Modelling urban heat island effects: a global analysis of 216 cities using machine learning techniques

Abstract

Urban areas globally have become home to over half of the world's population, leading to the intensification of the urban heat island (UHI) effect, where cities experience higher temperatures than their rural counterparts. The current study develops a new model predicting UHI intensity for 216 cities across all climate zones for both the Global North and Global South using machine learning techniques, focusing on the years 2019 to 2023. Utilising a novel dataset, integrating climate, economic, population, and land use data from 216 cities worldwide, the model, trained using Support Vector Regression (SVR), demonstrates a mean absolute error (MAE) of 0.86 °C. Results reveal that wind speed significantly mitigates UHI intensity, while cities in temperate climates exhibit more pronounced UHI effects compared to those located within tropical climbs. Additionally, results show the crucial role of coastal proximity in reducing UHI intensity and find no significant differences in UHI intensity between cities in the Global North and Global South. Findings offer important empirical actionable insights alongside a robust tool for urban planners and policymakers to measure, map, and monitor the UHI effect, contributing to the development of more liveable and sustainable urban environments.

Keywords

Urban Heat Island / Machine Learning / Predictive Modelling / Urbanisation / Climate Zones / Global Cities

References

[1] Australian Bureau of Statistics. (2021). Brisbane City 2021 Census All Persons QuickStats.https://www.abs.gov.au/census/find-census-data/quickstats/2021/305011105

[2] Adithya, S., Menon, K. P., & Anjana, M. (2023). A Machine Learning Approach to Self-Adjust Optimal Air Conditioning Temperature using Differential Air Velocity. 2023 International Conference on Power, Instrumentation, Control and Computing (PICC)

[3] Al SmadiA, MehmoodA, AbugabahA, AlmekhlafiE, Al-SmadiAM. Deep convolutional neural network-based system for fish classification. International Journal of Electrical and Computer Engineering, 2022, 1222026.

[4] Amir Forouhar, Chapple, K., Allen, J., Jeong, B., & Greenberg, J. (2024). Assessing downtown recovery rates and determinants in North American cities after the COVID-19 pandemic. Urban Studies.

[5] AnibabaB, DurowojuO, AdedejiO. Assessing the significance of meteorological parameters to the magnitude of urban heat island (UHI). Ann. Univ. Oradea Geogr. Ser, 2019, 29: 30-39

[6] Australian Bureau of Statistics. (2022). Population by Capital City. https://www.abs.gov.au/statistics/people/population/regional-population/latest-release#data-downloads

[7] Awad, M., spsampsps Khanna, R. (2015). Efficient learning machines: theories, concepts, and applications for engineers and system designers. Springer nature.

[8] BanerjeeS, SingalG, SahaS, MittalH, SrivastavaM, MukherjeeA, MahatoS, SaikiaB, ThakurS, SamantaS. Machine Learning approach to Predict net radiation over crop surfaces from global solar radiation and canopy temperature data. International Journal of Biometeorology, 2022, 66122405-2415.

[9] Barai, A., & Dey, L. (2017). Outlier detection and removal algorithm in k-means and hierarchical clustering.

[10] BuyantuyevA, WuJ. Urban heat islands and landscape heterogeneity: Linking spatiotemporal variations in surface temperatures to land-cover and socioeconomic patterns. Landscape Ecology, 2010, 25: 17-33.

[11] CaiY, ZhangH, ZhengP, PanW. Quantifying the impact of land use/land cover changes on the urban heat island: A case study of the natural wetlands distribution area of Fuzhou City, China. Wetlands, 2016, 36: 285-298.

[12] Chen, L., Jiang, R., & Xiang, W.-N. (2016). Surface heat island in Shanghai and its relationship with urban development from 1989 to 2013. Advances in Meteorology, 2016.

[13] Chen, X.-L., Zhao, H.-M., Li, P.-X., & Yin, Z.-Y. (2006). Remote sensing image-based analysis of the relationship between urban heat island and land use/cover changes. Remote sensing of environment, 104(2), 133–146. City of Chicago. (2023). Boundaries - Central Business District.

[14] City of Chicago. (2023). Boundaries - Central Business District. https://data.cityofchicago.org/Facilities-Geographic-Boundaries/Boundaries-Central-Business-District/tksj-nvsw

[15] DeilamiK, KamruzzamanM, HayesJF. Correlation or causality between land cover patterns and the urban heat island effect? Evidence from Brisbane Australia. Remote Sensing, 2016, 89716.

[16] Drucker, H., Burges, C. J., Kaufman, L., Smola, A., & Vapnik, V. (1996). Support vector regression machines. Advances in neural information processing systems, 9.

[17] Duarte, F., & Álvarez, R. (2019). The data politics of the urban age. Palgrave Communications, 5(1).

