Scholarlink

<p>Urban growth (UG) prediction is essential for risk assessment and urban planning. This paper proposes a probabilistic hexagonal cellular automata model integrating land-cover and socio-economic indicators to predict UG over five-year intervals. Indicators are defined on a regional scale using a hexagonal grid, with time-varying values derived from satellite imagery and publicly available databases. Cells are classified by urbanization level. The model is calibrated using predictor values from 2001, 2006, and 2011, with growth predicted in 2006, 2011, and 2016; performance is validated on a held-out test set of 200 cells from 2006, 2011, and 2016. Applied to the Houston urban area, the model achieves 81% average accuracy. Results indicate that both land-cover and socio-economic indicators significantly influence UG, with relevance shifting across urbanization levels. Distinct growth mechanisms are revealed, including greater neighborhood influence in low-density areas and the dominant role of employment and income in highly urbanized zones.</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

The importance of socio-economic indicators in predicting urban growth

  • 期刊论文

  • 出版模式：

    开放获取

  • 发表日期：

    2026年6月18日

查看全文
  • Alessandro Contento,
  • Jessica Boakye,
  • Lorena Fiorini,
  • Alessandro Marucci,
  • Paolo Gardoni

摘要

Urban growth (UG) prediction is essential for risk assessment and urban planning. This paper proposes a probabilistic hexagonal cellular automata model integrating land-cover and socio-economic indicators to predict UG over five-year intervals. Indicators are defined on a regional scale using a hexagonal grid, with time-varying values derived from satellite imagery and publicly available databases. Cells are classified by urbanization level. The model is calibrated using predictor values from 2001, 2006, and 2011, with growth predicted in 2006, 2011, and 2016; performance is validated on a held-out test set of 200 cells from 2006, 2011, and 2016. Applied to the Houston urban area, the model achieves 81% average accuracy. Results indicate that both land-cover and socio-economic indicators significantly influence UG, with relevance shifting across urbanization levels. Distinct growth mechanisms are revealed, including greater neighborhood influence in low-density areas and the dominant role of employment and income in highly urbanized zones.