Open access

Evaluation of Regional Emission Control Based in Photochemical Air Quality Modelling

Written By

Ángel Rodríguez,1Santiago Saavedra, María Dios, Carmen Torres, José A. Souto, Juan Casares, Belén Soto and José L. Bermúdez

Submitted: October 21st, 2010 Published: August 17th, 2011

DOI: 10.5772/17536

Chapter metrics overview

1,904 Chapter Downloads

View Full Metrics

1. Introduction

The growing use of natural gas as fuel both forin industrial and domestic purposescombustions implies an increment in the emissions of potential photochemical oxidants species in the troposphere, particularly, nitrogen oxides (NOx). This effect can be higher in regions with a significant use of other fossil fuels for electricity production, such as in coal fired power plants.

Therefore, even though the air quality levels of typical primary pollutants (as SO2 and coarse particles) can keep to acceptable values, synergies between additional NOx emissions, biogenic emissions and the meteorological conditions can lead to derive in the production of photochemical oxidants, such as O3. The application of meteorological and air quality modelingmodelling techniques to evaluate the impact of new NOx sources in the O3 levels over a region can achieve a quantitative estimation of thesuchat impact, in terms of the variation of both NOx and O3 levels in the region. This approach tofor the impact of new sources in air quality, nowthat is common for primary pollutants, requires the use of comprehensive air quality modelingmodelling in the evaluation of photochemical oxidants., as O3.

There are severalDifferent comprehensive air quality models are currently available that can be applied toin this approach. However, the characteristics of the new emissions sources, regional conditions (i.e., terrain, sea influence) and pollutants of interest (i.e., photochemical oxidants) result indrive to the selection of the most appropriate model for a particular problem (Chang et al., 1987; Song et al., 2010).

In this paperwork, comprehensive air quality modelling is applied to evaluate the effectiveness of emissions control policy in a European Atlantic coastal region, with complex terrain and mixed land use, andwith a significant area of forestry. This region, located in the northwest of the Iberian Peninsula, includes both medium-sized cities and large coal fired power plants. Therefore, changes in both industrial and domestic emissions due to an increase in the use of the natural gas use can affect the generation of photochemical oxidants, depending on the biogenic emissions (forestry) and typical meteorological conditions.


2. A case study

Air pollution onat local and regional scales is conditioned by a significant number of parameters grouped into three classes (Seinfeld and Pandis, 1998): topography, meteorology and precursors sources. Meteorology, which is implicitly linked to topography, plays an important role in air quality. Topography defines air flow paths and pollutants depositions. The influence of eEmissions sources influence depends on both the amount of pollutants and the location of thesuchamount and sources location (Beck et al., 1998). Therefore, the description of the area under study must include these three groups of parameters.

Figure 1.

Regional domain under study, including the Galician region. In addition, the location of the five O3 ground level concentration stations around As Pontes power plant, to be used identifying considered for the identification of air quality episodes, is shown.

2.1. Topography

The area under study (figure 1) is characterized by a complex terrain environment, with largegreat differences. The central part is defined by the As Pontes valley, from west to east, with theflowing river Eume flowing along it. This valley is surrounded by the smooth Atlantic coast inat the west and the more complex Cantabrianc coast inat the nNorth; there are several mountains in the north-east along the Northeast several Northern mountains are located ( Xistral), and the valley boundaries are closed inat the East east bywith the Dorsal Gallega mountain range, with heights up to 1000 m high. Along the sSouth, the topography displaysis smooth, and uniform,, as elevated plains. Therefore, it is a complex terrain, combiningwith several granitic mountains and valleys mixed in the same environment.

2.2. Meteorology

This region is characterized (Martínez Cortizas et al, 1999) by rains distributed throughout along the year, with an annual precipitation rate between 1000 and 1600 mm, mainlymore usual during autumn and spring time, and sporadic induring summertime (but not unusual, as there are isolated storms in the afternoons). Summers are mildsoft, as the sea breeze refreshes the coastal areas and the height of the mountains regulates the temperature in the inner areas. Summer days are usually sunny, with low moisture and temperatures from 20 to 30 ºC. On the other hand, heat waves are not usual in summer, and they only lastspend a few days, as the proximity of the coast keeps the average temperature at 20-25 ºC, with higher temperatures (up to 30 ºC) inat the interiorner valleys.

