TRANSPHORM Transport related Air Pollution and Health impacts – Integrated Methodologies for Assessing Particulate Matter Collaborative project, Large-scale Integrating Project SEVENTH FRAMEWORK PROGRAMME

ENV.2009.1.2.2.1 Transport related air pollution and health impacts

Deliverable D2.4.3

Interaction between pollution and climate for future transport scenarios (Report on the application of meteorology with climate change for future transport scenarios)

Due date of deliverable:

project month 24

Actual submission date:

project month 24

Start date of project: 1 January 2010

Duration:

48 months

Organisation name of lead contractor for this deliverable: DMI Scientists responsible for this deliverable:

Prof. A. Baklanov

Authors: Baklanov A., Collins B., Folberth G., Gauss M., Gonzalez-Aparicio I., Jonson J.E., Haugen J. E., Kong X., Mahura A., Nuterman R., Nyiri A., Simpson D., Soares J., Sofiev M., Sokhi R., Theloke J., Tsyro S., Zakey A. Revision: [1]

D2.4.3

TRANSPHORM

Deliverable

Contents 1. Introduction

3

2. Climate-Chemistry Interaction Studies in MEGAPOLI Relevant to TRANSPHORM

5

2.1. Modelling tools for the simulation of multi-scale city air quality - climate interactions

5

2.2. Key feedbacks between air quality, local climate and global climate change relevant to cities and transport emissions

7

2.3. Current impact of Megacities on regional and global climate

10

2.4. Effects of growing Megacities on future climate at global and regional scales

12

2.5. Policy options to reduce transport emissions of air pollutants and GHGs in cities

14

2.6. Further research needs

22

3. Climate-Chemistry Interaction Studies in CityZen Relevant to TRANSPHORM

23

3.1. CTMs driven by GCM meteorology

23

3.2. Traffic mitigation scenario

24

3.3. Studies of the Eastern Mediterranean ozone levels and its precursors in summer

26

4. EMEP Study of Meteorology and Climate Effects on Air Pollution

28

4.1. Meteorological data from HIRHAM

28

4.2. Climate effects on air pollution

30

5. ENSCLIMA Study on Impact of Climate Change on Surface Ozone over Europe

33

5.1. Models employed and methods applied

33

5.2. DMI modelling system setup

36

5.3. EnvClimA and ensemble models results

36

6. PM10 Future Projections for a South-Western European Medium Size City

38

6.1. Urban climate system for generation of air pollution scenarios

38

6.2. Bilbao metropolitan area

39

6.3. HIRHAM climatic future scenarios for modelling PM10 impacts

40

6.4. Future projections of climate change impacts on PM10

41

7. Application of meteorology with climate change for future transport scenarios

44

7.1. HadGEM-WRF-CMAQ modelling system and selected case studies

44

7.2. Contributions of climate change and emission change to meteorology and PM

45

8. Main Conclusions and Recommendations

2 of 52

50

Deliverable

TRANSPHORM

D2.4.3

1. Introduction Emissions from ships, aircrafts and land transport, in particular from road transport, can have significant impacts on the atmosphere and on climate change. The EC FP7 project TRANSPHORM brings together internationally leading air quality and health researchers and users to improve the knowledge of transport related airborne particulate matter (PM) and its impact on human health and to develop and implement assessment tools for scales ranging from city to Europe. Several previous studies in Europe already considered some issues of transport emission impacts on the atmosphere pollution and on climate change. For example, Uherek et al. (2010) gives a nice overview of past, present and future emissions from land transport, of their impacts on the atmospheric composition and air quality, on human health and climate change and on options for mitigation. In the past vehicle exhaust emission control has successfully reduced emissions of nitrogen oxides, carbon monoxide, volatile organic compounds and particulate matter. This contributed to improved air quality and reduced health impacts in industrialised countries. In developing countries however, pollutant emissions have been growing strongly, adversely affecting many populations. In addition, ozone and particulate matter change the radiative balance and hence contribute to global warming on shorter time scales. Latest knowledge on the magnitude of land transport's impact on global warming is reviewed here. In the future, road transport's emissions of these pollutants are expected to stagnate and then decrease globally. This will then help to improve the air quality notably in developing countries. On the contrary, emissions of carbon dioxide and of halocarbons from mobile air conditioners have been globally increasing and are further expected to grow. Consequently, road transport's impact on climate is gaining in importance. The expected efficiency improvements of vehicles and the introduction of biofuels will not be sufficient to offset the expected strong growth in both, passenger and freight transportation. Technical measures could offer a significant reduction potential, but strong interventions would be needed as markets do not initiate the necessary changes. Further reductions would need a resolute expansion of low-carbon fuels, a tripling of vehicle fuel efficiency and a stagnation in absolute transport volumes. Land transport will remain a key sector in climate change mitigation during the next decades. Earlier Brasseur et al. (1998) published an extensive European scientific assessment of the atmospheric effects of aircraft emissions. They stressed that, in spite of considerable progress made to better quantify the atmospheric impact of present and future fleet of aircraft, many uncertainties remain. The conclusions presented by Brasseur et al. (1998) should therefore be regarded as interim findings, which will be susceptible of changing as new scientific information becomes available. As indicated there, more reliable predictions of future changes in the concentration of ozone in the lower stratosphere and upper troposphere will only be possible if fundamental processes occurring in the atmosphere are better understood through carefully designed field and laboratory experiments. Among the questions that require urgent attention they mentioned, in particular, the following: the impact of soot and acids on particle formation and related changes in cloudiness and chemistry; evaluations of the climatic impact of the changes in composition and cloudiness relative to the impact of CO2; the precise composition of aircraft exhaust as regard to particle formation processes. This report for the TRANSPHORM deliverable D.4.3 ‘Report on the application of meteorology with climate change for future transport scenarios’ is realised within the Task 2.4.3: Interaction between pollution and climate for future transport scenarios (DMI, TNO, UH, Met.no, FMI, USTUTT). This task is focused on an assessment that describes the influence of climate change on the transport scenarios carried out in the impact assessment SP4. Changes in climate and its relationship to air quality were carried out in the two FP7 EC projects MEGAPOLI and CITYZEN. 3 of 52

D2.4.3

TRANSPHORM

Deliverable

Though TRANSPHORM will not directly study this relationship, it will integrate results from these two projects and, in so doing, will provide information on the effects of climate change on the policy measures designed for reducing the health effects of transport emissions. Analyses of possible AQ-Climate interactions and mechanisms of climate change influence on transport-related pollution were initiated and reported on the TRANSPHORM First Annual Meeting in Amsterdam, 1-3 December 2010. Studies of changes in climate and its relationship to air quality are carrying out in several collaborating projects: first of all MEGAPOLI and CITYZEN, PEGASOS, EnsCLIMA, etc. In particular within the Nordic EnsClima network (lead by SMHI) DMI, FMI and Met.no are realising simulations of climate change effects on atmospheric pollution over Northern Europe for 2010-2050 (see ACPD paper Langner et al., 2012). TRANSPHORM will not directly study this relationship, but will integrate results from these projects and provide information on the effects of climate change on the policy measures designed for reducing the health effects of transport emissions (realizing in collaboration with SP5). Within the framework of EnviroHIRLAM modelling system a new version of the online coupled model suitable for climate time-scale long-term simulations (EnvClim), is developed; it is under testing and evaluation at the current stage. So, we can consider the ACPD paper by Langner et al., 2012 as the D2.4.3 report for the EC, however we received additional contributions from several partners, so this extended report for D2.4.3 includes most of them as well. Concerning scenarios, at least 3 datasets have been analysed in the beginning of the project: the RCPs, the GEA scenarios and the HTAP scenarios (knowing that almost the same people are developing GEA and HTAP scenarios). But some sectors are not really well considered, such as shipping (no scenario include the opening of the Arctic passages) and biomass burning (the RCPs are quite questionable for biomass burning), so the main focus should be done on these gaps and uncertainties in the scenarios elaborated for TRANSPHORM. References Brasseur, G.P., R.A. Cox, D. Hauglustaine, I. Isaksen, J. Lelieveld, D.H. Lister, R. Sausen, U. Schumann, A. Wahner, P. Wiese. (1998) European scientific assessment of the atmospheric effects of aircraft emissions. Atmospheric Environment, Vol. 32, No. 13, p. 2329-2418. Langner, J., M. Engardt, A. Baklanov, J. H. Christensen, M. Gauss, C. Geels, G., B. Hedegaard, R. Nuterman, D. Simpson, J. Soares, M. Sofiev, P. Wind and A. Zakey (2012) A multi-model study of impacts of climate change on surface ozone in Europe. ACPD, Special Issue: EMEP an integrated system of models and observations in support of European air quality and policy, http://www.atmos-chem-phys-discuss.net/12/4901/2012/ Uherek, E., T. Halenka, J. Borken-Kleefeld, Y. Balkanski, T. Berntsen, C. Borrego, M. Gauss, P. Hoor, K. Juda-Rezler, J. Lelieveld, D. Melas, K. Rypdal, S. Schmid (2010) Transport impacts on atmosphere and climate: Land transport. Atmospheric Environment 44: 4772-4816

4 of 52

Deliverable

TRANSPHORM

D2.4.3

2. Climate-Chemistry Interaction Studies in MEGAPOLI Relevant to TRANSPHORM By A. Baklanov (DMI) and MEGAPOLI team (see the Final Report on megapoli.info) Several studies regarding climate-chemistry interactions have been performed within the FP7 EC MEGAPOLI project. Results from some of these studies of relevance for the TRANSPHORM project are described briefly below. 2.1. Modelling tools for the simulation of multi-scale city air quality - climate interactions Processes involving nonlinear interactions and feedbacks between emissions, chemistry and meteorology require coherent and robust modelling approaches. This is particularly important where multiple spatial and temporal scales are involved with a complex mixture of pollutants from large sources, as in the case of Megacities. Numerical weather and air pollution prediction models are now able to approach urban-scale resolution, as detailed input data are becoming more often available. MEGAPOLI suggested a comprehensive integrated modelling framework which was tested for a range of Megacities within Europe and across the world to increase our understanding of how large urban areas and other hotspots affect air quality and climate on multiple scales. The integration strategy in MEGAPOLI was not focused on any particular meteorological and/or air pollution modelling system. The approach considered an open integrated framework with flexible architecture and with a possibility of incorporating different meteorological and chemical transport models. Baklanov (2010) has explained the levels of integration and orders of complexity (temporal and spatial scales and ways of integration) considered in MEGAPOLI:  Level 1 – Spatial: One-way (Global to regional to urban to street); Models: All.  Level 2 – Spatial: Two-way (Global from and to regional from and to urban); Models: UMWRF-CMAQ, SILAM, M-SYS, FARM.  Level 3 – Time integration: Time-scale and direction; Direct and Inverse modelling. o Order A – off-line coupling, meteorology / emissions to chemistry; Models: All. o Order B – partly online coupling, meteorology to chemistry and emissions; Models: UKCA, DMAT, M-SYS, UM-WRF-Chem, SILAM. o Order C – fully online integrated with two-way feedbacks, meteorology from and to chemistry and emissions; Models: UKCA, WRF-Chem, Enviro-HIRLAM, EMAC (former ECHAM5/MESSy). A multi-scale modelling framework for global to street scale includes nesting of the land-use characteristics and scenarios, anthropogenic heat fluxes, emission inventories and scenarios, and the representation of atmospheric processes using two-way nesting, zooming, nudging, parameterizations and urban increment methodology. The new or improved interfaces for coupling (direct links between emissions, chemistry and meteorology at every time step) can be implemented or developed and common formats for data exchange can be defined to ease the implementation and to help combine the different models via conventional data exchange protocols. The current chemistry schemes were examined for their suitability to simulate the impact of complex emissions from megacities. The coupled model systems were applied to different European megacities during the project. The framework is used and demonstrated for selected models including UKCA (MetO), WRF-CMAQ (UH-CAIR), PMCAMx (FORTH), Enviro-HIRLAM (DMI), 5 of 52

D2.4.3

TRANSPHORM

Deliverable

STEM/FARM (ARIANET), M-SYS (UHam) and EMAC (MPIC) on different scales. This part of the studies was linked also to the requirements and use of simpler tools for assessing air quality impacts within megacities (OSCAR - UH-CAIR, AIRQUIS - NILU, URBIS - TNO, EcoSence UStutt). The focus on integrated systems is timely, since recent research has shown that meteorology/climate and chemistry feedbacks are important in the context of many research areas and applications, including numerical weather forecasting, climate modelling, air quality forecasting, climate change, and Earth system modelling. The prediction and simulation of the coupled evolution of atmospheric transport and chemistry will remain one of the most challenging tasks in environmental modelling over the next decades. Many of the current environmental challenges in weather, climate, and air quality involve strongly coupled systems. It is well accepted that weather is of decisive importance for air quality, or for the aerial transport of hazardous materials. It is also recognized that chemical species will influence the weather by changing the atmospheric radiation budget as well as through cloud formation. Until recently however, because of the complexity and the lack of appropriate computer power, air chemistry and weather forecasts have developed as separate disciplines, leading to the development of separate modelling systems that are only loosely coupled (offline). In NWP, the dramatic increase in computer power enables us to use higher resolution to explicitly resolve fronts, convective systems, local wind systems, and clouds, or to increase the complexity of the numerical models. Additionally we can now directly couple air quality forecast models with numerical weather prediction models to produce a unified modelling system – online – that allows two-way interactions. While climate modelling centres have gone to an Earth System Modelling approach that includes atmospheric chemistry and oceans, NWP centres as well as entities responsible for Air Quality forecasting are only beginning to discuss whether an online approach is important enough to justify the extra cost. NWP and AQ forecasting centres may have to invest in additional computer power as well as additional man power, since additional expertise may be required. We are in favour of integrating weather and chemistry together, for both NWP and air quality and chemical composition forecasting. For NWP centers, an additional attractiveness of the online approach is its possible usefulness for meteorological data assimilation (Hollingsworth et al., 2008), where the retrieval of satellite data and direct assimilation of radiances will likely improve – assuming that the modelling system can beat climatology when forecasting concentrations of aerosols and radiatively active gases. The application of integrated modelling systems for the megacity air quality modelling was implemented for different megacities in Europe as we all other cities (Schlunzen, 2011). In MEGAPOLI (Francis, et al., 2011) describes a number of different integrated modelling systems applied to assess the impact of air quality of European and other megacities. These include RAMS-FARM, WRF-Chem, WRF-CMAQ, LOTOS-EUROS, M_SYS-METRAS-MECTM, MEMO-MARS, GRECAPS, Enviro-HIRLAM, MESO-NH, and ENSEMBLE. Examples of the use of integrated modelling systems over the European megacities (London, Paris, Po-Valley, and Rhine-Ruhr) and other cities (Lithuania, St. Petersburg metropolitan area-Russia) have been discussed. WRF-CMAQ simulations over London under estimated the PM measurements from air quality stations. There are a number of reasons that may explain this disagreement. These include the underestimation of certain PM emissions, such as coarse dust fractions and non-exhaust sources in the emission inventories. The sensitivity of PM2.5 and PM10 to boundary conditions used in the regional simulation could also provide insight to the model underestimation. LOTUS- EUROS, MEMO-MARS, WRF-CMAQ, PMCAMx, and MESO-NH modelling system have been applied to study the air quality over Paris. The results from LOTUS-EUROS suggested that the model simulation around the Paris area clearly benefit from emission inventories based on 6 of 52

