Land-use/cover change and future prediction by integrating the ML techniques of random forest and CA-Markov chain model of the Ganges alluvial tract of Eastern India.
Saved in:
| Title: | Land-use/cover change and future prediction by integrating the ML techniques of random forest and CA-Markov chain model of the Ganges alluvial tract of Eastern India. |
|---|---|
| Authors: | Roy, Kailash Chandra1 (AUTHOR), Soren, David Durjoy Lal1 (AUTHOR), Biswas, Brototi1 (AUTHOR) brototibiswas@gmail.com |
| Source: | Environment, Development & Sustainability. Jul2026, Vol. 28 Issue 7, p15635-15662. 28p. |
| Subject Terms: | *Land use, *Random forest algorithms, *Alluvial plains, *Land use planning, *Cellular automata, *Watersheds, *Ecological impact |
| Geographic Terms: | India |
| Abstract: | Land-use-land-cover (LULC) transformations are a major worldwide challenge, thereby LULC study and future prediction is an essential part of decision-making for sustainable land use planning and extenuating environmental impacts. The watershed area of the Ajay River, which is located on the eastern fringe of the Chotanagpur plateau, belonging to the Ganges alluvial tract of India has been selected for the present study of LULC dynamics. The purpose of this study is to investigate changes in LULC in the Ajay River basin between 1992 and 2022 using the Random Forest (RF) classifier algorithm on the Google earth engine (GEE) environment and simulate future (LULC) maps. Landsat-5 (TM) in 1992, 2002, and 2011 and Landsat-8 (OLI) in 2022 were used for preparing the LULC maps. Five major land-use categories were classified: water body, vegetation, agriculture, bare-land/sandbar, and built-up. For enhanced accuracy in LULC classification, spectral indicators like NDVI, NDBI, NDWI, NDMI, and BSI were also computed. The RF classifier achieved overall accuracy of 88.29%, 90.24%, 90.73%, and 89.27% for LULC maps in 1992, 2002, 2011, and 2022 respectively, with kappa values of 0.85, 0.87, 0.88, and 0.86, respectively. According to the findings of the study the highest increase was found in agriculture land (18.39%) while the highest decrease was found in the bare land or sandbars (16.29%) class between the periods of 1992 and 2022. By utilizing previous (LULC) maps, the hybrid cellular automata-Markov chain model was applied to simulate the (LULC) trends in 2032, and 2042 with the help of a transition probability matrix using altitude, closeness to the river, road, settlement, agricultural, and barren terrain as driver variables. The validity of the prediction was proved by comparing the simulated and observed (LULC) maps of 2022 and it showed excellent results. It showed values for K no of 0.92, K location of 0.88, K location Strata of 0.88, and K standard of 0.85. Based on the findings of the LULC simulation conducted between 2022 and 2042, it has been found that the highest land use increase will be in the built-up areas category (347. 55 km2) while barren land category will face the highest decrease (696.93 km2). It is hoped that the results of this research will ensure sustainable development growth with a balance between conservation and land-use for future planning and management, especially in the watershed area of the Ajay River. [ABSTRACT FROM AUTHOR] |
| Database: | Energy & Power Source |
|
Full text is not displayed to guests.
Login for full access.
|
|
| FullText | Links: – Type: pdflink Text: Availability: 1 |
|---|---|
| Header | DbId: enr DbLabel: Energy & Power Source An: 194936938 AccessLevel: 6 PubType: Academic Journal PubTypeId: academicJournal PreciseRelevancyScore: 0 |
| IllustrationInfo | |
| Items | – Name: Title Label: Title Group: Ti Data: Land-use/cover change and future prediction by integrating the ML techniques of random forest and CA-Markov chain model of the Ganges alluvial tract of Eastern India. – Name: Author Label: Authors Group: Au Data: <searchLink fieldCode="AR" term="%22Roy%2C+Kailash+Chandra%22">Roy, Kailash Chandra</searchLink><relatesTo>1</relatesTo> (AUTHOR)<br /><searchLink fieldCode="AR" term="%22Soren%2C+David+Durjoy+Lal%22">Soren, David Durjoy Lal</searchLink><relatesTo>1</relatesTo> (AUTHOR)<br /><searchLink fieldCode="AR" term="%22Biswas%2C+Brototi%22">Biswas, Brototi</searchLink><relatesTo>1</relatesTo> (AUTHOR)<i> brototibiswas@gmail.com</i> – Name: TitleSource Label: Source Group: Src Data: <searchLink fieldCode="JN" term="%22Environment%2C+Development+%26+Sustainability%22">Environment, Development & Sustainability</searchLink>. Jul2026, Vol. 28 Issue 7, p15635-15662. 