[18] Fung, K.Y., Yang, Z.-L., Martilli, A., E Scott Krayenhoff and Dev Niyogi (2024). Prioritizing social vulnerability in urban heat mitigation. PNAS Nexus, 3(9).

[19] FuruyaMTG, FuruyaDEG, de OliveiraLYD, da SilvaPA, CicerelliRE, GonçalvesWN, JuniorJM, OscoLP, RamosAPM. A machine learning approach for mapping surface urban heat island using environmental and socioeconomic variables: A case study in a medium-sized Brazilian city. Environmental Earth Sciences, 2023, 8213325.

[20] GiridharanR, LauS, GanesanS, GivoniB. Urban design factors influencing heat island intensity in high-rise high-density environments of Hong Kong. Building and Environment, 2007, 42103669-3684.

[21] GobakisK, KolokotsaD, SynnefaA, SaliariM, GiannopoulouK, SantamourisM. Development of a model for urban heat island prediction using neural network techniques. Sustainable Cities and Society, 2011, 12104-115.

[22] GreenstoneC. Digging up the Past: Re-Visiting the Potential Role Green Roofs can Play in Mitigating the Urban Heat Island (UHI) Effect: A Case Study of the Central Business District (CBD), Durban, South Africa. African Journal of Inter/multidisciplinary Studies, 2023, 511-13.

[23] GrunwaldL, SchneiderA-K, SchroederB, WeberS. Predicting urban cold-air paths using boosted regression trees. Landscape and Urban Planning, 2020, 201: 103843.

[24] Gyasi-Addo, J. A. (2021). Evaluation of the effect of urbanization on urban thermal behaviour using urban heat island indicators: the case of the CBD of Accra

[25] Hao, S., Zhou, X., & Song, H. (2015). A new method for noise data detection based on DBSCAN and SVDD. 2015 IEEE International Conference on Cyber Technology in Automation, Control, and Intelligent Systems (CYBER),

[26] HarmayNSM, ChoiM. The urban heat island and thermal heat stress correlate with climate dynamics and energy budget variations in multiple urban environments. Sustainable Cities and Society, 2023, 91: 104422.

[27] HeT, ZhouR, MaQ, LiC, LiuD, FangX, HuY, GaoJ. Quantifying the effects of urban development intensity on the surface urban heat island across building climate zones. Applied Geography, 2023, 158: 103052.

[28] HeinlM, HammerleA, TappeinerU, LeitingerG. Determinants of urban–rural land surface temperature differences–A landscape scale perspective. Landscape and Urban Planning, 2015, 134: 33-42.

[29] HuangS, RenG, ZhangP. Urbanization Effects in Estimating Surface Air Temperature Trends in the Contiguous United States. Land, 2024, 133388.

[30] ImhoffML, ZhangP, WolfeRE, BounouaL. Remote sensing of the urban heat island effect across biomes in the continental USA. Remote Sensing of Environment, 2010, 1143504-513.

[31] Joshi, M. Y., Aliaga, D. G., & Teller, J. (2023). Predicting urban heat island mitigation with random forest regression in Belgian cities. International Conference on Computers in Urban Planning and Urban Management.

[32] Jie, N., Cao, X., & Zhuo, L. (2024). Identifying the Central Business Districts of global megacities using nighttime light remote sensing data. International Journal of Digital Earth, 17(1).

[33] Kahraman, Emine & Kubat, Ayse. (2015). In the effects of accessibility factors on land values in the CBD of Izmir.

[34] KarpinskaL, ŚmiechS. Invisible energy poverty? Analysing housing costs in Central and Eastern Europe. Energy Research & Social Science, 2020, 70: 101670.

[35] L’heureux, A., Grolinger, K., Elyamany, H. F., & Capretz, M. A. . Machine learning with big data: Challenges and approaches. Ieee Access, 2017, 5: 7776-7797.

[36] LevermoreG, ParkinsonJ, LeeK, LaycockP, LindleyS. The increasing trend of the urban heat island intensity. Urban Climate, 2018, 24: 360-368.

[37] LiJ-J, WangX-R, WangX-J, MaW-C, ZhangH. Remote sensing evaluation of urban heat island and its spatial pattern of the Shanghai metropolitan area China. Ecological Complexity, 2009, 64413-420.

[38] LiJ, SunR, LiuT, XieW, ChenL. Prediction models of urban heat island based on landscape patterns and anthropogenic heat dynamics. Landscape Ecology, 2021, 36: 1801-1815.

[39] LiuW, YouH, DouJ. Urban-rural humidity and temperature differences in the Beijing area. Theoretical and Applied Climatology, 2009, 96: 201-207.

[40] LiuY, LiQ, YangL, MuK, ZhangM, LiuJ. Urban heat island effects of various urban morphologies under regional climate conditions. Science of the Total Environment, 2020, 743: 140589.

