Audio files require the Windows Media Player.
Moderator
At this time, I would like to welcome everyone to the Talking Freight Conference Call. All lines have been placed on mute to prevent any background noise. After the speakers remarks, there will be a question and answer session. If you would like to ask a question during this time, simple press star and then the number 1 on the telephone key pad. If you would like to withdraw you question, press the pound key. Thank you. Nicole Coene, you may begin your conference.
Nicole Coene
Good afternoon, or good morning for those of you on the West Coast. Welcome to the talking freight seminar serious. May name is Nicole Coene and I will moderate today's seminar. Today's topic is Measuring Lifecycle Impacts of Freight Transportation Emissions.
Before I go any further, I do want to let those of you who are calling into the teleconference for the audio know that you need to mute your computer speakers or else you will be hearing your audio over the computer as well.
Today we'll have three presentations, given by:
Dr. Mikhail Chester is an Assistant Professor in Civil, Environmental, and Sustainable Engineering at Arizona State University where he runs a research laboratory focused on transportation life cycle assessment and infrastructure resilience to climate change. Dr. Chester has worked with a variety of public and private passenger and freight agencies to develop energy and environmental assessments of transportation systems including infrastructure, vehicle, and energy production processes, in addition to vehicle operation. He is currently leading 3 National Science Foundation projects to study how the design of infrastructure is vulnerable to extreme heat and flooding.
John Davies is an Environmental Protection Specialist with FHWA's Office of Natural Environment. His work focuses on tools, methods and research supporting greenhouse gas analysis and policy development. Before coming to FHWA, John worked on climate issues with EPA's Office of Transportation and Air Quality. John holds a Master's in Regional Planning from the University of North Carolina.
Frank Gallivan is a transportation and sustainability planner with ICF International. He specializes in the design and implementation of analytical tools to promote better decision making in transportation planning, design, maintenance, and operations. Frank holds a Master of City Planning from UC Berkeley.
Jeff Houk is an Air Quality Specialist with FHWA's Resource Center in Lakewood, Colorado. Jeff works on issues related to highway project air quality analysis, transportation conformity, and climate change. He has 30 years of experience with transportation/air quality issues at EPA and FHWA, and has a degree in Chemical Engineering from Michigan State.
Chris Porter is a Principal of Cambridge Systematics with 18 years of experience in air quality, energy and greenhouse gas analysis, transportation and land use, and performance measurement. He led a study for the National Renewable Energy Laboratory on the effects of travel demand management and the built environment on energy use, and was a lead author of the U.S. DOT Report to Congress on transportation's contribution to reducing greenhouse gas emissions. He has supported state and regional GHG inventories and reduction plans, including both passenger and freight strategies, in Connecticut, Florida, Maryland, Massachusetts, New Jersey, and Oregon.
Today's seminar will last 90 minutes, with 60 minutes allocated for the speakers, and the final 30 minutes for audience Question and Answer. If during the presentations you think of a question, you can type it into the chat area. Please make sure you send your question to "Everyone" and indicate which presenter your question is for. Presenters will be unable to answer your questions during their presentations, but I will start off the question and answer session with the questions typed into the chat box. If we run out of time and are unable to answer all questions we will attempt to get written responses from the presenters to the unanswered questions.
The PowerPoint presentations used during the seminar are available for download from the file download box in the lower right corner of your screen. The presentations will also be available online within the next few weeks, along with a recording and a transcript. I will notify all attendees once these materials are posted online.
Talking Freight seminars are eligible for 1.5 certification maintenance credits for AICP members. In order to obtain credit for today's seminar, you must have logged in with your first and last name or if you are attending with a group of people you must type your first and last name into the chat box. I have included more detailed instructions in the file share box on how to obtain your credits after the seminar.
For those of you who are not AICP members but would like to receive PDH credits for this webinar, please note that FHWA does not formally offer PDHs, however, it may be possible to receive PDHs for your participation in Talking Freight if you are able to self-certify. To possibly receive PDHs, please download the agenda from the file download box and submit this agenda to your respective licensing agency.
Finally, I encourage everyone to please also download the evaluation form from the file share box and submit this form to me after you have filled it out.
I'm now going to turn it over to Mikhail Chester of Arizona State University to get started.