Main winds come from the SW and NW during winter and autumn time, when pressure is lowwith low pressure conditions,; whereason the other hand, high pressure conditions, typical induring summertime, usually corresponds to NE winds. The atmosphere-topography interaction iss are very important in this environment, especially the mountains, which affect the production ofwith their effects in rain production, and the complex coastal topography, which that increases the sea breeze.

2.3. Sources of ozone precursors

TheAs the main ozone precursors are taken to be, oxidized nitrogen and volatile organic compounds are considered. Sources of these precursors located inat the area under study (Casares et al., 2005) can be divided into point sources (usually, industrial sources) and diffuse sources. From the iIndustrial sources include, the Artabric Arc Industrial Belt (located onat the Wweest coast, including the cities of A Coruña and Ferrol) is considered, where there are variouswith different industrial sectors such as petroleum refining, energy (thermal power plants and cogeneration units), wood, food and metallurgy. Other sparse emissions sources, such as industrial waste plants, are also included.

As tThe main diffuse sources are, traffic and biogenic emissions, from cattle farming and forest cultivationures (mainly, eucalyptus and pines). are considered. However, as ozone can be transported over long distances,a long distance transport of ozone is possible, it is necessary to take into account not only the local sources, but also sources as in neighboringneighbouringhood regions (Asturias, Castilla-León and Portugal), and even further afield. This It was done by the analyzingsis of air quality levels in those regions, as a characteristic of air masses that can be carried to the area under study. In order to evaluateFor the evaluation of the impact of emissionemissions changes on impact in air quality, three typical meteorological conditions which are favorablefavourable to increasing the increment of tropospheric ozone levels are identified in the NW of the Iberian Peninsula. From these typical conditions, three different periods were selected as reference episodes to apply a comprehensive air quality modelling of different emissions scenarios: Episode 1 (14-23 July 2002), episode 2 (16-24 March 2003), and episode 3 (12-22 September 2003). The characterization of these episodes shows a different principal origin of ozone peaks depending on the meteorological conditions: episode 1, North of Portugal; episode 2, Castilian plateau, ; and episode 3, and Galicia.


3. Air pollution episodes

3.1. Air quality data

Hourly ground level concentration registers of O3, NOx and NO2, were collected for the 3-years period 2002-2004. First of all, validated data were obtained from 10 air quality stations included in the Galician Air Quality Network, in order to identify the O3 episodes, and to obtainget a first approach toof them. These stations are mainly rural, with only background influence or, sometimes, influence from distantfar industrial sources. After identifying the main O3 episodes, in order to study the possible influence from surrounding regions was studied by looking at, more air quality stations were considered: 19 air quality stations from the National Portuguese Air Quality Network, 18 from the Castilla-León Regional Air Quality Network, and daily average data from the Asturias Regional Air Quality Network (as hourly data were not available).

Figure 2.

An example of a synoptic chart provided by Wetterzentrale, applied in the classification of ozone air pollution episodes in the region under study.

3.2. Meteorological dataset

The regional scale meteorological analysis of the episodes was done by means of synoptic maps (Font-Tullot, 1983; Castell et al., 2004), in order to identify the synoptic conditions with more influence in the tropospheric ozone episodes over Galician region.

Synoptic charts (figure 2) were collected from the National Center for Environmental Prediction (NCEP, USA) reanalysis. These charts include synoptic observations which are interpolated and recalculated by diagnostic models, so they are more trustworthy thanconfident that forecast charts. In these NCEP charts, several variables were included, over geopotentials at 500 mb and, sea level pressure; in addition, temperature maps at 500 and 850 mb are included. These charts are routinely collected by the Wetterzentrale (German Meteorological Office) on its web page (

Regional meteorology was completed by observations of surface temperature and wind, precipitations and other typical meteors (sky cloudsiness, storms, fogs, etc.)…) from Wetterzentrale maps based in European meteorological stations with 6 hr observations.

After identifying the ozone episodes, local meteorological observations were collected, in order to obtainget a more comprehensive analysis of the meteorological influence over local pollution episodes, taking into account local phenomena, such as katabatic and anabatic winds, sea breeze and, thermal inversions, which are not represented on ain synoptic scale. This is an important issue in a complex terrain like domain as Galician region.