[41] LouisSHL, ZhangH, KathyTKN, LoSC, AlanSLY. Analysing urban trees on verges and slopes along a highway using machine learning methods. Urban Forestry & Urban Greening, 2022, 78: 127786.

[42] MacEachernSJ, ForkertND. Machine learning for precision medicine. Genome, 2021, 644416-425.

[43] Marcílio, W. E., & Eler, D. M. (2020, November). From explanations to feature selection: assessing SHAP values as feature selection mechanism. In 2020 33rd SIBGRAPI conference on Graphics, Patterns and Images (SIBGRAPI) (pp. 340–347). IEEE.

[44] MathewA, KhandelwalS, KaulN. Investigating spatial and seasonal variations of urban heat island effect over Jaipur city and its relationship with vegetation, urbanization and elevation parameters. Sustainable Cities and Society, 2017, 35: 157-177.

[45] MathewA, SreekumarS, KhandelwalS, KumarR. Prediction of land surface temperatures for surface urban heat island assessment over Chandigarh city using support vector regression model. Solar Energy, 2019, 186: 404-415.

[46] MokarramM, TaripanahF, PhamTM. Investigating the effect of surface urban heat island on the trend of temperature changes. Advances in Space Research, 2023, 7283150-3169.

[47] MyrupLO. A numerical model of the urban heat island. Journal of Applied Meteorology and Climatology, 1969, 86908-918.

[48] NawaratK, ReynsJ, VousdoukasMI, DuongTM, KrasE, RanasingheR. Coastal hardening and what it means for the world’s sandy beaches. Nature Communications, 2024, 1511-8.

[49] Office for National Statistics. (2023). A Beginner's Guide to UK Geography. https://geoportal.statistics.gov.uk/datasets/c0db0e8c67d04935bcf1749ca6027fef

[50] O’BrienRM. A caution regarding rules of thumb for variance inflation factors. Quality & Quantity, 2007, 415673-690.

[51] OgashawaraI, BastosV, d. S. B. . A quantitative approach for analyzing the relationship between urban heat islands and land cover. Remote Sensing, 2012, 4113596-3618.

[52] OkeTR. The energetic basis of the urban heat island. Quarterly Journal of the Royal Meteorological Society, 1982, 1084551-24

[53] OkumusDE, TerziF. Evaluating the role of urban fabric on surface urban heat island: The case of Istanbul. Sustainable Cities and Society, 2021, 73: 103128.

[54] OukawaGY, KreclP, TarginoAC. Fine-scale modeling of the urban heat island: A comparison of multiple linear regression and random forest approaches. Science of the Total Environment, 2022, 815: 152836.

[55] PalS, ZiaulS. Detection of land use and land cover change and land surface temperature in English Bazar urban centre. The Egyptian Journal of Remote Sensing and Space Science, 2017, 201125-145.

[56] PeelMC, FinlaysonBL, McMahonTA. Updated world map of the Köppen-Geiger climate classification. Hydrology and Earth System Sciences, 2007, 1151633-1644.

[57] PengJ, HuY, DongJ, LiuQ, LiuY. Quantifying spatial morphology and connectivity of urban heat islands in a megacity: A radius approach. Science of the Total Environment, 2020, 714: 136792.

[58] PengS, PiaoS, CiaisP, FriedlingsteinP, OttleC, BréonF-M, NanH, ZhouL, MyneniRB. Surface urban heat island across 419 global big cities. Environmental Science & Technology, 2012, 462696-703.

[59] PirachaA, ChaudharyMT. Urban air pollution, urban heat island and human health: A review of the literature. Sustainability, 2022, 14159234.

[60] ReimannL, VafeidisAT, HonselLE. Population development as a driver of coastal risk: Current trends and future pathways. Cambridge Prisms: Coastal Futures, 2023, 1: e14

[61] Sadik-Zada, E. R., & Gatto, A. (2022). Vulnerability to the urban heat islands effect in the Global North and the Global South: assessment of the drivers and mitigation strategies. In Global urban heat island mitigation (pp. 29–45). Elsevier.

[62] SanaB, CastiglioneJ, CooperD, TischlerD. Using google’s passive data and machine learning for origin-destination demand estimation. Transportation Research Record, 2018, 26724673-82.

[63] Sanchez, L. and Reames, T.G. (2019). Cooling Detroit: A socio-spatial analysis of equity in green roofs as an urban heat island mitigation strategy. Urban Forestry & Urban Greening, 44(126331).

[64] SarkerIH. Machine learning: Algorithms, real-world applications and research directions. SN Computer Science, 2021, 23160.

