Estimation of Mean Radiant Temperature in Urban Canyons Using Google Street View: A Case Study on Seoul
Abstract
:1. Introduction
2. Study Area and Data Collection
2.1. Study Area
2.2. Data Collection Area
3. Method
3.1. Schematic Framework
3.2. View Factor Calculation and Shadow Detection
3.3. Calculation of Total Shortwave Radiation in Street Canyon
3.3.1. Calculation of Street-Level Shortwave Radiation
3.3.2. Calculation of Street-Level Longwave Radiation
3.3.3. Calculate Mean Radiation Temperature
4. Results and Discussion
4.1. Verification of Shortwave Radiation Estimated at Street-Level
4.1.1. Validation of Shortwave Radiation
4.1.2. Validation of Longwave Radiation
4.2. Comparison of Estimated MRT with Other Models
4.3. Comparison between LST and Estimated MRT
4.4. Limitations and Future Developments
5. Conclusions
Author Contributions
Funding
Institutional Review Board Statement
Informed Consent Statement
Data Availability Statement
Conflicts of Interest
Appendix A
Measured | Sensor | Unit | Accuracy |
---|---|---|---|
Air temperature | S-thb-m002 | °C | ±0.21 °C |
Relative Humidity | % | ±2.5% | |
Wind speed | S-wcf-m003 | m/s | ±1.1 m/s |
Shortwave radiation | CNR4 | W/m² | ±10% |
Appendix B
Appendix C
Appendix D
Appendix E
Urban Morphology | Location | Lat | Lon | LST | SVF | TVF | BVF | Street Orientation | MRT | |
---|---|---|---|---|---|---|---|---|---|---|
High LST | compact low-rise density building | 205-422, Cheongnyangni-dong, Dongdaemun-gu | 37.5895 | 127.0414 | 31.627 | 0.75 | 0 | 0.25 | N-S | 66.1 |
Anam-ro 24-gil, Jegi-dong, Dongdaemun-gu | 37.5882 | 127.0362 | 33.015 | 0.64 | 0.01 | 0.35 | E-W | 62.1 | ||
Changsin 1-dong, Jongro-gu | 37.5718 | 127.0139 | 32.141 | 0.41 | 0 | 0.59 | N-S | 61.3 | ||
977-18, Bangbae-dong, Seocho-gu | 37.4815 | 126.9923 | 32.542 | 0.6 | 0.02 | 0.38 | E-W | 61.8 | ||
Munrae-dong 4-ga, Yeongdeungpo-gu | 37.5147 | 126.8906 | 33.771 | 0.7 | 0 | 0.3 | E-W | 65.4 | ||
bare paved area | Suseo station parking lot | 37.4854 | 127.1056 | 34.783 | - | |||||
735, Suseo-dong, Gangnam-gu | 37.4878 | 127.0998 | 35.232 | |||||||
Ilwonbon-dong, Gangnam-gu | 37.4874 | 127.0801 | 35.547 | |||||||
compact mid-rise density building | 279-47 Sangdo 4-dong, Dongjak-gu | 37.4957 | 126.9374 | 29.341 | 0.42 | 0 | 0.58 | NE-SW | 43.9 | |
41-5, Hwayang-dong, Gwangjin-gu | 37.5451 | 127.0666 | 29.997 | 0.38 | 0 | 0.62 | N-S | 42.2 | ||
254-239, Daehak-dong, Gwanak-gu | 37.4649 | 126.9359 | 31.011 | 0.3 | 0 | 0.7 | E-W | 40.1 | ||
9-34, Suyu3-dong, Gangbuk-gu | 37.6383 | 127.0205 | 30.014 | 0.55 | 0 | 0.45 | E-W | 44.5 | ||
Low LST | dense tree | Nakseongdae park | 37.4719 | 126.9599 | 18.354 | 0.03 | 0.97 | 0 | E-W | 33.8 |
Janggunbong Sports Park | 37.4787 | 126.9384 | 18.997 | 0.01 | 0.99 | 0 | E-W | 35.1 | ||
44-3 Ogeum-dong, Songpa-gu | 37.5051 | 127.1277 | 19.584 | 0.3 | 0.6 | 0.1 | NE-SW | 32.4 | ||
low plants | Montmartre park | 37.4954 | 127.0038 | 21.711 | 0.99 | 0 | 0.01 | NE-SW | 70.2 | |
Yeouido hangang park | 37.5293 | 126.9326 | 22.667 | 0.96 | 0 | 0.04 | N-S | 69.1 | ||
pyeonghwaui park | 37.5618 | 126.8907 | 23.421 | 0.95 | 0 | 0.05 | E-S | 68.8 | ||
high density building | 460 Hongje-dong, Seodaemun-gu | 37.5854 | 126.9506 | 24.145 | 0.55 | 0.03 | 0.42 | E-W | 47.1 | |
140 Garak-dong, Songpa-gu | 37.4956 | 127.1278 | 25.245 | 0.4 | 0 | 0.6 | N-S | 43.8 | ||