Mikhail Chester
Hi guys. Thank you. Today, I would like to kick this off by giving an overview of environmental lifecycle assessment of freight transportation services, which is a field that has slowly emerged over the last 10 years. The objective here is to give you an overview of how we are thinking about developing lifecycle assessments about freight, and what we can define as significant in the lifecycle and what we suggest to do about it. So I would like to start maybe 10 years ago roughly when freight LCA started to emerge and then move into today and the types of analyses that we are doing today focused on vehicle infrastructure and energy production components.
To get you thinking about lifecycle assessment, I would like to start by making sure that we all are thinking about transportation as a service. So I think we often think about transportation as the movement of vehicles, which is obviously a necessity, but when we think about it as a service, we tend to think about the ancillary systems that are needed to support the movements of vehicles. So transportation systems, whether passenger or freight, all require three fundamental elements which are: vehicles, right-of-way, and terminal capacity.
Those translate into lifecycle processes that I would categorize as either vehicle manufacturing and maintenance, infrastructure or construction, rehabilitation, maintenance and operation, or energy production. And from a lifecycle perspective, what we are trying to do is connect the cradle-to-grave, that is: starting with the processes that are needed to extract raw materials; followed by the processing of the raw materials and the movement of the raw materials; then turning them into something like a vehicle or infrastructure and ultimately using those materials or those vehicles and infrastructure components; and finally, disposing of them at the end.
So the three major categories of lifecycle processes that we are interested in are: vehicle, infrastructure, and energy production. Vehicle and infrastructure are pretty intuitive because we see those on a day-to-day basis. We are surrounded by them. But, energy production is a significant component. And I will show you results in a little bit. You can see how significant this is. To use energy, you need to invest energy. Meaning that in order to have a gallon of gasoline or diesel or any other form of energy in a vehicle, additional energy is triggered in the supply chain; and that is for the extraction of those fuels, the processing of those fuels, and the movement of those fuels to the final destination where they are put into the vehicle.
When I started doing LCA about 10 years ago, the rough number was that for every 100 gallons of gasoline in a car, about 16 gallons of gasoline equivalent energy were triggered in the supply chain. Now that number has moved closer to about 30. So, for every hundred gallons of gasoline in a passenger automobile, about 30 are triggered in the supply chain. That is because we have moved toward heavier oils, that is, crude oils and things like that.
So it is not just that we have the manufacturing and processing of materials into vehicle, infrastructure, and the energy processes, it is also that you have supply chain process that exist to support those activities as well. And it turns out that each of these supply chain processes is also significant. For each of these processes, what we do in lifecycle assessment (you see the process on the right there in the white box) is we evaluate the energy input and the air emission outputs. So energy input includes both electricity as well as fuel. Those go into the process and out of process come emissions. Today I will focus largely on air emissions (conventionally the precursors to pollutants) and greenhouse gases. We can certainly use a lifecycle of framework to evaluate a variety of other things. Water, toxic plumes and so on. We tend to be focused on greenhouse gases. Lately there has been a lot of LCA work on air emissions and greenhouse gases. Ultimately, we connect those pollutants to human environmental impact, translating them from quantity of pollutants into their potential for causing environmental and human health impacts.
So a bit more specifics on how we do this. I will go through each of these processes and major categories of lifecycle processes and give you more background. For vehicles, we tend to use modeling platforms like FEMA Pro or the GREET model, in the case of passenger. In the case of freight, we tend to use a European lifecycle assessment model called FEMA Pro where we actually model the manufacturing and maintenance of vehicles under U.S. conditions. Sometimes the manufacturing occurs outside the U.S. For example, most of the ocean going vessel fleet is primarily manufactured in Asia. We are able to model that. The vehicle then comes to the U.S. and it is used and maintained in the U.S.
For infrastructure, you will hear more about this today. I won't go into too many specifics. But for infrastructure, we are generally thinking about modeling the manufacturing of materials, the construction activities, and the maintenance and rehabilitation cycles, as well as the energy and air emissions associated with each of these.
PALATE it was one of the first pavement and lifecycle assessment tools for energy and economic impacts developed by UC Berkeley in the mid- 2000. From what I understand, the FHWA Infrastructure Carbon Estimator builds somewhat on that knowledge. So I won't go into details their sense it will be covered. And then my lab here has been developing a city and regional network lifecycle assessment tool to be able to evaluate these materials, construction activities, and maintenance rehabilitation cycles for an entire roadway network. You can do this for state, region, or a city.
