Scholarlink

Green spaces are critical for mitigating urban heatwaves and enhancing urban climate adaptation. Urban parks play an important role in reducing the urban heat island (UHI) effect by providing localized cooling. This study investigates the impact of various land cover types on air temperature in Queen Elizabeth Olympic Park (QEOP) using data from 15 bespoke IoT real-time heat sensors and machine learning model XGBoost. Open Street Map data and QGIS were used to classify land cover and determine land cover fractions around each sensor location. XGBoost models were used to quantify the relative impact of different land cover types on local air temperature. The integration of IoT sensor data with machine learning model enabled detailed microclimate assessments, identifying how different land cover types (paved areas, buildings, grass, bare soil, and water) affect local temperature patterns. This study provides insights for the intelligent operation and maintenance of green spaces and informs urban design strategies aimed at reducing heat exposure and enhancing environmental sustainability.

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

Evaluating Urban Green Spaces with IoT Sensors and Machine Learning: Land Cover Impacts on Microclimate for Sustainable City Planning

  • 会议论文

  • 发表日期：

    2026年

查看全文
  • Dongyi Ma,
  • Andrew Hudson-Smith,
  • Martin de Jode,
  • Valerio Signorelli

摘要

Green spaces are critical for mitigating urban heatwaves and enhancing urban climate adaptation. Urban parks play an important role in reducing the urban heat island (UHI) effect by providing localized cooling. This study investigates the impact of various land cover types on air temperature in Queen Elizabeth Olympic Park (QEOP) using data from 15 bespoke IoT real-time heat sensors and machine learning model XGBoost. Open Street Map data and QGIS were used to classify land cover and determine land cover fractions around each sensor location. XGBoost models were used to quantify the relative impact of different land cover types on local air temperature. The integration of IoT sensor data with machine learning model enabled detailed microclimate assessments, identifying how different land cover types (paved areas, buildings, grass, bare soil, and water) affect local temperature patterns. This study provides insights for the intelligent operation and maintenance of green spaces and informs urban design strategies aimed at reducing heat exposure and enhancing environmental sustainability.