Entrainment of pollutants into the urban canopy from outside sources, or from material originally released within the canopy (re-entrainment), is a little-understood aspect of urban dispersion. We analyse data from direct numerical simulations (DNS) over an idealised street network to shed light on entrainment, and compare it with the vertical detrainment of material out of the street network and with horizontal dispersion processes. This gives insight into differences in the resulting plume patterns within and above the street network. The transient evolution of the concentration through the network is interpreted in terms of timescales linked with different processes. Lagrangian simulations are performed to shed light on associated dispersion pathways. We then use a simple street network approach to model entrainment and re-entrainment and apply it to model localised releases within and above an urban area. The results will help to inform the representation of entrainment processes in dispersion models.

Episodes in summer with an extremely high near-surface air temperature lasting several days or longer are termed as heat waves (Robinson, 2001; Lau and Nath, 2012). Under a meteorological point of view they are generally associated with quasi-stationary anticyclonic circulation anomalies, which produce subsidence, clear skies, warm-air advection and prolonged hot conditions in the near-surface atmosphere (Fischer et al ., 2007; Barriopedro et al ., 2011).

HWs represent a natural hazard and have significant impacts on wellbeing, efficiency and health of humans, which can lead to marked short-term increases of morbidity and mortality (Kovats and Ebi, 2006; Basu, 2009), particularly in cities, where most humans are living. From WMO Report (2013), the number of casualties of HW was increased by 2298% in 2001-2010 as compared with 1991-2000.

Our study of HWs in Ukraine (Shevchenko et al., 2014) for the period 1911-2011 indicate, that in contrast to other decades, the number of HW episodes was highest for almost all stations in the decade 2001–2010. For many stations, the longest HW duration occurred in the first two decades of August 2010, i.e. in the period of the extremely severe HW in Western Russia.

HW’s duration in Kiev was 18 days (from 31 July to 17 August 2010). During this period excess of the average daily temperature was at least 5 ?C; about 10 days this excess reached 8-10 ?C; maximal days temperature reached 38,2?C (8.08.2010).

During periods of HWs in urban areas created ideal conditions for the accumulation of a number of pollutants and formation of photochemical smog. Hot weather tends to acceleration of photochemical processes and increase particulate matters (PM), ozone (O3), formaldehyde and other pollutants concentrations.

In Kiev have seen during last 20 years a significant increase in the concentration of ozone precursors, especially NOx (by 50%) and non-methane VOCs, for example formaldehyde ( by 200%) due to the increasing number of road transport in the city.Mean annual concentration of PM10 (calculation method) increased by 150% for the central part of the city Kiev and 6 times exceeds European threshold value 40 œm (Snizhko, Shevchenko,2013).

Usually, during HW periods air quality gets worse. Concentration of O3, the photochemical air pollutant, can rise dramatically during the periods of warm and sunny weather that is characteristic for prolonged HWs. Hot weather also tends to the increase in the concentrations of PM and other pollutants.

To estimate the impact of HW on air pollution we have performed a comprehensive analysis of climatic characteristics and some air pollutants, in particular NO2, formaldehyde and AOT-index (Aerosol Optical Thickness). Statistical analysis shoved high positive correlation between temperature indexes and concentrations of pollutants. Significant values coefficients of correlations reached for connection between Tmax daily and concentrations of formaldehyde 0,56; for Tmax daily and NO2 - 0,62.

The accumulation in the urban atmosphere during HW period significant amount of NO2, as main precursor of photochemical smog, led to production of enormous amount of formaldehyde. Maximal concentration of formaldehyde reached 0,037 mg/m3 on 8 August 2010. This amount exceeds 12 times of threshold values and exceeds 2 times pre-HW’s level of concentration. At the same day concentration of aerosols reached maximal level too (AOT=1,31) and exceeds 4 times of its pre-HW’s level.

Concentration of NO2 was changed from 0,2 to 0,25 mg/m3, its sufficient increasing during HW period was not detected. During HWs periods by accelerating photochemical processes it take part as ozone-precursor in photochemical reactions.

Using statistical multivariate methods we made approximation of multivariate regression equation for short-time forecast of critical level of formaldehyde in urban air during HW taking into account both precursor concentration of NO2 and maximal daily temperature in the city.

Until 2050, the fraction of global urban population will increase to over 69%, which means that about 6.3 billion people are expected to live in urban areas. Although covering less than 3 % of the land surface, cities are the main contributor to global greenhouse gas emissions. With 78 % of total global carbon emissions, they are heavily implicated in global climate change. The Urban Heat Island (UHI) describes the tendency of an urban area to remain warmer than its rural surroundings. Urban planning strategies such as urban greening and bright building materials help to mitigate UHI formation but can also promote secondary effects on air quality.

