EMS Annual Meeting Abstracts
Vol. 21, EMS2024-364, 2024, updated on 05 Jul 2024
https://doi.org/10.5194/ems2024-364
EMS Annual Meeting 2024
© Author(s) 2024. This work is distributed under
the Creative Commons Attribution 4.0 License.
Poster | Thursday, 05 Sep, 18:00–19:30 (CEST), Display time Thursday, 05 Sep, 13:30–Friday, 06 Sep, 16:00|

Mapping urban heat islands in Pärnu/Estonia by leveraging satellite data

Amir Dehkordi1,3, Andreas Hoy1, Kelli Marie Jama2, and Heidi Tuhkanen2
Amir Dehkordi et al.
  • 1Stockholm Environment Institute (Tallinn Centre), Climate Systems and Energy Policy Unit, Tallinn, Estonia (andreas.hoy@sei.org)
  • 2Stockholm Environment Institute (Tallinn Centre), Green and Circular Economic Transformation Unit, Tallinn, Estonia
  • 3Polytechnic University of Turin, Interuniversity Department of Regional and Urban Studies and Planning, Turin, Italy

Urban heat islands (UHIs) pose significant challenges for cities of various sizes worldwide. Extreme summer temperatures strongly increased in frequency and intensity in all parts of Europe including Estonia, especially since the millennium. This development increased exposure and vulnerability of people and infrastructure to heat impacts. This study contributes to the Horizon Europe funded Regions4Climate project by investigating the characteristics of land surface temperatures (LST) in Pärnu city and county, located in Estonia (north-eastern Europe). We employ remote sensing data from satellites to identify areas experiencing the most intense heat in recent summers.

We studied available satellite data for the region via a web and literature research, focussing on the May-September timeframe. Considering available spectral bands and despite a comparably rare revisit frequency, Landsat-8 emerged the most applicable choice, offering comparably high-resolution data and an available data range of >10 years (2013-2023). Data are available for midday (around 12:30 local time), when the solar altitude is highest. We are additionally investigating into alternative remote sensing options are well (Modis, Ecostress). Furthermore, we obtained necessary ancillary data including city/county borders, building structure data, and open/green space maps.

To identify relevant days, we analyse both available Landsat-8 data (around 16 days revisit cycle) and local weather station records. We use days with maximum air temperatures exceeding 25°C at Pärnu station, combined with minimal cloud cover. The chosen days are processed in a GIS environment to derive LST values for Pärnu city/county. We employ an intuitive colour scale and legend to represent absolute temperatures alongside anomalies from a predefined baseline. Finally, we apply threshold values to identify local areas experiencing overheating, by classifying pixels exceeding specific LST thresholds or exhibiting significant temperature anomalies.

Our final results, tailored for integration into Pärnu city/county's urban planning processes, will be developed through reiterative consultation to ensure relevance. This information will inform strategies for mitigating UHI effects, improving thermal comfort, and promoting equitable and just climate adaptation. These results will serve as the foundation for a digital decision-making tool by Pärnu City, empowering stakeholders in urban development and land-use planning.

How to cite: Dehkordi, A., Hoy, A., Jama, K. M., and Tuhkanen, H.: Mapping urban heat islands in Pärnu/Estonia by leveraging satellite data, EMS Annual Meeting 2024, Barcelona, Spain, 1–6 Sep 2024, EMS2024-364, https://doi.org/10.5194/ems2024-364, 2024.