Deliverable

TRANSPHORM

D2.4.3

nesting information from local inventories for the simulation of particulate matter. However, the average concentrations of PM predicted by LOTUS-EUROS over Paris are much lower compared to available measurements. Simulations with ‘nested’ emissions represented better diurnal cycles for PM when compared with observations. The online coupled modelling system MEMO-MARS has been used to study the feedbacks of direct aerosol effect on PM10 simulations over Paris. The difference field was produced by removing the control simulations from the online results. PM 10 concentrations revealed a clear increase over almost the entire domain, reaching up to 2.5 μg m-3 in the central Paris area and the southern part of the computational domain, where higher pollution loads occur due to the prevailing wind flow during the simulation period. This study indicates that feedback processes can be significant for air quality assessment and should form the basis of future studies. Similarly, PMCAMx has been employed to predict concentrations of PM and ozone during July 2009 and winter of 2010. The use of the volatility basis approach in PMCAMx has resulted in significant improvements in the ability to reproduce organic aerosol levels over Paris. MESO-NH has been used to predict primary and secondary aerosols components over Paris region. Analysis of air quality over the Po valley with RAMS-FARM modelling system has shown that the large urbanised areas contribute significantly to regional levels of pollutants such as PM 2.5. The study also showed that the Po valley footprint, under prevailing anti-cyclonic circulation conditions, can extend well beyond the immediate region and can affect southern parts of continental Europe and northern Italy. On urban scales, a study with Enviro-HIRLAM for Vilnius has indicated that modifications of the surface parameters can have significant impact on meteorological fields that affect air pollution. For example, it was found that the air temperature at 2 m height is typically higher in modified simulation runs. As an extension of this study, the effect of modified roughness, anthropogenic heat fluxes and the albedo (urbanization) in Enviro-HIRLAM modelling system has been studied for St. Petersburg (Russia). Results showed that for winter the differences between control vs. urbanized runs over the metropolitan area and surroundings were the following: wind at 10 m up to 2 m/s (with a maximum up to 2.9 m/s, at nighttime) and air temperature at 2 m is more than 1ºC (with a maximum up to 2.7oC, at nighttime). The air quality over other non-European megacities like Mexico (PMCAMx and extended version of WRF-Chem), New York (PMCAMx), Phoenix (MM5-CMAQ), Shanghai (WRF-Chem and WRF-CMAQ) and New Delhi (SAFAR) has been also studied. The regional CTMs PMCAMx and an extended version of WRF-Chem have also been used successfully to Mexico City for the periods of April 2003 and March 2006. As mentioned earlier summer levels of PM10 are underestimated as was observed for Shanghai and European cities. An example of air quality forecast (SAFAR) for New Delhi has been provided and was found to be within 10% to 20% confidence limit of observations. The forecast model was also able to resolve the diurnal pattern. For this case study sensitivity tests revealed that windblown dust from paved and unpaved roads and construction activities are major contributors to PM and on occasion can be more than the levels due to fossil fuel combustion from transport sector. Although these examples show successful application of integrated models to investigate air quality affecting megacities, a number of research developments are required. These include greater consistency in the use of data for megacities, moving towards online or coupled approaches and the need for dedicated and targeted data sets for model evaluation purposes. Prediction of PM remains a challenge for several models and is an important area for continued research. See more details in the MEGAPOLI deliverable reports D7.1, D7.2 and D.7.3. 2.2. Key feedbacks between air quality, local climate and global climate change relevant to cities and transport emissions

7 of 52

D2.4.3

TRANSPHORM

Deliverable

Figure 2.1 shows the main linkages between megacities, air quality and climate and the main feedbacks, ecosystem, health and weather impact pathways, and mitigation routes which were investigated in MEGAPOLI. The relevant temporal and spatial scales are also depicted in Figure.

Figure 2.1: Connections between Megacities, air quality and climate.

Direct impact of climate change on air quality in Megacities can be significant due to temperature (BVOC fluxes, wild fires, deposition, O3, CH4, SOA, pSO4, pNO3), radiation (photolysis), clouds and precipitation change. As climate changes ozone concentrations will further increase, if no emission reduction measures take place. However, the expected emissions reduction during the next several decades is much stronger than the ozone increase due to climate change expected in the most megacities regions studied within the project. Only a more frequent appearance of extremes like heat waves under the climate change might bring air quality problems with e.g. exceedances of limits of ozone concentrations on local or regional scales. The global climate change will also affect on the megacity climate, UHI and urban-induced weather, however it could be different in different climate zones around the world. In general the diurnal temperature range as well as UHI of megacities may amplify as a result of climate change. However, for coastal Megacities climate change-induced increases in land temperature can lead to an increase in the temperature gradient between land and sea resulting in more intensive and frequent sea breeze events and associated cooler air and fog. Potential impacts of urban aerosol feedbacks include:  a reduction of downward solar radiation (direct effect);  changes in surface temperature, wind speed, relative humidity, and atmospheric stability (semi-direct effect);  a decrease in cloud drop size and an increase in drop number by serving as cloud condensation nuclei (first indirect effect);  an increase in liquid water content, cloud cover, and lifetime of low level clouds, and suppression or enhancement of precipitation (second indirect effect). The urban effects of Megacities together with the direct and indirect aerosol feedbacks constitute an important aspect of the multiple interactions between urban air quality and climate. The aerosol 8 of 52

Deliverable

TRANSPHORM

D2.4.3

indirect effects can lead to changes in daytime temperatures and PBL height, while pollutant concentrations are moderately affected (Figure 2.4). The radiative forcing associated with the aerosol direct effect has a significant impact on average surface temperature and wind speed, as well as on the development of a lower inversion layer. Compared to the direct and indirect aerosol feedbacks, urban feedbacks exhibit the same order of magnitude effects on mixing height, but with strong sensitivity of chemistry and a strong non-linearity. On the local and meso-scale, an observable Megacity effect on meteorology is expected to occur, both via the influence of the Urban Heat Island (UHI) and the city plume. On the regional scale such effects can also be expected since the urban plume can extend up to a few hundred kilometers (Figure 2.3). On the global scale, the effect of UHI is expected to be negligible.

(a)

(b)

(c)

Figure 2.2: Comparison time-series of (a) temperature, (b) wind speed, and (c) PM10 concentrations calculated by taking into account (“coupled”) or without (“baseline”) the direct aerosol effect.

(a)

(b)

(c)

Figure 2.3. Difference maps demonstrating the modeled influence of the aerosol direct effect on (a) wind speed, (b) turbulent kinetic energy and (c) temperature, for the greater Paris area (Douros et al., 2011).

The direct aerosol effect on mesoscale meteorological and dispersion fields over the urban area of Paris, France has been studied by modelling the response of the primary meteorological variables governing the transport and dispersion of pollutants, as well as trends in particulate matter concentrations. The impact of the direct aerosol effect was found to be substantial with regard the turbulent characteristics of the flow near the surface. Nevertheless, the performance of the modelling tools in predicting urban meteorology and air quality in the specific case was only improved marginally, in this way suggesting a stronger influence of the indirect and higher-order aerosol effects that the particular models did not take into account.

9 of 52

D2.4.3

TRANSPHORM

Deliverable

Figure 2.4: Indirect aerosol effects simulated by Enviro-HIRLAM: Monthly averaged CCN number concentration (x107 m3) at 850 hPa level (Korsholm et al., 2010)

Simulations of indirect effects of urban plume aerosols on meteorological fields showed that aerosol feedbacks through the second indirect effect induce considerable changes in meteorological fields and large changes in chemical composition, in particular NO2, in a case of convective cloud cover and little precipitation. However, monthly averaged changes in surface temperature due to aerosol indirect effects of aerosol emissions in Western Europe on surface temperature are about 0.5°C. It also shows that modifications of cloud properties due to anthropogenic aerosols (Figure 8.4) may take place through modification of cloud reflectance and precipitation development, referred to as the first and second aerosol indirect effects respectively. In particular the indirect effects led to stronger convection and heavier precipitation in some places and suppression of precipitation in other places. Given the very small global climate impact of the air quality effects of Megacities it is expected that any feedbacks on global climate will be very small. On regional scale the effect of Megacities on air-quality is high and locally in vicinity of Megacities studied it can achieve a contribution of a few tens percent in concentrations of pollutants like ozone (mainly in summer) or some aerosols (especially in winter). This suggests the Megacities potential to contribute eventually up to a few degrees (given theoretical complete emission reduction) in surface temperature, but rather locally in space and time, mainly due to semi-direct and indirect effects. This effect will be in order of magnitude weaker at climate time scales. It should be mentioned that these conclusions come from one year simulation only, longer (at least, one decade) simulations or ensembles runs can provide more comprehensive information. The direct impact of climate change on air quality in megacities is significant due to changes in temperature (BVOC fluxes, wild fires, deposition, O3, CH4, SOA, pSO4, pNO3), radiation (photolysis), clouds, and precipitation. As climate changes, ozone concentrations will further increase, unless emission reduction measures are implemented. However, the expected emissions reduction during the next several decades is much stronger than the ozone increase due to climate change expected in the most megacities regions studied within the project. Only a more frequent appearance of extremes like heat waves under the climate change might bring air quality problems with them, e.g. exceedances of limits of ozone concentrations on local or regional scales. See more details in MEGAPOLI deliverable reports D4.1, D4.3, and D.6.2. 2.3. Current impact of Megacities on regional and global climate Megacities have strong so-called urban heat islands (UHI), due to differences in surface properties and waste heat from anthropogenic activity. The effects of UHI can be substantial. Anthropogenic heat fluxes for Megacities can be up to 50-500 W m-2, locally (e.g., in Tokyo) reaching 1600 W m-2. Hence Megacities can be warmer than surrounding rural environments by several degrees Celsius 10 of 52

Deliverable

TRANSPHORM

D2.4.3

(see e.g. Figure 2.5). This heating obviously impacts the local environment directly, but also affects the regional air circulation. The local and partly regional wind patterns are also affected by an increase in the so-called roughness length due to the Megacity building height and density. High resolution simulations show that the boundary layer height (with is extremely important for air pollution) can be increased due to the UHI up to several hundred meters (this is a considerable effect especially for the nocturnal conditions).

Figure 2.5. Difference plots for 2 m air temperature (ºC) for Paris metropolitan area between outputs of the urbanized vs. control runs of Enviro-HIRLAM on 21.07.2009 at 06 UTC (Gonzalez-Aparicio et al., 2011)

On regional scale, clearly, the contribution of Megacities to air-quality is high and locally it can achieve order of tens percent in concentrations of pollutants like ozone and other products of photochemical processes in summer, or aerosols, mainly in winter due to the stability conditions in mixing layer. Based on one year simulation, this results in the Megacities potential to modify surface temperature eventually up to a few degrees (given theoretical complete emission reduction), actually only in local episodes, moreover mainly due to semi-indirect effects. In reality, this effect will be an order of magnitude weaker and at climate time scales probably even lower. Longer (at least one decade) simulations or ensembles runs can provide more comprehensive information. MEGAPOLI studies investigating the possible impacts of European megacities aerosol emissions on regional climate have been conducted for the 2001–2010 decade considering base year 2005 emissions. Results indicate that regional [typically 10–1000 km scale] megacity aerosol radiative forcings can be ten to hundred times larger than at global scale. The balance between aerosol radiative cooling and warming and the net sign of TOA forcing depends strongly on the nature of aerosol mixing varying from one megacity to another with the nature of emissions (e.g. sulfur vs black carbon) and local climatic conditions. Due to aerosol aging processes, aerosol absorption and radiative warming is more intense in the immediate vicinity of megacities compared to surrounding regions. As a result megacity aerosols could enhance locally GHG and urban heat island radiative warming, but they however generally tend to mitigate this warming on the regional to continental scale. Aerosol surface megacity radiative forcing is negative due to incoming shortwave radiation dimming which can be significant at regional scale (up to –[1-5] W/m2). Over the European domain, radiative forcing patterns show an important spatial and seasonal variability, with maximum regional forcings being observed in the summer as a result of reduced ventilation and aerosol removal by precipitation. Beyond radiative forcing the question of whether megacity aerosols are able to trigger significant change in regional climate patterns was also addressed. Using regional climate models, it is very challenging to quantitatively assess the effect of a European megacities aerosol radiative forcing which still represents perturbation of rather limited intensity at the regional to continental scale. Statistical treatments based on ensembles simulations are required methodologies to produce robust results against model internal variability. Qualitatively, a possible mesoscale climatic response to aerosol forcing not only depend on megacity emission intensity, but also on the climatic context. Regions (e.g. the Po Valley for Europe) where meso-scale gradients are important factors in shaping meteorology and climate (e.g. through convection, sea and slope breezes, etc) are more 11 of 52

D2.4.3

TRANSPHORM

Deliverable

likely to be sensitive to regional aerosol radiative perturbation compared to region where synoptic circulation dominates. Finally, sensitivity studies showed that modification of local and regional climate through megacity aerosol emission is potentially of less importance than the co-induced land-use perturbation associated to urbanization (compared to a rural state).

Figure 2.6. Global distribution of - (left) short-wave, SW all-sky and (right) long-wave, LW clear sky - topof-atmosphere (TOA) radiative forcing due to aerosols from megacities denoted in W/m2 (Folberth et al., 2011)

For studies of megacity effects on the global climate, using a simple analytical technique based on the climate sensitivity computed by complex models, it was estimated that Megacities contribute due to long-living GHGs emission a warming of over 0.2 K after 100 years, with nearly 90% of this being due to carbon dioxide emissions, and most of the rest due to methane. However, most of this effect would be present even if people did not leave in Megacities but elsewhere in the same country. MEGAPOLI has also examined the impact of pollutants such as NOx, VOC and aerosols that are emitted from Megacities on global climate under present-day conditions (base year 2005). Generally, the contribution of Megacities to global pollutant emissions is of the order of 2% to 6% of the total global annual anthropogenic emission flux (see also the answer to Q2 and Q4). There are four main direct radiative forcing impacts of Megacity emissions (Figure 2.6):  Ozone production: +5.7±0.02 mW/m2  Reduction of the methane lifetime due to OH radical production: -2.1±0.13 mW/m2  Short-wave direct forcing from aerosols: -6.1±0.21 mW/m2  Long-wave direct forcing from aerosols: +1.5±0.01 mW/m2 The combined effect of all of these individual terms is a rather small negative forcing, that is a cooling, of -1.0±0.32 mW/m2 under present-day conditions. As a result the effect of Megacities on global climate (especially when compared to the same population living elsewhere) is very small. In addition, see more details in MEGAPOLI deliverable reports D6.1, D6.2, D6.3, and D9.6. 2.4. Effects of growing Megacities on future climate at global and regional scales Two main types of the mechanisms of growing Megacities on the future climate at global, regional and local scales were considered. Changes to: - Urban features including the urban heat island, land-use, albedo, roughness, moisture, etc. - Emissions of atmospheric pollutants and their subsequent effects feedbacks on climate. The growth of Megacities will considerably affect future urban climate, including increasing urban heat islands, altering the formation and evolution of precipitation, increasing thunderstorm intensity and frequency, etc. For example global climate change (e.g. 2 degrees warming) will increase air 12 of 52

Deliverable

TRANSPHORM

D2.4.3

conditioners use in Megacities. Simulations showed that it can lead to increasing atmospheric temperature in the most populated districts of a Megacity by up to one additional degree. While the growth of (at least, some) Megacities might be expected, the reduction as well as increase of some emissions of anthropogenic origin might occur. Thus, the total effect of the development is not clear. However, in most pollutant concentrations the contribution of the Megacities looks to be consistent and similar across the next decades, especially in winter as seen in Figure 2.7 for ozone and primary organic aerosol. For the estimate of climate effect potential of these changes see Q5.

