 Remotely Sensed Derived Land Surface Temperature (LST) as a Proxy for Air Temperature and Thermal Comfort at a Small Geographical ScaleRan Goldblatt,
Abdullah Addas,
Daynan Crull,
Ahmad Maghrabi,
Gabriel Gene Levin and
Steven Rubinyi
Land, 2021, vol. 10, issue 4, 1-24 Abstract: Urban Heat Islands (UHIs) and Urban Cool Islands (UCIs) can be measured by means of in situ measurements and interpolation methods, which often require densely distributed networks of sensors and can be time-consuming, expensive and in many cases infeasible. The use of satellite data to estimate Land Surface Temperature (LST) and spectral indices such as the Normalized Difference Vegetation Index (NDVI) has emerged in the last decade as a promising technique to map Surface Urban Heat Islands (SUHIs), primarily at large geographical scales. Furthermore, thermal comfort, the subjective perception and experience of humans of micro-climates, is also an important component of UHIs. It remains unanswered whether LST can be used to predict thermal comfort. The objective of this study is to evaluate the accuracy of remotely sensed data, including a derived LST, at a small geographical scale, in the case study of King Abdulaziz University (KAU) campus (Jeddah, Saudi Arabia) and four surrounding neighborhoods. We evaluate the potential use of LST estimates as proxy for air temperature (T air ) and thermal comfort. We estimate LST based on Landsat-8 measurements, T air and other climatological parameters by means of in situ measurements and subjective thermal comfort by means of a Physiological Equivalent Temperature (PET) model. We find a significant correlation (r = 0.45, p < 0.001) between LST and mean T air and the compatibility of LST and T air as equivalent measures using Bland-Altman analysis. We evaluate several models with LST, NDVI, and Normalized Difference Built-up Index (NDBI) as data inputs to proxy T air and find that they achieve error rates across metrics that are two orders of magnitude below that of a comparison with LST and T air alone. We also find that, using only remotely sensed data, including LST, NDVI, and NDBI, random forest classifiers can detect sites with “very hot” classification of thermal comfort nearly as effectively as estimates using in situ data, with one such model attaining an F1 score of 0.65. This study demonstrates the potential use of remotely sensed measurements to infer the Physiological Equivalent Temperature (PET) and subjective thermal comfort at small geographical scales as well as the impacts of land cover and land use characteristics on UHI and UCI. Such insights are fundamental for sustainable urban planning and would contribute enormously to urban planning that considers people’s well-being and comfort. Keywords: urban heat island; urban cool island; LST; Landsat; thermal comfort; T air; physiological equivalent temperature (PET); random forest classifier (search for similar items in EconPapers) Downloads: (external link) Related works: Export reference: BibTeX RIS (EndNote, ProCite, RefMan) HTML/Text Persistent link: https://EconPapers.repec.org/RePEc:gam:jlands:v:10:y:2021:i:4:p:410-:d:535527 Access Statistics for this article Land is currently edited by Ms. Carol Ma More articles in Land from MDPI |
Is your work missing from RePEc? Here is how to contribute.
Questions or problems? Check the EconPapers FAQ or send mail to .
EconPapers is hosted by the
School of Business at Örebro University.