On the energy production side, I would say that the state-of-the-art is the U.S. DOE Argonne GREET model, which is the fuel cycle model that we typically rely on for U.S. energy use and transportation. Then, the supply chains have typically been evaluated with Carnegie Mellon's economic input/output lifecycle assessment model. This allows us to look upstream into the supply chain and evaluate the energy and environmental impacts associated with so much economic activity.
From these sorts of approaches, you early on see results like this. This is showing grams per ton mile of CO2, NOx, PM10, CO and SO2 across Rail, Air, and Road modes. You see a consistent pattern emerge which is: Rail is the lowest, followed by Road, followed by Air. The uncertainty bounds capture a variety of different vehicles and different parameter uncertainties associated with this model. This came out in 2007. It was the first rigorous freight LCA performed. Interestingly, when you flip the results (i.e., instead of normalizing per ton mile, you normalize per dollar of value of the goods that are transported), you see the opposite results: Rail is the largest, Road is in the middle, and Air is the lowest. So Air moves high value goods and Rail moves very low value goods.
So this is where we started with freight LCA about eight years ago. Those were lifecycle results. Here you see the lifecycle results broken down into more detail. Now you are seeing the fuel combustion as light gray; the vehicle manufacturing and maintenance is black; the infrastructure is white; and the energy production as the darker gray. The bars are showing you the percentage of the bars on the previous slide and what is contributing to them. For example, if you look at PM-10 emissions for Air or for Road, you see that a lot of those emissions are associated with the infrastructure. So the infrastructure is contributing a large portion of PM emissions. This is the insight that we desire to acquire using lifecycle assessment. It gives us a bit more information about where major impacts are occurring from these ancillary services that are associated with freight transport. Looking at this, we can say, "There are a lot of PM emissions coming out of roads. What do we do about this?" And I will get to that.
More recently, we have been much more rigorous about the components that we are including in the lifecycle assessment. Here is a table showing you all the different lifecycle processes (denoted by bullets) that we are assessing, associated with Trucking, Air, Marine, and Rail. You have far more detail in terms of what is going in, everything from:
For each of these bullets, we are trying to understand the energy requirements and greenhouse gas and other air emissions associated with them.
Here are some results for California that use that more robust framework. So this is a lifecycle energy use in the top left and emissions as metric tons per kilometer in California. This is a study that I lead for the California air resources Board as part of AB-32. We tried to characterize what the lifecycle emissions were with this more robust framework. Again you see here the gray being the operation, the tailpipe of the vehicle, so to speak. It constitutes a large portion of the lifecycle results, but not necessarily the entire portion of it.
What you see is that certain pollutants tend to appear fairly significantly in the lifecycle. So the SO2 tends to be one that we see in both passenger and freight. That is largely because we have done a really good job cleaning up the tailpipe vehicles, by doing low carbon fuels. But we haven't done nearly as much cleaning up of the emissions from electricity generation. Electricity appears everywhere in the lifecycle of these modes. So those are the sorts of insights that we are getting with the more robust framework that we are using now.
What we find in California is that the infrastructure tends not to be a huge piece of the puzzle. It tends to be largely vehicle manufacturing and energy production in the lifecycle in addition to the tailpipe. That leads us to questions about, not only the per kilometer or mile results which I have been showing you, but what does this mean systematically across the region.
Now I am showing you the lifecycle results for California. And what you will see -- as many of you are experts and may know this -- the heavy duty and medium heavy duty trucks are responsible for the vast majority of miles of goods moved. Therefore they are responsible for the vast majority of greenhouse gas emissions in that state which is probably true in other states as well. So rail and rail truck intermodal modes are very small fractions. So this leads to a question of -- what do we do about this? Obviously we should be targeting trucks. And now when we target trucks we should be thinking about doing it across the lifecycle processes. It is not that we have to specifically focus on the tailpipe of vehicles. But we can move into not only strategies for the tailpipe of vehicles but what those strategies that reduce the emissions at the tailpipe of vehicles, due to other lifecycle processes, particularly the infrastructure and energy production.