28p. – Name: Subject Label: Subject Terms Group: Su Data: *<searchLink fieldCode="DE" term="%22Land+use%22">Land use</searchLink><br />*<searchLink fieldCode="DE" term="%22Random+forest+algorithms%22">Random forest algorithms</searchLink><br />*<searchLink fieldCode="DE" term="%22Alluvial+plains%22">Alluvial plains</searchLink><br />*<searchLink fieldCode="DE" term="%22Land+use+planning%22">Land use planning</searchLink><br />*<searchLink fieldCode="DE" term="%22Cellular+automata%22">Cellular automata</searchLink><br />*<searchLink fieldCode="DE" term="%22Watersheds%22">Watersheds</searchLink><br />*<searchLink fieldCode="DE" term="%22Ecological+impact%22">Ecological impact</searchLink> – Name: SubjectGeographic Label: Geographic Terms Group: Su Data: <searchLink fieldCode="DE" term="%22India%22">India</searchLink> – Name: Abstract Label: Abstract Group: Ab Data: Land-use-land-cover (LULC) transformations are a major worldwide challenge, thereby LULC study and future prediction is an essential part of decision-making for sustainable land use planning and extenuating environmental impacts. The watershed area of the Ajay River, which is located on the eastern fringe of the Chotanagpur plateau, belonging to the Ganges alluvial tract of India has been selected for the present study of LULC dynamics. The purpose of this study is to investigate changes in LULC in the Ajay River basin between 1992 and 2022 using the Random Forest (RF) classifier algorithm on the Google earth engine (GEE) environment and simulate future (LULC) maps. Landsat-5 (TM) in 1992, 2002, and 2011 and Landsat-8 (OLI) in 2022 were used for preparing the LULC maps. Five major land-use categories were classified: water body, vegetation, agriculture, bare-land/sandbar, and built-up. For enhanced accuracy in LULC classification, spectral indicators like NDVI, NDBI, NDWI, NDMI, and BSI were also computed. The RF classifier achieved overall accuracy of 88.29%, 90.24%, 90.73%, and 89.27% for LULC maps in 1992, 2002, 2011, and 2022 respectively, with kappa values of 0.85, 0.87, 0.88, and 0.86, respectively. According to the findings of the study the highest increase was found in agriculture land (18.39%) while the highest decrease was found in the bare land or sandbars (16.29%) class between the periods of 1992 and 2022. By utilizing previous (LULC) maps, the hybrid cellular automata-Markov chain model was applied to simulate the (LULC) trends in 2032, and 2042 with the help of a transition probability matrix using altitude, closeness to the river, road, settlement, agricultural, and barren terrain as driver variables. The validity of the prediction was proved by comparing the simulated and observed (LULC) maps of 2022 and it showed excellent results. It showed values for K no of 0.92, K location of 0.88, K location Strata of 0.88, and K standard of 0.85. Based on the findings of the LULC simulation conducted between 2022 and 2042, it has been found that the highest land use increase will be in the built-up areas category (347. 55 km2) while barren land category will face the highest decrease (696.93 km2). It is hoped that the results of this research will ensure sustainable development growth with a balance between conservation and land-use for future planning and management, especially in the watershed area of the Ajay River. [ABSTRACT FROM AUTHOR] |
| PLink | https://search.ebscohost.com/login.aspx?direct=true&site=eds-live&db=enr&AN=194936938 |
| RecordInfo | BibRecord: BibEntity: Identifiers: – Type: doi Value: 10.1007/s10668-024-05545-x Languages: – Code: eng Text: English PhysicalDescription: Pagination: PageCount: 28 StartPage: 15635 Subjects: – SubjectFull: Land use Type: general – SubjectFull: Random forest algorithms Type: general – SubjectFull: Alluvial plains Type: general – SubjectFull: Land use planning Type: general – SubjectFull: Cellular automata Type: general – SubjectFull: Watersheds Type: general – SubjectFull: Ecological impact Type: general – SubjectFull: India Type: general Titles: – TitleFull: Land-use/cover change and future prediction by integrating the ML techniques of random forest and CA-Markov chain model of the Ganges alluvial tract of Eastern India. Type: main BibRelationships: HasContributorRelationships: – PersonEntity: Name: NameFull: Roy, Kailash Chandra – PersonEntity: Name: NameFull: Soren, David Durjoy Lal – PersonEntity: Name: NameFull: Biswas, Brototi IsPartOfRelationships: – BibEntity: Dates: – D: 01 M: 07 Text: Jul2026 Type: published Y: 2026 Identifiers: – Type: issn-print Value: 1387585X Numbering: – Type: volume Value: 28 – Type: issue Value: 7 Titles: – TitleFull: Environment, Development & Sustainability Type: main |
| ResultId | 1 |