Figure 2.7: Megacities contribution (assessed by the difference in simulations with reduction of emission by 5% and rescaled linearly to 100%) in winter for ozone (upper panels, ppbv) and primary organic aerosol (bottom panels, µg/m3) under emission scenarios for 2005, 2020, 2030 and 2050 (from the left to the right) (Solmon et al., 2011).

On the global scale given the relatively small current signal of Megacity CO2 (around 0.1 W/m2, equivalent to about 0.2 K by the end of the century), its future impact will simply scale approximately linearly with the evolution of the future emissions of CO2 from megacities. The impact of pollutants such as NOx, VOC and aerosols on climate under future conditions (for the base year of 2050) has been analyzed in MEGAPOLI, showing that compared to present-day total forcing of these atmospheric pollutants of -1.0±0.32 mW/m2, the overall forcing reduces substantially, to less than half of the magnitude of present-day conditions (due to the competition of megacity growth with overall urbanization and meso-city growth), and it changes sign, becoming slightly positive at +0.4±0.11 mW/m2; thus, pollutant emissions from megacities are found to have a very small, slightly warming impact on future climate, adding to the somewhat larger warming impact from CO2. Over the European domain, regional climatic simulation including climate change effects (A1B) and emission scenario has also been conducted for the 2040-2051 decade. Result show that, in term of regional radiative forcing, the reduction of emission (result of cleaner technologies) impact is dominant despite climatic conditions more favourable to pollution accumulation (increased temperatures and anticyclonic stagnation conditions). As a result aerosol radiative forcing induced by European megacities slightly decrease in future conditions and according to scenarios produced by MEGAPOLI. Summarizing the possible effects of growing Megacities on climate change on different scales, we can answer that: • On the city- and meso-scales definitely both the urban heat island and megacity emissions are significantly modify the urban climate and this processes will be increasing for growing megacities, 13 of 52

D2.4.3

TRANSPHORM

Deliverable

•

On regional and continental scale: the effect is modest but it can extends up to one thousand kilometres, • On global scale: the answer probably ‘No’ due to the urban features and pollutant emissions, but there is a small effect due to greenhouse emissions. • One should note however that the regional and global effect would be similar or larger if the same population did not live in Megacities but in other areas in the same countries. In addition, see more details in MEGAPOLI deliverable reports D6.6 and D9.6. 2.5. Policy options to reduce transport emissions of air pollutants and GHGs in cities By Jochen Theloke and USTUTT team The policy options for mitigating emissions of air pollutants and greenhouse gases have been developed in the MEGAPOLI project for the first level MCs. The summarization of the abatement measures descriptions for the road transport, off-road transport, large and small combustion plants, and industrial processes is shown in the following tables. Table 2.1: Road transport abatement measures descriptions # RT 1 RT 2

RT 3

RT 4

RT 5

RT 6

RT 7

RT 8

RT 9

Measure Enhanced use of bicycles in cities Enhanced use of public traffic

Short description Shift from passenger cars to bicycles in cities. It is assumed that 10% of those car trips could be shifted to bicycling. Shift from passenger cars to public traffic. The aim is to make public transport more attractive. This includes to open up new routes and to improve the quality of the connections (low waiting and transfer times) and to improve the vehicle comfort Promotion of low emis- One strategy to improve the air quality in cities is to promote low sion vehicles (E-cars, emission vehicles (hybrid and electric cars) hybrid vehicles) Traffic management Traffic management, such as Green wave, is an system in which a (Green Wave, im- series of traffic lights are coordinated to allow continuous traffic provement of the opera- flow over several intersections on a main road. tion of signalized intersections) Low emission zone is a geographically defined area, which seeks to Low emissions zones restrict or prohibit access by specific polluting vehicles or only allow access for low or zero emission vehicles. To overcome the burden caused by traffic, the city of London introCity toll duced its congestion charging zone in 2003. Drivers entering the zone have to pay a fee. Costs for fuelling can be an important aspect to influence the driving Increase in fuel costs behaviour. An increase in fuel prices (e.g. by raising the fuel taxes) can be an instrument to decrease the amount of kilometres travelled in private transport. A national toll for passenger cars (as in the Netherlands) can support Passenger car toll the reduction of vehicle mileage. A differentiation in toll prices depending on the time of the day and the road type and section can additionally help to relieve densely trafficked roads. The EU Directive 2009/28/EC (Renewable Energy Directive) (sucUse of biofuels cessor to Directive 2003/30/EC) describes and regulates the use of biofuels in Europe. An incorporation rate of 10% to fossil fuels by 2020 is to be made. This measure will have a higher incorporation rate to fossil fuels.

14 of 52

Deliverable

TRANSPHORM

D2.4.3

The Reduction Potentials are given related to sectoral emissions, in this case related to European (EU 29) road transport emissions on base of the MEGAPOLI reference scenario (Theloke et al., 20101). For the road transport sector the measures which have impacts only in the Megacities and measures which affect the whole EU 29 domain has been distinguished. The measure with the highest abatement potential for 1st level Megacities is the measure: “Enhanced use of public traffic”. There you have a PM abatement potential of 0.04%, CO2 abatement potential of 0.2 related to sectoral emissions in the EU27+CH+NO (EU29) in 2030. This corresponds to a PM abatement potential of 0.0004% and a CO2 abatement potential of 0.05% related to the pollutant specific emissions in EU29 in 2030. These abatement potentials are very dependent of the composition of the bus fleet. Later it will be shown that effective measures are not certainly efficient on base of a cost-benefit analysis. The measure with the highest abatement potential for the whole European domain is: “Promotion of low emission vehicles (E-cars, hybrid vehicles)”. With this measure impacts a PM abatement potential of 6.5% and a CO2 abatement potential of 17.1% related to sectoral emissions in 2030, corresponding to a PM abatement potential of 0.4% and a CO2 abatement potential of 4.5% related to the pollutant specific emissions in EU29 in 2030. Table 2.2: Off-road transport short description # OT 1

OT 2

OT 3

OT 4

Measure Short description Differentiation of track access Distinction of track access charges for rail transport on the emission standards leads to an accelerated introduction of new charges for rail transport technologies and reduces the specific NOx emissions from diesel-powered rail transport. The long life of locomotives (in average 32 years) affects new technologies in a fleet with a high delay. Furthermore, it can lead to an increase in the share of electric traction. Further development of emis- The emission limit II leads to significant reductions in NOx, sion limits in inland waterway CO, HC and PM emissions for inland water vessels. More stringent emission limits (e.g. emission limit III) can help to transport further reduce tailpipe emissions in the whole fleet. Likewise to other transport modes, damages to health and enKerosene tax for aviation vironment caused by aviation are so far not internalized. A kerosene tax could help to internalize some of those damages. Low emission zones for con- In analogy to the low emission zones for on-road vehicles, this policy targets high emitting diesel fuelled construction mastruction equipment chinery. Those would be banned from the city if they are not retrofitted with a particulate filter or even replaced by machines with a higher emission standard. The emission standard IIIb, valid from 2012 on, will lead to a 90% reduction compared to level II. The abatement potential is thus higher the sooner this measure is implemented.

The measure with the highest abatement potential for off-road transport is the measure: “Kerosene tax for aviation”. This measure has a CO2 abatement potential of 18.3% related to sectoral emissions in 2030. The CO2 abatement potential related to the pollutant specific emissions in EU29 in 2030is 1.9%.

1

Theloke J. and M.Blesl, D. Bruchof, T.Kampffmeyer, U. Kugler, M. Uzbasich, K. Schenk, H. Denier van der Gon, S. Finardi, P. Radice, R. S. Sokhi, K. Ravindra, S. Beevers, S. Grimmond, I. Coll, R. Friedrich, D. van den Hout (2010): European and megacity baseline scenarios for 2020, 2030 and 2050. Deliverable D1.3, MEGAPOLI Scientific Report 10-23, MEGAPOLI-26-REP-2010-12, 57p, ISBN: 978-87-92731-04-3 http://megapoli.dmi.dk/publ/MEGAPOLI_sr10-23.pdf

15 of 52

D2.4.3

TRANSPHORM

Deliverable

Table 2.3: Abatement measures for large combustion plants (LCP) #

Measure

Short description

LCP 001 Modernisation of the existing coal-fired Large combustion plants The large number of coal-fired power plants in Europe were built in the 60 years and don’t represent the state of the art. The average efficiencies of the current power plant parks system are about 36%. In the modern hard coal plants can archive the efficiency rates of about 46%, the lignite-fired installations of about 43%. The further development of combustion technologies will allow the efficiency rates higher than 50%. In the IGCC (integrated gasification combined cycle) plants without CCS (Carbon capture and storage) is possible to archive the efficiency rates of about 55%. By application of CCS technology the plant efficiency decrease up to 10%. LCP 002 Expansion of electricity generation from renewable resources in LCP In order to prevent the negative impacts of climate change, the EU has set several targets aiming at the reduction of greenhouse gas (GHG) emissions, an intensified use of renewable energy and an increase of energy efficiency. The underlying goal is to keep global warming below 2°C. In order to achieve the 2°C targets renewable energies for electricity generation play an extraordinary role for achieving the climate target. Compared to the Reference scenario the electricity production from renewable energies increases to a maximum of about 1800 TWh in 2050, reaching a share of 45 % of gross electricity consumption in 2050.Wind energy ramps up to 400 TWh in 2020 and further on to 900 TWh in 2050 with an offshore share of about 50 % in 2050. But also electricity from solar photovoltaic and solar thermal power plants contribute with 8 % (260 TWh) in 2050 to total electricity generation.

The Reduction Potentials are given related to sectoral emissions, in this case related to European (EU 29) emissions from the Public Electricity and Heat Production. By implementation of Measure LCP 001 mainly NOx, SO2 and particle/PM emissions will be reduced. The measure LCP 002 reduces, in particular, the CO2 emissions. Table 2.4: Abatement measures for small combustion plants (SCP) # SCP 001

Name Short description Replacement of solid fuels fired small combustion plants with efficient combustion techniques Generally, the solid fuels (coal, biomass) fired small combustion plants cause higher specific emissions of incomplete combustion (e.g. particulate matter (PM), CO, VOC) compared to the gas or oil combustion plants. The high emissions are result of inefficient combustion techniques especially of older installations. The significantly emission reduction of PM, CO and VOC can be archived by replacing of old inefficient installation with new biomass fired small combustion plants (e.g. wood pellet boilers, automatic wood boilers) or with gas installations.

SCP 002

Replacement of old gas/oil boilers with modern condensing boilers The modern condensing boilers have the thermal efficiency up to 98%; compared with conventional boilers 30% higher energy yield can be archived

SCP 003

Energy-efficient modernisation of old buildings “Insulation is the most effective way to improve the energy efficiency of the buildings. Insulation of the building envelope helps keep heat in during the winter, but let heat out during summer to improve comfort and save energy. Insulating a home can save 45–55% of heating and cooling energy”. By the energy-efficient modernisation of old buildings about 100 kWh/m²/a can be saved

SCP 004

Switch to renewable heat supply in residential sector

16 of 52

Deliverable

TRANSPHORM

D2.4.3

#

Name

Short description In order to prevent the negative impacts of climate change, the EU has set several targets aiming at the reduction of greenhouse gas (GHG) emissions, an intensified use of renewable energy and an increase of energy efficiency. The underlying goal is to keep global warming below 2°C.

SCP 005

Expansion of district heating network The district heat has its largest distribution in metropolitan areas and contributes significantly to reducing emissions. By use of district heat about 50% of primary energy saving is possible. The current share of the district heat supply in Germany is about 12%, in France and Norway by about 5%. Poland, Iceland and Lettland have a highest share on district heat supply (more than 50%). This would mean a substantial expansion potential in Europe .

The measures SCP 001 - SCP 004 have impact to the whole European domain. The measure SCP 005 is relevant only for Megacities. The measure ”Expansion of district heating network”(SCP 005) reduces in particular PM10, PM2.5, CO, NMVOC, which are mostly caused by incomplete combustion. An overview on megacity policies and measures for industrial processes is given in Kampffmeyer et al. (2011)2. Table 2.5: Abatement measures for Industrial Processes #

Measure

IND 001

Combined climate protection measures in cement industry The overall output of thermal CO2 can be reduced (“CO2 neutrality”) with the substitution of fossil fuels by (renewable) alternative fuels. In a cement plant with an annual production of 1 million tons of clinker, a thermal substitution rate of 40% reduces the net CO2 generation by about 100,000 t. Regarding the Power generation, cement kilns can contribute with appropriate waste heat utilization to the supply of thermal/electrical power thus saving natural resources and reducing CO2 emissions. The use of waste instead of fossil fuel may reduce CO2 emissions by 0.1 to 0.5 kg/kg cement (varying from 20 to 40%). On average blended cements may reduce carbon emissions from 0.81 kg to 0.64 kg per kg cement (20%). An end-of-pipe technology to reduce carbon emissions may be CO2 removal. Probably the main technique is combustion under oxygen while recycling CO2. However, considerably research is required to all unknown aspects of this technique. Summing up, the use of raw materials, alternative fuels and mineral additions are relevant for the reduction of greenhouse gases. Reducing CO2 emissions it is possible to reduce in the same time emissions like NOx, SOx, NMVOC and PM.

IND 002

Iron making – Blast furnace - Injection of pulverized coal (PCI) The steel industry is the second largest energy consumer of all producing sectors. Iron making is the most energy-intensive step in integrated steel making. One of the main energy efficiency measures is the injection of fuels into the blast furnace, especially the injection of pulverized coal. Pulverized coal injection replaces the use of coke, reducing in the same time coke production and hence saving energy consumed in coke making (above) and reducing emissions of coke ovens and associated maintenance costs.