Hopefully you can that the lifecycle gives us a sense of what a strategy does in the interdependence of the system. That is, if we make trucks more energy efficient, if we improve their miles per gallon, then we are also going to get benefits on the energy production side. You will have to produce less energy. You will get less impact up stream. Similarly, if you reduce the impacts associated with infrastructure construction or if you reengineer infrastructure to be slightly different (through reduced material use, smoother surfaces, more stiffness, and those sorts of strategies) you might have positive benefits on the vehicle operation side which then would translate to the energy production side. Those are the connections we are trying to make across the lifecycle. It is not that those components I showed you before are interdependent. They are very much dependent on each other.
Going forward (my last slide here and I will finish up), we have been asking what it would take to get vehicles down to policy goals or to certain levels that we are starting to think about, e.g., the California 1990 levels of greenhouse gas emissions or below that, and then by 2050, down to 20% of 1990 greenhouse gas levels. So, we're positioning the lifecycle results.
Here I am showing annual greenhouse gas emissions over time and what would happen under different technology changes and different fuel changes, and whether or not that gets us to those policy goals by a certain year. What we are seeing in our analysis is that you basically need to have zero emission vehicle technologies or zero emission fuels to get close to the 2050 policy goals of 80% below 1990 levels. And while we do this with CO2, we can do this with other pollutants as well. Connecting this back to the lifecycle gives us a more robust way of thinking about this because we are starting to piece together that the strategies that occurs within the state, to meet these policy goals have affects beyond the state. This is the result of lifecycle processes that don't necessarily followed geopolitical boundaries.
With that, I will end. All of this information is available on our lab's website: www.transportationLCA.org. I am happy to take questions by email in addition to the questions in the seminar later today.
Nicole Coene
Thank you Mikhail. Now we will move on to John Davies of the Federal Highway Office of Natural Environment.
John Davies
Thanks so much, Nicole. And thanks to Mikael for providing a great overview of the whole lifecycle. I am going to drill down and hopefully talk a bit more about a tool that FHWA has developed to specifically address the infrastructure piece. One point to make that I think is really important is that greenhouse gases are different from criteria pollutants in that the impacts are cumulative and not location specific. Recognizing this, FHWA started thinking about this issue and we recognize that there is value in developing a tool that would help FHWA evaluate the emissions associated with projects that we helped fund as well as transportation agencies to evaluate the projects that they plan and build.
So anyway, this is all to recognize that one of the reasons that agencies are building projects is to help offset the impact of operational emissions. Part of the entire author of lifecycle analysis is being able to capture the emissions that are associated with infrastructure itself.
In terms of just the state of practice, FHWA has not thought a lot about this issue until a few years ago. Greenhouse gas analysis is not an uncommon thing. A majority of states have climate action plans of some sort. And most of the MPOs are involved in emissions analysis. And as part of that, many of them do greenhouse gas analysis. Some states have a requirement to analyze greenhouse gas emissions. But almost the vast majority of all of that analysis is the focus on the tailpipe. Only one state has the requirement or analysis of the infrastructure piece. Recognizing that, we thought it seemed like a blind spot that transportation agencies don't really have a really good and tool to address the infrastructure that they are ultimately responsible for. And maybe just as critically, we noticed that there was growing interest in this administration in the ability to address greenhouse gas emissions as part of the NEPA process.
The Council on Environmental Quality, which is responsible for NEPA, has been talking about this for some time. They have developed a draft guidance. The early versions of this have envisioned an implied requirement for certain large Federal actions to be able to characterize the emissions that are associated.
So, this is all to say that we recognize it is likely to be a requirement at some point. Again, there is not a really good way to be able to address that. About well over two and a half or three years ago, we partnered with ICF international to start developing a tool that would be simple and hopefully useful for practitioners to start addressing emissions at a pre-engineering level. We looked around and we found that there were a number of tools available. A lot of them required a fairly detailed input. We wanted to come up with something that planners and analysts would be able to potentially use when really detailed data wasn't available. ICF did some really thoughtful work in coming up with this. Now, I want to turn it over to Frank Gallivan to talk about what his team was able to come up with.
Frank Gallivan
Okay. Thanks, John. I will be going through the capabilities of the tool and showing you some screenshots and getting you familiar with what the tool looks like and how it functions. I'm not going spend any time on the data sources or that kind of detailed analyses that went into creating the tool. I will just say that we did rely heavily on the Palate model that Mikhail mentioned as well as NCHRP fuel factors data and a couple of databases of road projects for quantities and materials. For those that are interested in more about that, there is more than you could ever want in the documentation of the tool on the FHWA website.
