DEVELOPING A MICROSCOPIC TRANSPORTATION EMISSIONS MODEL TO ESTIMATE CARBON DIOXIDE EMISSIONS ON LIMITED ACCESS HIGHWAYS

0 0 0 DEVELOPING A MICROSCOPIC TRANSPORTATION EMISSIONS MODEL TO ESTIMATE CARBON DIOXIDE EMISSIONS ON LIMITED ACCESS HIGHWAYS Hatem Abou-Senna, PhD., P.E. * Research Associate Center for Advanced Transportation Systems Simulations (CATSS) Department of Civil, Environmental and Construction Engineering (CECE) University of Central Florida Orlando, Florida Phone (0) -00 E-mail: Hatem.Abou-Senna@ucf.edu Essam Radwan, PhD., P.E., F. ASCE Professor and Executive Director: Center for Advanced Transportation Systems Simulations (CATSS) Department of Civil, Environmental and Construction Engineering (CECE) University of Central Florida Orlando, Florida -0 Phone (0) - E-mail: Ahmed.Radwan@ucf.edu Submitted for Presentation and Publication at the rd TRB Conference Washington D.C., January 0 Date of Submission: August 0 Revised Submission: November 0 * Corresponding Author 0 Abstract,,0 Text, (*0) Tables, (*0) Figures Total length =,0 words

Abou-Senna/Radwan 0 0 ABSTRACT This paper presents an optimal design approach for developing a microscopic transportation emissions model (Micro-TEM). The main purpose of Micro-TEM is to serve as a surrogate model for predicting carbon dioxide (CO ) transportation emissions on limited access highways in lieu of running micro-simulations using a traffic model and integrating the results in an emissions model to an acceptable degree of accuracy. Key parameters that are traffic-related (volume, truck percentage, speed limits); geometry-related (road grade) and environment-related (temperature) are selected for detailed evaluation. Estimating vehicle emissions based on secondby-second vehicle operation created the opportunity to integrate a microscopic traffic simulation model (VISSIM) with the latest Environmental Protection Agency s (EPA) mobile source emissions model (MOVES) for higher precision and accuracy. VISSIM/MOVES integration software (VIMIS) was developed to facilitate running the multilevel factorial experiment on a test bed prototype of the I- urban limited access highway corridor located in Orlando, Florida. The analysis identified the optimal settings for CO emissions reduction based on two main parameters; traffic volume and speed. Volume correlation with CO emission rates revealed an exponentially decaying function towards a limiting value expressed in the freeway capacity. Moreover, speeds between and 0 mph showed significant emission rate reduction effect while maintaining up to 0% of the freeway s capacity. The results demonstrated that active speed management does have a significant impact on CO emissions provided that detailed and microscopic analysis of vehicle operation of acceleration and deceleration is achieved. This approach can provide environmental decision makers with practical guidelines when deciding on environmental transport policies. Keywords: Optimal Design; Transportation CO Emissions; Limited Access Highways; MOVES; VISSIM; Vehicle Specific Power (VSP);, Micro-TEM

