Abstract. For the purpose of qualifying and quantifying the climate impact of cities and urban surfaces in general on climate of central Europe, the surface parameterization in regional climate model RegCM4 has been extended with the Single-layer Urban Canopy Model (SLUCM). A set of experiments was performed over the period of 2005-2009 for central Europe, either without considering urban surfaces or with the SLUCM treatment. Results show a statistically significant impact of urbanized surfaces on temperature (up to 1.5 K increase in summer) as well as on the boundary layer height (increases up to 50 m). Urbanization further influences surface wind with a winter decrease up to −0.6 m s −1 , though both increases and decreases were detected in summer depending on the location relative to the cities and daytime (changes up to 0.3 m s −1 ). Urban surfaces significantly reduce the humidity over the surface. This impacts the simulated summer precipitation rate, showing a decrease over cities of up to −2 mm day −1 . Significant temperature increases are simulated over higher altitudes as well, not only within the urban canopy layer. With the urban parameterization, the climate model better describes the diurnal temperature variation, reducing the cold afternoon and evening bias of RegCM4.Sensitivity experiments were carried out to quantify the response of the meteorological conditions to changes in the parameters specific to the urban environment, such as street width, building height, albedo of the roofs and anthropogenic heat release. The results proved to be rather robust and the choice of the key SLUCM parameters impacts them only slightly (mainly temperature, boundary layer height and wind velocity).Statistically significant impacts are modelled not only over large urbanized areas, but the influence of the cities is also evident over rural areas without major urban surfaces. It is shown that this is the result of the combined effect of the distant influence of the cities and the influence of the minor local urban surface coverage.

Abstract. Cities are characterized by different physical properties of surface compared to their rural counterparts, resulting in a specific regime of the meteorological phenomenon. Our study aims to evaluate the impact of typical urban surfaces on the central European urban climate in several model simulations, performed with the Weather Research and Forecasting (WRF) model and Regional Climate Model (RegCM). The specific processes occurring in the typical urban environment are described in the models by various types of urban parameterizations, greatly differing in complexity. Our results show that all models and urban parameterizations are able to reproduce the most typical urban effect, the summer evening and nocturnal urban heat island, with the average magnitude of 2–3 °C. The impact of cities on the wind is clearly dependent on the urban parameterization employed, with more simple ones unable to fully capture the wind speed reduction induced by the city. In the summer, a significant difference in the boundary-layer height (about 25 %) between models is detected. The urban-induced changes of temperature and wind speed are propagated into higher altitudes up to 2 km, with a decreasing tendency of their magnitudes. With the exception of the daytime in the summer, the urban environment improves the weather conditions a little with regard to the pollutant dispersion, which could lead to the partly decreased concentration of the primary pollutants.

Abstract. It is well known that the urban canopy (UC) layer, i.e., the layer of air corresponding to the assemblage of the buildings, roads, park, trees and other objects typical to cities, is characterized by specific meteorological conditions at city scales generally differing from those over rural surroundings. We refer to the forcing that acts on the meteorological variables over urbanized areas as the urban canopy meteorological forcing (UCMF). UCMF has multiple aspects, while one of the most studied is the generation of the urban heat island (UHI) as an excess of heat due to increased absorption and trapping of radiation in street canyons. However, enhanced drag plays important role too, reducing mean wind speeds and increasing vertical eddy mixing of pollutants. As air quality is strongly tied to meteorological conditions, the UCMF leads to modifications of air chemistry and transport of pollutants. Although it has been recognized in the last decade that the enhanced vertical mixing has a dominant role in the impact of the UCMF on air quality, very little is known about the uncertainty of vertical eddy diffusion arising from different representation in numerical models and how this uncertainty propagates to the final species concentrations as well as to the changes due to the UCMF. To bridge this knowledge gap, we set up the Regional Climate Model version 4 (RegCM4) coupled to the Comprehensive Air Quality Model with Extensions (CAMx) chemistry transport model over central Europe and designed a series of simulations to study how UC affects the vertical turbulent transport of selected pollutants through modifications of the vertical eddy diffusion coefficient (Kv) using six different methods for Kv calculation. The mean concentrations of ozone and PM2.5 in selected city canopies are analyzed. These are secondary pollutants or having secondary components, upon which turbulence acts in a much more complicated way than in the case of primary pollutants by influencing their concentrations not only directly but indirectly via precursors too. Calculations are performed over cascading domains (of 27, 9, and 3 km horizontal resolutions), which further enables to analyze the sensitivity of the numerical model to grid resolution. A number of model simulations are carried out where either urban canopies are considered or replaced by rural ones in order to isolate the UC meteorological forcing. Apart from the well-pronounced and expected impact on temperature (increases up to 2 ∘C) and wind (decreases by up to 2 ms−1), there is a strong impact on vertical eddy diffusion in all of the six Kv methods. The Kv enhancement ranges from less than 1 up to 30 m2 s−1 at the surface and from 1 to 100 m2 s−1 at higher levels depending on the methods. The largest impact is obtained for the turbulent kinetic energy (TKE)-based methods. The range of impact on the vertical eddy diffusion coefficient propagates to a range of ozone (O3) increase of 0.4 to 4 ppbv in both summer and winter (5 %–10 % relative change). In the case of PM2.5, we obtained decreases of up to 1 µg m−3 in summer and up to 2 µg m−3 in winter (up to 30 %–40 % relative change). Comparing these results to the “total-impact”, i.e., to the impact of all meteorological modifications due to UCMF, we can conclude that much of UCMF is explained by the enhanced vertical eddy diffusion, which counterbalances the opposing effects of other components of this forcing (temperature, humidity and wind). The results further show that this conclusion holds regardless of the resolution chosen and in both the warm and cold parts of the year.