Getting into the capabilities of the tool, we are talking about freight, but I wanted to make it clear that the tool does much more than freight. It is designed to address every type of surface-based transportation. So there are modules that calculate greenhouse gas emissions from construction of roadways and parking facilities, public transportation, infrastructure, bridges, and bike and pedestrian facilities as well.
A little bit about the way that we address the lifecycle calculations and the tool. As Mikael said, we are accounting for emissions from materials, from construction activities, and for maintenance and rehabilitation. We just categorize those a little bit differently in the tool in terms of upstream energy and emissions, and then direct energy and emissions. The upstream energy and emissions are those that come from the materials that go into those facilities: the raw materials extraction, the materials production, transportation to sites, as well as chemical reactions in producing cement and CO2 emissions.
Then with direct emissions, we have the construction equipment, that is, fuel used in transportation of materials to the site, fuel used in the construction equipment on site, and then fuel used in the routine maintenance of those facilities: snow removal, vegetation and other types of routine maintenance activities like striping and sweeping. All of these are incorporated into the calculations of the tool.
We also have a range of mitigation strategies that can be applied in the tool. So the tool inventories the emissions that occur from your basic action of constructing or maintaining facilities. We also allow you to experiment with several mitigation strategies to reduce the energy and greenhouse gas emissions associated with those activities. There are mitigation strategies in terms of alternative fuels that you can use in construction vehicles as well as maintenance vehicles. There are different types of materials for recycling, like in place roadway recycling. You can also incorporate recycled materials from other sites into your construction as well as use a preventive maintenance routine that can reduce the overall amount of energy that you have to put into maintaining your roadway every time.
Using the tool breaks down into six basic steps. These are actual screen shots from the tool. This is an Excel-based tool. It is fairly simple to navigate and to use. There are only three or four sheets that you need to interact with. Basic inputs to start with: location of your state, number of years you want to consider for the lifecycle emissions of your particular project, and information about the extent of your transportation project or your network. The tool works at any geographical scale that you want to analyze. So you could be doing a project on a 10-mile roadway or look at entire regional transportation systems.
In the next step, you input information about the construction and maintenance activities that you foresee as part of your project. The most basic input there is lane miles by facility type. So if you were building a new roadway, you would input the number of lane miles of new roadway in each of those seven facility types that you see there on the left, divided by rural and urban arterials and interstates.
In the next step, we ask for information about construction delay. The tool includes calculations to estimate the amount of greenhouse gas emissions that could occur additionally because of roadway delay caused by construction activity. If you want to look at that you can input a couple of simple data points about project days of lane closure and the extent of the facility.
Step four would be to input mitigation strategies. And you can look at both a baseline and a planned level of deployment. Many States' Department of Transportation are already using a high portion of recycled materials in their projects. You may input what you are currently doing in baseline deployment and then increase that in the planned deployment, in order to understand how increasing the level of deployment can affect greenhouse gas emissions.
In step five, we're showing you the results. So the results are presented both in terms of energy use in BTU and in greenhouse gas emissions. They are broken down, in terms of the different basic facility types, roadway, bridges, rail, bus, bicycle, and pedestrian as well as by lifecycle emissions category. We also got some nice bar charts in the tool to help you visualize that.
And then the final step, if you did want to look at the construction delay components as well as calculations that will estimate the impact of pavement smoothness, as Mikael talked about. Pavement smoothness can contribute to increasing vehicle fuel economy and therefore reduce greenhouse gas emissions. These are the only components of the tool that get at the vehicle operations part of the transportation system. But because they are so intrinsically linked to the construction activity, we did want to include those.
And that is it. With that, I will turn it over to Jeff Houk to talk about a case study of the tool used.
Jeff Houk
Thanks, Frank. As we were developing and testing the tool, we also developed a few case studies. This one happens to represent a hypothetical port project. The project is to improve access to a small port that sees about 500 trucks a day. I have a couple of screenshots coming up so that you can visualize what this looks like but I will describe it here. There are deficient bridges on one roadway going into the port, so trucks have to take a roundabout route through town to access the port. In addition, because of space restrictions at the port, the trucks have to line up and idle along city streets while they wait their turn to go to the port. So this project is designed to fix those two problems. The project would widen the southern access roadway, reconstruct two bridges to be able to handle the weight of heavy trucks, and also build truck parking at the ports, so the trucks would no longer have to idle along city streets. That in turn would require relocation of about a half a mile of rail access lines to the port.