Abou-Senna/Radwan 0 0 0 INTRODUCTION Emissions of greenhouse gases (GHGs), primarily carbon dioxide (CO ), are contributing to global climate change, which is one of the most critical environmental issues facing the world this century. CO from transportation is expected to remain the major source of total U.S. GHG emissions (). The emission factors (EFs) for highway vehicles used in the U.S. GHG inventory are based on laboratory testing of vehicles. Although the controlled testing environment simulates actual driving conditions, the results can only approximate real world vehicle activity and interactions. For some vehicle and control technology types, the testing did not yield statistically significant results within the percent confidence interval, requiring reliance on expert judgment when developing the EFs (). Since % of transportation GHG emissions are in the form of CO (), uncertainty in the CO estimates has a much greater effect on the transportation sector estimates of GHG than uncertainty associated with nitrous oxide (N O), methane (CH ), or hydro fluorocarbons (HFC) emissions. Transportation agencies and researchers have a long history of implementing techniques to calculate transportation-related emissions. Traditional methods for creating emission inventories utilized annual average estimates. One comparison of annual estimates with monthly estimates of vehicular emissions provided similar results, implying that detailed calculations were not necessary for annual emissions inventories (). Travel demand models have been utilized to provide an intermediate level of detail using daily volumes. However, static planning models were found to ignore individual vehicle activity, link capacity and other dynamic variables leading to underestimation of pollutant emissions and significant biases in different traffic conditions (). Currently, more accuracy has been established using microscopic analyses through the reduction of time and distance scales while splitting the network links into sub-links and utilizing second-by-second operations to calculate emissions. Despite all the past studies conducted on GHG and CO emissions, there s still a crucial need to identify and analyze, microscopically, the effect of major transportation-related key parameters on CO emissions. Since the transportation system encompasses different aspects of traffic, geometric, and environmental factors, key parameters that are traffic-related (volume, truck percentage, speed limits); geometry-related (road grade) and environment-related (temperature) were selected for detailed evaluation. An experimental custom design approach was developed to study the effect of each of these parameters on CO emissions. A limited access urban highway corridor in Orlando, Florida was used as the test bed in this paper. The network was modeled using the latest microscopic traffic simulation model VISSIM coupled with EPA s mobile source emissions model, MOVES00a. Detailed traffic operations generated from VISSIM on a second-by-second basis were input into the MOVES model to quantify emission rates for each parameter.

Abou-Senna/Radwan 0 0 0 0 BACKGROUND Several studies have related the increase in CO emissions to the increase in vehicle miles travelled (VMT) (). Other studies considered managing VMT growth through land use decisions (), compact developments () and increasing modal choices (). Barth et al. (0) indicated that CO emissions can be lowered by improving traffic operations and reducing traffic congestion. Consideration was also given to the characteristic variation of emissions at different vehicle speeds. Marsden et al. () demonstrated an approach in microscopic traffic exhaust emissions modeling for carbon monoxide (CO) emissions based on vehicle speed and classification of vehicle acceleration, deceleration, cruising and idle inputs. They showed that vehicle-exhaust emissions depend on the fuel to air ratio. Sturm et al. () described three approaches of compiling emission inventories based on actual driving behavior, specific streets and vehicle-miles traveled. Emissions also vary with respect to drivers behavior. Aggressive driving increases emissions compared to normal driving where % of total driving time in aggressive mode contributed about 0% of total emissions (). In an effort by Panis et al., () to determine PM, NO x and CO emission reductions from speed management policies in Europe, they concluded that emissions do not rise or fall dramatically due to specific speed reduction schemes but lower maximum speed for trucks consistently resulted in lower emissions of CO and lower fuel consumption. In an earlier attempt by Panis et al., () to model instantaneous traffic emissions, they concluded that the frequent acceleration and deceleration movements in the network significantly reduced the benefit of changing the overall speed. The conclusion from that study was that active speed management had no significant impact on total pollutant emissions. Boriboonsomsin and Barth () showed that road grade have significant effects on CO emissions and fuel economy of light-duty vehicles both at the roadway link level and at the route level. Papson et al. () integrated SYNCHRO with MOVES and calculated emissions at congested and uncongested intersections using a time-in-mode (TIM) methodology that combines emission factors for each activity mode (i.e., acceleration, deceleration, cruise, idle). They suggested opportunities for control strategies with the potential to affect intersection emissions. Chamberlin et al., () developed a best-practices guide for conducting a MOVES Project-Level analysis. They concluded that greater resolution in link geometry (i.e. shorter links) closer to the intersection center will capture the greater emissions generated and suggested that if micro-simulation models are used to provide traffic activity input into MOVES, the vehicle trajectory outputs of the traffic model needs to be pre-processed into Operating Mode Distributions for running MOVES. The literature shows that the most studied yet indistinct factor in quantifying CO emissions was the speed followed by traffic composition expressed in truck percentages and VMT. Other factors included the geometry (grade). It was apparent that most studies included several factors without studying the specific effect of each factor on emissions. Research underlying CO emissions is still in its infancy especially when compared to other pollutants. Furthermore, few studies were found to integrate the latest mobile source emissions model MOVES with a microscopic traffic simulation model. This paper will address those findings.