Rising emissions of anthropogenic greenhouse gases (GHG) have led to tropospheric warming and stratospheric cooling over recent decades. As a thermodynamic consequence, the troposphere has expanded and the rise of the tropopause, the boundary between the troposphere and stratosphere, has been suggested as one of the most robust fingerprints of anthropogenic climate change. Conversely, at altitudes above ∼55 km (in the mesosphere and thermosphere) observational and modeling evidence indicates a downward shift of the height of pressure levels or decreasing density at fixed altitudes. The layer in between, the stratosphere, has not been studied extensively with respect to changes of its global structure. Here we show that this atmospheric layer has contracted substantially over the last decades, and that the main driver for this are increasing concentrations of GHG. Using data from coupled chemistry-climate models we show that this trend will continue and the mean climatological thickness of the stratosphere will decrease by 1.3 km following representative concentration pathway 6.0 by 2080. We also demonstrate that the stratospheric contraction is not only a response to cooling, as changes in both tropopause and stratopause pressure contribute. Moreover, its short emergence time (less than 15 years) makes it a novel and independent indicator of GHG induced climate change.

<p><strong>Abstract.</strong> Urban surfaces due to specific geometry and physical properties bring modified transport of momentum, moisture and heat between them and the air above and perturb the radiative, thermal and mechanical balance resulting in changed meteorological condition (e.g. the UHI – urban heat island phenomenon). From an air quality perspective, many studies argue that one of the most important changes is the increased turbulence enhancing vertical mixing of pollutants above cities, although increased temperatures and wind stilling play an important role too. Using the regional climate model RegCM4 coupled to chemistry transport model CAMx over central Europe we study how urban surfaces affect the vertical turbulent transport of selected pollutants through modifications of the vertical eddy diffusion coefficient (<i>Kv</i>). For the period of 2007–2011 and over central Europe numerous experiments are carried out in order to evaluate the impact of six different methods for <i>Kv</i> calculation on the surface concentrations as well as vertical profiles of ozone and PM<sub>2.5</sub> over selected cities (Prague and Berlin). Three cascading domains are set up at 27 km, 9 km and 3 km resolutions, which further enables to analyze the sensitivity to model grid resolution. Numerous experiments are performed where urban surfaces are considered or replaced by rural ones in order to isolate the urban canopy meteorological forcing. Apart from the well pronounced and expected impact on temperature (increases up to 2 °C) and wind (decreases up to −2 m s<sup>−1</sup>) there is strong impact on vertical eddy diffusion in all of the six <i>Kv</i> methods. The <i>Kv</i> enhancement ranges from a few 0.5 up to 30 m<sup>2</sup> s<sup>−1</sup> at the surface and from 1 to 100 m<sup>2</sup> s<sup>−1</sup> at higher levels depending on the methods, while the turbulent kinetic energy (TKE) based methods produce the largest impact.</p> <p>The range of impact on the vertical eddy diffusion coefficient propagates to a range of ozone (O<sub>3</sub>) increase of 0.4 to 4 ppbv near the surface in both summer and winter, while at higher levels, decreases occur from a few −0.4 ppbv to as much as −2 ppbv. In case of PM<sub>2.5</sub>, enhanced vertical eddy diffusion leads to decrease of near surface concentrations ranging from almost zero to −1 μg m<sup>−3</sup> in summer and to decreases from −0.5 to −2 μg m<sup>−3</sup> in winter. Comparing these results to the <q>total-impact</q>, i.e. to the impact of all considered urban meteorological changes, we can conclude that much of the overall urban meteorological forcing is explained by acting of the enhanced vertical eddy diffusion, which counterweights the opposing effects of other components of this forcing (temperature, humidity and wind impact). The results further show that this conclusion holds regardless of the resolution chosen and in both the warm and cold part of the year. Our study demonstrates the dominant role of turbulent transport of pollutants above urban areas and stresses the need for further investigation how variation of urban land-use influence the pollutant transport from the urban canopy.</p>