Local surface meteorological measurements were takenincluded, in on an hourly basis and included, surface temperature (2 m) and surface wind speed and direction (10 m) from five stations of the As Pontes Power Plant air quality network, integrated in the Galician regional air quality network. These data were extended with hourly measurements from A Mourela station (D1 station, close to the Power Plant), providing temperatures at 2, 10, 30 and 80 agl-m (which is useful for a stability classification), wind at 10 and 80 agl-m, solar radiation and precipitation.

Episode period Maximum hourly O3 glc date/time Maximum hourly O3 glc (μg/m3)
Year 2002
14-23 July 18 July - 16 UTC 201
11-17 August 14 August - 16 UTC 142
23-04 September 02 September - 16 UTC 157
09-18 September 14 September - 14 UTC 136
21-30 September 27 September - 17 UTC 145
Year 2003
16-24 March 21 March - 16 UTC 148
26-01 June 29 May - 05 UTC 161
18-24 June 20 June - 17 UTC 210
06-13 July 08 July - 15 UTC 139
30-16 August 07 August - 17 UTC 174
12-22 September 16 September - 15 UTC 193
Year 2004
14-22 May (12) 19 May - 15 UTC 170
31-11 June (13) 05 June - 15 UTC 156
12-19 June (14) 17 June - 16 UTC 196
28-05 August (15) 01 August - 00 UTC 181

Table 1.

O3 episodes identified in the region under study, from the glc measurements available, takingconsidering the dual criteria into account: either 150 μg/m3 threshold or significant rise of glc in a short period.

3.3. Air pollution episodes

A coarse selection of episodes with tropospheric ozone levels higher than usual (risky episodes) was made with reference todone considering the air pollution measurements dataset inat the domain for 2002, 2003 and 2004. years. These measurements corresponds to the five air quality stations located mainly in rural areas around As Pontes Power Plant (figure 1), and distributed overfrom 2 to 30 km faraway from theis industrial site. These stations are integrated intoby the Galician air quality network, controlled by the regional government.

In order to select O3 risky episodes, two criteria were adopted: (a) tropospheric ozone threshold of 150 μg/m3, as an hourly average (20% below first European Union legal threshold); (b) persistent rise of tropospheric ozone levels in several stations, independently on the absolute values achieved.

The period under study for each episode was established as at least 7 days around the maximum concentration, which can be extended depending onconsidering the concentration tendency (persistence either in ascent or descentrise or low). This period is important because of the time required by ozone production and destruction, taking into account the influence of meteorological conditions in these chemical phenomena. In fact, the duration of an episode was extended until it was clear that clean air masses with low ozone levels were achieved in the domain, enassuring that the episode hadwas ended.

With these criteria, 15 tropospheric ozone air pollution episodes (table 1) were identified in the domain, 5 in the year 2002, 6 induring year 2003, and 4 induring year 2004. All the episodes were observed between March and September. Episodes lastedduration varied from 7 to 18 days. These episodes can be represented by three typical meteorological conditions which are favourable to the increase in ment of tropospheric ozone levels, and are identified in the NW of the Iberian Peninsula. Thereforen, three different periods were selected as reference episodes to apply a comprehensive air quality modelling toof different emissions scenarios: Episode 1 (14-23 July 2002), episode 2 (16-24 March 2003), and episode 3 (12-22 September 2003). The characterization of these episodes shows a different main origin of ozone peaks depending on the meteorological conditions: episode 1, North of Portugal; episode 2, Castilian plateau, ; and episode 3, and Galicia.


4. Air quality modelling

In this workpaper, the CAMx model (Environ International Co., 2010; Song et al., 2010) was selected because of its capacity to integrate large aloft emissions from power plants stacks in the model grid with area emissions representing sparse sources (domestic and transport). This approach is essential to get obtain a good representation of the impact from large point sources impact in the regional air quality.

The CAMx model is coupled withto the PSUN/NCAR MM5 meteorological model (Grell et al., 1995) in order to provide the required meteorological conditions for both dispersion and chemical transformation of pollutants.

4.1. Simulation domains

The area under study is coversing Galicia, a region in the north-west of Spain,; however,although the simulation domain has to cover a larger areaextension in order to minimize the boundary effect onover the model results. However, because of the high resolution required in the complex region under study, to applying a high resolution to the larger simulation domain leadsdrives to heavya high computingational workeffort. Therefore, the useapplication of the nesting technique is highly recommended.