Abou-Senna/Radwan 0 0 0 TEST BED MODELING The test bed network under study is a prototype of the Interstate (I-) downtown corridor located in Orlando, Florida. I- is a primary east-west transportation corridor between Tampa and Daytona cities, serving commuters, commercial and recreational traffic. I- is known to have severe recurring congestion during peak hours. The congestion spans about miles in the evening peak period in the central corridor area as it is the only non-tolled limited access facility connecting the Orlando Central Business District (CBD) and the tourist attractions area (Walt Disney World). The traffic on the I- freeway section is collected from double inductance loops embedded in the pavement every 0. miles, which extends from the Walt Disney World area on the west side of the corridor to Lake Mary area on the east side, for a total length of miles. The interstate carries an average annual daily traffic of 00,000 vehicles on segments in Orlando. It was imperative to evaluate the environmental impacts of this corridor especially during the peak hour. The modeled section is composed of approximately 0-mile stretch with three lanes in each direction. For modeling purposes, the freeway section links were aggregated into links ( of the segments are approximately -mile in length including horizontal curves and the first and last links are 0. mile each), main on-ramps and main off-ramps as shown in Figure. Traffic composition, as obtained from FDOT traffic information, included 0% passenger cars, % passenger trucks and % heavy-duty diesel trucks which correspond to MOVES vehicle types, and, respectively. The analysis was conducted for the eastbound direction and the experimental design encompassed all peak and off-peak conditions. Because the collection of a representative second-by-second vehicle operation dataset for every traffic circumstance is not realistic, the use of microscopic traffic simulation models to replicate real world travel patterns second-by-second for thousands of vehicles is essential. Therefore, a calibrated model of the subject test bed was created for the evening peak hour, using the latest VISSIM micro-simulation software. The calibration involved accurately modeling drivers driving behavior and ensuring reasonable throughput values on the network, travel times, delays and queuing conditions. FIGURE I- test bed corridor (Urban limited access highway).

Abou-Senna/Radwan 0 0 0 OPTIMAL DESIGN APPROACH Standard experimental designs are known as factorial designs. However, if a non-standard model is required to adequately explain the response or the model contains a mix of factors with different levels resulting in an enormous number of runs, the requirements of a standard experimental design will not fit the research requirements (). Under such conditions, optimal custom designs are recommended as the design approach. Choosing an optimality criterion to select the design points is another requirement. The custom design module in JMP (statistical software created by SAS) generates designs using a mathematical optimality criterion. Accordingly, D-optimality and I-optimality criteria were the two custom designs employed for this research (). These are non-regular orthogonal designs that avoid any confounding or correlation between main effects (0). The multilevel factorial design consisted of five () quantitative factors and one quantitative response (CO emissions in kg). The factors and levels ranges included: ) Volume (from,000 vph to,000 vph) ) Speed (from 0 mph to 0 mph) ) Trucks (from 0% to %) ) Grade (from 0% to %) ) Temperature (from 0 F to 00 F) The optimal custom design resulted in 0 runs (0 for each design). The factors levels were designated from (-) as the low setting and (+) as the high setting. The resulting factor settings and levels are summarized in Table. TABLE Factors and Levels Setting Volume Speed Truck % Grade % Temperature - 000 0 0 0 0-0. 000 0 0-0. 00.. 0 00.. 0. 00 0.. 0. 000 0 0 000 0 00 MOVES PROJECT LEVEL DATA The MOVES model is different from previous EPA mobile source emissions models in that it was purposely designed to be flexible with databases which allows and facilitates the import of data specific to a user's unique needs (). The project level database is specified in separate files for each input parameter and on a link-by-link basis. For example, link data included volume, speed, length, and grade, traffic composition with fuel type, operating mode data and driving

