On-board emission measurement of high-loaded light-duty vehicles in Algeria.(TECHNICAL PAPER)

ABSTRACT

A sample of eight private gasoline and diesel conventional light-duty vehicles (LDVs) in use with various ages, carrying a load of 460 kg, were tested on a representative trip in the traffic flow of the city of Blida to obtain emission factors representing the actual use conditions of Algerian LDVs. The gas sampling system (mini-constant volume sampling) as well as the analyzers are carried on-board the vehicle. Around 55 tests were conducted during 3 months covering more than 480 km under various real driving conditions. The mean speed downtown is about 16.1 km/hr with a rather low acceleration, an average of 0.60 m/sec (2). For each test, kinematics are recorded as well as the analysis of the four emitted pollutants carbon dioxide, carbon monoxide, oxides of nitrogen, and total hydrocarbons. Emission factors were evaluated according to speed for each category of gasoline and diesel engines. The influence of some parameters such as cold/hot start, age of vehicle and its state of maintenance are discussed. Results are compared with the European database ARTEMIS for comparable vehicles. These measurements contribute to the development of unit emission of the vehicles used in Algeria, which are necessary for the calculation of emission inventory of pollutants and greenhouse gases from the road transportation sector. The unit emissions constitute a tool of decision-making aid regarding the conception of new regulations of vehicle control and inspection in Algeria and even in similar developing countries.

INTRODUCTION

All the Maghreb countries are sensitive to the air quality deterioration in urban cities and the side effects on human health and the environment due to road traffic emissions. (1-3) As a result, many stations of air pollution measurement and control have been installed in Algeria, Tunisia, and Morocco. (4-6) Vehicle pollutant emissions constitute not only a problem of air quality in big cities such as Algiers, (7-10) but also a source of a constant growing greenhouse gases (GHGs), particularly in developing countries. (11-15) Thus, unit emission factors of each vehicle category become crucial for all new policies of road transportation, traffic management, and any use of air quality models.

The default emission factor of the various international databases like IPCC (www.ipcc-nggip.iges.or.jp/EFDB), CORINAIR (www.air.sk/projects/corinair/e_reporter.html), COPERT (lat.eng.auth.gr/copert), ARTEMIS (www.trl.co.uk/artemis, www.inrets.fr/ur/lte/publi-autresactions/fichesresultats/cadrefichesresuvert.html), and MOBILE (www.epa.gov/otaq) are not realistic for developing countries and are the source of over- or underestimations of pollutant emissions. (13,16) Emission factors coming from a given country could be representative on a regional scale when there are strong socioeconomic, environmental, and car fleet similarities as it is for the three countries of the Maghreb: Algeria, Morocco, and Tunisia. The development of specific emission factors on the basis of national data is encouraged by the Intergovernmental Panel on Climate Change (IPCC, 2000) to reduce uncertainties on the GHG inventory to comply with the United Nations Framework Convention on Climate Change (UNFCCC). After the Kyoto protocol entered into force in February 2005, potential GHG emission reduction is being sought where its estimation is based on the calculation of a baseline using emission factors. The standardized methodology used to evaluate the vehicle emission factors is based on constant volume sampling (CVS) using the chassis dynamometer for measurement in a laboratory (18) by reproducing driving cycles legislated or representative of the real situations. On-road remote sensing (19,20) is also carried out with vehicle identification by their license plate and access to central vehicle databases. This technique allows the measurement of a large number of vehicles but cannot evaluate the vehicle load rate with precision. Both of these two and similar methods are not adapted to many developing countries because of lack of equipped laboratories with CVS and chassis dynamometers, which are expensive, and a database for vehicle identification is not easily accessible.

