**Author details**

spectral bands. This approach leads to the determination of the thermodynamic temperature and emissivity per surface element of type n. Then by considering the other spectral bands, if the satellite image has more than 2 thermal infrared bands, it leads to the determination of the emissivity of each type of surface element in those spectral bands. In the URBANFLUXES project, we will work with satellite images that have several thermal infrared bands, as shown

An innovative methodology using a 3D RT modeling and satellite reflectance data to derive urban surface albedo images has been designed and implemented. The method was tested using an atmospherically corrected Landsat‐8 multispectral image of the city of Basel. Although the preliminary results are encouraging, the newly presented methods have to be still properly validated. Different validation approaches are foreseen for testing their robust‐

**•** Application of the methodology on other satellite images acquired over the city of Basel. Using images of other satellite platforms with a different spatial resolution within relative‐ ly short time intervals (e.g., Landsat‐8 images acquired earlier or later and/or images from Sentinel‐2) will allow us to check how much the derived optical properties vary in time (urban surfaces elements' optical properties are expected not to vary significantly in time,

**•** Application of the methodology over additional cities, in particular the cities of Heraklion and London. These two other cities are indeed of scientific interest in the URBANFLUXES project. This will allow us to test the robustness of the method for different climatic conditions and urban structures, and to study the impact of those differences in terms of

A comprehensive comparison of the direct and iterative method technical performances is also required. It is especially important for future transformation of the methodology into a

Another important development lies in finalization of the thermal infrared spectral domain calibration. Combination of both calibrations in the visible and thermal part of the electro‐ magnetic spectrum would lead to a complete 3D radiative budget produced by DART at the

whereas optical properties of vegetation should be more variable).

method accuracy as well as technical constraints.

standardized operational approach.

same time as the other RT products.

**Upper**

**wavelength [nm]**

**Spatial resolution [m]**

**Lower**

**wavelength [nm]**

1 10,895 10,600 11,190 100 2 12,005 11,500 12,510 100

in **Table 2**.

244 Sustainable Urbanization

**Central**

**wavelength [nm]**

**Table 2.** Landsat‐8 thermal infrared band specifications.

**4. Conclusions and future perspectives**

**Thermal band [nm]**

ness:

Lucas Landier1\*, Nicolas Lauret1 , Tiangang Yin1,2, Ahmad Al Bitar<sup>1</sup> , Jean‐Philippe Gastellu-Etchegorry<sup>1</sup> , Christian Feigenwinter<sup>3</sup> , Eberhard Parlow<sup>3</sup> , Zina Mitraka4 and Nektarios Chrysoulakis4