Abou-Senna/Radwan 0 cycle data. This information was updated for each run. The main project data are described in Table. TABLE Project-Level Parameters Location County Orange County, Florida Calendar Year 00 Month November Time :00 PM to :: PM (one hour) Weekday/Weekend Weekday Temperature 0 F - 00 F Humidity 0.0 % Roadway type Urban Restricted Access represents freeway urban road with lanes in each direction Types of Vehicles Passenger cars, Passenger trucks & Long haul combination diesel trucks Type of Fuel Gasoline for cars and diesel for trucks Roadway Length ( links - Total) Approx. mile/link Total of 0 miles Link Traffic Volume,000 -,000 vehicles per hour Link Truck traffic 0 % Trucks Average road grade 0 % upgrade Link Average Speed 0 0 miles per hour Operating Mode Running Exhaust Emissions Output Atmospheric CO, Total Energy Consumption & CO Equivalent OPERATING MODES & VEHICLE SPECIFIC POWER MOVES calculates total emissions as weighted average of emissions by operating mode. Operating mode is the amount of travel time spent in various driving conditions including: braking, idling, coasting, cruising, and accelerating within various speed ranges and at various ranges of vehicle specific power (VSP). VSP represents the power demand placed on a vehicle when the vehicle operates in various modes and at various speeds, a measure that has been shown to have better correlation with emissions than average vehicle speeds (). VSP is calculated based on a vehicle s instantaneous speed and acceleration. Hence, the simulated vehicle driving cycle data from VISSIM was input into the MOVES model based on the above mentioned project-level traffic conditions. However, processing the VISSIM output for input into MOVES required the calculation of the operating mode for each vehicle in the network. There are "speed bins" in MOVES which describe the average driving speed on a road type or link. Use of an average link speed input and link road grade allows MOVES to create an

Abou-Senna/Radwan 0 0 0 operating mode distribution from two built-in driving schedules whose average speeds bracket the given speed. However, it may not be representative of the actual driving schedules of the modeled corridor or the specific vehicle trajectories generated from VISSIM on a second-bysecond basis which emphasizes the importance of VIMIS. VISSIM/MOVES INTEGRATION SOFTWARE (VIMIS) Estimating vehicle emissions based on second-by-second vehicle operation encourages the integration of microscopic traffic simulation models with more accurate vehicle activity-based regional mobile emissions models. VIMIS is custom software developed to integrate between VISSIM and MOVES in order to automate the design of experiment portion and facilitate the conversion process of VISSIM files into MOVES files. The program consisted of four () main modules which were compiled using Microsoft Visual Studio 00. The modules included () generating the Design Cases (developed from JMP ) in VISSIM file format, () VISSIM which automated the simulation runs for each case with different seed numbers to account for the randomness and variability of the simulation output, () OPMODE which converted the trajectory output file from VISSIM into an operating mode distribution for input into MOVES. In most cases, when the output is set on a second-by-second basis, the file size can reach 0 gigabytes and cannot be accessed by a conventional program. The magnificence of this module is that it converts this 0 gigabyte file into a 00 kilobyte file, and in a MOVES input format containing all the necessary links to be analyzed, types of pollutants and emission processes (running, extended idle, etc., ) to be executed by () MOVES the fourth module that calculates emissions. DESIGN SETTINGS VERSUS ACTUAL SETTINGS Because we were performing the experiment in a simulation environment, there were a lot of combined effects and interactions between specific variables which demanded careful consideration, specifically Volume and Speed factors. Random arrival of vehicles, dynamic network loading, stochasticity of the traffic system, and unexpected traffic demand variations due to the freeway capacity clarified that the design setting of the volume and/or speed will not always be the same. In other words, setting the volume input at,000 vph resulted in an actual volume output less than or greater than the designed,000 vph. Likewise were the speed settings. Furthermore, increasing the volume level more than a specific threshold reduces the speed. Compounding this complexity of the main effects identified that we should be analyzing the actual output values instead of the design values as shown in Table (sample of 0 runs). 0

