Thermal Remote Sensing

In contrast to the labour intensive field measurements approach, with advancements in sensor technology, Thermal Remote Sensing (TRS) has become a very comprehensive alternative featuring large scale thermal observation through the use of satellite and aircraft platforms. 

TRS has the ability to make use of all of the missing parameters that needed to be inferred or ignored during field measurements, which include albedo, surface emissivity, irradiative input to the system and surface moisture content in addition to the plethora of atmospheric dynamics and properties as suggested by Becker and Li (1995). 

Unfortunately, despite the added benefits to using TRS over field measurements, it has been met with an abundance of criticism as highlighted by Mirzaei and Haghighat (2010). TRS is very expensive, meaning experiments cannot be run for extended periods of time without large-scale funding, which isn't often available. Additionally, TRS can only measure Surface UHI (Figure 1) as it measures surface radiation, which means the atmospheric UHI would need to be inferred using mathematical models.  

There also exists the issue of unstable images due to being taken from constantly moving platforms and the interference of cloud coverage on readings. Considering cloud coverage plays a strong role in UHI formation, taking UHI readings during overcast where a strong UHI is expected isn't even an option. 

TRS has the benefit of covering a much wider observational range than field measurements and can be a very viable solution under favourable conditions but due to the position of the sensors, vertical profiles of temperature cannot be measured and large 3D structures could interfere with readings particularly because shadows will lower the surface UHI but the buildings themselves will be absorbing the thermal irradiance and re-radiating the heat back into the system warming the atmospheric UHI. This is not accounted for in TRS readings as atmospheric UHI can only be estimated from Surface UHI. This is concerning as the atmospheric UHI can vary greatly from surface UHI in some instances and may be underestimated for UHI intensities lower than 1 ºC, particularly due to the inaccuracies that are to be expected of thermal measurements taken from such a large distance.

Figure 1: Thermal Remote Sensing of surface thermal emissivity over a variety of cities.