percent of India's yearly rainfall is brought on by the monsoons, making India a prime example of a country with a monsoonal climate. The monsoon season is the apex of all erosional rainfall. Therefore, changes in the frequency and volume of the monsoon's rain will result in corresponding shifts in the season's erosivity patterns. So, identifying vulnerable regions through proper modeling is one of the essential tasks in soil erosion research. In this study, the MaxEnt, RF, SVM, and SLR model has been considered for estimating erosion susceptibility. The overall result has been confirmed, and we found that the MaxEnt model is optimal to other models in this study in terms of performance. This kind of research is helpful to the decision-maker in determining whether or not to use these strategies for preventing soil erosion and achieving the desired results. The main novelty of this work is the use of an appropriate machine learning method to evaluate soil erosion susceptibility for the first time in the study area, taking into account the most significant feasible number of causal factors.
Bibliografie:ObjectType-Article-1
SourceType-Scholarly Journals-1
ObjectType-Feature-2
content type line 14
ISSN:2363-6203
2363-6211
DOI:10.1007/s40808-022-01681-3