Abou-Senna/Radwan 0 Run# TABLE Design Settings versus Actual Settings Input Volume (vph) Posted Speed Limit (mph) Truck% Grade% Output Volume (vph) Output Speed (mph) Temp (F) CO (kg) 000 0 0 0 000 0 0 00 0 00 0 0 0 0 000 0 0 0 0 000 000 0 0 0 0 0 000 0 00 0 00 000 0 0. 0 0 000 0 0.. 0 0 000 0 0 0 0 0 00 0 000 0 0 0 0 0 0 000 0 0 0 000 0 0 00 000 0 0 0 000 0 0 0 000 0 00 00 0 0 000 0 0 0 000 0. 0 00 000 0 0 0 000 0 0 00 ANALYSIS OF RESULTS The results were analyzed using JMP forward stepwise regression approach. The CO emissions output values were transformed to log space for better correlation and improved presentation. Preliminary analysis showed an initial model including all main effects of Volume, Trucks, Grade and Temperature except for the Speed. However, two-factor interactions included Volume, Speed, Speed*Grade and Trucks*Grade. There was no confounding between any of the main effects and two-way or three-way factor interactions that were aliased with each other. However, the lack of fit was significant which implied that this was not the correct form of the model. An attempt to add the main effect of the speed and the stepwise regression was done for both cases. However, the analysis showed no significant difference between the two cases. Finally, we concluded that the Speed showed a strong correlation with CO emission rates instead of total emissions. Therefore, CO emissions were normalized with respect to their VMT.

Abou-Senna/Radwan 0 0 Moreover, the Temperature factor in the model represented the effect of air condition (AC) being turned on in the vehicles during high or low ambient temperatures which was an indirect measure of the emission rates, thus statistically insignificant. This improved final form of the model included the Volume, Volume, Speed, Speed, Trucks, Grade, Speed*Grade and Trucks*Grade interaction terms. Since the traffic network consisted of links, further model validation was conducted on link by link basis. That is each run of the experimental design resulted in outputs (one output for each link). Therefore, total sample size of 0 points were analyzed using the stepwise regression and the same model was valid across all corridor links. JMP has an interactive capability of fitting a separate prediction equation for each dependent variable such as Volume or Speed to the observed response CO. This feature enabled to predict all different combination of parameter levels on the response variable. The prediction profiles shown in Figure a display which levels of the predictor variables produce the most desirable responses on the dependent variable. Through the desirability function, the graph displays the optimal settings at which CO would be minimized. Figure b shows the response variable in log space with prediction traces for each factor at optimal settings. The analysis of the experiment determined the lowest settings for volume, trucks and grade while the speed setting between and 0 mph; optimum speed for engines that requires the least amount of work to complete fuel combustion. These settings yielded approximately 00 kg of CO emissions almost matching the modeled data. FIGURE a Prediction profiles for factors at center points.

Abou-Senna/Radwan 0 0 0 FIGURE b Prediction profiles for factors at optimal settings (log space). DISCUSSION The main objective of this paper is to study each of the transportation parameter s correlation with CO emissions, determine its significance and effect at each level in a microscopic and stochastic manner. The optimal design approach along with the prediction profilers enabled this process and resulted in a microscopic transportation emissions model (Micro-TEM). Several inferences were derived from the developed model. First, the results showed that there is significant potential for emissions rate reductions at certain travel speeds; specifically between about and 0 mph while maintaining volume levels up to 0% of the freeway capacity which correspond to level of service (LOS D) in the Highway Capacity Manual (HCM). It was observed from the runs that at certain volume levels and speeds ranging approximately from to mph; CO emission rates were minimized (0 grams) resulting in efficient operation with higher miles per gallon as shown in Table. Second, the modeled curves on Figure a showed an enormous increase in emission rates (from 0. to. kg/veh-mile) due to the effect of grade (0-)% and truck (0-)% amounting to approximately 0%. Third, the gradual and indirect effect of temperature appeared with the increase in truck and grade % and illustrated the speed-truck-grade interactions. The practicality of the developed speed curves lie in the determination of an instantaneous speed on a link or an average speed for a route, where emission rates can be predicted, thus total emissions at specific volume, grade%, truck% and ambient temperature levels. On the other hand, general correlation between the volume and CO was found to be quadratic as was the case with the speed. Figure b showed an exponentially decaying function towards a limiting value which matched the freeway capacity. The increase in CO emissions was limited by the capacity of the roadway and the amount of traffic that it can handle in our