In this example, we do two types of analysis. We use the Infrastructure Carbon Estimator to account for the construction and maintenance emissions associated with the project itself. We also use EPA's MOVES emissions model to calculate the changes in truck emissions, because we will have less idling and a shorter access route into the port, we can capture the change in truck emissions into the port using the EPA MOVES model.
So here is the no-build scenario. In this aerial shot, you can see a major highway coming up from the South. It basically dead ends at another highway going into town. This dashed red line shows the route trucks currently take going into the port. The solid yellow line shows the area where trucks are idling alongside the road while they wait their turn to get into the port. You can see another highway coming into the port area from the South. I pointed out where the two substandard bridges are located that would need to be rebuilt as part of the project. Finally, the dashed blue line going kind of across the middle of the picture shows the rail access line into the port.
So this next photograph shows the project. We are widening the southern access road and rebuilding those bridges, so the trucks can come into the port area from that direction instead of looping up and coming through town. We are building a truck parking lot at the port. And because of that, we are slightly relocating the rail terminus of the port. So that is our hypothetical project. It is a real place but it is just a fictitious project that we used to illustrate the capabilities of the tool.
So using the tool and going through the spreadsheet and filling out the different cells that account for the roadwork, bridge work, rail work we are going to do, along with maintaining the existing systems, for the no build a scenario, there are about 56 tons of carbon dioxide a year associated with maintaining the existing roadway and rail system. In the build scenario, there are about 223 tons of carbon dioxide a year accounting for constructing the new infrastructure, plus maintaining the new and existing infrastructure. In using the tool for this kind of build/no-build comparison, the no action alternative only includes maintenance of the roadways being used and the build scenario includes construction of the new stuff plus maintenance of the new stuff and the old stuff.
So the net difference between those two scenarios is 167 tons a year. Using the EPA MOVES model to estimate truck emissions, the no build emissions are close to 5000 tons a year of carbon dioxide, that is an average of 2020 and 2040. The build scenario goes down to 723 tons. That is because the trucks are taking a shorter route into the port and also we have eliminated all of the idling by building a truck parking area. So using the information about the construction load and the maintenance load, along with the emission savings from the trucks, we can calculate the payback period: whether the emissions savings from the trucks are enough to offset the emissions accrued in building the infrastructure. So the total change in emissions over the 20 year life of the project is 160 tons times 20 years or about 3300 tons. The on-road emissions benefit is about 4000 tons a year. So the payback period for this project is about 10 months. After 10 months, we start to see a benefit from this construction project at the port. So this is just one example of how this type of tool can be used.
I will close out this session of the presentation. If you want more information about the tool, here is the link where you can download the spreadsheet tool. There is also a document that is a combined user guide and technical report that includes the information Frank discussed about the sources of information behind the tool and the calculations within the tool and so on. At FHWA, John Davies and I are the contacts for this tool. And then here is Frank's contact information at ICF. That is it for us.
Nicole Coene
Thank you, Jeff. We will now move on to Chris Porter of Cambridge Systematics. Chris.
Chris Porter
Great. Thank you. My title is about lifecycle emissions assessment from freight vehicles but I'm not really going to be talking explicitly about the lifecycle components as our previous presenters did. I am really just going to be talking about freight contributions to emissions and strategies for reducing those emissions and really try to paint a picture from a number of recent studies on freight and emissions. So I will start off with some background about freight contribution and forecast and then I will talk about strategies related to vehicle and fuel technology, operational strategies, and also demand management and mode shift. And finally draw a few conclusions according to the different scopes that different agencies or actors have control over.
So this is from the EPA national emissions inventory. It shows surface transportation emissions including light-duty vehicles which are the light blue or turquoise components, and then heavy duty vehicles. So these are contributions to fine particulate matter and oxides of nitrogen in 2011. This shows that trucks are responsible for about half of all PM2.5 emissions and just over a third of NOx emissions. Rail and marine collectively make up about 20% of each pollutant emissions. Collectively, freight is responsible for the majority of both pollutants from surface transportation. Nearly all of this comes from diesel engines.