Thusen, a coarse resolution grid (with a resolution of 27x27 km2) covering the Iberian Peninsula (see figure 3a) is defined. Simulation results over this larger domain were as applied as input to a nested inner grid (with a resolution of 9x9 km2, see figure 3b), following a one-way nesting approach; i.ethat is, it is thoughtconsidered that the meteorology over the inner grid doesis not substantially affect to the meteorology over the outer grid.

Figure 3.

Simulation grids covering (a) the Iberian Peninsula, and (b) the Northwest northwest of the Iberian Peninsula, including the Galician regioón.

An additional problem in the application of air quality models is relatesd to the geographical projections. These models are usually solved over plane grids, even though the Earth surface is curved. In order to define the plane representation of the Earth surface, the following parameters are defined: (a) geodesic datum, that isi.e., the shape and dimensions of the ellipsoid to be represented; in our case, it is the the Earth planet. In this work, tThe European Datum 50 (ED50) wasis adopted; (b) geographical projection, that isi.e., a sorted relationship between the locations over the curved surface (Earth’s surface) and the plane surface over the simulation grid is defined. In this case, UTM (“Universal Transverse Mercator”) projection referringed to 29 North Time Zone is usedconsidered.

As a result, the Iberian Peninsula domain is represented by a grid (figure 3a) with 44x47 cells. The inner domain is represented by a grid (figure 3b) with 41x41 cells. However, the computingational effort work required to solve the inner grid is higher, because of the Courant number; that isi.e., the smaller is the size of the grid cells is, the shorter is the time step to be applied in the numerical integration of the model equations during the same simulation period. ThereforeSo, in the inner grid, it is necessary to solve the equations more oftentimes to simulate the same period. This is a significant limitation inof the application of the Eulerian models (such as the CAMx model) in the simulation of the air quality with high resolution grids.

4.2. Emissions inventories

As the simulation domains cover the Iberian Peninsula, including the Galician region,. For these domains, a 2001 annual emissions inventory was developed by the combiningation of the top-down inventory from EMEP (Iberian Peninsula) and the bottom-up industrial emissions EPER inventory from REGADE (Casares et al., 2005). SMOKE (The Institute for the Environment, 2003) was applied for speciation, time and spatial segregation of industrial emissions.

In addition, a 2010 annual emissions projection (table 2) was carried outdone usingconsidering changes in industrial emissions, due to the application of Best Available Techniques (BATs) to reduce emissions from the old power plants, and the installation of new combined cycles in the region. This new scenario follows the legal restrictions currently approved by the Government, so changes should leaddrive to a significant reduction ofin primary pollutants levels (SO2, PM10); however, secondary pollutants, such as ozone, are affected by extremely non-linear physical and chemical processes that should be considered whento evaluatinge the effect of emissions changes over their levels. These non-linear processes can be simulated by a comprehensive air quality model.

Source NOx (Tm/year) Change
2001 inventory 2010projection
Sabón 833 738 +320 %
- 2764
Meirama 9059 1604 -83 %
As Pontes 20035 6863 -52 %
- 2764

Table 2.

NOx emission changes between year 2001 (reference inventory) and 2010 (projected inventory), in the three power plants locations.

As the main differences in the industrial emissions of ozone precursors between 2001 inventory and 2010 projection, NMVOCs shows an increment of 6.6% and NOx a reduction of 27 %. In addition, a significant spatial redistribution of NOx industrial sources is observed (table 2), due to: (a) fuel changes in the three Galician high stacks power plants, which hasve reduced in 50% their NOx emissions by 50%; (b) two planned combined cycles, with NOx emissions at lower height (less than 200 m).

In addition to the anthropogenic contribution, the role of volatile organic compounds (VOCs) in the tropospheric ozone production is also taken into accountconsidered, especially for biogenic organic species (BVOCs), such as isoprene and terpenes (Makar et al., 1999; Geron et al., 2000). These BVOCs emissions were estimated by using MEGAN model (Guenther et al., 1995), as this model usesconsider a high resolution functional plant distribution (1x1 km2) and their results can be processed asto input (such as time-varying emissions) to the CAMx model.