467-7 Dogok-dong, Gangnam-gu | 37.4882 | 127.0519 | 26.114 | 0.42 | 0.21 | 0.37 | N-S | 44.2 | ||
27-45 Sangdo 2-dong, Dongjak-gu | 37.5043 | 126.9433 | 25.773 | 0.52 | 0.02 | 0.46 | E-W | 46.8 | ||
Medium of heat risk level | ||||||||||
No | Urban morphology | Land Surface Temperature (mean/sd) | Mean Radiant Temperature (mean/sd) | |||||||
1 | CLDB | 3.871/0.336 | 4/0.12 | |||||||
2 | CMDB | 3.38/0.486 | 1.157/0.364 | |||||||
3 | DT | 1/0.05 | 1/0.04 | |||||||
4 | HDB | 2.501/0.5 | 1.352/0.478 | |||||||
5 | LP | 1.75/0.434 | 4/0.23 |
References
- Dousset, B.; Gourmelon, F.; Laaidi, K.; Zeghnoun, A.; Giraudet, E.; Bretin, P.; Mauri, E.; Vandentorren, S. Satellite Monitoring of Summer Heat Waves in the Paris Metropolitan Area. Int. J. Climatol. 2011, 31, 313–323. [Google Scholar] [CrossRef]
- Gabriel, K.M.A.; Endlicher, W.R. Urban and Rural Mortality Rates during Heat Waves in Berlin and Brandenburg, Germany. Environ. Pollut. 2011, 159, 2044–2050. [Google Scholar] [CrossRef]
- Thorsson, S.; Rocklöv, J.; Konarska, J.; Lindberg, F.; Holmer, B.; Dousset, B.; Rayner, D. Mean Radiant Temperature - A Predictor of Heat Related Mortality. Urban Clim. 2014, 10, 332–345. [Google Scholar] [CrossRef]
- Bonamente, E.; Rossi, F.; Coccia, V.; Pisello, A.L.; Nicolini, A.; Castellani, B.; Cotana, F.; Filipponi, M.; Morini, E.; Santamouris, M. An Energy-Balanced Analytic Model for Urban Heat Canyons: Comparison with Experimental Data. Adv. Build. Energy Res. 2013, 7, 222–234. [Google Scholar] [CrossRef]
- Cohen, S.; Palatchi, Y.; Palatchi, D.P.; Shashua-Bar, L.; Lukyanov, V.; Yaakov, Y.; Matzarakis, A.; Tanny, J.; Potchter, O. Mean Radiant Temperature in Urban Canyons from Solar Calculations, Climate and Surface Properties – Theory, Validation and ‘Mr.T’ Software. Build. Environ. 2020, 178. [Google Scholar] [CrossRef]
- Varquez, A.C.G.; Darmanto, N.S.; Honda, Y.; Ihara, T.; Kanda, M. Future Increase in Elderly Heat-Related Mortality of a Rapidly Growing Asian Megacity. Sci. Rep. 2020, 10. [Google Scholar] [CrossRef]
- Wichmann, J. Heat Effects of Ambient Apparent Temperature on All-Cause Mortality in Cape Town, Durban and Johannesburg, South Africa: 2006–2010. Sci. Total Environ. 2017, 587–588, 266–272. [Google Scholar] [CrossRef] [Green Version]
- Wright, C.Y.; Garland, R.M.; Norval, M.; Vogel, C. Human Health Impacts in a Changing South African Climate. South Afr. Med. J. 2014, 104, 579–582. [Google Scholar] [CrossRef] [Green Version]
- Chen, L.; Yu, B.; Yang, F.; Mayer, H. Intra-Urban Differences of Mean Radiant Temperature in Different Urban Settings in Shanghai and Implications for Heat Stress under Heat Waves: A GIS-Based Approach. Energy Build. 2016, 130, 829–842. [Google Scholar] [CrossRef]
- Eckstein, D.; Winges, M.; Kunzel, V.; Schafer, L. Germanwatch Korperschaft Global Climate Risk Index 2020 Who Suffers Most from Extreme Weather Events? Wether-Related Loss Events in 2018 and 1999 to 2018; Briefing Paper; Germanwatch: Bonn, Germany, 2019. [Google Scholar]
- Van, T.T.; Duong, H.; Bao, X.; Thi, N.; Mai, T. Urban Thermal Environment and Heat Island in Ho Chi Minh City, Vietnam from Remote Sensing Data. Preprints 2017. [Google Scholar] [CrossRef]
- Yu, B.; Liu, H.; Wu, J.; Lin, W.M. Investigating Impacts of Urban Morphology on Spatio-Temporal Variations of Solar Radiation with Airborne LIDAR Data and a Solar Flux Model: A Case Study of Downtown Houston. Int. J. Remote Sens. 2009, 30, 4359–4385. [Google Scholar] [CrossRef]