2

Short description

Kampffmeyer T. and U. Kugler, M. Uzbasich, J. Theloke, R. Friedrich, D. van den Hout (2011): Short, Medium and Long Term Abatement and Mitigation Strategies for Megacities. Deliverable D8.1, MEGAPOLI Scientific Report 11-06, MEGAPOLI-32-REP-2011-05, 40p, ISBN: 978-87-92731-10-4,

17 of 52

D2.4.3

TRANSPHORM

Deliverable

#

Measure

Short description

IND 003

Sinter plant – Heat recovery A possibility to improve the efficiency of sinter making is the heat recovery at the sinter plant. The recovered heat can be used to preheat the combustion air for the burners and to generate high pressure steam. This can be used in steam turbines to generate electricity. Existing plants can be retrofit and there are various systems t for new sinter plants (e.g. Lurgi EOS process).Fuel savings were estimated, based on a retrofitted facility in The Netherlands, to be 0.47 MMBtu/t (0.55 GJ/t) of sinter. The increased electricity generation was estimated to be 1.4 kilowatt hour per ton (kWh/ton) (0.0056 GJ/t) of sinter. The payback time was estimated as 2.8 years. Emissions NOx, SOx and PM are expected to be reduced using this system.

IND 004

Coking plant – Coke dry quenching Dry quenching of the coke, in place of wet quenching, can be used to recover heat that would otherwise be lost from the coke while reducing dust. Dry coke quenching is typically implemented as an environmental control technology. Various systems are used in Germany, Brazil, Finland, Japan, and Taiwan. All essentially recover the heat in a vessel where the coke is quenched with an inert gas (nitrogen).The steam recovery rate with this equipment is about 0.5 MMBtu/t (0.55 GJ/t) coke. In addition, Nippon Steel's performance record shows that the use of coke manufactured by dry quenching reduces the amount of coke consumption in the blast furnace by 0.24 MMBtu/ton (0.28 GJ/t) molten iron. NOx, SOx and particulate emissions are also reduced with this system.

The measure IND 001 “Combined climate protection measures in cement industry” has the largest mitigation potentials for GHG as well as for AP. The assessment of the health impacts (benefits) of mitigation measures follows the ‘full chain’ or ‘impact pathway’ approach (www.integrated-assessment.eu; www.externe.info) shown in Figure 2.8.

Figure 2.8: The Full chain approach of integrated assessment modelling developed by the USTUTT team.

18 of 52

Deliverable

TRANSPHORM

D2.4.3

At first, a reference scenario is constructed that describes activities and emissions for the considered future year (2030 or 2050) without additional mitigation measures. Then, the simulated implementation of the identified measures leads to measure based emission scenarios with reduced emissions. In a next step, for all scenarios (reference and measure based) the concentration of several pollutants (PM10, PM2.5, O3, NO2, etc.) is calculated throughout Europe using an atmospheric model. The concentration increase in cities, and especially in the 1st level Megacities (London, Paris, RhineRuhr area and Po Valley) is taken into account by applying a new method to estimate the urban increment to be added to the regional concentrations (Torras Ortiz, 2010). Measures, which have only impact to the 1st level MCs, have been distinguished from measures having impacts to the overall European domain. Concentration–response–functions developed within the EC FP6 HEIMTSA3 project by carrying out a meta-analysis of various epidemiological studies are then applied to convert the concentration together with the population density into health impacts (e.g. years of life lost, cases of chronic bronchitis, respiratory hospital admissions and many others). The health endpoints are then aggregated to DALYs (disability adjusted life years). In order to be able to compare health endpoints with costs within a cost-benefit analysis, health impacts have to be converted into monetary values. It has been done by determining stated or revealed preferences of the affected population. One of the methods to estimate these monetary values is to ask for the willingness to pay to avoid a certain health risk. Usually the numerous existing contingent valuation studies are analysed to derive monetary values per health endpoint. The measures with the most avoided DALYs for the whole European domain for 2030 are the following:  Replacement of solid fuels fired small combustion plants with efficient combustion techniques (SCP1). Significant reductions of pollutants from incomplete combustion (PM, CO and VOC) processes can be achieved by replacing of old manually operated biomass boilers by modern automatically operated utilities, for example pellet or wood chip boilers and also by replacement with gas installations.  Combined climate protection measures in cement industry (IND1).  Energy-efficient modernization of old buildings (SCP3). For 2050 the energy efficient modernization of old building becomes more effective than the cement industry measures. Table 2.6. Ranking of Measures for the European Megacities by DALYs. Measure Year Replacement of solid fuels fired small combustion plants with efficient combustion techniques Combined climate protection measures in cement industry Expansion of district heating networks Kerosene tax for aviation

Rhine-Ruhr 2030 2050 1 1

2

3

Megacities Po Valley London 2030 2050 2030 2050 2 2 1 1

1

1

2

3

Paris 2030 2050 1 1

2

3 2

3

2

3

3

3

2

3

The ranking of measures for the selected European Megacities for 2030 and 2050 is given in Table 2.6. While the rankings change from area to area and depend on the period considered, the following measures lead to the most avoided DALYs in the 1st level Megacities: 3

http://www.heimtsa.eu/

19 of 52

D2.4.3

TRANSPHORM

Deliverable

 Replacement of solid fuels fired small combustion plants with efficient combustion techniques;  Combined climate protection measures in cement industry;  Expansion of district heating networks;  Kerosene tax for aviation. An example of results of the cost-benefit analyses (taking into account monetized health impacts, crop damages, material damages, climate damages, costs for implementing the measures and monetized utility losses) is shown in Figure 2.9. The differences calculated on a base of the following formula: (Health Impacts avoidance costs + Crop damage avoidance costs + material damage avoidance costs + CO2 avoidance costs) [Euro2010]) ((Measure costs +Utility Losses) [Euro2010])

have been ranked for identifying the most efficient measures for the selected 1st level Megacities. Four most efficient measures for Paris such as (i) district heating networks, (ii) kerosene tax for aviation; (iii) traffic management, and (iv) promotion of low emission vehicles (E-cars, hybrid vehicles) are presented in Figure 2.9:

Figure 2.9: Result of the cost-benefit analysis showing the most efficient measures for Paris (France) for 2030 and 2050.

Measures which only have impacts in the considered Megacities themselves (e.g. enhancement of bicycle use in cities), were found only have negligible cost efficiency, because the health benefits are very small. An energy efficient modernisation of old buildings (insulation) (SCP003) is also very cost efficient, but if the indoor exposure are also taken into account the effect of this measure turn over to additional health impacts due to accumulation of indoor air sources because air-tighter building envelopes the DALYs caused by indoor and outdoor air sources increase drastically. This example, despite all the involved uncertainties, shows that it is important to ensure a sufficient air exchange rate for buildings, especially when they are renovated. Figure 2.10 shows the effects, if taken into account the indoor exposure related health impacts.

20 of 52

Deliverable

TRANSPHORM

D2.4.3

Figure 2.10: Impact of the SCP003 mitigation measure (Energy efficient modernization of old buildings) in the residential sector: avoided DALYs.

So, MEGAPOLI applied a complex scenario approach with an energy model and emission inventory to perform an integrated assessment for European megacities. The multi-pollutant, multi-source and multi-scale impacts of measures were taken into account, as well as the relationship between climate change and air pollution. A major finding is that several European countries will not achieve the future threshold requirements of the Air Quality Directive (2008/50/EC) or the National Emission Ceilings (NEC) defined in the NEC directive (2001/81/EC), even with additional measures, especially for particulate matter (PM10 and PM2.5); the climate policy scenario in combination with the maximum feasible reduction scenario for 2020 comes closest to these requirements. The most efficient measures for improving air quality and climate change impacts in European Megacities by 2030 and 2050 are: - Switching to renewable heat supply in residential sector [2030/2050] - Implementing a European-wide passenger car toll [2030/2050] - Expanding electricity generation from renewables in large combustion plants [2030/2050] - Replacing solid fuel fired small combustion plants with efficient combustion techniques [2030/2050] - Replacing old gas/oil boilers with modern condensing boilers in small combustions [2030] - Implementing combined climate protection measures in the cement industry [2030] - Promoting low emission vehicles (E-cars, hybrid vehicles) [2050] Finally, there are many complexities involved in the interpretation, for instance the difference between developed and developing countries, and the accumulation of indoor pollutants, which strongly influences the impact on human health; as an example, the measure "energy-efficient modernisation of old buildings" has been identified also as very cost efficient, but it is not recommended due the non-negligible additional health impacts from accumulated indoor pollutants in insulated buildings.

21 of 52

D2.4.3

TRANSPHORM

Deliverable

2.6. Further research needs * Emissions: - There is a need for "shadow" scientific emission inventories. These should not compete with official reported data, with arguments about which is right, but rather should be the test-bed for new insights, delivered to modellers in a consistent manner across Europe to test their validity. If proven more reliable, certain aspects can be communicated to countries and/or find their way into the emission inventory guidebook. In the current set-up, new insights find their way into official reporting only very slowly, causing them to lag behind what modellers are using and what source-receptor insights we have. The simple example is spatial resolution. Since about 2008 we have been delivering EIs at ~ 7 x 7 km. It is widely recognized that 50 x 50 km is no longer acceptable, yet it is expected that official reporting at 10 x 10 km may take another 2-4 years, and still be rather incomplete. For a "new" or revisited source this process will be equally time consuming. By designing and orchestrating a good marriage between scientific and official reported EIs, much can be gained and both will profit. - There is also a need for periodic intercomparison and integration of emission inventories produced at different levels: Pan European inventories, member states national top-down inventories, subnational regions and local bottom-up inventories. Relevant differences have been highlighted and the integration work performed in MEGAPOLI should be updated periodically on a regular basis. * Evaluation of the population exposure to air pollution and the adverse health effects in megacities, and cost-effective ways to reduce the adverse health effects: Issue: The previous projects have focused on the air quality and climatic effects of megacities, but haven't taken the next step to assessing the health effects. The major problem that has not yet been directly studied in the EU programmes is the mortality and morbidity of populations in megacities caused by poor air quality. Method: this should be studied by constructing chains of models from emissions to air quality to exposure to health effects. Currently, exposure modules are included only in a small fraction of the integrated modelling systems, and simplified health effects modules in very few of them. In addition to modelling, what is also needed is complementary focused measurements campaigns. Influence on EU policy: The application of such integrated modelling chains would allow for the study of cost-effective measures, policies and abatement strategies in order to ensure a sustainable development of megacities, including the health effects on the populations. * Additionally to the existing linearised and simplified integrated assessment tools a new generation of integrated meteorology/climate and atmospheric chemical transport models with twoway feedbacks are needed to analyse non-linear interactions of climate change and air quality. * Better coordination of climate change and (urban) air quality policies is needed. Stricter regulations of biomass burning for domestic heating and leisure uses within and around urbanized and critical areas are desirable. MEGAPOLI showed a great complexity of the problem for megacities, which is reflected in several ways: - multi-component (numerous gas and aerosol phase interacting constituents) - multi-aspect (air quality, climate, health, agriculture, tourism, economics) - multi-disciplinary (natural and social scientists and humanities working together on "integrated research/assessments"). Important topic for further studies: Integrated assessment and reduction strategies of both the air quality and climatic effects which could be achieved by various climatic and air quality measures, regulations and policies.

22 of 52

Deliverable

3.

Climate-Chemistry TRANSPHORM

TRANSPHORM

Interaction

Studies

D2.4.3

in

CityZen

Relevant

to

By M. Gauss (Met.no) and CityZen team Several studies regarding climate-chemistry interactions have been performed within the FP7 EC CityZen project. Those of relevance to the TRANSPHORM project are described briefly below, with references to more detailed delivery reports and publications in the open literature. As part of the CityZen deliverable report D4.2.1, several studies on the impact of megacities on climate change by 2050 have been published in the report series “Schriften des Forschungszentrums Jülich”, Reihe Energie & Umwelt / Energy & Environment Band / Volume 116, of Forschungszentrum Julich (FZJ). In CityZen deliverable D2.4.1b (https://wiki.met.no/_media/cityzen/docs1/ dm/cityzen-d-2-4-1b.pdf), the effect of megacity emissions on climate is discussed, based on ECHAM5 calculations performed at Forschungszentrum Julich. 3.1. CTMs driven by GCM meteorology The Norwegian Meteorological Institute (met.no) has used the EMEP/MSC-W model, driven by climate data for present and future time slices, to calculate the response on ozone and particulate matter (PM) to changes in climate and emissions. The results are summarized in Chapter 2 of the CityZen deliverable report D2.3.1 (https://wiki.met.no/_media/cityzen/docs1/dm/cityzen-d-2-31.pdf). An example of modelling results for ozone is shown in Figure 3.1. Ozone tends to increase during the first half of this century in the Mediterranean area, when only climate change is taken into account. However, an overall decrease is modeled when projected emission reductions are implemented in addition. Exceptions are seen within some confined, heavily polluted areas, e.g. Southern England. Similar calculations have been made for CityZen for the year 2030 time frame, see CityZen deliverable report D3.5.1 (https://wiki.met.no/_media/cityzen/docs1/dm/cityzen-d-3-51.pdf). The results are similar to the 2050 results, although the signal from climate change is weaker. The same deliverable reports also present studies that were performed with the University of Oslo / Forschungszentrum Julich model setup, running the global Oslo-CTM2 model with climate data from a global GCM, the ECHAM5 model. The results are in qualitative agreement with the met.no results.

23 of 52

D2.4.3

TRANSPHORM

Deliverable

Figure 3.1: Upper left: Daily maximum near-surface ozone, averaged over the 11-year period 2000 to 2010, as modeled by the EMEP/MSC-W model. The three other plots show modeled changes in daily maximum surface ozone. Upper right: effect of climate change from 2000s to 2040s, lower left: effect of climate change and emission change - current legislation scenario, and lower right: effect of climate change and emission change – stringent climate policy. Unit: ppb. For details see CityZen deliverable report D2.3.1.

3.2. Traffic mitigation scenario Met.no also performed a more focused study for traffic mitigation, described in CityZen deliverable report D4.1.1 (https://wiki.met.no/_media/cityzen/docs1/dm/cityzen-d-4-1-1.pdf). IIASA provided a list of climate-friendly air quality measures, i.e. emission reduction policy that is designed to reduce air pollution while at the same time not contributing to climate change. A well-known example is black carbon (BC) which is an air pollutant but also contributes to climate change: Reductions of BC can thus, be considered as win-win measures, i.e. beneficial to both air quality and climate change policies. However, few sources emit BC only; they rather emit other species in addition (coemitted species). SO2 may be such a component. It has an overall cooling effect, due to its contribution to sulphate aerosols. Climate-friendly air quality measures aim to reduce air pollutants, and those species that contribute to climate warming. In the case of road traffic, the introduction of diesel particle filters (DPF) is among the list of promising measures. The EMEP/MSC-W model calcu24 of 52

Deliverable

TRANSPHORM

D2.4.3

lated the effect of DPF in the year 2030. As many countries in Western Europe will have introduced DPF by then even in the current legislation scenario, the effect of this measure is limited to areas outside Western Europe. Figure 3.2 shows surface concentrations and of BC and sulphate aerosols as modeled by the EMEP/MSC-W model, while Figure 3.3 shows changes due to the additional DPF measures in 2030. Reductions in BC are clearly seen in the hotspots, where there is still potential for further introduction of DPF. Sulphate aerosols are hardly changed, with slight increases in some hotspot areas. Air quality will thus be improved, without concurrent climate warming.