4.3. Emissions integration

4.3.1. Simulation grids

For a realistic representation of the anthropogenic emissions in the region, different emission inventories were combined: EMEP for the Peninsula domain, and EPER-REGADE for the Galician domain. Both inventories were adapted to the corresponding nested grids using SMOKE, in order to provide the emissions input to the CAMx model.

  1. Peninsula domain

In the Peninsula domain, only the EMEP emissions inventory was applied. However, this inventory is aggregated in a 50 km resolution grid with the origin in the North Pole; therefore, it must be adapted to adaptation to the Peninsula domain simulation grid is required. This adaptation has been done by applyingwith the application of the GIS extensiónextension of the PortgreSQL database, with a distribution weighted to the surface area (figure 4).

  1. Galician domain

In the Galician domain, EMEP emissions will bewere applied for the CORINAIR sources, otherdifferent than S1 and S3, following the same process as in the Peninsula domain.

Figure 4.

Adapting the EMEP emissions grid to the simulation grid, over the Peninsula domain.

With the EMEP inventoryadapted to both simulation grids, all data have been integrated into SMOKE in order to combine the different anthropogenic emissions and to provide the emissions input to the CAMx model.

For CORINAIR S1 and S3 sources, the EPER-REGADE Galician inventory was also integrated into SMOKE, as a bottom-up point sources inventorys. In pParticularly, the REGADE inventory applies to the detailed SCC classification from EPA, with equivalences in the coarser CORINAIR sources classification. In addition, this procedure is important in order to obtain an accurate geographical location of these industrial sources and, particularly, to take into account their different stack heights in the CAMx algorithm for elevated point sources.

4.3.2. Time-varying anthropogenic emissions

Depending on the type of emissions sources, the variation inon time of their emissions is very different. Usually, iIndustrial emissions are normally quite steady-state. However, non- industrial emissions present varyingdifferent behaviorbehaviour depending on their specific activity; therefore, more complex temporal profiles for the non- industrial emissions have been usedconsidered, depending on the simulation domain, which are:that is,

Peninsula domain: The industrial emissions from S1 and S3 CORINAIR sectors are appliedconsidered as steady state annually, weekly and daily. For the remainingrest of emissions (sectors), no annual or weekly variations were usedconsidered, but a daily profile with daytime emissions three times higher than night-time emissions wasis applied.

Galician regional domain: As the industrial emissions for this domain are available from the EPER-REGADE inventory with SCC EPA classification, temporal profiles following EPA recommendations were applied. These profiles are included in the SMOKE package for processing SCC classified emissions.

4.3.3. Aloft anthropogenic emissions distribution

The height of emissions height is a significant factor in the results of air quality models (Kumar and Russell, 1996), especially for large point sources with stacks. In fact, CAMx model considers separately point sources and surface sources separately, as the model includes a plume rise algorithm to estimate the increment of the emissions height for elevated point sources. However, this algorithm requires from complex datasets that are not always available. Therefore, depending on the precision required in eachevery of the simulation grids, different approaches to enteringinput the emissions height in the model are usedconsidered, as follows.

Peninsula domain: As the EMEP inventory only includes the geographical location of emissions location, their emissions height is assumed fromdepending on the group of emissions sources. Thusen, S1 and S3 industrial sectors emissions from every grid cell are converted to one virtual point source by cell, and every virtual point is considered to be emitted at 200 agl-m (as an average value); the rest of the emissions over every cell are takenconsidered as surface level emissions.

Galician regional domain: The EPER-REGADE inventory includes all the dataset required to define the stacks heights and their plume rises. This dataset is included in the SMOKE package to distinguish surface and elevated sources and to provide the CAMx with the required information for the elevated sources.

4.3.4. Chemical speciation

Emission inventories include a chemical speciation which is usually less detailed than is required by the chemical mechanisms applied in the air quality models (Carter, 1990). Therefore, it is necessary to segregate some groups of emissions in order to obtain an approximation to the actual chemical mechanism speciation. In this work, Carbon Bond IV (CBIV) chemical mechanism included in the CAMx model is applied. Therefore, for an initial emissions speciation to this mechanism, the Spcmat tool from SMOKE is usedapplied.


