The methodology for conducting this research is focused on obtaining pollutant emissions produced by the vehicular operation of a roadway. Currently, different software programs apply these models to calculate vehicle emissions. Some of these were tested for this research. A description of them is provided below:
3.1. Input Data
Based on the above, the minimum input variables necessary to be introduced into the calculation tool are as follows:
Identification of emission types.
Traffic volume on the road section, refers to the annual traffic volume in each flow period, i.e., vehicles per year.
Vehicle speeds, operational speed of vehicles when traveling on the road.
Fuel consumption, pertaining to the instantaneous fuel consumption of each vehicle type, in each traffic intensity period.
Vehicle lifespan and model parameters.
Characteristics of the road section, such as section length, slopes, and road surface.
Maximum and minimum temperatures of the study area.
Through a literature review, it was possible to identify the types of pollutant emissions generated by vehicles in road operations. It is important to note that the case study is located in Mexico, a developing country. Consequently, there are challenges in obtaining data for the necessary variables to create a traditional inventory. In this context, to conduct this research, aside from exploring various databases, fieldwork was essential to gather and verify statistical data about vehicles using the road and existing geometric data. The consulted databases included traffic data from the Ministry of Infrastructure, Communications, and Transportation (Secretaria de Infraestructura, Comunicaciones y Transportes, SICT, Mexico City, Mexico) [
77] and data from the Public Trust for Road Funds and Investment Administration of the Centinela-Rumorosa Highway Section (Fideicomiso Público de Administración de Fondos e Inversión del Tramo Carretero Centinela-Rumorosa, FIARUM, Mexicali, Mexico) [
78], as well as environmental information data from the Ministry of Environment and Natural Resources (SEMARNAT, Mexico City, Mexico) [
79].
Traffic composition is defined as the proportions of different types of vehicles using the road [
80]. To obtain the annual average daily traffic and vehicle characterization, the configuration scheme of the main vehicles circulating on the national network published in the Mexican official standard by the SICT was used [
81] (
Table 1).
The first inputted data pertain to traffic and vehicle composition. This information was taken from the year 2018 as a complete count was available for this period and to avoid biased data due to the reduction in traffic in subsequent years caused by the COVID-19 pandemic. Additionally, it is important to mention that these data correspond to the typical highway traffic. On the other hand, it should be noted that the case study has specific characteristics along its route. Therefore, the segment was divided into three subsections. The first segment, from kilometers 0 to 18, is located in an urban area. The second subsection is situated around the Laguna Salada, spanning kilometers 18 to 42, characterized by a non-urban environment with a completely flat and straight terrain. The third subsection covers the range from kilometer 42 to 64, as it traverses a mountainous area with a series of curves and significant slopes. Furthermore, the entire route has two separate lanes of traffic. In this context,
Table 2 presents the average annual daily traffic for both lanes, categorized within the three subsections.
In addition to the above,
Table 3 shows the vehicular classification of the road section; it is noteworthy that the highest traffic is recorded in the urban area. Vehicles of type A2 (light vehicles) account for 75% on uphill and 73% on downhill.
In the same way,
Table 4 displays the vehicle classification for the section located at Laguna Salada from kilometer 18 to 42. It is worth noting that this is no longer an urban area, but it does have recreational areas. The most prevalent vehicle with the highest traffic is the A2 light vehicle, accounting for 72% on the uphill and 70% on the downhill.
Table 5 presents the vehicle classification for the section located in the mountainous area from kilometers 42 to 64. This section represents lower traffic. Nevertheless, it accounts for a daily circulation of four thousand vehicles, with 65% being light vehicles on the uphill and 68% on the downhill.
With HDM-4 software, in addition to traffic and its characterization, vehicle information is inputted, including operating speeds, types of fuel used, and specific vehicle characteristics. As established previously, the emission models in HDM-4 are based on vehicle-specific fuel consumption; additionally, fuel consumption is dependent on vehicle characteristics, speed, and road conditions [
82]. A travelling vehicle is associated with two types of speeds, namely the vehicle free speed and operating speed; free speed can be achieved on an uncongested road, and this is predicted using mechanistic models, and operating speed includes uphill speed, downhill speed, and the average round trip speed [
83]. The equations for estimating this are described in [
84].
