The logarithmic law is used widely in engineering practice to describe wind profiles. Vertical profiles of horizontally averaged wind velocity can be fitted by the logarithmic law even in built-up areas, allowing estimation of urban surface roughness based on parameters such as roughness length and displacement height. Such roughness parameterization of urban surfaces is very important, particularly for mesoscale weather simulation, because it is difficult to implement calculations incorporating fully resolved roughness elements in real urban areas.

Conventional roughness parameterizations for urban surfaces have typically used urban morphological parameters such as the building plan area fraction (i.e., the ratio of the plan area occupied by buildings to the total surface area) and the frontal area index (i.e., the total area of buildings projected into the plane normal to the approaching wind direction). It was noted in the previous research that the roughness height is related primarily to the frontal area index. However, the value of the frontal area index for a real urban area depends on the wind direction, implying that there is potential for the roughness height to vary with the wind direction. Although several previous studies have considered computational fluid dynamical or wind tunnel approaches to urban roughness parameterization, few have attempted to consider real urban areas, fewer still have considered the role of wind direction. In the present study, we conducted observation of the wind profile for 1 year above a high-density built-up area in Tokyo, Japan, using a Doppler lidar system. The observation results provide a database of the horizontally averaged turbulent statistics, from which we obtained roughness parameters for each wind direction. We compared our results with the roughness parameterizations of previous studies. The wind profiles were influenced not only by the above mentioned urban morphological parameters, but also by the maximum building height, the standard deviation of building height, and the plan area-weighted average building height. Future works will focus primarily on an empirical roughness parameterization based on these observation results.

Developed by Hoffman and his group, the CTTC model is capable of calculating dynamic change of outdoor air temperature for architectural complex and the Green CTTC model is capable for complex with vegetations. However, the Green CTTC model was only tested in hot-dry cliamte in middle or high lattitude areas, and its validality for hot-humid climate in low lattitude areas is still remained unknown. Moreover, the Green CTTC model counted the convective heat flow of trees into ground, which is inconsistent with the fact that any convective heat flow can have direct impact on air temperature. Facing the above problems, this paper is aimed to propose and test an improved Green CTTC model in hot-humid area of China. The new model was established by directly counting the impact of convective heat flow of trees on rising of air temperature. In order to achieve the unknown factor, that is the ratio of convective heat flow to absorbed solar radiation for trees, several field experiments were conducted on various communities with trees in Guangzhou. The average outdoor air temperatures were measured in sunny summer day and the ration was determined for the improve Green CTTC model. The model was tested by using the experimental results of a commnunity and the results show that the improved model is more acurate and applicable for hot-humid area of China than the original model.

Nocturnal urban heat island (UHI) and intra-urban heat island (IUHI) mainly develop through differences in cooling rates. The cooling process consists of two distinctive phases. In the first phase, around sunset, dense urban areas cool more slowly than more open sites, creating large intra-urban temperature differences that are preserved during the whole night. The intensity of this differential cooling is mainly determined by surface characteristics (geometry and material), prevailing weather conditions and season. On the other hand, the cooling during the rest of night, in the second phase, is independent of the surface characteristics.

In this study, we investigated how intra-urban cooling rates in the two phases are statistically related to prevailing weather conditions, season, and sky view factor using observation data from Gothenburg, Sweden. Based on the results, a simple statistical intra-urban nocturnal cooling rate model was developed. The model requires only commonly-used meteorological variables and sky view factor.

It was shown that the most intensive cooling rate at an open site, in the first phase, was chiefly dominated by the clearness of the sky and wind speed, i.e. the weather conditions. The cooling rate also had a linear relationship with maximum daily air temperature, which can be treated as the seasonal effect. Under clear sky condition, the magnitude of the cooling rate significantly decreased with lower sky view factor, but, under cloudy conditions, the cooling rate varied less or little. In the second phase, cooling rate seemed to linearly decrease as the night progressed and the slope of the decrease was determined by the clearness of the sky.

The model was evaluated using three additional datasets, one from Gothenburg, one from London, UK and one from Ouagadougou, Burkina Faso. Gothenburg and London are classified to have a marine temperate climate (Cfb) and Ouagadougou has a tropical steppe climate (BSh) according to Köppen climate classification. The model simulated cooling rates along a smooth profile statistically determined, while observed cooling rates often fluctuated through night. Nevertheless, the model estimated well the total amount of cooling during the whole night. This resulted in the well-simulated nocturnal air temperature. Modeled cooling rates were deviated from the observation at the sites where the large effects of anthropogenic heat and evapotranspiration were present. The effects were not included in this model yet but were found to be significant.

This model can be used for multiple applications such as nocturnal human thermal comfort estimation and climate-sensitive urban planning and design.

Carbon fluxes represent the net exchange between an ecosystem and the overlying atmosphere. In urban areas, it is due to the sum of biogenic components (ecosystem respiration and photosynthesis processes) and anthropogenic contributes (human and animal respiration, transportation, domestic heating/cooling).

Suburban areas, with typically a higher vegetation surface fraction than city centers, show a similar behavior to natural ecosystems and the diurnal carbon uptake by plants and lawns can help in reducing CO2 emitted by human activities. This effect is particularly evident during the growing season.

Recent studies have shown that the natural cover fraction (obtained as the difference between total land cover and urban fraction) can be used as a proxy to estimate annual carbon exchange in urban sites.

Natural (vegetation and bare soils) and urban (buildings and impervious) cover fraction can therefore address to an estimation of carbon emissions.

The general aim of this work is the development of an empirical model to simulate the biogenic components of the vertical CO2 flux based on environmental variables and vegetation cover fraction that can be applied over different ecosystems (natural, agricultural, urban, and suburban). In particular, in urban and suburban ecosystems it can allow to investigate the role of vegetation in acting as a carbon sink or source.

The advantage of such a model is that the estimation of CO2 flux is based on a few and commonly measured variables, such as global radiation and vegetation cover fraction.

The empirical model simulates the soil and vegetation respiration and photosynthesis. It is developed using CO2 flux measurements over different ecosystems, obtained with Eddy Covariance observations both in experimental sites and from literature.

A first validation, both over natural and suburban sites, is proposed here, and results will be shown.