Computer Science School
Technical University of Madrid
|
|
Computer Science School Technical University of Madrid |
The ANA system is applied to Madrid domain (80 x 100 km). The emission model is called EMIMA 3.0 and it is implemented in C. This model follows the CORINE methodology. Pollutants included are SO2, NOx, and anthropogenic and biological volatile organic compounds and particles. This inventory provides a emission module into ANA model. Sources are split into line, point and area types, and considers emissions from small industries, domestic comsumptions, biogenic sources and surface traffic. Line emission sources include six roads and two city Rings (M-30 and M-40). Six types of vehicles and five types of fuel are considered. Point emissions consider those from big industries with more than 40 TM SO2 per year. Emissions depends on the fuel consumption and process type.
The time resolution is one hour while the grid spacing is 250 x 250 m2. The model is flexible and allows for different grid spacing to interface with the meteorological, transport and deposition modules. The model uses the "collective way" to stimate the emissions. The approach uses estimated emissions factors that represent emission rates produces as a result of one activity. The method allows the grouping of different sources such as vehicles, heaters and biogenic emissions. The point industrial sources are individually recorded (INYPSA).
These activities involve heat and hot water production. The inventory accounts for combustion sources for heating purposes by collective inventorying on the basis of population density and housing. Heating accounts for the 85% of fuel usage, hot water accounting for the rest. The emissions of substance 's' in a grid cell from domestic fuel consumption is calculated as follows:
where Gi is the consumption of fuel per unit of time in the area, and feis s the emission factor for the fuel and for the substance s.
Exhaust emissions from road traffic can be computed by the following expression:
E = n r c f
where E is the amount of the emitted pollutant, n is the number of travelling vehicles of each type per hour, r is the mean distance travelled per vehicle in km., c is the mean consumption per vehicle per km., and f is the emission factor for each pollutant in g/l. A yearly average amount is used to evaluate emissions per day and then daily emission is modelled by using hour factors from the Traffic Department of Madrid City Municipality. Vehicle type density for different land use types (urban and suburban) was obtained from INYPSA. The number of vehicles per inhabitant is 0.3 and holiday traffic is 65% from work day traffic. The following expression was used to evaluate the number of vehicles per cell:
n = (inhab/cell) (0.3) (0.65)(fcnv/hour)(NVehic/type,landuse)
Biogenic emissions have been also included in the model, because the importance of these emissions compared to those from anthropogenic sources. Biogenic emissions have been restricted to alfa-pinene, monoterpene, and the isoprene, the most active volatile organic compounds with biogenic origin. The Lamb algorithm is the base of our modelling approach. Isoprene emissions have a strong dependence on the sun light and leaf temperature. When leaf temperature have increased from 25 C to 30 C, we have an increase on isoprene emissions up to a 70%. Doubling the photosynthetically active radiation (PAR), also doubles isoprene emissions. South European ecosystems, with their higher solar radiation and ambient temperature, enhance extraordinary the VOC's budget in urban areas. The model in the present state runs with average hourly temperatures from the Meteorological Institute. A further improvement will link these emission temperatures to the meteorological module.
The isoprene emissions are calculated as follows:
where
ET(30,400) = 20.27
Ef is a specified emission factor at 30 C and 400*10-6 Em-2s-1, which depends on the corresponding class of vegetation, ET(T, PAR) is the isoprene emission rate obtained from Tingley's curves at the specified conditions. Coefficients are those shown in the next table.
| PAR
|
a
|
b
|
c
|
d
|
| 800 | 1.200 | 0.400 | 28.30 | 0.800 |
| 400 | 0.916 | 0.239 | 29.93 | 0.845 |
| 200 | 0.615 | 0.696 | 32.79 | 0.843 |
| 100 | 0.437 | 0.312 | 31.75 | 0.406 |
In the case of alfa-pinene and for the sum of monoterpenes, emission rate does not depend on the sun light (emission occurs all day), the only dependence is on temperature. Following the satellite classification, 22% of the domain emits isoprene and 4% emits monoterpenes. These percentages include "mixed" landuse formed by 95% caduceus and 5% perenne landuse.
 
|
|
|