This is actually something that Dr. Chester talked about in much more detail. This just shows operating or direct emissions, and then upstream emissions from fuel production and delivery which are generally directly proportional to the direct emissions from the actual operations of the transport vehicles and those upstream commissions are about 22% in our calculations using the GREET model in 2011. This can vary depending on the specific production pathway of the fuel. Carbon dioxide emissions, the primary greenhouse gas component, will be proportional to energy use.
This chart also shows energy use by mode. On average, a truck is about 10 times more intensive per ton mile than rail or water. This is a little misleading since most rail and water movements also involve truck and intermodal handling emissions at each end of the trip. They also may be more circuitous than on-road truck routes. When you average that all together, overall rail and water are about three times more efficient than trucking. This varies depending on the distance that the freight moves. You are going to have more relative benefits over a longer distance where you get more of the benefits of the line haul of the rail. You can also see that air is very energy intensive per ton mile. But it carries only a small fraction of ton miles.
This is showing direct and upstream emissions. This is comparing fuel types for common fuels and showing relative emissions from conventional gasoline. You can see that diesel is about 15% less carbon intensive than gasoline or corn ethanol. Soy-based bio diesel, with a 20% blend of biofuel, is about 25% less intensive than gasoline on the lifecycle basis. For compressed natural gas, the direct emissions are even lower but there are more upstream emissions due largely to methane leakage. So overall, the fuel cycle emissions are pretty comparable to diesel. I also want to point out that in these numbers there is a fair amount of uncertainty in some of the estimates of the upstream admissions especially for biofuels, the contributions of land use change, as well as methane leakage in the natural gas production cycle. So, these numbers may be changing and probably will change in current and future versions of the lifecycle model.
This paints a picture of where freight is projected to go in terms of overall growth which is one of the drivers of energy and emissions. This is from the Annual Energy Outlook. The DOE projects truck vehicle miles to grow at roughly 2% a year. Rail freight is expected to remain level and domestic shipping to decline lightly. Of course these could change due to various changes in economic activity and to the structure of freight and manufacturing industries and other factors.
Counteracting these trends, truck fuel efficiency is expected to improve by 14% in the future due to the 2014 to 2018 heavy duty vehicle greenhouse gas emissions standards, adopted by EPA and NHTSA. Rail and shipping efficiency are also expected to improve by about 15 to 20% in the next three decades.
You can see that the efficiency improvements are not going to be enough to counteract the forecast growth in terms of overall energy use. So we don't have a national emissions forecast, aside from the energy and greenhouse gas emissions, of criteria pollutants. So this is an example from Connecticut's inventory that we did with our partners. You can see that, comparing 2020 versus 2009, we projected a 25% increase in truck VMT compared to a 50% reduction in NOx and over 80% reduction in PM. That is reflecting the admission standards phased-in over the 2007 to 2010 period for EPA, for trucks, as well as the new Tier lll and tier four emissions standards for locomotives. You can also see that over the long run, through 2040, the increase in VMT is expected to start to level out the emissions benefits over time.
So next, I am going to talk about some control strategies. Again talking about technology, operations, and the potential for mode shift.
This table lists some of the more effective technology strategies and provides an assessment of their effectiveness and cost-effectiveness, which we typically measure in dollars per ton of pollutants reduced. Just for reference, $10,000 per ton is a pretty common reference point. If it costs less than that, it is considered to be cost-effective and more than that, not so. That is for NOx. It would be a different scale for a particulate matter.
Extended idling, which occurs when the driver is resting but needs to keep running the truck to run the power systems, is actually responsible for a large portion of emissions from long-haul trucks. It is expected to get larger, making up maybe half of all NOX emissions from long-haul trucks by 2020. So some of the strategies to control extended idling include truck stop electrification or vehicle based strategies including installation of auxiliary power units. And those can be pretty effective in addition to saving fuel, potentially leading to a net cost savings to the vehicle operator.
Inspection and maintenance programs are basically a way of making sure that vehicles continue to meet the emission standards that they were built to, which is going to be really important for the new clean vehicles. All 2013 and later model trucks will be equipped with a so-called OBD 2 systems that will identify trucks needing maintenance through remote monitoring. So that is going to be an important benefit. This can be a cost-effective way of controlling emissions.