[65] Sartina, S., Zhiddiq, S., Suprapta, S., Maddatuang, M., Basram, N., Rusdi, R., & Maru, R. (2023). Urban Heat Island Mitigation for CBD Areas With City Form and Ciy Function Approaches. E3S Web of Conferences,

[66] SharmaR, GhoshA, JoshiPK. Spatio-temporal footprints of urbanisation in Surat, the Diamond City of India (1990–2009). Environmental Monitoring and Assessment, 2013, 185: 3313-3325.

[67] SherafatiSA, SaradjianM, NiazmardiS. Urban heat island growth modeling using artificial neural networks and support vector regression: A case study of Tehran. Iran. the International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences, 2013, 40: 399-403

[68] ShiZ, LiX, HuT, YuanB, YinP, JiangD. Modeling the intensity of surface urban heat island based on the impervious surface area. Urban Climate, 2023, 49: 101529.

[69] Smith, G. P., Sarricolea, E. P., Peralta, T. O., spsampsps Toro, C. (2022). Surface urban heat islands in 33 medium-sized cities across different climates in Chile. In Global Urban Heat Island Mitigation (pp. 323–334). Elsevier.

[70] StraubA, BergerK, BreitnerS, CyrysJ, GeruschkatU, JacobeitJ, KühlbachB, KuschT, PhilippA, SchneiderA. Statistical modelling of spatial patterns of the urban heat island intensity in the urban environment of Augsburg. Germany. Urban Climate, 2019, 29: 100491.

[71] TranH, UchihamaD, OchiS, YasuokaY. Assessment with satellite data of the urban heat island effects in Asian mega cities. International Journal of Applied Earth Observation and Geoinformation, 2006, 8134-48.

[72] UNCTAD. (2023). UNCTAD Handbook of Statistics. https://hbs.unctad.org/classifications/

[73] VarentsovM, KrinitskiyM, StepanenkoV. Machine Learning for Simulation of Urban Heat Island Dynamics Based on Large-Scale Meteorological Conditions. Climate, 2023, 1110200.

[74] WangJ, HuangB, FuD, AtkinsonPM. Spatiotemporal variation in surface urban heat island intensity and associated determinants across major Chinese cities. Remote Sensing, 2015, 743670-3689.

[75] WickiA, ParlowE, FeigenwinterC. Evaluation and modeling of urban heat island intensity in Basel. Switzerland. Climate, 2018, 6355

[76] WongNH, YuC. Study of green areas and urban heat island in a tropical city. Habitat International, 2005, 293547-558.

[77] World Bank. (2024). Urban Development. https://www.worldbank.org/en/topic/urbandevelopment/overview#:~:text=Today%2C%20some%2056%25%20of%20the,people%20will%20live%20in%20cities.

[78] World Population Review. (2024). Largest Cities by Population. https://worldpopulationreview.com/cities

[79] YamashitaS. The urban climate of Tokyo. Geographical Review of Japan, Series b., 1990, 63198-107.

[80] Yogi, K. S., Gowda, D., Mouna, K., Sujithra, L., Prasad, K., & Midhun, P. (2024). Scalability and Performance Evaluation of Machine Learning Techniques in High-Volume Social Media Data Analysis. 2024 11th International Conference on Reliability, Infocom Technologies and Optimization (Trends and Future Directions)(ICRITO),

[81] YusufYA, PradhanB, IdreesMO. Spatio-temporal assessment of urban heat island effects in Kuala Lumpur metropolitan city using landsat images. Journal of the Indian Society of Remote Sensing, 2014, 42: 829-837.

[82] ZhangK, CaoC, ChuH, ZhaoL, ZhaoJ, LeeX. Increased heat risk in wet climate induced by urban humid heat. Nature, 2023, 6177962738-742.

[83] Zhang, X., Friedl, M. A., Schaaf, C. B., Strahler, A. H., & Schneider, A. (2004). The footprint of urban climates on vegetation phenology. Geophysical research letters, 31(12).

[84] ZhaoJ, ZhaoX, LiangS, ZhouT, DuX, XuP, WuD. Assessing the thermal contributions of urban land cover types. Landscape and Urban Planning, 2020, 204: 103927.

[85] ZhaoL, LeeX, SmithRB, OlesonK. Strong contributions of local background climate to urban heat islands. Nature, 2014, 5117508216-219.

[86] ZhouD, ZhaoS, LiuS, ZhangL, ZhuC. Surface urban heat island in China's 32 major cities: Spatial patterns and drivers. Remote Sensing of Environment, 2014, 152: 51-61.

[87] ZhouD, ZhaoS, ZhangL, SunG, LiuY. The footprint of urban heat island effect in China. Scientific Reports, 2015, 5111160.

[88] ZongL, LiuS, YangY, RenG, YuM, ZhangY, LiY. Synergistic influence of local climate zones and wind speeds on the urban heat island and heat waves in the megacity of Beijing. China. Frontiers in Earth Science, 2021, 9: 673786.

RIGHTS & PERMISSIONS

The Author(s)