This study presents a numerical modelling approach to analyse the effect of urban planning strategies on the urban heat island (UHI) intensity and further the feedback on the chemical composition of the urban atmosphere. The urban area of Stuttgart acts as test bed for the modelling.

The mesoscale chemical transport model WRF-Chem is used to investigate the effect of these urban heat island mitigation strategies on the surface concentration of primary (CO, NO, PM10) and secondary pollutants (O3).

Known mitigation strategies such as bright roofs and façades, urban greening and modification of the building density are in the focus. All these measures are able to reduce the urban temperature and thus mitigate urban heat island intensity.

Model results reveal that the most efficient way to cool down urban areas is the increase in the surface reflectivity. Changing the building albedo in the model from 0.2 to 0.7, lead to a reduction of the urban heat island by about 2 °C. The effect of urban greening and decreased building density is less.

The mitigation strategies which have been mentioned before promote changes in energetic and radiative properties of urban surfaces modifying the chemical nature of the urban atmosphere with regard to both primary and secondary compounds. A temperature reduction of 1 °C leads to an increase of NO and CO by 5-25 %, whereas the mean ozone concentration is projected to decrease by 5-8 %.

Reduced temperature on the surface and in the urban canopy layer influences the dynamical structure of the atmosphere, which leads to a reduction in turbulent mixing. The depth of the mixing layer is decreased accordingly. As a result, an increase of the near surface concentration of primary compounds is projected. Additionally, temperature directly controls the reactivity of chemical reactions, which explains the reduction of ozone concentration.

It has to be pointed out however, that different measures can generate secondary effects. The increased portion of short wave radiation due to a reflexion from white roofs for instance can promote photochemical reactions, leading to an increase of peak ozone levels although temperature has been reduced.

The additional emission of biogenic compounds (BVOCs) coming along with urban greening is also covered in this work. Besides the positive effect of urban greening on temperature and carbon sequestration it is important to discuss the BVOC emission potential of certain tree species. Especially in urban environments with considerable amounts of NOx, these biogenic compounds can act as precursor substances to the formation of secondary pollutants. This work tries to ponder the benefits and dangers from urban greening with regard to air quality.

Whereas in earlier studies the main effort had been put on the positive effect of temperature dependent reduction of urban ozone concentration, this work analyses a complete air chemistry, being able to show negative effects on primary compounds like CO, NO and PM10 as well. The main result of this work indicates the dominating role of atmospheric dynamics when analysing the impacts of urban heat island mitigation strategies on urban air quality.

An accurate understanding of urban air quality requires consideration of the coupled behavior between dispersion of reactive pollutants and atmospheric dynamics. Currently, urban air pollution is mostly dominated by traffic emissions. Only fast chemical reactions have influence on street pollutant concentration due to the short distances between sources and receptors. Therefore, some low reactive traffic-related pollutants like CO can be considered as practically inert species at microscale. However, nitrogen oxide (NO) and nitrogen dioxide (NO2) reacts extremely fast (time scales of the order of tens of seconds), besides the fact that Volatile Organic Compounds (VOCs) are involved in this complex chemical mechanism. Usually, NO and NO2 are modeled as passive tracer or using a steady state photochemistry (NO-NO2-O3) at microscale. But to properly account for VOCs, a more complex chemical mechanism is needed with more chemical reactions and more required computational time. Including the vast number of chemical species and reactions that occur in the urban atmosphere is not possible, and it is necessary to choose the most suitable mechanism for each scenario in terms of accuracy and CPU time.

The aim of this work is to investigate the flow and dispersion of NO and NO2 with different chemical approaches in simple urban configurations using a CFD-RANS model, in order to quantify the errors linked with the use of a simplified chemistry. Three types of simulations are performed: a) NO and NO2 as passive tracer (no-reactive), b) steady state NOx-O3 photochemistry (3 reactions and 3 species), and c) a more complex chemical scheme based on the RACM (‘Regional Atmospheric Chemistry Mechanism’) in the case of an urban atmosphere, developed to reduce the chemical system to 23 species and 25 reactions using CHEMATA software (Kirchner, Atmospheric Environment 2005). The influence of different parameters (zenith angle, VOCs-to-NOx emission ratio or temperature) on the errors in NO and NO2 concentration is investigated. The main conclusion is that errors induced by the use of the simple photochemistry state are larger when the VOCs-to-NOx emission ratio increases. Considering VOCs chemical reactions, lower NO and higher NO2 concentrations are obtained in comparison with steady state photochemistry. In this work, some criteria about what type of chemical mechanism is necessary to reproduce NO-NO2 concentration within streets are provided for different scenarios.