- Qin, Z.; Karnieli, A.; Berliner, P. A Mono-Window Algorithm for Retrieving Land Surface Temperature from Landsat TM Data and Its Application to the Israel-Egypt Border Region. Int. J. Remote Sens. 2001, 22, 3719–3746. [Google Scholar] [CrossRef]
- de Castro Pena, J.C.; Martello, F.; Ribeiro, M.C.; Armitage, R.A.; Young, R.J.; Rodrigues, M. Street Trees Reduce the Negative Effects of Urbanization on Birds. PLoS ONE 2017, 12. [Google Scholar] [CrossRef] [Green Version]
- Park, C.Y.; Lee, D.K.; Krayenhoff, E.S.; Heo, H.K.; Ahn, S.; Asawa, T.; Murakami, A.; Kim, H.G. A Multilayer Mean Radiant Temperature Model for Pedestrians in a Street Canyon with Trees. Build. Environ. 2018, 141, 298–309. [Google Scholar] [CrossRef]
- Rohat, G.; Flacke, J.; Dosio, A.; Dao, H.; van Maarseveen, M. Projections of Human Exposure to Dangerous Heat in African Cities Under Multiple Socioeconomic and Climate Scenarios. Earth’s Future 2019, 7, 528–546. [Google Scholar] [CrossRef] [Green Version]
- Perini, K.; Chokhachian, A.; Dong, S.; Auer, T. Modeling and Simulating Urban Outdoor Comfort: Coupling ENVI-Met and TRNSYS by Grasshopper. Energy Build. 2017, 152, 373–384. [Google Scholar] [CrossRef]
- Chow, A.; Fung, A.S.; Li, S. GIS Modeling of Solar Neighborhood Potential at a Fine Spatiotemporal Resolution. Buildings 2014, 4, 195–206. [Google Scholar] [CrossRef]
- Matzarakis, A.; Rutz, F.; Mayer, H. Modelling Radiation Fluxes in Simple and Complex Environments: Basics of the RayMan Model. Int. J. Biometeorol. 2010, 54, 131–139. [Google Scholar] [CrossRef] [Green Version]
- Fröhlich, D.; Matzarakis, A. Modeling of Changes in Thermal Bioclimate: Examples Based on Urban Spaces in Freiburg, Germany. Theor. Appl. Climatol. 2013, 111, 547–558. [Google Scholar] [CrossRef]
- Krüger, E.; Drach, P.; Emmanuel, R.; Corbella, O. Urban Heat Island and Differences in Outdoor Comfort Levels in Glasgow, UK. Theor. Appl. Climatol. 2013, 112, 127–141. [Google Scholar] [CrossRef]
- Kwon, Y.J.; Lee, D.K.; Kwon, Y.H. Is Sensible Heat Flux Useful for the Assessment of Thermal Vulnerability in Seoul (Korea)? Int. J. Environ. Res. Public Health 2020, 17, 963. [Google Scholar] [CrossRef] [Green Version]
- Gong, F.Y.; Zeng, Z.C.; Zhang, F.; Li, X.; Ng, E.; Norford, L.K. Mapping Sky, Tree, and Building View Factors of Street Canyons in a High-Density Urban Environment. Build. Environ. 2018, 134, 155–167. [Google Scholar] [CrossRef]
- Gong, F.Y.; Zeng, Z.C.; Ng, E.; Norford, L.K. Spatiotemporal Patterns of Street-Level Solar Radiation Estimated Using Google Street View in a High-Density Urban Environment. Build. Environ. 2019, 148, 547–566. [Google Scholar] [CrossRef]
- Li, X.; Ratti, C.; Seiferling, I. Quantifying the Shade Provision of Street Trees in Urban Landscape: A Case Study in Boston, USA, Using Google Street View. Landsc. Urban Plan. 2018, 169, 81–91. [Google Scholar] [CrossRef]
- Cândido, R.L.; Steinmetz-Wood, M.; Morency, P.; Kestens, Y. Reassessing Urban Health Interventions: Back to the Future with Google Street View Time Machine. Am. J. Prev. Med. 2018, 55, 662–669. [Google Scholar] [CrossRef]
- Liang, J.; Gong, J.; Zhang, J.; Li, Y.; Wu, D.; Zhang, G. GSV2SVF-an Interactive GIS Tool for Sky, Tree and Building View Factor Estimation from Street View Photographs. Build. Environ. 2020, 168. [Google Scholar] [CrossRef]
