Urban Energy Balance Obtained from the Comprehensive Outdoor Scale Model Experiment. Part II: Comparisons with Field Data Using an Improved Energy Partition

The objective of this study is to examine the differences and similarities in the annual trends of the urban surface energy balance (SEB) among long-term field measurements. Four datasets analyzed for the study were collected in the following experiments or observational sites: Comprehensive Outdoor Scale Model experiments (COSMO), the Kugahara site in Tokyo, Japan (Ku04), and the Sperrstrasse and Spalenring sites in Basel, Switzerland (BuU1 and BuU2). A new variable, the forcing radiation Q(FR), has been proposed to replace the conventional net radiation Q* for the normalization of the SEB components. Here, Q(FR) is defined as the sum of net shortwave radiation and downward longwave radiation. Because Q(FR) does not include the upward longwave radiation, it is independent of the surface temperature, which is determined by the energy partitioning process. Therefore, Q(FR) is independent of the energy partitioning process itself. With the use of Q(FR), the characteristics of the daytime normalized components of the SEB equation (i.e., upward longwave radiation, Q(L)up arrow/Q(FR), turbulence fluxes (Q(H) + Q(E))/Q(FR), and heat storage Delta Q(S)/Q(FR)) were investigated. The above energy fluxes normalized by the forcing radiation depended on the friction velocity u(*). An increase of u(*) predominantly enhanced (Q(H) + Q(E))/Q(FR) and reduced both Q(L)up arrow/Q(FR) and Delta Q(S)/Q(FR). When this u(*) dependency on the SEB was taken into consideration, the annual variations of these three flux ratios from cities located in similar latitudes and longitudes (i.e., BuU1 and BuU2, Ku04, and COSMO) were very similar. At BuU1 and BuU2, Q(L)up arrow/Q(FR) showed larger seasonal amplitudes than at COSMO and Ku04. The seasonal variations of (Q(H) + Q(E))/Q(FR) were roughly out of phase with respect to those of Q(L)up arrow/Q(FR), resulting in relatively small seasonal variations of Delta Q(S)/Q(FR). Furthermore, the effects of urban canyon geometry on the SEB were examined by comparing the SEB for the roofs to that for the canyon. The three-dimensional urban canyon geometry enhanced the heat storage efficiency relative to the forcing radiation Delta Q(S)/Q(FR) of the canyon in comparison with that of roofs or flat surfaces. This observation was explained by the continuous movement of sunlit areas on the walls and streets.