5. Results

Simulations were performed with CAMx, with two nested grids covering the Iberian Peninsula (27x27 km2 resolution) and Galicia (9x9 km2 resolution), using chemical boundary conditions provided by GOCART and MOZART data. Meteorological inputs were provided by PSUN/NCAR MM5 one-way nesting simulations at 27x27 km2 and 9x9 km2 resolutions, using NCEP re-analysis as initial and boundary conditions.

Figure 5.

Relative differences (%) in nocturnal NO2 (left) and O3 (right) glc over Galicia between 2001 emissions inventory and 2010 emissions projection simulations at the central date of the episodes under study. From above to below: 18/July/2002 – 04 UTC, 21/March/2003 – 03 UTC, and 17/September/2003 – 03 UTC, respectively.

Figure 6.

Relative differences (%) in diurnal NO2 (left) and O3 glc (right) over Galicia between 2001 emissions inventory and 2010 emissions projection simulations at the central date of the episodes under study. From above to below: 18/July/2002 – 16 UTC, 21/March/2003 – 14 UTC, and 17/September/2003 – 14 UTC, respectively.

Results for the three different reference episodes (figure 5) show that changes in the industrial emissions will produce a significant reduction in the nocturnal ozone glc around the new combined cycles (As Pontes and Sabón) and their surrounding areas sporadically affected by their plumes; this reduction is not observed during the diurnal period (figure 6). At least for episodes 1 and 2, nocturnal reduction is due to the significant increment of NOx glc around both sources (because their stacks are lower than the old coal fired power plant stacks). IOn episode 3, the ozone reduction is more sparse, even in an opposite direction (Easteast) to the NOx industrial plumes; this maycan be due to lower winds and higher dispersion during this episode.

No significant changes (figure 6) are observed in the diurnal ozone glc; only a small reduction over the local areas affected by the plumes from the aloft sources (As Pontes and Meirama), as the combined effect of the reduction of NOx emissions aloft (reducing diurnal ozone production) and the increment of NOx surface emissions (increasing diurnal ozone production and consuming nocturnal ozone).

Figure 7.

Relative differences (%) in nocturnal O3 concentration over the maximum glc location along the episode 9-22 September 2003, between 2001 emissions inventory and 2010 emissions projection simulations.

However, ozone glc reduction is accompanied byto an increment in the aloft O3 levels; as an example, figure 7 shows an increment inof aloft O3 concentration and, simultaneously, a decrease in O3 concentration at lower levels (including ground level). At the same time, there is no significant changes in nocturnal ozone glc in the rest of the region, even though a significant reduction of NOx emissions aloft is expected from the old high stack coal fired power plants (see table 2); this is because aloft emissions are dispersed above the nocturnal stable layer.

Changes in air quality are more significant in NO2 glc: there is an increment of NO2 glc around Sabón and As Pontes sources (figures 5 and 7), due to the contribution of surface emissions around these areas.


6. Conclusions

From the comparison of 2001 emissions inventory and 2010 emissions projection at Galician region, simulations results from the CAMx model show a change ion ozone distribution. Surprisingly,, remarkably decreasing its concentration downwind from the combined cycles is decreased because of their lower stacks (with respect to the old coal and fuel oil power plants), and slightly decreaseding it in remote areas. This effect is especially significant during nocturnal periods.

On the other hand, simulation results show an increment of O3 aloft concentration, which can be transported to neighbouring regions and increase their glc. This effect is not usually taken into accountconsidered in the current impact assessment, which is based onin either estimated or measured pollutants glc close to the sources. Therefore, although the strategy of emissions reduction in this region can achieve better results on aat local scale, it could leaddrive to an increase in the contribution of this region to the O3 levels in the vicinity.

At the same time, an increment in the NOx ground level concentration is expected both due to the rise ining of the surface NOx emissions and the lower stack height of the new industrial sources (combined cycles). Therefore, although ozone glc achieves local reductions along the NOx industrial plumes, the increment of NOx levels can leaddrive to worse air quality in other areas not affected by these industrial plumes (i.e. urban and rural areas far away from these sources). However, simulations results show no significant effect due to NOx in the glc values.



This work has been financially supported by the Spanish Research & Development Programme, Ministry of Science and Technology, under project CTQ15481-PPQ, and Endesa Generación S.A.. Research grants of the “María Barbeito” Programme (Xunta de Galicia) to Angel Rodríguez and María Dios, are acknowledged.