The characterization of the vehicles and consumption and operating expenses were obtained from [
85]. In order to calibrate the data related to the speed of the vehicles, a point speed study on the highway was carried out considering the methodology described in [
86]. With this methodology, it is possible to obtain a representative sample of the vehicles that travel at a given point, as well as its speed characteristics under prevailing conditions of traffic and weather conditions at the time of carrying out the study, allowing speeds by user groups to be obtained. It is based on obtaining the arithmetic mean of the point speeds of the vehicles passing through a specific point. That is, a distribution of point velocities is obtained and, with it, the observed and accumulated frequency distribution. Finally, the standard deviation, the standard error of the means, and a reliability level of 95.5% are taken into account. To validate the results of the point speed study, data from the Ministry of Communications and Transportation of Mexico were used [
87]. The speeds considered in the present study are presented in
Table 6.
As detailed previously, road conditions affect the generation of polluting emissions. since they are closely linked to the operating speed and therefore to fuel consumption, mostly geometric features, such as distances, widths, and slopes, as well as climatic conditions of the area and road pavement deterioration in form of the roughness index (IRI).
For the geometric conditions, the conventional criterion is based on the use of project standards, which dimension the components of the road, separately and together, to achieve an adequate balance between the desirable attributes. It is worth mentioning that the downhill section was a two-lane highway before the uphill section was built since 2000, so the downhill section design has inconsistencies and is inappropriate for use in the present days. Meanwhile, the rest of the highway is made up of flat and straight sections. The case study has 81 horizontal curves on the uphill and 82 on the downhill. For this research, the data obtained by [
17] of the horizontal and vertical alignments were used.
The IRI refers to an indicator that directly represents the functional condition of a pavement, and at the same time, it constitutes a complementary indicator to divide the road network according to its structural capacity. For this study, a measurement of performance indicators was carried out, including the IRI. The data collected, as well as deterioration information, are classified according to the criteria established by ASHTOO to measure the pavement surface condition. For IRI Roughness (m/km), New = 0–2, Good = 2–4, Fair =. 4–6, Poor = 6–8, and Bad ≥8. Highway pavement conditions are shown in
Figure 4. It is important to mention that the analysis of the conditions was carried out based on a calculation per km. However, it is averaged for a better appreciation of the conditions of each section of the highway.
One aspect that characterizes the highway is the variability of climatic conditions and environmental conditions (
Figure 5). The “Eastern” part of the road is located in the municipality of Mexicali, and it has a very dry desert climate (BWh). The maximum temperatures recorded exceed 54 degrees Celsius, while the minimum temperatures drop −7 degrees Celsius. In the central and flat area corresponding to the Laguna Salada section, a very dry semi-warm climate occurs (BWh), while in the mountainous area of the section that is located within the municipality of Tecate in the town of La Rumorosa, there is a cold desert climate (BWks) where they reach minimum temperatures of −10 degrees Celsius in winter and temperatures maximums of up to 40.3 degrees Celsius in summer. Additionally, in said area, in the rainy season, there are considerable snowfalls. For this study, data from the National Institute of Statistics and Geography of Mexico were used, as well as information from the climatological stations of the National Water Commission [
79,
88].
Criterion contaminant data used in this investigation were obtained from historical records and were available at air quality monitoring stations. Databases originate from stations in the United States of America and Mexico. The World Air Quality Project [
89] has stations near the case study for which records date from 2008 to date. Stations closest to the case study are those of Otay Mesa Donovan Correctional Facility and Calexico-Ethel Street. On the other hand, in Mexico, the air quality monitoring stations belong to the National Air Quality Information System (SINAICA) [
90]; of the three nearby stations of this network, the so-called COBACH is on the periphery the urban city and shows data closer to the traffic conditions and interurban emissions that are perceived on road sections, such as the highway. There are no air quality monitoring stations directly in the case study, so the data used correspond to these stations. The reliability between the data from the air quality monitoring stations was contrasted, finding that the variation in the records of criterion pollutants between stations does not exceed 5%.
3.2. Data Processing
Subsequently, once the input data are obtained, the model proposed by [
91] is applied. This model predicts vehicle exhaust emissions based on the fuel consumption and speed. Similarly, fuel consumption is influenced by vehicle speed, which in turn depends on road characteristics and the vehicle itself. This approach allows for the analysis of changes in emission levels as a result of implementing various road maintenance and improvement strategies or when significant changes occur in the vehicle fleet on the road network [
82]. On the other hand, the coefficients and constants mentioned in the formulas are derived from various studies under controlled conditions, which have enabled the creation of tables with recommended values for use in the model [
1].