- Kamal-Chaoui, L.; Grazi, F.; Joo, J.; Plouin, M. The Implementation of the Korean Green Growth Strategy in Urban Areas; OECD Publishing: Paris, France, 2011. [Google Scholar] [CrossRef]
- Kipp & Zonen, B.V. Net Radiometer CNR 4 Instruction Manual. Kipp & Zonen, B.V.: Delft, The Netherlands, 2003; pp. 1–183. [Google Scholar]
- He, K.; Zhang, X.; Ren, S.; Sun, J. Deep Residual Learning for Image Recognition. arXiv 2015, arXiv:1512.03385. [Google Scholar]
- Holtslag and Ulden. A Simple Scheme for Daytime Estimates of the Surface Fluxes from Routine Weather Data. J. Clim. Appl. Meteorol. 1983, 22, 517–529. [Google Scholar] [CrossRef]
- Offerle, B.; Grimmond, C.S.B.; Oke, T.R. Parameterization Net All-Wave Radiat. Urban Areas; J. Appl. Meteorol. 2003, 42, 1157–1173. [Google Scholar] [CrossRef]
- Reda, I.; Andreas, A. Solar Position Algorithm for Solar Radiation Applications (Revised); Nrel/Tp-560-34302; National Renewable Energy Laboratory: Golden, CO, USA, 2008; pp. 1–56. [Google Scholar] [CrossRef] [Green Version]
- Loridan, T.; Grimmond, C.S.B.; Offerle, B.D.; Young, D.T.; Smith, T.E.L.; Järvi, L.; Lindberg, F. Local-Scale Urban Meteorological Parameterization Scheme (LUMPS): Longwave Radiation Parameterization and Seasonality-Related Developments. J. Appl. Meteorol. Climatol. 2011, 50, 185–202. [Google Scholar] [CrossRef] [Green Version]
- Barad, M.L. Project Prairie Grass, a Field Program. In Diffusion; Geophysical Research Papers No. 59; Air Force Cambridge Research Center: Dayton, OH, USA, 1958; Volume 2. [Google Scholar]
- Brostow, G.J.; Fauqueur, J.; Cipolla, R. Semantic Object Classes in Video: A High-Definition Ground Truth Database. Pattern Recognit. Lett. 2009, 30, 88–97. [Google Scholar] [CrossRef]
- Chen, L.-C.; Zhu, Y.; Papandreou, G.; Schroff, F.; Adam, H. Encoder-Decoder with Atrous Separable Convolution for Semantic Image Segmentation. arXiv 2018, arXiv:1802.02611. [Google Scholar]
- Oke, T.R. Boundary Layer Climates; Routleege: London, UK, 1987. [Google Scholar]
- Allen, R.G.; Pereira, L.S.; Raes, D.; Smith, M. FAO Irrigation and Drainage Paper No. 56—Crop Evapotranspiration; FAO: Rome, Italy, 1998. [Google Scholar]
- Singh, U.P.; Johnson, N.; Govindasamy Tamizhmani, B.R. Diffuse Radiation Calculation Methods; Aizona State University: Tempe, AZ, USA, 2016. [Google Scholar]
- Kasten, F.; Czeplak, G. Solar and terrestrial radiation dependent on the amount and type of cloud. Sol. Energy 1980, 24, 177–189. [Google Scholar] [CrossRef]
- Jamei, E.; Rajagopalan, P. Urban Development and Pedestrian Thermal Comfort in Melbourne. Sol. Energy 2017, 144, 681–698. [Google Scholar] [CrossRef]
- de Bruin, H.A.R.; Holtslag, A.A.M. A Simple Parameterization of the Surface Fluxes of Sensible and Latent Heat during Daytime Compared with the Penman-Monteith Concept (Netherlands). J. Appl. Meteorol. 1982, 21, 1610–1621. [Google Scholar] [CrossRef] [Green Version]
- Matzarakis, A.; Programming, F.R.; Rutz, L.F.; Chen, Y.-C.; Fröhlich, D.; Fröhlich, M.D. RayMan Pro—A Tool for Applied Climatology Modelling of Mean Radiant Temperature and Thermal Indices; RayMan Manual Version 0; Research Centre Human Biometeorology: Freiburg, Germany, 2017. [Google Scholar]