Hourly ground level concentration measurements were provided by the Portuguese Ministry of Environment, the Spanish Ministry of Environment, the Galician Regional Departament of Environment, the Castilla-León Regional Department of Environment, and Endesa Generación, S.A..


  1. 1. Beck J. P. Krzyzanowski M. Koffi B. 1998 Tropospheric Ozone in the European Union- The consolidated report, Topic Report nº 8 1998, European Environment Agency.
  2. 2. Carter W. P. L. 1990 A detailed mechanism for the gas-phase atmospheric reactions of organic compounds, Atmospheric Environment, 24, 481 518 .
  3. 3. Casares J. J. Rodríguez R. Maceira P. Souto J. A. Ramos S. Costoya M. A. Sáez A. Vellón J. M. 2005 Inventario, Análisis y Proyección de las Emisiones Atmosféricas Industriales de Galicia, Ed. Servizo de Publicacións e Intercambio Científico, Universidade de Santiago de Compostela, Santiago de Compostela.
  4. 4. Castell N. Mantilla M. Fernández F. López E. 2004 Distribución espacial y temporal de los episodios de contaminación por ozono y su relación con las situaciones sinópticas en la vertiente mediterránea, Proyectos y métodos actuales en Climatología, Asociación Española de Climatología, 2004, Serie B, 1, 600 606 .
  5. 5. Chang J. S. Brost R. A. Isakson I. S. A. Madronich S. Middleton P. Stockwell W. R. Walcek C. J. 1987 A three-dimensional Eulerian acid deposition model: physical concepts and formulation, Journal of Geophysical Research, 92, 14651 14700 .
  6. 6. Environ International. Co 2010 CAMx: Comprehensive Air Quality Model with Extensions- User’s Guide. Novato (Ca), USA.
  7. 7. Font-Tullot I. 1983 Climatología de España y Portugal, Instituto Nacional de Meteorología, Madrid.
  8. 8. Geron C. Guenther A. Sharkey T. Arnts R. R. 2000 Temporal variability in basal isoprene emission factor, Tree Physiology, 20(12), 799 805 .
  9. 9. Grell G. A. Dudhia J. Stauffer D. R.,199 description A. of the. Fifth-Generation Penn. State . N. C. A. R. Mesoscale Model. . M. 5 Technical Report NCAR/TN-398+STR, NCAR, Boulder (Co), USA.
  10. 10. Guenther A. Hewitt C. N. Erickson D. Fall R. Geron C. Graedel T. Harley P. Klinger . L. Lerdau M. Mckay W. A. Pierce T. Scholes B. Steinbrecher R. Tallamraju R. Taylor J. Zimmerman P. A. 1995 Global-Model of Natural Volatile Organic-Compound Emissions, Journal of Geophysical Research, 100(D5), 8873 8892 .
  11. 11. Kumar N. Russell A. G. 1996 Development of a computationally efficient, reactive subgrid-scale plume model and the inpact impact in the northeastern United States using increasing levels of chemical detail, Journal of Geophysical Research, 101, 16737 16744 .
  12. 12. Makar P. Fuentes J. Wang D. Staebler R. Wiebe 1999 1999: Chemical processing of biogenic hydrocarbons within and above a temperate deciduous forest, Journal of Geophysical Research, 104 D3), pp. 3581- 3603.
  13. 13. Seinfeld J. H. Pandis S. N. 1998 Atmospheric Chemistry and Physics, From Air Pollution to Climate Change, John Wiley and Sons, Inc., New York.
  14. 14. Song J. et al. 2010 Ozone response to emission changes: a modeling study during the MCMA-2006/MILAGRO Campaign, Atmospheric Chemistry and Phys.ics, 10, 3827 3846
  15. 15. Sparse Matrix Operator Kernel Emissions (SMOKE), 2006 SMOKE 2 3 User’s Manual, Carolina Environmental Program, University of North Carolina at Chapel Hill
  16. 16. The Institute for the Environment, 2003 SMOKE 2 0 User Manual. CEP (Carolina Environmental Program). The University of North Carolina at Chapel Hill.

Written By

Ángel Rodríguez,1Santiago Saavedra, María Dios, Carmen Torres, José A. Souto, Juan Casares, Belén Soto and José L. Bermúdez

Submitted: October 21st, 2010 Published: August 17th, 2011