On-board emission measurement of high-loaded light-duty vehicles in Algeria.

The mass difference of pollutant emitted by cold and hot-start tests for a given trip with a same average speed represents absolute cold excess emission (Table 4). Rated by the unit hot emission, this absolute emission is expressed in an equivalent distance run with a hot engine. In other words, it is the distance necessary for a hot engine to emit the same mass of pollutant as the cold-start excess emission. This distance is not related to the cold distance, which explains the distance necessary to reach a hot engine. On the other hand, it expresses the weight of cold emission compared with hot emission. This equivalent distance is variable according to the pollutant and the vehicle: the maximum value reached is 83 and 90 km for the cold emission of CO and THC, respectively, for a diesel engine. The C[O.sub.2] emission is equivalent to 6.2 km for a gasoline engine and to 4.5 km for a diesel engine; for CO it is 28.2 km for a gasoline engine and 23.7 km for a diesel engine; for THC it is 22.7 km for a gasoline engine and 27.7 km for a diesel engine; and for the N[O.sub.x] it is -1.7 km for the gasoline and 4.1 km for a diesel engine. A negative distance suggests an underemission. The cold-start emission is thus particularly important for CO and THC, moderately important for C[O.sub.2], and of little importance for N[O.sub.x]. On average for all pollutants, the cold emission represents the equivalent of 13.8 km for the gasoline engine and 15.4 km for the diesel engine, which is thus of comparable relative importance for the two engines.

Fuel Consumption

The average fuel consumption of gasoline and diesel vehicles pseudo-Euro I and pseudo-Euro II are reported in Table 2. The consumption of gasoline vehicles remains lower than that of ARTEMIS, which is 68.9 g/km. On the other hand, the consumption of the diesel vehicles in the pseudo-Euro I group is comparable with ARTEMIS, which is 54.5 g/km whereas the pseudo-Euro II group has a higher consumption. These consumption variations can be due to weak acceleration but also to the repair of the pseudo-Euro I vehicles. Fuel consumption comparison for cold and hot-starts reported in Table 4 shows an average excess of consumption of 67 and 49% for gasoline and diesel engines, respectively. For comparison purposes, an extracted subsample from the ARTEMIS database (51,60,61) of eight LDVs vehicles from category N1-I that have been tested at both cold and hot-starts were used: three gasoline pre-Euro vehicles, one diesel pre-Euro vehicle, three diesel Euro I vehicles, and one diesel Euro II vehicle. Only data from cold starts with similar parameters to our sample such as load rate, ambient temperature, and speed were used to calculate the percentage of excess of fuel consumption at cold start to hot start during tests conducted on different urban cycles in Europe. The excess obtained varied in the range of 25.7-31.5% for an average speed of 18 km/hr and an ambient temperature of 18.4 [degrees]C and a load rate of 16% for all technologies of diesel vehicles. The range of excess for pre-Euro gasoline vehicles was 27.5-29.4% for an average speed of 41.5 km/hr, a load rate of 10.4%, and an ambient temperature of 18 [degrees]C. Our results compared with ARTEMIS data show a fuel consumption excess 20.4% higher for diesel engines, which could be attributed to the difference in load rate of 33%. In the case of gasoline vehicles, the excess is approximately 38.2% higher for a difference in load rate of 49.6%. According to these results, it seems that high load rate has an important influence on fuel consumption at cold start for diesel and gasoline LDVs. Further investigation should be conducted to explore the influence of the driving conditions, the ambient temperature, and high load on cold-start emissions and fuel consumption.

DISCUSSION AND CONCLUSIONS

For the first time in Africa, emission factor measurements were carried out on high loaded LDVs with diesel and gasoline engines. Similar to many developing countries, one of five vehicles in Algeria is a LDV, largely used for transportation of goods and people with an average annual mileage of 32,000 km (higher than the mileage in European countries) and an average lifespan of 10 yr. In addition, the number of LDVs is increasing. Although the results are representative of a local situation, they could also be valuable in other developing countries where similar conditions are occurring. The results should lead to more investigations in Africa on vehicle emissions in the future.

Because on-board emission measurement is a suitable technique for developing countries to conduct measurements at a low cost with affordable instrumentation, we used mini-CVS equipment. The conclusions are based on a small test fleet of eight vehicles conducted in the city of Blida, Algeria, but remain comparable with samples tested within different European studies. In addition, each vehicle was tested five times to reduce the uncertainty per vehicle. Results are representative of real in-use LDVs with high mileage in Algeria.