- Naboni, E.; Meloni, M.; MacKey, C.; Kaempf, J. The Simulation of Mean Radiant Temperature in Outdoor Conditions: A Review of Software Tools Capabilities. In Proceedings of the Building Simulation Conference Proceedings, International Building Performance Simulation Association, Rome, Italy, 2–4 September 2019; Volume 5, pp. 3234–3241. [Google Scholar]
- Gál, C.V.; Kántor, N. Modeling Mean Radiant Temperature in Outdoor Spaces, A Comparative Numerical Simulation and Validation Study. Urban Clim. 2020, 32. [Google Scholar] [CrossRef]
- Kwok, Y.T.; Schoetter, R.; Lau, K.K.L.; Hidalgo, J.; Ren, C.; Pigeon, G.; Masson, V. How Well Does the Local Climate Zone Scheme Discern the Thermal Environment of Toulouse (France)? An Analysis Using Numerical Simulation Data. Int. J. Climatol. 2019, 39, 5292–5315. [Google Scholar] [CrossRef]
- Zhou, X.; Okaze, T.; Ren, C.; Cai, M.; Ishida, Y.; Watanabe, H.; Mochida, A. Evaluation of Urban Heat Islands Using Local Climate Zones and the Influence of Sea-Land Breeze. Sustain. Cities Soc. 2020, 55. [Google Scholar] [CrossRef]
- Dervishi, S.; Mahdavi, A. Computing Diffuse Fraction of Global Horizontal Solar Radiation: A Model Comparison. Sol. Energy 2012, 86, 1796–1802. [Google Scholar] [CrossRef] [Green Version]
Object | Location | Lat. | Log. | Data Collection Date | Input Data | No. of Panorama Images | |
---|---|---|---|---|---|---|---|
Google Street View | Validation sites | Low building | 37.457391 | 126.948493 | 18.04.16~18 | Radiation (shortwave, longwave), air temperature, relative humidity | 7 |
Park | 37.495193 | 127.003546 | 18.04.19~21 | ||||
Commercial area | 37.521532 | 126.927314 | 18.04.28~30 | ||||
Apartment | 37.503094 | 126.943548 | 18.05.04~07 | ||||
River | 37.528474 | 126.934370 | 18.05.10~13 | ||||
Narrow alley | 37.482026 | 126.929579 | 18.05.31~06.03 | ||||
Residential area | 37.469727 | 126.942584 | 18.06.01~04 | ||||
Mapping | Seoul | 37.34~37.5666 | 126.584~126.978 | 2014~2020 (4~10) | Air temperature Relative humidity | 58,794 | |
Satellite image | Attribute | ||||||
Date | Satellite image | Projection | Datum | Cloud cover (in %) | Sensor | Time | |
2018. 06. 19 | Landsat 8 | UTM zone52 | WGS84 | 1.79 | OLI_TIRS | 09:52 |
Publisher’s Note: MDPI stays neutral with regard to jurisdictional claims in published maps and institutional affiliations. |
© 2022 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (https://creativecommons.org/licenses/by/4.0/).
Share and Cite
Kim, E.-S.; Yun, S.-H.; Park, C.-Y.; Heo, H.-K.; Lee, D.-K. Estimation of Mean Radiant Temperature in Urban Canyons Using Google Street View: A Case Study on Seoul. Remote Sens. 2022, 14, 260. https://doi.org/10.3390/rs14020260
Kim E-S, Yun S-H, Park C-Y, Heo H-K, Lee D-K. Estimation of Mean Radiant Temperature in Urban Canyons Using Google Street View: A Case Study on Seoul. Remote Sensing. 2022; 14(2):260. https://doi.org/10.3390/rs14020260
Chicago/Turabian StyleKim, Eun-Sub, Seok-Hwan Yun, Chae-Yeon Park, Han-Kyul Heo, and Dong-Kun Lee. 2022. "Estimation of Mean Radiant Temperature in Urban Canyons Using Google Street View: A Case Study on Seoul" Remote Sensing 14, no. 2: 260. https://doi.org/10.3390/rs14020260
APA StyleKim, E. -S., Yun, S. -H., Park, C. -Y., Heo, H. -K., & Lee, D. -K. (2022). Estimation of Mean Radiant Temperature in Urban Canyons Using Google Street View: A Case Study on Seoul. Remote Sensing, 14(2), 260. https://doi.org/10.3390/rs14020260