Scholarlink

<p>Rapid urbanization is a growing challenge for sustainable land management. Continued conversion of agricultural and forest lands to urban areas may negatively affect ecological stability, food availability, and long-term environmental integrity. For this reason, it is critical to understand historical changes in land use/land cover (LULC) and to predict future trends to enable informed decision-making. This study used the Landsat image dataset to analyze LULC dynamics for 1985, 2005, and 2025, and then used the CA–Markov model to project future land-use patterns to 2045. LULC classifications were generated using supervised Random Forest (RF) classification, and the accuracy assessments indicated very strong performance, with overall accuracies ranging from 88.5% to 94.4% and Kappa coefficients greater than 0.86. Change detection analyses identified continued increases in urban land, with an expansion from approximately 1.01% in 1985 to 5.25% in 2025. Conversely, forest and cropland areas experienced a gradual decline during the study period. Validation of the simulated 2025 LULC map showed excellent correspondence with the actual 2025 map (overall accuracy = 0.911; Kappa = 0.837), providing assurance of the modeling approach’s reliability. The projected 2045 LULC map indicates additional urban expansion to 9.52% of the study area, primarily at the expense of forest and cropland. These results indicate that urban expansion will continue to drive landscape transformations in the region. The results provide valuable information for land-use planners and environmental managers on LULC changes, and future studies should incorporate socioeconomic factors and scenario-based modeling to support sustainability planning initiatives. Additionally, applying machine-learning classification in conjunction with CA-Markov modeling provides a practical methodology for long-term LULC projections. Future applications of these methodologies could help decision-makers identify areas susceptible to rapid urban encroachment and ecosystem disruption.</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

Economic implications of urban expansion and land use transformation: a remote sensing and CA–Markov approach

  • 期刊论文

  • 发表日期：

    2026年7月26日

查看全文
  • Xiangmei Xue,
  • Zhen Bai

摘要

Rapid urbanization is a growing challenge for sustainable land management. Continued conversion of agricultural and forest lands to urban areas may negatively affect ecological stability, food availability, and long-term environmental integrity. For this reason, it is critical to understand historical changes in land use/land cover (LULC) and to predict future trends to enable informed decision-making. This study used the Landsat image dataset to analyze LULC dynamics for 1985, 2005, and 2025, and then used the CA–Markov model to project future land-use patterns to 2045. LULC classifications were generated using supervised Random Forest (RF) classification, and the accuracy assessments indicated very strong performance, with overall accuracies ranging from 88.5% to 94.4% and Kappa coefficients greater than 0.86. Change detection analyses identified continued increases in urban land, with an expansion from approximately 1.01% in 1985 to 5.25% in 2025. Conversely, forest and cropland areas experienced a gradual decline during the study period. Validation of the simulated 2025 LULC map showed excellent correspondence with the actual 2025 map (overall accuracy = 0.911; Kappa = 0.837), providing assurance of the modeling approach’s reliability. The projected 2045 LULC map indicates additional urban expansion to 9.52% of the study area, primarily at the expense of forest and cropland. These results indicate that urban expansion will continue to drive landscape transformations in the region. The results provide valuable information for land-use planners and environmental managers on LULC changes, and future studies should incorporate socioeconomic factors and scenario-based modeling to support sustainability planning initiatives. Additionally, applying machine-learning classification in conjunction with CA-Markov modeling provides a practical methodology for long-term LULC projections. Future applications of these methodologies could help decision-makers identify areas susceptible to rapid urban encroachment and ecosystem disruption.