Abou-Senna/Radwan 0 0 case,00 vph; capacity for a -lane freeway. Although it appeared as if the speed was not taken into account; when examining the data points thoroughly, a V-like shape was observed as shown on Figure c. The effect of the speed was incorporated in the spectrum created by the V-shape, ranging from speeds of 0 mph to 0 mph with emission rates higher at the 0 mph curve. Furthermore, there was another observed deviation in the data points closer to the 0 mph curve. These data points reflect higher emission rates for speeds higher than 0 mph which matched the speed curves shown on Figure b. To demonstrate the observed Speed Spectrum on the volume curves, more detailed data points were plotted at different speeds as shown on Figure d. The volume-speed curves can be used to predict emissions per mile, thus total emissions for a link or group of links using specific volume and speed at different parameter settings. These results proved the microscopic nature of Micro-TEM. The prediction expression of the model equation can be represented as: Ln (CO ) = 0.0 0. (Volume ) + 0.0 (Speed ) + 0. (Volume) 0.0 (Speed) + 0. (Truck%) + 0. (Grade%) + 0.0 (Speed*Grade) + 0.0 (Truck%*Grade%)------------(Eq.) The above mentioned parameters should be utilized as follows: Volume = {(Volume 00) / 00} Speed = {(Speed ) / } Truck% = {(Truck% 0.0) / 0.0} Grade% = {(Grade% 0.0) / 0.0} 0

Abou-Senna/Radwan TABLE CO Emission Rates by Link at Zero Truck and Zero Grade Levels VMT Length Volume Speed Truck% Grade% CO (ft) (vph) (mph) (kg) Miles/Gallon CO/veh-mi 0.0 0 0.0.0 0.. 0 0.. 0. 0. 0 0.. 0. 0.0 0 0 0.0.0 0.. 0 0.. 0. 0 00.0 0 0..0 0. 0.0 0 0..0 0..0 0 0.0.0 0.. 0 0.0.0 0. 0 0.0 0 0.. 0. 0 0 0. 0 0.. 0. 00 0 0. 0 0.. 0..00 0 0 0..0 0.. 0 0 0.. 0. 0.0 0 0 0..0 0.. 0 0..0 0.. 0 0..00 0..0 0 0..0 0.. 0 0 0..0 0..0 0 0.0. 0. 0. 0 0.. 0. 0. 0 0.. 0.. 0 0.. 0. 0. 0 0.0.0 0.. 0 0.. 0..0 0 0 00.0. 0.. 0 0 00.. 0. 0. 0 0 0.0. 0.. 0 0 0.. 0. 0. 0 0.. 0. 0 0. 0 0.. 0. 0. 0 0..0 0.. 0 0.. 0.00 00. 0 0 0.. 0.0. 0 0 0.. 0.0 0.0 0 0.. 0.0 0 0. 0 0 00.. 0.

Abou-Senna/Radwan VMT Length Volume Speed Truck% Grade% CO (ft) (vph) (mph) (kg) Miles/Gallon CO/veh-mi. 0 0 00.. 0.0. 0 0.0. 0. 0. 0 0.. 0.. 0 0.. 0. 0. 0 0 00.. 0. 0. 0 0 0.. 0. 0. 0 0.. 0. 0. 0 0.. 0. 0. 0 0.. 0. 0. 0 0 0.0. 0.. 0 0.. 0. 00 0. 0 0.. 0. 0. 0 0.. 0. 0. 0 0.. 0.. 0 0.. 0. 0.0 0 0.0. 0. 0. 0 0.. 0. 0. 0 0 0..0 0.0 0.0 0 0.. 0. 0. 0 0.. 0. 0

Abou-Senna/Radwan FIGURE a Speed CO emission rates at different temp., truck & grade levels. FIGURE b Volume CO emissions rates at different temp., truck% & grade% levels.