During the data processing, an adjustment of the Calibration Factors of Equations (1)–(7) was carried out, which by default in HDM-4 is equal to 1. However, it is important to mention that this calibration was performed only on the calibration factor k0, which responds to the calibration factor for a given emission. This adjustment consists of a relationship between the criterion pollutant value reported by the air quality monitoring station and the maximum value allowed by the reference standard, and in the present investigation, these were the Official Mexican Standards: NOM-020-SSA1-2014 [
92], NOM-022-SSA1-2019 [
93], NOM-021-SSA1-2021 [
94], NOM-023-SSA1-2021 [
95], NOM-025-SSA1-2021 [
96], NOM-026-SSA1-2021 [
97]. An example is the case of the calibration factor for NOM-021-SSA1-2021 is presented, which establishes a maximum hourly concentration of carbon monoxide (CO) of 26 ppm in the standard and 21 ppm, which was recorded at the monitoring stations, and the ratio of 21 ppm/26 ppm is equal to 0.8076; this value replaces the calibration factor k0 = 1.0 that was predefined in HDM4, and in this way, the information on criterion contaminants from the monitoring stations to be considered to calibrate parameters in the equations is indicated in HDM4.
This study was conducted with level 1 and 2 calibrations. This includes a desk study based on data collected from secondary sources, such as publications from government agencies and reports from previous studies, and also direct measurements of local conditions to verify and adjust the prediction capacity of the model, mainly in determining local conditions, such as traffic characterization, speeds, geometry, pavement conditions, weather, and air quality monitoring. Level 3 is outside the scope of this study as it involves significant field studies and historic real-time data, so they were not possible at this stage.
Next, the equations applied for calculating the various emissions produced by road operations are presented.
EHC: Hydrocarbon Emissions (g/veh-km)
IFC: Instantaneous Fuel Consumption (mL/s)
LIFE: Vehicle Lifetime (years)
SPEED: Vehicle Speed (km/h)
A0 a A2: Model Parameters
Kehc0: Calibration Factor (predefined = 1.0)
Kehc1: Calibration Factor (predefined = 1.0)
Carbon Monoxide (CO)
where:
ECO: Carbon Monoxide Emissions (g/veh-km)
IFC: Instantaneous Fuel Consumption (mL/s)
LIFE: Vehicle Lifetime (years)
SPEED: Vehicle Speed (km/h)
A0, A1, A2: Model Parameters
Kec0: Calibration Factor (predefined = 1.0)
Kec1: Calibration Factor (predefined = 1.0)
Nitrogen Oxide (NOx)
where:
ENOX: Nitrogen Oxide Emissions (g/veh-km)
IFC: Instantaneous Fuel Consumption (mL/s)
LIFE: Vehicle Lifetime (years)
SPEED: Vehicle Speed (km/h)
A0 a A2: Model Parameters
Kenox0: Calibration Factor (predefined = 1.0)
Kenox1: Calibration Factor (predefined = 1.0)
ESO2: Sulfur Dioxide Emissions (g/veh-km)
IFC: Instantaneous Fuel Consumption (mL/s)
SPEED: Vehicle Speed (km/h)
A0, A1: Model Parameters
Keso0: Calibration Factor (predefined = 1.0)
Carbon Dioxide (CO
2)
where:
ECO2: Carbon Dioxide Emissions (g/veh-km)
IFC: Instantaneous Fuel Consumption (mL/s)
SPEED: Vehicle Speed (km/h)
A0: Model Parameters
Keco0: Calibration Factor (predefined = 1.0)
Particulate Matter (PM)
where:
EPM: Particulate Matter Emissions (g/veh-km)
IFC: Instantaneous Fuel Consumption (mL/s)
SPEED: Vehicle Speed (km/h)
A0, A1: Model Parameters
Kepar0: Calibration Factor (predefined = 1.0)
Kepar1: Calibration Factor (predefined = 1.0)
EPB: Lead Emissions (g/veh-km)
A0, A1: Model Parameters
Kepb0: Calibration Factor (predefined = 1.0)
On the other hand, to obtain the results, it is necessary to apply the basic equation used to estimate emissions from motor vehicles, which involves vehicle activity data and an emission factor, where the factor is provided by the aforementioned models [
10].
where:
Ep = Total emissions of pollutant p (Ton)
KRV = Kilometers traveled by the vehicle
FEp = Emission factor of pollutant p
The data used, as well as the data obtained from the calculation tool, are applied to the basic equation. However, because the software provides results per thousand vehicles, the emission generated by each vehicle is calculated using the following equation:
where:
EV = Emissions per vehicle
HDM-4 results = Outputs of SW HDM-4
Finally, with the results obtained from the calculation tool, it is necessary to adjust for the case study. To do this, the average annual daily traffic is obtained, along with its classification and the total distance of the road section under analysis. Thus, the following equation is derived:
where:
TEOC = Total emissions from road operation (Ton)
AADT = Average annual daily traffic
CLV = Vehicle classification
EV = Emissions per vehicle
D = Distance in kilometers