Figure 3.2: Monthly-mean concentrations of black carbon (top panels) and sulphate aerosols (bottom panels) at the surface for January (left) and July (right), as modeled by the EMEP/MSC-W model, averaged over the 2025-2035 period. Unit: μg/m3.

25 of 52

D2.4.3

TRANSPHORM

Deliverable

Figure 3.3: Change in monthly-mean concentrations of black carbon (top panels) and sulphate aerosols (bottom panels) at the surface for January (left) and July (right), due to further introduction of DPF, as modeled by the EMEP/MSC-W model for the period 2025-2035. Unit: μg/m3.

Climate-air quality interactions and co-benefits in emission reduction policy have been one of the focus areas within CityZen. The traffic mitigation study, performed for the CityZen project, demonstrates how climate-friendly air quality mitigation measures applied to traffic can improve air quality, while not contributing to climate warming. 3.3. Studies of the Eastern Mediterranean ozone levels and its precursors in summer The impact of temperature changes on summer time ozone and its precursors in the Eastern Mediterranean has been studied by Im et al. (2011). Changes in temperature due to variability in meteorology and climate change are expected to significantly impact atmospheric composition. The Mediterranean is a climate sensitive region and includes megacities like Istanbul and large urban agglomerations such as Athens. The effect of temperature changes on gaseous air pollutant levels and the atmospheric processes that are controlling them in the Eastern Mediterranean are here investigated. The WRF/CMAQ mesoscale modeling system is used, coupled with the MEGAN model for the processing of biogenic volatile organic compound emissions. A set of temperature perturbations (spanning from 1 to 5 K) is applied on a base case simulation corresponding to July 2004. The results indicate that the Eastern Mediterranean basin acts as a reservoir of pollutants and their precursor emissions from large urban agglomerations. During summer, chemistry is a major sink at these urban areas near the surface, and a minor contributor at downwind areas. On average, the atmospheric processes are more effective within the first 1000 m above ground. Temperature increases lead to increases in biogenic emissions by 9±3% K−1. Ozone mixing ratios increase almost linearly with the increases in ambient temperatures by 1±0.1 ppb O3 K−1 for all studied urban and receptor stations except for Istanbul, where a 0.4±0.1 ppb O3 K−1 increase is calculated, which is about half

26 of 52

Deliverable

TRANSPHORM

D2.4.3

of the domain-averaged increase of 0.9±0.1 ppb O3 K−1. The computed changes in atmospheric processes are also linearly related with temperature changes. A model study of the Eastern Mediterranean ozone levels during the hot summer of 2007 was done by Hodnebrog et al. (2011). The hot summer of 2007 in southeast Europe has been studied using two regional atmospheric chemistry models; WRF-Chem and EMEP MSC-W. The region was struck by three heat waves and a number of forest fire episodes, greatly affecting air pollution levels. We have focused on ozone and its precursors using state-of-the-art inventories for anthropogenic, biogenic and forest fire emissions. The models have been evaluated against measurement data, and processes leading to ozone formation have been quantified. Heat wave episodes are projected to occur more frequently in a future climate, and therefore this study is a contribution to climate change research. The plume from the Greek forest fires in August 2007 is clearly seen in satellite observations of CO and NO2 columns, showing extreme levels of CO in and downwind of the fires. Model simulations reflect the location and influence of the fires relatively well, but the modelled magnitude of CO in the plume core is too low, presumably due to underestimation of CO in the emission inventories. Moreover, higher maximum values are seen in WRF-Chem than in EMEP MSC-W, indicating differences in plume rise altitudes in the two models and more rapid dilution in the latter model. The model results are also in fairly good agreement with surface ozone measurements. Biogenic VOC emissions reacting with anthropogenic NOX emissions are calculated to contribute significantly to the levels of ozone in the region, but the magnitude and geographical distribution depend strongly on the model and biogenic emission module used. During the July and August heat waves, ozone levels increased substantially due to a combination of forest fire emissions and the effect of high temperatures. We found that the largest temperature impact on ozone was through the temperature dependence of the biogenic emissions, closely followed by the effect of decreased dry deposition. The impact of high temperatures on the ozone chemistry was much lower. To conclude this study we suggest that forest fire emissions, and the temperature effect on biogenic emissions and dry deposition, will potentially lead to substantial ozone increases in a warmer climate. Other studies have been or will be published in the scientific literature. They are given in the list below. They include the response of Air Quality to climate change but also the response of climate to emissions from megacities.

References Im, U., Markakis, K., Poupkou, A., Melas, D., Unal, A., Gerasopoulos, E., Daskalakis, N., Kindap, T., and Kanakidou, M. (2011): The impact of temperature changes on summer time ozone and its precursors in the Eastern Mediterranean, Atmos. Chem. Phys., 11, 3847-3864, doi:10.5194/acp-11-3847-2011, 2011; http://www.atmos-chem-phys.net/11/3847/2011/acp-11-3847-2011.pdf Hodnebrog, Ø., S. Solberg, F. Stordal, T. M. Svendby, D. Simpson, M. Gauss, A. Hilboll, G. G. Pfister, S. Turquety, A. Richter, J. P. Burrows, (2011): A model study of the Eastern Mediterranean ozone levels during the hot summer of 2007, submitted to ACPD. Schultz, Martin (Ed.) (2011); CITYZEN Climate Impact Studies. Band/Volume 116, 45 Seiten; ISBN 978-389336-729-0, http://wwwzb.zb.kfa-juelich.de/publikationen/redirect.asp? id_schriften= 47998 &getInfo=http%3A%2F%2Fhdl%2Ehandle%2Enet%2F2128%2F4476&online=online

27 of 52

D2.4.3

TRANSPHORM

Deliverable

4. EMEP Study of Meteorology and Climate Effects on Air Pollution Contribution by Jan Eiof Jonson and Met.no team 4.1. Meteorological data from HIRHAM Meteorological input data to the EMEP model for this experiment have been provided to MSC-W through collaboration with the Meteorology Section at the Research and Development Department of the Norwegian Meteorological Institute. The data were generated using a dynamic downscaling approach. The regional climate change scenario was produced within the Nordic Climate and Energy Systems (CES) project and the European FP6 project ENSEMBLES (van der Werf et al., 2009). The joint ENSEMBLES/CES regional climate model (RCM) ensemble consists of more than 25 RCM simulations (see http://ensemblesrt3.dmi.dk/extended_table.php) covering most of Europe at 25 km horizontal resolution. All RCM scenarios are based on the SRES A1B emission scenario (Nakićenović and Swart, 2000). The particular ensemble member used to generate data for the Unified EMEP model runs was produced by the HIRHAM RCM (Haugen et al. 2006; 2008). This HIRHAM run was forced by data from the Hadley Centre global climate model HadCM3. The HadCM3 data was run at the Hadley Centre on 3.75 degrees (lat) x 2.5 degrees (lon) resolution with emissions from SRES A1B. The HIRHAM RCM was run on a rotated spherical projection with 0.22 x 0.22 degrees horizontal resolution for the period 1950-2050. All meteorological parameters required by the Unified EMEP model were extracted from this 101-year HIRHAM data set, interpolated into the vertical grid of the Unified EMEP model and converted into netCDF format. No horizontal interpolation was necessary, thanks to the high flexibility of the Unified EMEP model code. In this preliminary study we have run three 10 (11)-year periods, 2001-2010 (2000–2010), 20212030 and 2040–2050 to look at future air quality, taking into account climate change. Figure 4.1 shows differences in temperature and surface precipitation averaged over the two 10-year periods, for the summer (June/July/August) and the winter (December/January/February) seasons, as seen in the HIRHAM data. As seen in Figures 1 there are marked increases in temperatures and changes in precipitation for the time span between decade 2001 to 2010 and 2020 to 2030. Even for the 20 year projection there increases in temperature is clearly seen in nearly all parts of Europe, North Africa and the European Arctic. For the 40 year projection the differences are even more pronounced. At high latitudes the temperature increase is most pronounced in winter, while in mid latitudes the warming is most pronounced during summer. The projected precipitation is reduced mainly in coastal regions of the Mediterranean, the Bay of Biscay and some regions in the Alps. Increases in precipitation are seen mainly in the winter in Western parts of the European continent (as the Western parts of Norway, Ireland, Portugal). These scenario results have in broad terms the same characteristics as other regional projections, reviewed by IPCC (Christensen et al., 2007). The summer temperatures in the 2040s are projected to be much higher than in the 2000s around the Mediterranean, while in the north in particular the winters will be much warmer. For precipitation, the projection is that there will be much more winter rain in western Norway and the western part of the British Isles, and drier, or much drier, than now both winter and summer around the Mediterranean. Both surface temperature and precipitation are important parameters for the physical exchange processes of trace gases and particles between the atmosphere and the surface. 28 of 52

Deliverable

TRANSPHORM

(a) Δ W temp. 2001–2010 + 20 years

(c) Δ W temp. 2000–2010 + 40 years

29 of 52

(b) Δ S temp. 2001–2010 + 20 years

(d) Δ S temp. 2000–2010 + 40 years

D2.4.3

D2.4.3

TRANSPHORM

(e) Δ W prec. 2001– 010 + 20 years

Deliverable

(f) Δ S prec. 2001–2010 + 20 years

(g) Δ W prec. 2000–2010 + 40 years

(h) Δ S prec. 2000–2010 + 40 years

Figure 4.1: Differences (a-d) in temperature (in degrees K) and (e-h) in precipitation (in %) for winter (left column) and summer (right column) between the decade 2000–2010 and projections for 20 and 40 years forward in time.

4.2. Climate effects on air pollution The Unified EMEP model was run with different sets of meteorological data generated by climate models for future time slices. The objective was to address changes in Air Quality in the future based on different emission scenarios and taking into account climate change. Figure 4.2 shows the basic experimental setup as described in EMEP report 1/2010 (Benedictow et al., 2010). The Unified EMEP model has been run calculating the average for the decades 2001 – 2010, 2021 – 2030 and 2040 - 2050 with present (year 2000) and future emission projections for 2040 with current legislation (see Benedictow et al., 2010), and the climate projection data as described above.

30 of 52

Deliverable

TRANSPHORM

D2.4.3

Figure 4.2: Experimental setup for studying effects of climate change on air quality in the EMEP (CM: Climate model, CTM Chemical Trace3r Model).

Figure 4.3a shows the annual average PM2.5 concentrations for the base calculation with emissions and meteorology for the decade 2000-2010 (2000s). Figures 4.3bc shows the calculated difference in calculated PM2.5 concentrations induced by changes in climate between the decade 2000–2010 and 2021–2030 and 2040–2050, respectively. Figure 4.3d shows the calculated change in PM2.5 including also the projected emission reductions in the 2040–2050 calculations. For PM2.5 climate change induced changes in concentrations increase from up to 0.4μgm -3 in central Europe for the decade 2021-2030 to more than 1 μgm-3 in the decade 2040–2050. For the latter decade the largest changes are calculated in and around the Mediterranean ocean. For northern Europe changes are small or even negative. The calculated changes in PM2.5 are mainly driven by changes in precipitation. Figure 4.3d shows the difference in PM2.5 concentrations assuming future emission projections. PM2.5 concentrations are lower or unchanged everywhere compared to the 2000s, reflecting lower emissions in this scenario in the 2040s than what was the case for the 2000s.

(a) PM2.5 (μgm-3), year 2000 em. and met.

(b) Δ PM2.5 2001 – 2010 + 20 met. years 31 of 52

D2.4.3

TRANSPHORM

(c) Δ PM2.5, 2001–2010 + 40 met. years

Deliverable

(d) Δ PM2.5, 2001–2010 + 40 met. and em. years

Figure 4.3: (a) Surface concentrations of PM2.5, averaged over the 11 year period 2000 to 2011; and Difference in PM2.5 from climate change (b) between the decades 2001–2010 to 2021–2030, (c) between the decades 2000–2010 to 2040–2050, and (d) as in (c) and emission projections for 2040.

References Christensen, J.H., B. Hewitson, A. Busuioc, A. Chen, X. Gao, I. Held, R. Jones, R.K. Kolli, W.-T. Kwon, R. Laprise, V. Magaña Rueda, L. Mearns, C.G. Menéndez, J. Räisänen, A. Rinke, A. Sarr and P. Whetton, (2007): Regional Climate Projections. In: Climate Change 2007: The Physical Science Basis. Contribution of Working Group I to the Fourth Assessment Report of the Intergovernmental Panel on Climate Change [Solomon, S., D. Qin, M. Manning, Z. Chen, M. Marquis, K.B. Averyt, M. Tignor and H.L. Miller (eds.)]. Cambridge University Press, Cambridge, United Kingdom and New York, NY, USA. Haugen, J.E. and H.Haakenstad (2006): The development of HIRHAM version 2 with 50km and 25km resolution, RegClim General Technical Report No. 9, 159-173. Haugen, J.E. and T. Iversen (2008): Response in extremes of daily precipitation and wind from a downscaled multi-model ensemble of anthropogenic global climate change scenarios. Tellus A, 60(3):411–426. doi: 10.1111/j.1600-0870.2008.00315x. URL http://www.blackwell-synergy.com/doi/abs/10.1111/j. 16000870.2008.00315.x. Nakićenović, N. and R. Swart; Eds (2000): Special Report on Emissions Scenarios. A Special report of Working Group III of the Intergovernmental Panel on Climate Change. Cambridge University Press, Cambridge, United Kingdom and New York, NY, USA. van der Werf, G.R., Randerson, J.T., Giglio, L., Collatz, G.J., Kasibhatla, P.S. and Arellano Jr., A.F. (2006): Interannual variability in global biomass burning emissions from 1997 to 2004. Atmos. Chem. Phys., 6, 3423-3441.

32 of 52

Deliverable

TRANSPHORM

D2.4.3

5. ENSCLIMA Study on Impact of Climate Change on Surface Ozone over Europe Contribution by Nordic EnsCLIMA team (DMI, FMI, Met.no partners, SMHI, DMU): Langner et al. (2012) The impact of climate change on surface ozone over Europe was studied using four offline regional chemistry transport models (CTMs) and one online regional integrated climate-chemistry model (CCM) driven by the same global projection of future climate from the ECHAM5 model under the SRES A1B scenario. Anthropogenic emissions of ozone precursors from RCP4.5 for year 2000 were used for simulations of both present and future periods in order to isolate the impact of climate change and to assess the robustness of the results across the different models. 5.1. Models employed and methods applied •

A global climate projection from ECHAM5 (Roeckner et al., 2005) has been the basis for all the simulations performed in this study. The meteorological data from the GCM were used to drive an offline hemispheric chemical transport model, DEHM, and as boundaries for a regional climate model, RCA3 (Kjellström et al., 2011). The output from DEHM and RCA3 were used to drive three regional offline CTMs for the European domain, EMEP, SILAM and MATCH. One online climate-chemistry model, EnvClimA, which simulated its own regional climate using the GCM climate and hemispheric CTM output on its boundaries was also included in the study. Emissions of O3 precursors from the same data base, RCP 4.5, are fed both into the hemispheric CTM and the regional CTMs. Two 10-year time periods were studied, a reference period, 2000-2009, and a future period, 2040-2049. Further details and model references are given in Langner et al. (2012).