Abou-Senna/Radwan 000 CO Emission Rate (kg/mile) 00 000 00 000 00 0 mph 0 mph 0 0 000 000 000 000 000 000 000 000 Volume (vph) FIGURE c Volume Speed CO emission rate relationship (0%Trucks-0%Grade).

Abou-Senna/Radwan 00 CO Emissions (kg/m) 000 00 000 00 000 00 R² = 0. R² = 0. T00, T0, G0, S 0 T00, T0, G0, S0 T0, T0, G0, S 0 T00, T0, G0, S 0 T00, T0, G0, S0 0 T00, T0, G0, S0 0 T0, T0, G0, S 0 0 0 0 0 000 000 000 000 000 000 000 000 Flow Rate (vph) FIGURE d Speed spectrum on volume CO emission rate curves. CONCLUSIONS This paper presented a detailed microscopic examination of major key parameters that contribute to the increase of CO emissions, specifically traffic volume, speed, truck percent, road grade and temperature on a limited access highway corridor in Orlando, Florida. VISSIM outputs (link volumes, speeds, and acceleration/deceleration profiles) within each specified link in the network were combined with the MOVES model which used VSP and instantaneous speeds to generate emission rates on a second-by-second basis. The same temporal resolution and level of detail in VISSIM and MOVES supported the integration of the two models. The OPMODE approach covered all the simulated combinations of instantaneous speeds and accelerations, and was used to develop precise emissions for all desired driving patterns. The results demonstrated that obtaining second-by-second vehicle operations from a traffic simulation model are essential to achieve the most accurate operating mode distributions and presumably the most accurate emissions estimates. Specifically, emission rates are found to be highly sensitive to the frequent acceleration events that occur at lower speeds. The results of this research also elucidated on some of the literature ambiguities regarding the effectiveness of implementing speed management policies to reduce CO emissions and demonstrated that the

Abou-Senna/Radwan 0 0 0 0 speed does have a significant impact on CO emissions, provided that detailed and microscopic analysis of vehicle operation of acceleration and deceleration is achieved. Significant emission rate reductions were observed on the modeled corridor especially for speeds between and 0 mph while maintaining up to 0% of the freeway s capacity. Emissions at a given speed appeared to be influenced by the vehicle s engine loading getting to that speed from previous speed, which depends primarily on the acceleration rate. The difference in CO emissions between speeds lower than mph and higher than mph were represented by rapid succession of high power events expressed by the VSP, which resulted in more aggressive driving cycles, likely resulting in higher emissions. The analysis of the experiment identified the optimal settings of the key factors and resulted in the development of Micro-TEM (Microscopic Transportation Emissions Model). The main purpose of Micro-TEM is to serve as a substitute model for predicting CO emissions on limited access highways in lieu of running simulations using a traffic model and integrating the results in an emissions model to an acceptable degree of accuracy. The predictive model consisted of a function of most of the estimated main effects and their two-way factor interactions. The developed model can predict emission rates using average speeds and flow rates, thus total emissions for a link or route at different parameter settings. Micro-TEM experiment was conducted for CO emissions; however, methods have been demonstrated that can be used for other mobile-source pollutants. Microscopic traffic simulation models like VISSIM can produce second-by-second vehicle operating mode data, which can be used directly in MOVES to obtain accurate emission estimates. Furthermore, CO, NOx and PM dispersion modeling analyses, which are often required for roadway projects, can use the resulting spatially-determinate EFs in roadway dispersion models such as CALQHC or AERMOD to predict concentrations of various pollutants near roadways, or in gridded ozone modeling. It is also recommended that future research expand on the developed Micro-TEM model to include arterials with signalized intersections as well as other emission processes such as extended idling, crankcase and start exhausts along with other criteria pollutants that was not studied in this research. REFERENCES. Intergovernmental Panel on Climate Change (IPCC)'s Fourth Assessment Report (00). 00. Habitat Australia ():-.. U.S. Environmental Protection Agency. March 00. Greenhouse Gas Emissions from the US Transportation Sector 0-00. ICF Consulting, VA: Office of Transportation and Air Quality, EPA-0-R-0-00.. U.S. Environmental Protection Agency. April 00. Inventory of U.S. Greenhouse Gas Emissions and Sinks: 0-00. DC: Office of Atmospheric Programs, EPA-0-R-0-00.. Cooper, C.D., Arbrandt, M., 00. Mobile Source Emission Inventories--Monthly or Annual Average Inputs to MOBILE? Journal of the Air & Waste Management Association (), 0-00.

