Exploring a new method for the retrieval of urban thermophysical properties using thermal infrared remote sensing and deterministic modeling

Abstract

[1] Increasingly, mesoscale meteorological and climate models are used to predict urban weather and climate. Yet, large uncertainties remain regarding values of some urban surface properties. In particular, information concerning urban values for thermal roughness length and thermal admittance is scarce. In this paper, we present a method to estimate values for thermal admittance in combination with an optimal scheme for thermal roughness length, based on METEOSAT-8/SEVIRI thermal infrared imagery in conjunction with a deterministic atmospheric model containing a simple urbanized land surface scheme. Given the spatial resolution of the SEVIRI sensor, the resulting parameter values are applicable at scales of the order of 5 km. As a study case we focused on the city of Paris, for the day of 29 June 2006. Land surface temperature was calculated from SEVIRI thermal radiances using a new split-window algorithm specifically designed to handle urban conditions, as described in Appendix A, including a correction for anisotropy effects. Land surface temperature was also calculated in an ensemble of simulations carried out with the ARPS mesoscale atmospheric model, combining different thermal roughness length parameterizations with a range of thermal admittance values. Particular care was taken to spatially match the simulated land surface temperature with the SEVIRI field of view, using the so-called point spread function of the latter. Using Bayesian inference, the best agreement between simulated and observed land surface temperature was obtained for the Zilitinkevich (1970) and Brutsaert (1975) thermal roughness length parameterizations, the latter with the coefficients obtained by Kanda et al. (2007) Citation: De Ridder, K., C. Bertrand, G. Casanova, and W. Lefebvre (2012), Exploring a new method for the retrieval of urban thermophysical properties using thermal infrared remote sensing and deterministic modeling,

Assessing the current and future urban heat island of Brussels

“…They found a strong correlation between the UHI and the cloud cover, wind speed, wind direction and surface temperature accomplished in a rather simple way, by representing the urban surface as a rough impermeable slab, with appropriate values for the albedo, emissivity, thermal conductivity and volumetric heat capacity. The main feature of the extension of the scheme is the inclusion of a parameterization of the inverse Stanton number, which is known to be much higher in urban areas (Kanda et al, 2007;De Ridder et al, 2012).…”

“…Many models, also recent ones, rely on fairly ancient parameterizations, e.g., by Jürges (1924) (used in, e.g., Ikeda and Kusaka, 2010), or Rowley et al (1930) (used in, e.g., Masson, 2000;Oleson et al, 2008). Whereas these wall transfer coefficients were established by means of scale experiments, the Stanton-based heat transfer coefficients in UrbClim are obtained from a series of real-world experiments that we conducted on actual cities, using remotely sensed surface thermal infrared temperature (De Ridder, 2006;De Ridder et al, 2008;De Ridder et al, 2012). While doing so entails disregarding certain physical processes occurring within the urban canopy, our approach is based on observations from actual urban areas, rather than that it has to rely on scale model experiments.…”

Detailed Urban Heat Island Projections for Cities Worldwide: Dynamical Downscaling CMIP5 Global Climate Models

“…This urbanization is accomplished in a rather simple way, by representing the urban surface as a rough impermeable slab, with appropriate values for the albedo, emissivity, thermal conductivity and volumetric heat capacity. The main feature of the extension of the scheme is the inclusion of a parameterization of the inverse Stanton number, which is known to be much higher in urban areas [15,16]. All of the details can be found in De Ridder et al [13].…”

“…Lauwaet et al [39] evaluated the simulated UHI effect for the city of Brussels (Belgium) during the summer of 2008 and found that the model was able to reproduce the observed differences in time series of 2-m air temperatures from 3 different stations, with very small positive biases, root mean square errors around 1 °C and correlation coefficients up to 0.7. Furthermore, the land surface temperatures from the UrbClim land surface scheme have already been validated in the past with satellite data for the city of Paris and the German Ruhr area, yielding good comparisons between simulated and observed land surface temperatures from thermal infrared satellite imagery [16,22,23].…”

“…Many models, also recent ones, rely on fairly ancient parameterizations, e.g., by Jürges [20] (used in, e.g., [17]) or Rowley et al [21] (used in, e.g., [18,19]). Whereas these wall transfer coefficients were established by means of scale experiments, the Stanton-based heat transfer coefficients in UrbClim are obtained from a series of real-world experiments that we conducted on actual cities, using remotely-sensed surface thermal infrared temperature [16,22,23]. While doing so entails disregarding certain physical processes occurring within the urban canopy, our approach is based on observations from actual urban areas, rather than it having to rely on scale model experiments.…”