Scholarlink

<p>Along with rapid economic growth, the fast pace of urbanization poses significant socio-economic and environmental challenges in many Indian cities. Leveraging geospatial tools and satellite datasets can aid in sustainable urban planning efforts. To this end, this study presents the application of robust machine learning techniques to analyze Nashik's urban growth dynamics from 1992 to 2042. Maximum Likelihood Classification algorithm is applied over multiple LANDSAT satellite datasets to develop Land Cover Maps for the&#xa0;Nashik region during 1992, 2003, 2013, and 2022. LC maps classified the regional landscape into Water body, Built-Up, Vegetation, and Bareland classes. The map’s accuracy is confirmed by Kappa statistics ranging from 0.89 to 0.91 across the four periods. LC change detection revealed a&#xa0;significant expansion of the built-up area, from 34.6 sq.km in 1992 to 115.5&#xa0;sq.km in 2022, with a peak growth rate of 80.36% during 1992–2003. In comparison, vegetation coverage decreased by 38.4% (52.9&#xa0;sq.km), and Barelands declined by 17.0% (27.4&#xa0;sq.km) from 1992 to 2022. Urban population also increased from 0.77 million to 2.18 million during this period. Further, the Multi-Layer Perceptron—Markov Chain Analysis (MLP-MCA) model, trained with 2002 and 2012 LC maps and urban growth drivers like ground elevation, slope, road, water, built-up, and vegetation proximity predicted the 2032 and 2042 LC Maps. With a validation accuracy of 94.33%, the MLP-MCA model predicts a 19.65% rise in Built-Up areas to 138&#xa0;sq.km by 2032 and a further 15.70% increase to 160 sq.km by 2042, signalling significant declines in Vegetation and Barelands coverage. Moreover, regression predicts a 51% rise in population to 3.29 million by 2032 and a 68% increase to 3.69 million by 2042 from 2022 levels. These findings are valuable for shaping sustainable urban growth and environmental management strategies in the region.</p>

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Analyzing the Spatiotemporal Urban Growth Dynamics in Nashik, India from 1992 to 2042 Using MLC and MLP-MCA Algorithms

  • 期刊论文

  • 发表日期：

    2025年5月27日

查看全文
  • Kratika Sharma,
  • Ritu Tiwari,
  • Arun Kumar Wadhwani,
  • Shobhit Chaturvedi

摘要

Along with rapid economic growth, the fast pace of urbanization poses significant socio-economic and environmental challenges in many Indian cities. Leveraging geospatial tools and satellite datasets can aid in sustainable urban planning efforts. To this end, this study presents the application of robust machine learning techniques to analyze Nashik's urban growth dynamics from 1992 to 2042. Maximum Likelihood Classification algorithm is applied over multiple LANDSAT satellite datasets to develop Land Cover Maps for the Nashik region during 1992, 2003, 2013, and 2022. LC maps classified the regional landscape into Water body, Built-Up, Vegetation, and Bareland classes. The map’s accuracy is confirmed by Kappa statistics ranging from 0.89 to 0.91 across the four periods. LC change detection revealed a significant expansion of the built-up area, from 34.6 sq.km in 1992 to 115.5 sq.km in 2022, with a peak growth rate of 80.36% during 1992–2003. In comparison, vegetation coverage decreased by 38.4% (52.9 sq.km), and Barelands declined by 17.0% (27.4 sq.km) from 1992 to 2022. Urban population also increased from 0.77 million to 2.18 million during this period. Further, the Multi-Layer Perceptron—Markov Chain Analysis (MLP-MCA) model, trained with 2002 and 2012 LC maps and urban growth drivers like ground elevation, slope, road, water, built-up, and vegetation proximity predicted the 2032 and 2042 LC Maps. With a validation accuracy of 94.33%, the MLP-MCA model predicts a 19.65% rise in Built-Up areas to 138 sq.km by 2032 and a further 15.70% increase to 160 sq.km by 2042, signalling significant declines in Vegetation and Barelands coverage. Moreover, regression predicts a 51% rise in population to 3.29 million by 2032 and a 68% increase to 3.69 million by 2042 from 2022 levels. These findings are valuable for shaping sustainable urban growth and environmental management strategies in the region.