Abou-Senna/Radwan 0 0 0. You, S., Lee, G., Ritchie, S., Saphores, J.-D., Sangkapichai, M., Ayala, R., 00. Air Pollution Impacts of Shifting Freight from Truck to Rail at California's San Pedro Bay Ports. Transportation Research Record: Journal of the Transportation Research Board (-), -.. Mui, S., Alson, J., Ellies, B., Ganss, D., 00. A Wedge Analysis of the US Transportation Sector. Environmental Protection Agency. EPA 0-R-0-00, pp.. Toth, G., 00. Reducing Growth in Vehicle Miles Traveled: Can we really Pull it off? Driving Climate Change: Cutting Carbon from Transportation, -.. Liu, H. F. December 00. Vehicle CO emissions and the Compactness of Residential Development. Special Studies, National Association of Home Builders (NAHB).. Burwell, D.G., 00. Beyond Congestion: Transportation s Role in Managing VMT for Climate Outcomes. Reducing Climate Impacts in the Transportation Sector. 0. Barth, M., Boriboonsomsin, K., 00. Real-world carbon dioxide impacts of traffic congestion. Transportation Research Record: Journal of the Transportation Research Board 0(-), -.. Marsden, G., Bell, M., Reynolds, S., 00. Towards a real-time microscopic emissions model. Transportation Research: Part D D(), -0.. Sturm, P.J., Kirchweger, G., Hausberger, S., Almbauer, R.A.,. Instantaneous emission data and their use in estimating road traffic emissions. International journal of vehicle design 0(), -.. De Vlieger, I., De Keukeleere, D., Kretzschmar, J.G., 000. Environmental effects of driving behaviour and congestion related to passenger cars. Atmospheric Environment (), -.. Int Panis, L., Beckx, C., Broekx, S., De Vlieger, I., Schrooten, L., Degraeuwe, B., Pelkmans, L., 0. PM, NO x and CO emission reductions from speed management policies in Europe. Transport Policy (), -.. Int Panis, L., Broekx, S., Liu, R., 00. Modelling instantaneous traffic emission and the influence of traffic speed limits. Science of the Total Environment (-), 0-.. Boriboonsomsin, K., Barth, M., 00. Impacts of road grade on fuel consumption and carbon dioxide emissions evidenced by use of advanced navigation systems. Transportation Research Record: Journal of the Transportation Research Board (-), -0.. Papson, A., Hartley, S. and Kuo, K. 0. Analysis of Emissions at Congested and Uncongested Intersections using MOVES00. Transportation Research Record: Journal of the Transportation Research Board 0. (0): -.

Abou-Senna/Radwan 0. Chamberlin, R., Swanson, B., Talbot, E., Sharma, S., and Crouch, P., 0. Toward Best Practices for Conducting a MOVES Project-Level Analysis. Transportation Research Board st Annual Meeting (No. -00).. Johnson, R., Montgomery, D., Jones, B., 0. An Expository Paper on Optimal Design. Quality Engineering (), -0. 0. Jones, B., Montgomery, D.C., 00. Alternatives to resolution IV screening designs in runs. International Journal of Experimental Design and Process Optimisation (), -.. U.S. Environmental Protection Agency. April 00. Technical Guidance on the Use of MOVES00 for Emission Inventory Preparation in State Implementation Plans and Transportation Conformity. DC: Transportation and Regional Programs Division, Office of Transportation and Air Quality, EPA-0-B-0-0.. U.S. Environmental Protection Agency. November 00. MOVES00 Highway Vehicle Population and Activity Data. DC: Assessment and Standards Division, Office of Transportation and Air Quality, EPA-0-R-0-0.