Abstract. This paper deals with the urban land-surface impact (i.e. the urban canopy meteorological forcing; UCMF) on extreme air pollution for selected central European cities for present-day climate conditions (2015–2016) using three regional climate-chemistry models: the regional climate models RegCM and WRF-Chem (its meteorological part), the chemistry transport model CAMx coupled to either RegCM and WRF and the chemical component of WRF-Chem. Most of the studies focused on the change of average conditions or only on a selected winter and summer air pollution episode. Here we extend these studies by focusing on long term extreme air pollution levels by looking at not only the change of average values but also their high (and low) percentile values and we combine the analysis with investigating selected high pollution episodes too. As extreme air pollution is often linked to extreme values of meteorological variables (e.g. low planetary boundary layer height, low winds, high temperatures), the extreme meteorological modifications will be analyzed too. The validation of model results shows reasonable model performance for regional scale temperature and precipitation. Ozone is overestimated by about 10–20 μg m−3, on the other hand, extreme summertime ozone values are underestimated by all models. Modeled nitrogen dioxide (NO2) concentrations are well correlated with observations, but results are marked with a systematic underestimation up to 20 μg m−3. PM2.5 (particles with diameter

Abstract. This paper deals with the urban land-surface impact (i.e., the urban canopy meteorological forcing; UCMF) on extreme air pollution for selected central European cities for present-day climate conditions (2015–2016) using three regional climate-chemistry models: the regional climate models RegCM and WRF-Chem (its meteorological part), the chemistry transport model CAMx coupled to either RegCM and WRF and the “chemical” component of WRF-Chem. Most of the studies dealing with the urban canopy meteorological forcing on air pollution focused on change in average conditions or only on a selected winter and/or summer air pollution episode. Here we extend these studies by focusing on long-term extreme air pollution levels by looking at not only the change in average values, but also their high (and low) percentile values, and we combine the analysis with investigating selected high-pollution episodes too. As extreme air pollution is often linked to extreme values of meteorological variables (e.g., low planetary boundary layer height, low winds, high temperatures), the urbanization-induced extreme meteorological modifications will be analyzed too. The validation of model results show reasonable model performance for regional-scale temperature and precipitation. Ozone is overestimated by about 10–20 µg m−3 (50 %–100 %); on the other hand, extreme summertime ozone values are underestimated by all models. Modeled nitrogen dioxide (NO2) concentrations are well correlated with observations, but results are marked by a systematic underestimation up to 20 µg m−3 (−50 %). PM2.5 (particles with diameter ≤2.5 µm) are systematically underestimated in most of the models by around 5 µg m−3 (50 %–70 %). Our results show that the impact on extreme values of meteorological variables can be substantially different from that of the impact on average ones: low (5th percentile) temperature in winter responds to UCMF much more than average values, while in summer, 95th percentiles increase more than averages. The impact on boundary layer height (PBLH), i.e., its increase is stronger for thicker PBLs and wind speed, is reduced much more for strong winds compared to average ones. The modeled changes in ozone (O3), NO2 and PM2.5 show the expected pattern, i.e., increase in average 8 h O3 up to 2–3 ppbv, decrease in daily average NO2 by around 2–4 ppbv and decrease in daily average PM2.5 by around −2 µg m−3. Regarding the impact on extreme (95th percentile) values of these pollutants, the impact on ozone at the high end of the distribution is rather similar to the impact on average 8 h values. A different picture is obtained however for extreme values of NO2 and PM2.5. The impact on the 95th percentile values is almost 2 times larger than the impact on the daily averages for both pollutants. The simulated impact on extreme values further well corresponds to the UCMF impact simulated for the selected high-pollution episodes. Our results bring light to the principal question: whether extreme air quality is modified by urban land surface with a different magnitude compared to the impact on average air pollution. We showed that this is indeed true for NO2 and PM2.5, while in the case of ozone, our results did not show substantial differences between the impact on mean and extreme values.

No abstract