• • • • • •

Model characteristics

Model

Type

Horizontal grid

Lowest model Model layer (m)

DEHM

Eulerian/Offline

150km

x

150km 60

#levels

Meteorological input data

16

20

ECHAM5-r3 (global data)

20

18

ECHAM5-r3 – RegCM4

16

20

ECHAM5-r3 - RCA3

10

9

ECHAM5-r3 - RCA3

5.5

15

ECHAM5-r3 - RCA3

top (km)

polar stereographic EnvClimA

Eulerian/online

50 km x 50 km 50 Lambert

EMEP

Eulerian/Offline

0.44°x0.44°

rotated 90

latitude longitude SILAM

Eulerian/Offline

0.44°x0.44°

rotated 50

latitude longitude MATCH

Eulerian/offline

0.44°x0.44°

rotated 60

latitude longitude

Table 5.1: Characteristics of the models used for ENSCLIMA runs (Langner et al., 2012). Observed and simulated monthly mean ozone 1

33 of 52

latitude longitude

D2.4.3 TRANSPHORM Deliverable Observed and simulated monthly mean ozone 1

Figure 5.1: Observed vs. simulated monthly mean ozone by a set of models used for ENSCLIMA runs for (a) NW, (b) NE, (c) SW, and (d) SE sectors (Langner et al., 2012).

Figure 5.2: EnvClimA: Monthly average of ozone concentration at the lowest model change between (2000-2009) and (2040-2049) (Zakey et al., 2012). 34 of 52

Deliverable

TRANSPHORM

D2.4.3

Figure 5.3: Ensemble simulated April–September change 2000–2009 to 2040–2049 in average O3 concentration at the first model level [in ppb(v)] (Langner et al., 2012).

Figure 5.4: Ensemble simulated April–September change 2000–2009 to 2040–2049 in average daily maximum O3 concentration at the first model level [in ppb(v)] (Langner et al., 2012). 35 of 52

D2.4.3

TRANSPHORM

Deliverable

5.2. DMI modelling system setup The Enviro-HIRLAM Integrated Online-Coupled Multi-Scale Meteorological-Chemical Transport Modelling System involves several models. The Enviro-HIRLAM model is an on-line coupled NWP (HIRLAM) and ACT model for research and short-term forecasting of meteorological as well as chemical weather (Baklanov et al., 2008; Korsholm, 2009; Korsholm et al., 2009). The system is developed by DMI and is included by the European HIRLAM consortium as the baseline system in the HIRLAM Chemical Branch (https://hirlam.org/trac/wiki). For long-term multi-year runs and climate studies a more economical version EnvClimA is used. The climate component of the EnvClimA version is the Regional Climate Model (version RegCM4), and the Environment component is the same as in the Enviro-HIRLAM model. The RegCM model, developed at the Abdus Salam International Centre for Theoretical Physics (ICTP), is a hydrostatic, sigma coordinate model (Pal et al., 2007). The chemistry component in EnvClimA includes the condensed gas-phase chemistry which is based on CBM-Z (Zaveri and Peters, 1999) and uses lumped species that represent broad categories of organics based on carbon bond structure. CBM-Z also defines a general organic category (PAR for paraffin) to represent miscellaneous carbon content to conserve carbon mass. The computationally rapid radical balance method (RBM) of (Sillman et al., 1991) and (Barth et al., 2002) is coupled as a chemical solver to the gas-phase mechanism to provides a solution to the tendency equation for photochemical production and loss. Photolysis rates are determined as a function of various meteorological and chemical inputs and interpolated from an array of pre-determined values based on the Tropospheric Ultraviolet-Visible Model (Madronich and Flocke, 1999) with cloud cover corrections by (Chang et al., 1987). Cloud optical depths and cloud altitudes from EnvClimA are used in the photolysis calculations, thereby directly coupling the photolysis rates and chemical reactions to meteorological conditions at each model time step. The modelling system is also coupled with aerosol modules and includes direct and indirect aerosol effects. In this study only gas-phases species and their direct effects on meteorological variables were considered. For multi-year simulations the EnvClimA version was used, short-term episodes and feedback mechanisms were analysed using the original Enviro-HIRLAM model. 5.3. EnvClimA and ensemble models results The signature of climate change on surface ozone over Europe was studied using the online integrated climate-chemistry model for Environmental applications (EnvClimA) (Zakey et al., 2012). The model domain has a horizontal resolution of 50 × 50 km and 18 levels in the vertical. In this study we preformed a 20 years simulation over Europe a reference period, 2000-2009, and a future period, 2040-2049. For present and future simulation, the initial and lateral boundary conditions for the meteorological fields are provided by global ECHAM5-r3 every six hours. The Chemical boundary conditions are provided by the Danish Eulerian Hemispheric Model (DEHM) every six-hours over Europe. The anthropogenic emissions of nitrogen oxides (NOx), sulfur dioxide (SO2), ammonia (NH3), nonmethane hydrocarbons (NMVOC) and carbon monoxide (CO) were taken from the IPCC-RCP4.5 scenario. In the current simulation of EnvClimA the biogenic isoprene emissions were not considered, because the MEGAN module, which is on-line coupled with the land surface model in EnvClimA, was found to overestimate the total emitted biogenic isoprene. Half of the emitted isoprene emission from MEGAN would give reasonable results for O3 concentrations.

36 of 52

Deliverable

TRANSPHORM

D2.4.3

The observations from EMEP stations are averaged over the period 1997–2003 while the model data are taken from simulations forced by climate model data covering the reference period, 2000– 2009. We have evaluated the models’ capabilities to reproduce diurnal average and average of daily maximum O3 concentration over the full year and during summer time (April–September). All comparisons are made with results from the lowest model level in each model. Generally this gives somewhat higher O3 concentrations compared to concentrations extrapolated to e.g. three meter level. Note also that the results for EnvClimA are based on six-hourly data while the results from all other models are based on hourly averages. In winter EnvClimA has a substantial negative bias for both mean and daily maximum O3 , this may be due to the underestimation in the winter temperature (not shown) over north-east Europe and due to feedback of ozone on the meteorological variables, which is included in EnvClimA. For the summer period the bias is reduced and is similar to the DEHM model which also underestimates the daily maximum concentration considerably. The negative bias in EnvClimA is also partly related to the use of six-hourly O3 output data and the omission of biogenic isoprene emissions. SILAM and EMEP overestimate the diurnal average concentration but this overestimation would be reduced if concentrations had been extrapolated to three meter level. The climate simulation (2000-2009) & (2040-2049) indicated zonal behavior of average daily maximum concentrations for the surface ozone (O3 ). In winter, model has a substantial negative bias for both mean and daily maximum O3 . This may be due to an underestimation of the winter air temperature over north-eastern Europe. Although the model spatial correlation is rather poor for diurnal average concentration, but for the average of daily maximum O3 concentrations the model showed correlation coefficients higher than 0.8 during summer. The model always showed the highest spatial correlation over central and southern Europe. The general pattern indicated an increase of surface ozone changes in southern Europe and a decrease in northern Europe for a chosen climate scenario. The sensitivity of the simulated surface O3 to changes in climate differ among different models in the EnsCLIMA ensemble (Langner et al., 2012), but the general pattern of change with an increase in southern Europe and decrease in northern Europe is similar across different models for the chosen climate projection, in particular for the subset of models using meteorological data downscaled using the same regional climate model. Emissions of isoprene differ substantially between different CTMs ranging from 1.8 to 8.0 Gt/year for the current climate. Also the simulated change in isoprene emissions varies substantially across models. Differences in horizontal model resolution and corresponding horizontal resolution in temperature fields are an important factor contributing to these differences. Ensemble mean changes between the periods 2000-2009 and 2040-2049 in summer (AprilSeptember) mean O3 and mean of daily maximum O3 exceed 1 ppb(v) in parts of the land area in southern Europe assuming no changes in anthropogenic air pollution emissions. In northern Europe ensemble mean changes both of these measures are mostly negative. In general, changes in surface O3 due to climate change presented here are much smaller than what can be expected from anthropogenic emission reductions over the same time period from previous studies. Future model evaluation tasks will include an assessment of the radiative forcing produced by ozone in these simulations, the impacts of online chemistry on the simulation of atmospheric aerosols in the EnvClimA aerosol tracer model, and the effects of using non-climatological chemical

37 of 52

D2.4.3

TRANSPHORM

Deliverable

boundary conditions for decadal scale simulations. Future model development tasks include the addition of a full thermodynamic aerosol model, new schemes for indirect aerosol feedbacks, etc. References Kjellström, E., Nikulin, G., Hansson, U., Strandberg, G. and Ullerstig, A. (2011): 21st century changes in the European climate: uncertainties derived from an ensemble of regional climate model simulations. Tellus 63A, 24-40, doi:10.1111/j.1600-0870.2010.00475.x, 2011 Langner, J., Engardt, M., Baklanov, A., Christensen, J. H., Gauss, M., Geels, C., Hedegaard, G. B., Nuterman, R., Simpson, D. Soares, J., Sofiev, M., Wind, P. and Zakey, A. (2012) A multi-model study of impacts of climate change on surface ozone in Europe. Atmos. Chem. Phys. Discuss., 12, 4901-4939. doi:10.5194/acpd-12-4901-2012 Roeckner, E., Brokopf, R. Esch, M. Giorgietta, M. Hagemann, S. and etc. (2006): Sensitivity of simulated climate to horizontal and vertical resolutions in the ECHAM5 atmosphere model, J. Clim., 19, 37713791, 2006. Zakey, A.S., A. Baklanov, F. Solmon, F. Giorgi, R. Nuterman, B. H. Sass, U. S. Korsholm, K. P. Nielsen, J. H. Sørensen, and A. Mahura (2012): ‘The signature of climate change on surface ozone: Using the Online integrated climate-chemistry model (EnvClimA)’. Abstract and presentation on EGU-2012, Vienna, Austria, April 2012; EGU2012-10887.

6. PM10 Future Projections for a South-Western European Medium Size City Contribution by DMI and TECNALIA team (by Iratxe Gonzalez-Aparicio et al, 2012.) 6.1. Urban climate system for generation of air pollution scenarios

Air quality is strongly dependent on weather, and it is therefore sensitive to climate change. Both concerns are associated with the urbanization and industrialization. Since the global population is becoming increasingly urbanised, interest has also emerged into the integrated impacts on the air pollution due to the climate change and urban development. Here, “Urban-Climate System” is defined as a system which includes the main derived currents affecting the present and hence, the future local air pollution (Figure 6.1): (1) regional and local climate under a framework of Climate Change and (2) urban features: the spatial (urbanization, land-use change, life style, etc.) and emission (energy consumption, agglomerations, traffic mobility, etc.) impacts. Urban development can have significant adverse effects on the budget of the pollution (Koracin et al., 2009; Southerland, 2004; Kepner et al., 2004) and on its interaction with the local atmospheric patterns. Developed urban scenarios vary in complexity and often rely on historic information to provide probabilities of land cover change (Koracin et al., 2009; Solecki and Oliveri, 2004). Recent studies have also estimated the effect of urbanisation on climate and the future climate change impact on the urbanisation. A detailed reviewed can be found in Wilby (2008). The Intergovernmental Panel on Climate Change (IPCC) has defined the potential conditions of future land use change (Solecki and Oliveri, 2004) under the regional greenhouse gas (GHG) emission scenarios (Special Report on Emissions Scenarios). In particular, the A1 scenario family describes a future world of very rapid economic growth, global population that peaks in middle of the 21st century and declines thereafter, and the rapid introduction of new and more efficient technologies. It shares the lowest trajectory, increasing to 8.7 billion by 2050 and declining toward 7 billion by 2100 (based on the low International Institute for Applied Systems Analysis (IIASA) 1996 projection). Further, it is expected that population in urban areas will continue increasing from 49% (3.2 billion people) of 38 of 52

Deliverable

TRANSPHORM

D2.4.3

the world’s population in year 2005 to 60% (4.5 billion people) in 2030 (UN, 2008:1). Regarding climatic variability under a framework of climate change, the 4th Assessment Report of the IPCC presents mean regional climate projections for the 21st century from an ensemble of about 20 global climate models (GCMs) (Christensen et al., 2007) based on the A1B scenario (Nakicenovic et al., 2000).

Figure 6.1: General scheme of the urban climate system to generate air pollution scenarios.

The results presented in this study are particulate matter (PM10) future scenarios based on regional meteorological fields in a context of climatic change, isolating the rest of the features in the Urban Climate-System. The meteorological fields are based on the A1B scenario at hourly frequency extracted from the HIRHAM-ARPEGE Regional Climate Model (RCM) (Christensen et al. 2007) and force the PM10 diagnosis model developed by González-Aparicio et al. (2012) to obtain the PM10 increase for the XXI century. 6.2. Bilbao metropolitan area The integrated system is applied to the Bilbao metropolitan area, Northern of Spain, located in a coastal complex terrain (Figure 6.2a), covering an area of 360 km2 with a population density of 3732.1 inhabitants/ km2 (Basque Statistical Institute, EUSTAT 2009). The geographical location and its natural resources made Bilbao one of the most industrialized cities in Europe since 1980s. The main local sources of air pollution are traffic and industrial activities, being the PM10 of the main concern. According to the characteristics of its urban environment, Bilbao is classified in four types of urban districts based on the land-use database UDALPLAN 2009 (González-Aparicio et al. 2010): (1) city centre; industrial commercial district; (3) residential low density district, and (4) residential high density district (Figure 6.2b). The city is characterized by the transition strip of Atlantic and Mediterranean climates. Its regional climate is influenced by the westerlies and the Polar front (Millán et al. 1987). Due to its complex terrain and the proximity to the sea, the local transport mechanisms are influenced by the valley breezes interactions. In particular, under anti-cyclonic conditions and clear skies, the thermal differences and pressure gradients produce own local circulation developing the urban heat island (UHI) phenomenon and causing the trapping of pollutants under the urban boundary layer and the stagnation of the air flow drainage. 39 of 52

D2.4.3

TRANSPHORM

(a)

Deliverable

(b)

Figure 6.2: (a) The Bilbao metropolitan area in the Basque Country Region (Northern Spain) with main four valleys and the heights of the mountain ranges along the river? estuary /four meteorological stations selected for the analysis of the urban heat island effect are also shown/. (b) Classification into four urban districts: (1) city centre – ?high buildings district? ( ); (2) industrial commercial district ( ); (3) residential low density district ( ) and (4) residential high density district ( ) based on the land use database UDALPLAN 2009 (figures extracted from Gonzalez-Aparicio et al., 2010).