Nevertheless, some interpretations remain limited, in particular the comparison with the European ARTEMIS database: the average vehicle mileage of our sample was up to 696,000 and 163,000 km, respectively for gasoline and diesel engines, i.e., much higher than in Europe. In addition, our load rate is higher than in ARTEMIS; and the ARTEMIS model is not validated for load rate over 25% for Euro I diesel LDV category N1-I. The comparison should be more accurate if more data were available. Research on the influence of high load (>50%) on the cold and hot emissions should be fruitful because there are very few related articles published, and the high loads are very common in developing countries for all LDV categories.

Measurements also enable evaluation of cold-start excess of emission and of fuel consumption. In the case of gasoline engines, cold start emissions were 50% higher than hot emissions for C[O.sub.2], 80% for CO and THC, and 44% lower for N[O.sub.x]. They were 43% higher for C[O.sub.2], 78% higher for both CO and THC, and 40% higher for N[O.sub.x] for diesel engines. Fuel consumption excess at cold start was 67 and 49% higher for gasoline and diesel engines, respectively. Cold-start emissions were higher for our high-loaded vehicles than for the partially loaded vehicles from the ARTEMIS model.

Most emission inventories for pollutants or GHGs and air quality models for the African road transportation sector are based on the emission unit factors of European vehicles as input. In the case of many countries such as Algeria, the absence of emission standards for vehicles causes the age of the vehicles to be the main parameter used to find the corresponding European standards. The comparison of our data with European data enables evaluation of the European standard to which emissions of each tested vehicle correspond; vehicles of the same age show a deviation of zero standards for vehicles before 2001 and to one standard for those after 2000. Therefore, the major finding out of this work is that vehicles used in Algeria should be characterized by their true corresponding standard instead of referring simply to the year of the vehicle, which does not give any information on the engine technology and leads to high uncertainty. This issue could be addressed if Euro standards are adopted in the future in all the countries of the Maghreb region so that the governments can adopt a strategy of control and reduction of road vehicle emissions. In addition to this misallocation of standards, the emission factors do not consider local parameters of vehicle use and environment.

New vehicles are usually automatically associated with low emission levels without referring to any technology or standard. They are totally managed by electronic processes on the basis of high technology rarely available in developing countries. In such a case, the vehicle maintenance for emission reduction is hardly applicable. Facing this situation, it is crucial for these countries to adapt their regulations and to develop capacity building projects to strengthen their ability to maintain such vehicles.

Future research tasks should use a larger vehicle sample more representative of the national and even regional state of the fleet and local traffic conditions including all LDV categories (N1-II and N1-III) to obtain more reliable results. A quantitative evaluation of the traffic and of the driving behavior could improve selection of the representative trip. Emissions from the transportation sector are growing fast in developing countries. Appropriate emission unit factors are crucial for an accurate emission estimation, which constitutes the basis for policy makers to address issues regarding air pollution from traffic and to adopt efficient environmental regulations.

REFERENCES

1. Azri, Ch.; Maalej, Ah.; Tlili, A.; Medhioub, Kh. Caracterisation du niveau de pollution atmospherique dans la ville de Sfax (Tunisie): Influence des sources et des facteurs meteorologiques [Physico-Chemical Characterization of the Urban Aerosol in the City of Sfax (Tunisia)]; Techniques-Sciences-Methodes, 2002, 1, 78-92.

2. Nejjari, C.; Filleul, L.; Zidouni, N.; Laid, Y.; Atek, M.; El Meziane, A.; Tessier, J.F. Air Pollution: a New Respiratory Risk for Cities in Low-Income Countries; Int. J. Tuberc. Lung Dis. 2003, 7, 223-231.

3. Atek, M.; Laid, Y.; Oudjehane, R.; Zidouni, N.; Filleul, L.; Tessier, J.F.; Boughedaoui, M.; Baough, L. Effects Sanitaires de la Pollution Atmospherique Urbaine en Afrique du Nord: Cas de la Ville d'Alger [Health Effects of Urban Air Pollution in North Africa: the Case of Algiers]; Int. J. Tuberc. Lung Dis. 2004, 8, S20.