Temperature Prediction at Street Scale During a Heat Wave Using Random Forest

The rising frequency of heatwaves, combined with the urban heat island effect, increases the population’s exposure to high temperatures, significantly impacting the health of vulnerable groups and the overall well-being of residents. While mesoscale meteorological models can reliably forecast temper...

Full description

Saved in:
Bibliographic Details
Main Authors: Panagiotis Gkirmpas, George Tsegas, Denise Boehnke, Christos Vlachokostas, Nicolas Moussiopoulos
Format: Article
Language:English
Published: MDPI AG 2025-07-01
Series:Atmosphere
Subjects:
Online Access:https://www.mdpi.com/2073-4433/16/7/877
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1849303484369731584
author Panagiotis Gkirmpas
George Tsegas
Denise Boehnke
Christos Vlachokostas
Nicolas Moussiopoulos
author_facet Panagiotis Gkirmpas
George Tsegas
Denise Boehnke
Christos Vlachokostas
Nicolas Moussiopoulos
author_sort Panagiotis Gkirmpas
collection DOAJ
description The rising frequency of heatwaves, combined with the urban heat island effect, increases the population’s exposure to high temperatures, significantly impacting the health of vulnerable groups and the overall well-being of residents. While mesoscale meteorological models can reliably forecast temperatures across urban neighbourhoods, dense networks of in situ measurements offer more precise data at the street scale. In this work, the Random Forest technique was used to predict street-scale temperatures in the downtown area of Thessaloniki, Greece, during a prolonged heatwave in July 2021. The model was trained using data from a low-cost sensor network, meteorological fields calculated by the mesoscale model MEMO, and micro-environmental spatial features. The results show that, although the MEMO temperature predictions achieve high accuracy during nighttime compared to measurements, they exhibit inconsistent trends across sensor locations during daytime, indicating that the model does not fully account for microclimatic phenomena. Additionally, by using only the observed temperature as the target of the Random Forest model, higher accuracy is achieved, but spatial features are not represented in the predictions. In contrast, the most reliable approach to incorporating spatial characteristics is to use the difference between observed and mesoscale temperatures as the target variable.
format Article
id doaj-art-4e5d56fdb3694bda9f5eb37b618dd11e
institution Kabale University
issn 2073-4433
language English
publishDate 2025-07-01
publisher MDPI AG
record_format Article
series Atmosphere
spelling doaj-art-4e5d56fdb3694bda9f5eb37b618dd11e2025-08-20T03:58:26ZengMDPI AGAtmosphere2073-44332025-07-0116787710.3390/atmos16070877Temperature Prediction at Street Scale During a Heat Wave Using Random ForestPanagiotis Gkirmpas0George Tsegas1Denise Boehnke2Christos Vlachokostas3Nicolas Moussiopoulos4Sustainability Engineering Laboratory, Aristotle University of Thessaloniki, GR-54124 Thessaloniki, GreeceSustainability Engineering Laboratory, Aristotle University of Thessaloniki, GR-54124 Thessaloniki, GreeceInstitute for Geography and Geoecology, Karlsruhe Institute of Technology, 76131 Karlsruhe, GermanySustainability Engineering Laboratory, Aristotle University of Thessaloniki, GR-54124 Thessaloniki, GreeceMain Campus, Aristotle University of Thessaloniki, GR-54124 Thessaloniki, GreeceThe rising frequency of heatwaves, combined with the urban heat island effect, increases the population’s exposure to high temperatures, significantly impacting the health of vulnerable groups and the overall well-being of residents. While mesoscale meteorological models can reliably forecast temperatures across urban neighbourhoods, dense networks of in situ measurements offer more precise data at the street scale. In this work, the Random Forest technique was used to predict street-scale temperatures in the downtown area of Thessaloniki, Greece, during a prolonged heatwave in July 2021. The model was trained using data from a low-cost sensor network, meteorological fields calculated by the mesoscale model MEMO, and micro-environmental spatial features. The results show that, although the MEMO temperature predictions achieve high accuracy during nighttime compared to measurements, they exhibit inconsistent trends across sensor locations during daytime, indicating that the model does not fully account for microclimatic phenomena. Additionally, by using only the observed temperature as the target of the Random Forest model, higher accuracy is achieved, but spatial features are not represented in the predictions. In contrast, the most reliable approach to incorporating spatial characteristics is to use the difference between observed and mesoscale temperatures as the target variable.https://www.mdpi.com/2073-4433/16/7/877low-cost sensorsurban heat islandNWP downscalingrandom forest regression
spellingShingle Panagiotis Gkirmpas
George Tsegas
Denise Boehnke
Christos Vlachokostas
Nicolas Moussiopoulos
Temperature Prediction at Street Scale During a Heat Wave Using Random Forest
Atmosphere
low-cost sensors
urban heat island
NWP downscaling
random forest regression
title Temperature Prediction at Street Scale During a Heat Wave Using Random Forest
title_full Temperature Prediction at Street Scale During a Heat Wave Using Random Forest
title_fullStr Temperature Prediction at Street Scale During a Heat Wave Using Random Forest
title_full_unstemmed Temperature Prediction at Street Scale During a Heat Wave Using Random Forest
title_short Temperature Prediction at Street Scale During a Heat Wave Using Random Forest
title_sort temperature prediction at street scale during a heat wave using random forest
topic low-cost sensors
urban heat island
NWP downscaling
random forest regression
url https://www.mdpi.com/2073-4433/16/7/877
work_keys_str_mv AT panagiotisgkirmpas temperaturepredictionatstreetscaleduringaheatwaveusingrandomforest
AT georgetsegas temperaturepredictionatstreetscaleduringaheatwaveusingrandomforest
AT deniseboehnke temperaturepredictionatstreetscaleduringaheatwaveusingrandomforest
AT christosvlachokostas temperaturepredictionatstreetscaleduringaheatwaveusingrandomforest
AT nicolasmoussiopoulos temperaturepredictionatstreetscaleduringaheatwaveusingrandomforest