Scholarlink

<p>Landscape architecture evolution presents challenges due to its varying temporal and spatial scales. To address this, a prediction method based on the MCE-CA–Markov model is proposed, aiming to analyze the evolution law of landscape patterns. This method integrates cellular automata for simulating spatial complexity, a Markov model for predicting landscape type transformation probabilities, and an MCE model for determining transformation rules and priorities. Using the transfer matrix, the dynamic attitude and pattern index of landscape land use were analyzed. Based on Landsat remote sensing images, the model was applied to study the evolution of landscape patterns in Suzhou, China, with forecasts extending to 2026. Results indicate significant changes in road landscapes from 2015 to 2023, followed by architectural and agricultural landscapes. By 2026, architectural, built, and green landscapes will increase, while agricultural and forest landscapes will decrease, trending towards stability. This study combines MCE and CA-Markov models for multi-standard, dynamic prediction, enhancing understanding of historical and cultural backgrounds, promoting sustainable ecological development, improving urban planning and design scientificity, and raising public awareness and protection of natural and cultural values.</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

Analysis of landscape pattern evolution based on MCE-CA–Markov model

  • 期刊论文

  • 发表日期：

    2025年6月6日

查看全文
  • Junli Sha

摘要

Landscape architecture evolution presents challenges due to its varying temporal and spatial scales. To address this, a prediction method based on the MCE-CA–Markov model is proposed, aiming to analyze the evolution law of landscape patterns. This method integrates cellular automata for simulating spatial complexity, a Markov model for predicting landscape type transformation probabilities, and an MCE model for determining transformation rules and priorities. Using the transfer matrix, the dynamic attitude and pattern index of landscape land use were analyzed. Based on Landsat remote sensing images, the model was applied to study the evolution of landscape patterns in Suzhou, China, with forecasts extending to 2026. Results indicate significant changes in road landscapes from 2015 to 2023, followed by architectural and agricultural landscapes. By 2026, architectural, built, and green landscapes will increase, while agricultural and forest landscapes will decrease, trending towards stability. This study combines MCE and CA-Markov models for multi-standard, dynamic prediction, enhancing understanding of historical and cultural backgrounds, promoting sustainable ecological development, improving urban planning and design scientificity, and raising public awareness and protection of natural and cultural values.