6.3. HIRHAM climatic future scenarios for modelling PM10 impacts Global Climate Models (GCM) represent the processes responsible for maintaining the general atmospheric/oceanic circulations and their variability. Given their coarse resolution, downscaling techniques are applied to capture the effect of fine scale forcing in areas characterized by variability of features such us topography and land surface conditions (Déqué, 2007; Hewitt and Griggs, 2004; Kjellström, 2007). In this study, climatic future scenarios are given by the HIRHAM regional climate model (RCM) driven by the ARPEGE GCM (Christensen et al, 2007). The resolution is 25x25 km covering the Bilbao metropolitan area. The simulations of the model run under the IPCC A1B (GHG) emissions scenario (Nakicenovic et al., 2000). The analysed variables are the hourly temperature (Temp), short-wave incoming radiation (solar), specific humidity (q) and wind speed and direction (wspd, wdir). The periods selected are summer (June, July, August - JJA) and winter (December, January, February - DJF) seasons during the years 2000-2100. The HIRHAM-ARPEGE output is used to force the statistical PM10 diagnosis model developed by González-Aparicio et al (2012). Additional variables are required to force the model: hourly NO2 emissions (µg/m3) and vehicle intensity (number of vehicles/hour, traffic). The main focus of this study is to estimate the PM10 impacts only due to climatic variability; these variables are considered the same as in 2010 for a standard street of the city centre (Eq 6.1). The hour of the day, month and day/night time (0420/21-03 UTC, respectively) are also included in the input of the model. The model is defined for two additional urban environments. The coastal sub-urban area characterised by residential neighbourhoods is considered to have 10 and 18% less of PM10 concentration in summer and winter, respectively. The inland sub-urban area is characterised by high traffic density and industrial activities and the PM10 concentrations are 35 and 20% higher than in the city centre. (Eq. 6.1) PM 10 j , k , h  K j , k , h  A  Temp j , k , h  B  q  j , k , h  C  solar  j , k , h  D  NO2  j , k , h  E  wspd  j .k .h  F  wdir  j , k , h  G  DayNight  j , k , h  H  traffic  j , k , h  I  month j , k , h  J  hour  j , k , h

where: j is summer or winter; k is the type of day (Working day, Saturday or Sundays), and h is the quarters of the wind direction (North, South, East, or West). 40 of 52

Deliverable

TRANSPHORM

D2.4.3

6.4. Future projections of climate change impacts on PM10 The analysis of the climatic influence on the PM10 variability is carried out at decadal based frequency (Figure 6.3). The 2 m air temperature has a positive trend and it is found a positive correlation with the PM10 levels. The PM10 highest concentrations occur at the ambient air temperature range from 15 to 25ºC. For summer season in the 2050’s decade it is expected an increment of 1.6ºC and 2.5ºC in 2100. For winter season, the anomaly is 0.2ºC less than in summer. The upwards trend of the specific humidity, 0.0020 kg kg-1 and 0.0008 kg kg-1 for summer and winter, respectively; and that means a higher water vapour content in the future climate. This could lead to higher H2O2 concentrations, the principal SO2 oxidant, and thus increasing sulphate aerosol concentrations. However, the wind speed is one of the main dispersive patterns of the air pollution. When the winds are weak, there is stagnation and the pollution levels increase. When the winds are higher, the pollutant dispersion is larger. It is observed that, for summer season, the wind speed has a downward trend. The wind speed changes with respect to the present period are -0.06 m/s in 2050 decade and 0.16 m/s at the end of the 21st century, which let the PM10 future concentrations to increase. The behaviour for winter season is more complex. At the first third of the century it showed an increase up to 0.5 m/s, causing lower PM10 concentrations. During 2040-2060 there is a substantial decrease up/down? to -0.2 m/s which lead to an increase in the concentration. The future scenarios do not estimate a significant change in the radiation and in the wind direction not influencing in/on? the PM10 anomalies.

(a1)

(b1)

(c1)

(a2)

(b2)

(c2)

Figure 6.3: Meteorological future scenarios of the decadal anomalies for the air temperature (in deg C; a1a2), wind speed (in m/s; b1-b2), and specific humidity (in g/kg?; c1-c2) increments with respect to the present period (2000-2010) for summer (a1-c1) and winter (a2-c2) seasons.

A statistical analysis was performed to assess the meteorological future projections and the impact of each variable on the PM10 variability. For the present-day period (2000-2010), mean observedPM10 concentrations are 29 and 32µg/m3 for summer and winter, respectively. The corresponding simulated are also 29 and 35 µg/m3. For summer, there is no difference between simulations and observations and for winter there is 3 µg/m3 overestimated. For summer future scenarios (Figure 6.4a), there is positive trend. The increment during the first half of the century is 0.15 µg/m3. The seasonal increment for the year 2050 is 1 µg/m3, and for 2100 it is 1.5 µg/m3 with respect to the present time. For winter (Figure 6.4b), it is found that seasonal mean concentrations for the period 2011-2050 show in general no significant change relative to present day. However, between the years 2050-2075 there is 0.6 µg/m3 and for the years 2076-2100 there is slight decrease but the increment is 0.3 g/m3 with respect to the present day. Extrapolating to the two additional urban environments for summer seasons, the increase of the PM10 is 1.35 µg/m3 at the coastal site and 2.1

41 of 52

D2.4.3

TRANSPHORM

Deliverable

µg/m3 at the inland site by the end of the century. For winter, it is 0.2 µg/m 3 at the coastal site and 0.36 µg/m3 at the inland site.

(a)

(b)

Figure 6.4: PM10 annual anomalies/ increments? scenarios for summer (a) and winter (b) seasons at the city centre.

For summer, the highest PM10 realisations by 2100 involve a rise around/about? 1.75 µg/m3. The interval of the future changes with respect to the present period ranges from -1.25 µg/m3 to 1.75 µg/m3 and the mean is 0.38 µg/m3 with the standard deviation (std.) 0.58 µg/m3. For the air temperature, the range of increase reaches 4ºC, and it is quite equally distributed along/within? the XXIst century. The mean is 1.25ºC (std. 1ºC). The wind speed future shift is also equally distributed with the mean value of -0.08 m/s (std. 0.20 m/s). For winter, the distribution does not show a clear trend: the range is from -1 to 1 µg/m3 with a mean of 0.2 µg/m3 (std 0.5 µg/m3). The temperature range is distributed with 15 times each bar ranging between -1 and 3ºC with a mean of 1.15ºC (std. 1.16ºC). The wind speed mean is 0.12 m/s (std 0.46 m/s). As the histogram shows (Figure 6.5), the slope is more pronounced during the last half of the century in summer and in the mid of the century for winter seasons.

(a)

(b)

Figure 6.5: Decadal anomalies distribution for the PM10 and the main variables influencing the PM10 for summer (a) and for winter (b).

Although the PM10 variability is known to have a larger range of uncertainties with respect to other pollutants, the analysis realized is in agreement with other similar studies. Liao et al. (2006) found that increased stagnation in the future climate causes PM to increase in polluted regions. Heald et 42 of 52

Deliverable

TRANSPHORM

D2.4.3

al. (2008) found also a positive response to rising temperature in continental regions due to increasing biogenic NMVOC emissions. Jacob et al. (2009) reviewed the studies focused on the impact of 21st century climate change on surface PM concentrations using Global Climate Models coupled with Chemical Transport Models (GCM-CTM) and they found that the projected changes are in a range ±0.1-1 µg/m3. This represent a potentially significant effect for polluted regions, but the differences in the regional precipitation response to climate change are the major cause of discrepancy in the PM10 response (Racherla and Adams, 2006; Pye et al., 2012; Jacob et al., 2009). Summarizing results of analysis it can be concluded that following the current trend the population is agglomerating in urban areas and hence, increases the size of cities. This concern together with future climatic variability turns against the improvement of the air quality and therefore, in detriment of the human health. In this context, the integrated system based on the urban, climatic and air quality scenarios is suggested for supporting decision making in the emission reductions. Additionally, it could also help in the urban planning as well as adaptation and mitigation strategies to combat the climate change on vulnerability basis. For a future perspective, the applicability of the system could be the foundations of a new sustainable city model. The results presented in this study are a part of the Urban-Climate System. It has been isolated the impact of the Climate Change at a Regional Scale and evaluated on the PM10 influence. It indicates that climate change could have substantial effects on air quality in urban areas and as a consequence on human health, agriculture, and natural ecosystems. Changes in local climatic variables such us temperature and circulation affect the different components of the pollutant life cycle and therefore affect the pollutant concentration. In addition, climate change can be highly variable from place to place, depending on the urban development as well, so that the effects of climate change on air quality need to be evaluated from case to case. Finally, consideration is that climate change is opposite in sign to that of currently envisioned pollution control measures. Although changes in pollutant emission might actually modify the influence of climate change, a global perspective of emission-control policies that consider both climate change and air quality should be integrated. References Christensen JH, Christensen OB. Summary of the PRUDENCE model projections of changes in European climate by the end of this century 2007. Springer Science + Business Media B.V. 7–30. Déqué M. 2007. Frequency of precipitation and temperature extremes over France in an anthropogenic scenario: model results and statistical correction according to observed values. Global planetary change 57: 16–26. González-Aparicio I. and Hidalgo J. (2011) Dynamically based future daily and seasonal temperature scenarios analysis for the northern Iberian Peninsula. International Journal of Climatology, Vol 32, issue 12, 1825-1833. González-Aparicio, I., Nuterman R., Korsholm U.S., Mahura A., Acero J.A., Hidalgo J. and Baklanov A. 2010. Land-Use Database Processing Approach for Meso-Scale Urban NWP Model Initialization. DMI Scientific Report 10-02, 34 pages. ISBN: 978-87-7478-593-4. González-Aparicio, Hidalgo J, Baklanov A, Korsholm U, Nuterman R, Mahura A, Santa-Coloma O (2012) Urban Boundary Layer Analysis in the Complex Coastal Terrain of Bilbao using Enviro-HIRLAM. Theoretical and Applied Climatology. DOI: 10.1007/s00704-012-0808-6 González-Aparicio I, Hidalgo J, Baklanov A, Padró A, Santa-Coloma O (2012) An hourly PM10 diagnosis model for the Bilbao metropolitan area using a linear regression methodology. Environmental Science and Pollution Research, DOI: 10.1007/s11356-012-1353-7 Heald, C. L., J. H. Kroll, J. L. Jimenez, K. S. Docherty, P. F. DeCarlo, A. C. Aiken, Q. Chen, S. T. Martin, D. K. Farmer, and P. Artaxo 2010. A simplified description of the evolution of organic aerosol composition in the atmosphere, Geophys. Res. Lett., 37, L08803, doi:10.1029/2010GL042737

43 of 52

D2.4.3

TRANSPHORM

Deliverable

Hewitt C, Griggs D. 2004. Ensembles-based predictions of climate changes and their impacts (ENSEMBLES) EOS 85 566. Technical Report. Jacob D, Bärring L, Christensen OB, Christensen JH, de Castro M, Déqué M, Giorgi F, Hagemann S, Hirschi M, Jones R, Kjellstrom E, Lenderink G, Rockel B, S´anchez E, Sch¨ar C, Sonia IS, Somot S, van Ulden A, van den Hurk B. 2009. An inter-comparison of regional climate models for Europe: model performance in present-day climate. Climatic Change 81: 31–52 Jülide Kahyaoglu-Koracin, Scott D. Bassett, David A. Mouat, Alan W. Gertler 2009. ation of a scenariobased modeling system to evaluate the air quality impacts of future growth. Atmospheric Environment 43 1021–1028 Kjellström E, Bärring L, Jacob D, Jones R, Lenderink G, Sch¨ar C 2007. Modelling daily temperature extremes: recent climate and future changes over Europe. Springer Climatic Change 81: 249–265 Kepner, W.G., D.J. Semmens, S.D. Basset, D.A. Mouat, and D.C. Goodrich. 2004. Scenario analysis for the San Pedro River, analyzing hydrological consequences for a future environment. Environmental Modeling and Assessment 94:115–127. Liao D, Peuquet DJ, Duan Y, Whitsel EA, Dou J, Smith RL, et al. 2006. GIS approaches for the estimation of residential-level ambient PM concentrations Environ Health Perspect 114:1374– 1380.10.1289/ehp.916916966091 Millán M., Otamendi E., Alonso L., Ureta I 1987. Experimental Characterization of Atmospheric Diffusion in Complex Terrain with Land-Sea Interactions. JAPCA 37:807-811. Nakicenovic N, Alcamo J, Davis G, de Vries B, Fenhann J, Gaffin S, Gregory K, Gr¨ubler A, Jung TY, Kram T, La Rovere EL, Michaelis L, Mori S, Morita T, Pepper W, Pitcher H, Price L, Riahi K, Roehrl A, Rogner HH, Sankovski A, Schlesinger M, Shukla P, Smith S, Swart R, van Rooijen S, Victor N, Dadi Z. 2000. IPCC Special Report on Emissions Scenarios, Cambridge University Press: Cambridge, United Kingdom. Pye, H.O.T., Seinfeld, J.H., Liao, H.,Wu, S., Mickley, L.J., Jacob, D.J. Effects of changes in climate and emissions on future sulfate-nitrate-ammonium aerosol levels in the United States. J. Geophys. Res., in press. Racherla, P.N., and P.J. Adams, 2006: Sensitivity of global tropospheric ozone and fine particulate matter concentrations to climate change. J. Geophys. Res., 111, D24103, doi:10.1029/2005JD006939 Sutherland, W.J., Pullin, A.S., Dolman, P.M. & Knight, T.M. 2004. The need for evidence-based conservation. Trends in Ecology and Evolution, 19, 305–308. Solecki D. and Oliveri C. Downscaling climate change scenarios in an urban land use change model, 2004. Journal of Environmental Management 72 105–115 Wilby Robert L. Constructing climate change scenarios of urban heat island intensity and air quality 2008. Environment and Planning B: Planning and Design volume 35, pages 902-919.

7. Application of meteorology with climate change for future transport scenarios UH-CAIR Contributions by Xin Kong and Ranjeet Sokhi UK Met Office Contributions by Gerd Folberth and Bill Collins In this study the interaction of climate change and the future air quality changes in Europe is examined and quantified. The regional WRF-CMAQ modelling system, driven by the global climate model HadGEM2-ES (RCP 8.5 scenario), has been applied for one summer month for the base year 2000 and the future year 2050. 7.1. HadGEM-WRF-CMAQ modelling system and selected case studies In UH-CAIR, the HadGEM-WRF-CMAQ modelling system (Figure 7.1) is used for air quality and climate change interactions studies. HadGEM2-ES is provided by the UK Met Office to drive the regional WRF-CMAQ modelling system. HadGEM2-ES is a fully coupled earth system model with 44 of 52

Deliverable

TRANSPHORM

D2.4.3

interactive chemistry from the UKCA model (UK Chemistry and Aerosols; http://www.ukca.ac.uk). HadGEM2-ES includes aerosol scattering and absorption of solar and terrestrial radiation (direct effect) and the cloud droplet number (indirect effect), and as such is capable of evaluating AQ (air quality) and CC (climate change) interactions at the global scale (Collins et al., 2008). An interface ‘UM-WRF’ has been developed by UH-CAIR to assimilate the HadGEM2-ES output data (in pp format) into WRF and vertically interpolate the HadGEM2 data from hybrid heights onto sigma pressure levels. The HadGEM2-ES data is downscaled from 1.25×1.875 degrees to 18 km over the European area and 6 km over the London area in the WRF model. The anthropogenic emission data for 2050 provided by TNO are for a moderate climate protection regime which assumes the adoption of efficient technologies and increased renewable energy usage (Theloke et al., MEGAPOLI report 2010), which leads to reductions in most air pollutant emissions (~20% reduction in PM2.5 and PM10 emissions by 2050). These data are processed using SMOKE (Houyoux et al., 2000), an emission preparation tool. The global model GEOS-Chem driven by global NASA GISS data (SRES A1B scenario) is used to drive the CMAQ chemistry at the boundaries. In order to assess the contributions of climate change and emission change separately, four cases have been considered: (case 1) current emission (Em2000) and current climate (Cli2000), which represents current baseline; (case 2) current emission (Em2000) and future climate (Cli2050), which represents changes due to climate effects and we define delta_C = Em2000_Cli2050 – Em2000_Cli2000; (case 3) future emission (Em2050) and current climate (Cli2000), which represents changes due to emission effects and we define delta_E = Em2050_Cli2000 – Em2000_Cli2000; (case 4) future emission (Em2050) and future climate (Cli2050), which represents changes due to combinations of climate effects and emission effects and we define delta_CE = Em2050_Cli2050 – Em2000_Cli2000.