However, several alternative measurement techniques with on-board instrumentation are developed for scientific purposes. (21-28) Results obtained using onboard measurements in real traffic showed emission differences between realistic emission factors compared with CVSs in Europe and the United States. (29,30) It is possible with this methodology to measure vehicle emissions in sites whose specific conditions are not easily reproducible in a laboratory. It also reduces the uncertainties caused by reproduction of real conditions in a laboratory like bench calibration, engine cooling, state of the roadway, real environment of the vehicle, ambient temperature, sunning, and especially driving behavior. However, the method must be rigorous for a representative driving pattern, which consequently requires a lot of experimentation time. Several studies showed the reliability of on-board measurement compared with tests in laboratories. (16,31) The mini-CVS has been developed by the Warren Spring laboratory in the United Kingdom (32) and its performance evaluated and validated against a full CVS system by three laboratories (Warren Spring Laboratory [WSL], Union Technique de l'Automobile du Motocycle et du Cycle [UTAC], and Institut National de Recherche sur les Transports et leur Securite) using ECE-15 hot cycle. (16,33,34) The results showed an average deviation between mini-CVS and CVS of -3.4% for carbon monoxide (CO), -2% for oxides of nitrogen (N[O.sub.x]), -9.7% for total hydrocarbons (THCs), and a variation for all pollutants and all tests range between -15 and +1%. The mini-CVS underestimates emissions by an average of -5% compared with CVS. These techniques are well adapted for developing countries and have a double advantage of being relatively inexpensive and not requiring a prerequisite driving cycle development, which takes a long time to obtain.

Several works were conducted on light-duty vehicles (LDVs) to characterize their emissions and the influence of many parameters such as load rate and age. (35-42) Zallinger and Hausberger (43) studied the effect of the load on personal cars with a maximum loading of 450 kg for an engine capacity between 1.2 and 2.1 L for gasoline and diesel. The effect of load was highlighted only for the diesel vehicles and only for N[O.sub.x] and particles. The emission variation of the other pollutants due to the load is on the same order as their standard deviation. Corvalan and Vargas (44) showed that degradation of pollutant emission is related to vehicle global mileage and speed, which depend on local traffic conditions. THC emission measurements of three-way catalyst equipped vehicles in Santiago (Chile) depict degradation factors 2-2.8 times higher than models such as COPERT III and EPA AP-42. This result points out the importance of local measurements in the development of national emission factors. ARTEMIS is a European database that gathers data from European vehicle emission measurements from various laboratories, including the majority of the LDV makes and models used in Algeria. However, Algerian vehicles do not comply with the European standards and are not driven in similar conditions. This permits access to carry out comparison studies on vehicles of the same origin but subjected to different use and environmental factors.

This work aims to measure the emission factors of high loaded LDV with a load rate more than 50%, which is largely reached in Algeria and in the Maghreb region. (45) It contributes to knowledge on vehicle pollutant emissions and gives decision makers realistic data for this category of vehicles often used with a high load rate.

TRAFFIC CHARACTERISTICS

Facing a low supply and high cost of goods transportation in Algeria, individuals and professionals prefer investing in LDVs given their flexibility and compactness. They are often equipped with back-folding seats are used for both private and commercial purposes, making this type of vehicle run a significant mileage with an average load rate of 20-30%, calculated with eq 1 according to surveys carried out.

Load rate (%) = [Weigh[t.sub.loaded vehicle] - Weigh[t.sub.empty vehicle]]/Weigh[t.sub.empty vehicle] * 100 (1)

LDV average annual mileage is 38,000 km for vehicles less than 5 yr old according to our survey carried out in 2001 in Algiers and Blida on a sample of 935 vehicles. The results compared with European data are shown in Figure 1. (46-48) The LDV life is longer in Algeria than in Europe, with an average age of 10 yr and a weak level of maintenance. The Algerian fleet consists of more than 3 million vehicles, the LDVs account for 21%, in which 79% are European make and only 17% are from Asia. According to the 2005 statistics of the Office National de Statistiques (ONS), (49) the mean life of the LDV administrative fleet, which includes all the recorded vehicles on the basis of their license, is 17.8 and 15.4 yr for Blida and Algiers, respectively. A vehicle on-road survey was carried out during 2005 by visual observation of more than 65,000 vehicles during 1 week including weekends in eight selected places in the cities of Blida and Algiers. In Algeria, the year of the first setting of a vehicle is indicated on the license plate, where the identification of its age is made easy. The results show that LDVs account for 11 and 15% of the vehicle flow with an average age of 9.3 and 10.2 yr in the two cities, respectively. This highlights the gap between the ONS administrative fleet and the running one, in which one can have a real image of the fleet composition on the road at a given time.

[FIGURE 1 OMITTED]

METHODOLOGY