Figure 7.1: Schematics of the HadGEM-WRF-CMAQ modelling system used in this study.

7.2. Contributions of climate change and emission change to meteorology and PM The simulated meteorological variables are shown in Figure 7.2. As seen in Figure 7.2a the mean air temperature in July increases by about 2.7 K in the European domain by 2050. As seen in Figure 45 of 52

D2.4.3

TRANSPHORM

Deliverable

7.2b the surface specific humidity increases by 1.5 g/kg over most of Europe (except Italy and Croatia) as the air temperature increases. The increased 10 m wind speeds (Figure 7.2c) in the future year generally improve air quality; however, there will be more wind blown dust leading to increased PM10 concentrations. The southern area shows a greater increase in temperature than the north, and reduced wind speeds and planetary boundary layer (PBL) heights compared to the base year (Figure 7.2d). The model predicts a more stable atmosphere in southern Europe that may affect air quality over this region. Figures 7.3a and 7.4a show that the pollution levels over the southern part of Europe is higher as compared to the northern part. The above findings re-emphasises the importance of emission controls in the countries in southern parts of the Europe. Figure 7.3b and 7.4b show the effects of changes in temperature, humidity, wind speed and PBL heights on future PM2.5 and PM10 concentrations. As the same anthropogenic emissions have been used for both the base and future year, the results only take climate change into account. The result indicates that climate change could cause increased ground-level PM2.5 and PM10 concentrations over southern and eastern areas of the domain (e.g., Portugal, Spain, Italy, Poland, Ukraine), while decrease ground-level PM2.5 and PM10 concentrations over the north of the domain (e.g., UK, France, Germany, Norway, Sweden and Finland). The result for the Portugal case is consistent with the recent paper by Carvalho et al. (2010). The simulations of PM2.5 and PM10 shown in Figures 7.3c and 7.4c illustrates an improvement in air quality almost everywhere over Europe under the TNO 2050 emission change scenario. Figures 7.3d and 7.4d shows the extent to which reduced anthropogenic emissions can compensate for the effects of climate change delta_C (negative effect over southern/ eastern part of domain and positive effect over northern part of the domain). The future year ground-level averaged PM2.5 and PM10 concentrations over the whole of the European domain reduced by 33 and 4%, respectively in 2050 compared to the base year. Therefore, we derived an emission control index –delta_C/delta_E, which can be used at the country level to provide information for policy makers. The countries with pink/red areas in Figures 7.3e and 7.4e would need to strengthen their emission controls in order to maintain their current level of air quality in the future climate scenario. Additionally, the results in Figures 7.3 and 7.4 showed that both PM2.5 and PM10 over the ocean increased in 2050 compared with the base year due to climate change. For example, future PM2.5 concentrations increase significantly over the north-east Atlantic (e.g., off the Portuguese coast). This could be caused by a more stable atmosphere (i.e., reduced wind speed and PBL heights) over the region. Similarly, future PM10 increase significantly over the both Norwegian and North Seas which might be due to increased wind speed. Further investigation is needed to look at the sea-salt emissions and how they are affected by climate change. Shipping emission controls over ocean areas with high climate change effects may need to be carefully considered as well. Model simulation results show that climate change alone will likely have increased ground-level PM2.5 and PM10 by 2050 over the European domain, but with large variations from region to region. The southern area of the domain is affected more by climate change due to the generally more stable atmosphere. However, the reduced anthropogenic emissions under the TNO future scenario can compensate for the effects of climate change. An emission control index –delta_C/delta_E has been derived at the country level to provide information for policy makers.

46 of 52

Deliverable

TRANSPHORM

D2.4.3

Figure 7.2: WRF simulated monthly-mean - (a) 2m air temperature; (b) 2m specific humidity; (c) 10m wind speed; (d) planetary boundary layer (PBL) heights - differences between the future (Jul 2050) and base (Jul 2000) years under RCP8.5 scenario.

47 of 52

D2.4.3

TRANSPHORM

Deliverable

Figure 7.3: CMAQ simulated monthly-mean PM2.5: (a) base run: Em2000_Cli2000; (b) Em2000_Cli2050 – Em2000_Cli2000, which represents changes due to climate effects delta_C; (c) Em2050_Cli2000 – Em2000_Cli2000, which represents changes due to emission effects delta_E; (d) Em2050_Cli2050 – Em2000_Cli2000, which represents changes due to combinations of climate effects delta_C and emission effects delta_E.; and (e) Emission control indicator: -delta_C/delta_E.

48 of 52

Deliverable

TRANSPHORM

D2.4.3

Figure 7.4: CMAQ simulated monthly-mean PM10:(a) base run: Em2000_Cli2000; (b) Em2000_Cli2050 – Em2000_Cli2000, which represents changes due to climate effects delta_C; (c) Em2050_Cli2000 – Em2000_Cli2000, which represents changes due to emission effects delta_E; (d) Em2050_Cli2050 – Em2000_Cli2000, which represents changes due to combinations of climate effects delta_C and emission effects delta_E, and (e) Emission control indicator: -delta_C/delta_E.

References Carvalho, A., Monteiro A., Solman S., Miranda, A.I., Borrego, C., 2010. Climate-driven changes in air quality over Europe by the end of the 21st century, with special reference to Portugal. Environmental Science and Policy 13 (2010) 445-458.

49 of 52

D2.4.3

TRANSPHORM

Deliverable

Collins, W.J., Bellouin, N., Doutriaux-Boucher, M., Gedney, N., Hinton,T., Jones, C.D., Liddicoat, S., Martin, G., O’Connor, F., Rae, J., Senior, C., Totterdell, I., Woodward, S., Reichler, T., Kim, J. and Halloran, P., 2008. Evaluation of the HadGEM2 model, Hadley Centre technical note 74, Nov 2008, the UK Met Office. Houyoux, M.R., Vukovich, J.M., Coats Jr., C.J., Wheeler, N.M., Kasibhatia, P.S., 2000. Emission inventory development and processing for the seasonal model for regional air quality (SMRAQ) project. Journal of Geophysical Research 105 (D7), 9079-9090. Theloke, J., M. Blesl, D. Bruchhof, T. Kampffmeyer, U. Kugler, M. Uzbasich, K. Schenk, H. Denier van der Gon, S. Finardi, P. Radice, R. S. Sokhi, K. Ravindra, I. Coll, R. Friedrich, D. van den Hout, 2010. European and Megacity Baseline Scenarios for 2020, 2030 and 2050. MEGAPOLI Project Scientific Report 10-23. MEGAPOLI-26-REP-2010-12.

8. Main Conclusions and Recommendations MEGAPOLI studies relevant to TRANSPHORM showed that: 1. Highly detailed local inventories would be needed in order to correctly quantify the emissions from megacties and the exposure to pollution in megacities. The higher total emissions in megacities lead to higher exposure levels, despite the lower emissions per capita, since the population is in the direct vicinity of the source, which could be responsible for many different health effects, such as asthma in children. 2. Pollutants from megacities are transported over regional and even intercontinental scales and contribute to the pollution levels in other specific regions. It is found that megacities in Europe are the main contributors to deposition of aerosols (especially absorbing aerosols like soot) in the Arctic, both for the annual and the wintertime deposition. 3. Dominant process reducing the concentration of air pollutants when pollutants are dispersed away from megacities was found to be dilution; the rate of dilution depends strongly on meteorological conditions. Air pollutants undergo physical and chemical changes as they are transported away from megacities; this contributes to increased levels of ozone and secondary particulate matter in nearby regions downwind of megacities. 4. Understanding of megacity emissions has improved, but there is still a need to determine these more accurately; so, more work is needed, such as establishing an inventory development level, improving knowledge about differences between the megacities and the remainder of the country, such as in fuels, fuel quality and appliance types. The megacity emissions analysis showed very distinct characteristics in the main sources of pollutants depending on the geographical regions where the megacities are located. 5. Megacities contribute a warming of over 0.2 K after 100 years, with nearly 90% of this due to carbon dioxide emissions, and most of the remaining 10% due to methane. It has been done using a simple analytical technique based on the climate sensitivity computed by complex models. 6. Contribution of megacities to global pollutant emissions is of the order of 2-6% of the total global annual anthropogenic emission flux; however, megacities have a disproportionately smaller impact on air pollutants, especially ozone, for which the global burden changes by < 1% when megacity emissions are removed from the simulations. 7. Growth of megacities will considerably affect future urban climate; the analysis of the impact of pollutants such as NOx, VOC and aerosols from megacities on climate under future conditions suggested that, compared to today, the total forcing of these atmospheric pollutants is slightly positive at +0.4±0.11 mW/m2 with a warming impact on future climate that adds to the much larger impact from CO2. 8. Large-scale circulation in the Northern Hemisphere causes megacities in Europe, especially Saint-Petersburg, Moscow and the Ruhr Valley to be the most significant contributors to 50 of 52

Deliverable

TRANSPHORM

D2.4.3

deposition of aerosols (especially absorbing aerosols like soot) in the Arctic; the strong winds in mid-latitude cyclonic and anti-cyclonic systems, along with suppressed vertical mixing, result in strong long-range, near-surface export of pollutants. 9. Indirect effects of aerosols significantly modify meteorological parameters, such as daytime temperatures and height of the planetary boundary layer. NO2 concentrations are moderately affected and the direct aerosol effect was found to have substantial impact on the turbulence of the flow near the surface. For coastal megacities, increases of land temperature induced by climate change can lead to more intensive and frequent sea breeze events and associated cooler air and fog. 10. Hierarchy of urban canopy models/parameterisations for different model types and scales is needed and was suggested. For urban air pollution, from traffic emissions and for the modelling of cases of emergencies (e.g., accidental releases), there is a great need of vertical profiles of the main meteorological parameters and the turbulence characteristics within the urban canopy. A large number of urban surface energy balance models now exist with different assumptions about the important features of the surface and exchange processes that need to be incorporated. 11. Integrated models framework was successfully applied to investigate air quality affecting megacites. But still a number of further research developments are needed. MEGAPOLI recommended working towards better consistency in the use of data for megacities, moving towards online or coupled approaches and addressing the need for dedicated and targeted data sets for model evaluation purposes. Prediction of PM remains a challenge and is an important area for continued research. The sensitivity of air quality to feedbacks from climate change interactions needs to be quantified; this will require online coupled models. 12. The direct impact of climate change on air quality in megacities is significant due to changes in temperature (BVOC fluxes, wild fires, deposition, O3, CH4, SOA, pSO4, pNO3), radiation (photolysis), clouds, and precipitation. As climate changes, ozone concentrations will further increase, unless emission reduction measures are implemented. However, the expected emissions reduction during the next several decades is much stronger than the ozone increase due to climate change expected in the most megacities regions studied within the project. Only a more frequent appearance of extremes like heat waves under the climate change might bring air quality problems with them, e.g. exceedances of limits of ozone concentrations on local or regional scales. 13. In order to reduce pollutant emissions and exposure efficient, MEGAPOLI recommends expansion of district heating networks in the residential sector (reducing especially domestic wood burning), the implementation of a kerosene tax for aviation activities, implementing traffic management measures in cities and the promotion of low emission vehicles (E-cars, hybrid vehicles) on a midterm (2030) as well a long term time scale (2050). The measure "energy-efficient modernisation of old buildings" has been identified also as very cost efficient, but is not recommended because of the non- negligible additional health impacts from accumulated indoor pollutants in insulated buildings. CityZen and EMEP studies relevant to TRANSPHORM showed that: 1. CityZen, in collaboration with the EUCAARI and ENSEMBLE projects, and the EMEP programme, has run the EMEP chemical transport model with climate data from the HIRHAM climate model. Projected emission reductions will improve air quality in terms of ozone, but climate change could reduce these benefits according to model results. 2. In particular, particle loadings in the Mediterranean Area will be lower in 2030 to 2050 due to reductions in emissions of particles and their precursors. Climate change, however, could lead to a small reduction of this benefit due to reduced washout of particles in a dryer climate.

51 of 52

D2.4.3

TRANSPHORM

Deliverable

3. In Europe a prominent win-win emission reduction measure is the introduction of Diesel particle filters, which will reduce particle loadings – a benefit for both climate change and air pollution mitigation. 4. Main uncertainties in today’s model frameworks are in the representation of climatechemistry couplings, e.g. between climate change and biogenic emissions. Additional TRANSPHORM studies showed that: 1. The differences in temperature and precipitation between the 2000 - 2010 and the 2020 2030 decade (and 2040 – 2050), and effects on PM2.5 have been shown by EMEP team. The effects on temperature and precipitation are larger, and hence the effects on PM2.5 are also larger. This has been included in the 2010 EMEP report and has also been reported to the EU project EUCAARI. 2. The UH-CAIR and UK-MetOffice model simulation results show that climate change alone will likely have increased ground-level PM2.5 and PM10 by 2050 over the European domain, but with large variations from region to region. The southern area of the domain is affected more by climate change due to the generally more stable atmosphere. However, the reduced anthropogenic emissions under the TNO future scenario can compensate for the effects of climate change. An emission control index –delta_C/delta_E has been derived at the country level to provide information for policy makers. 3. The TECNALIA-DMI results presented also PM10 future scenarios based on regional meteorological fields in a context of climatic change, isolating the rest of the features in the Urban Climate-System on example of the Bilbao region. The meteorological fields are based on the A1B scenario from the HIRHAM-ARPEGE RCM (Christensen et al. 2007) and force the PM10 diagnosis model developed by González-Aparicio et al. (2012) to obtain the PM10 increase for the XXI century. Quantitative (climatic and urban) and qualitative (temporal and spatial) variables have been used to build the PM10 diagnosis model. Three different types of linear models have been tested: simple linear regression techniques; linear regression with interaction terms and linear regression with interaction terms following the Sawa’s Bayesian Information Criteria. Each type of model is calculated using training and a testing dataset from the period 2005-2011. The results of each type of model show that the linear regression with interactions following the BIC criterion is the best with a R2 of 0.42 and 0.40 for the training and testing period, respectively.

52 of 52