Another strategy is conversion to natural gas, which some people have been doing recently just because of the cost difference between the fuels. The pollutant benefits are pretty limited given the vehicle controls. But you do get some greenhouse gas savings as I showed a couple of slides ago.
Drayage trucks are trucks used for short-haul goods movements around ports and terminals. These are typically older and higher admitting than average vehicles and often can be cost-effectively controlled through accelerated retirement or retrofitting or a rebuild of engines. Since these are local routes, they are also favorable to alternative fuel applications. Over the long run, this is going to be a less effective strategy as the new trucks become cleaner and they get phased in with the older operations.
You can also retrofit trucks in use in higher-speed applications with products to improve aerodynamic and rolling resistance, such as a smart way certified product: fairings, low rolling resistance tires, etc. There are numerous technologies for retrofitting diesel engines with pollution control. The reduction potential and cost-effectiveness really varies a lot depending on the technology, the vehicle characteristics, how it is used. So it is difficult to summarize except to say that, in certain applications, where there is a lot of emissions and a lot of exposure to local population, that these can be very effective and cost-effective control strategies.
Some strategies are also direct at Rail and Marine emissions. For example, operations at Rail switching yards actually make up a pretty significant fraction of the real emissions inventory. These emissions can be reduced by up to 90% by repowering older locomotives with new engines or by replacing with GENSET locomotives which use hybrid drivetrains for much better efficiency. There is also potential for fuel savings from long-distance locomotives although it is somewhat less cost-effective.
Then around ports, if you have a port in your area, there are probably fewer options for reducing emissions from the Marine vessels: reducing speeds is one strategy; clean fuel for auxiliary engines; and cold ironing or providing electricity for the operation of the vessel while in the port, although that can be a fairly expensive strategy.
On the operations side, this table lists some strategies that have been considered. I would say in general, the potential is fairly small, localized in nature, or hasn't been well quantified. It can really differ a lot from situation to situation. One point to note is, in terms of empty trucks and filling them, in Connecticut, we found about seven empty backhauls for every 10 deliveries. In theory, there is a lot of potential there. However there is already a lot of load matching that goes on through the private sector through optimization software, brokerages, etc. So it is not clear what we can do as public sector agencies to promote this further. It is also difficult to completely eliminate backhauls because flows tend to be unbalanced. In Connecticut, a lot of consumer goods going in and a lot of trash waste going out. Some vehicles, such as fuel tankers, are incompatible with back hauls.
Electronic screening allows trucks to bypass inspection stations. It makes sense economically but is pretty much a drop in the bucket and already being done in most places. Information on efficient routing and truck parking. Probably, most truckers already use some form of navigational aid and may have little ability to change routes. So this is another area where probably the overall potential appears modest from a global perspective. Other strategies like delivery restrictions and queue management at ports may actually have important localized benefits for controlling emissions in city centers or around ports or terminals, but from a regional or global perspective, they may have fairly modest impacts.
Finally, moving to demand management and mode shift. So this table shows the maximum estimated effectiveness, in terms of mode shift percentage, if you implemented this strategy all the way, as well as the cost to the public sector. So infrastructure improvement, such as bringing railroads up to double stack clearance and weight standards and improving access to intermodal terminals, potentially could have a pretty significant impact in some markets, but of course they are also very costly. There are a number of regulatory measures that could encourage mode shift. For example, making tracking more expensive or less attractive. But this might have other undesirable impacts and it is not likely that a lot of them will be implemented since it is re-regulating freight, or size and weight regulations, or hours of service regulations.
Pricing can also have a modest impact as well as generating revenue to pay for other improvements. Fuel costs do make an impact on manufacturers and shipper decisions. But the impact of a unit price increase on freight traffic will be smaller then it will be on passenger traffic because the shipping cost is just one component, usually a lot less than about 10% of the cost of producing a good. So you need a pretty large price signal to really get much of a mode shift.
Short sea shipping is something asked about frequently in coastal environments, and has a lot of challenges including high-cost, low freight density in most markets, and limited congestion along surface roadways. Some tests of it in the U.S. have not had a lot of viability.
One of the key factors in mode shift is what types of commodities you are moving and how far you are using them. So this shows some data that we developed for the National Renewable Energy Laboratory as part of the freight analysis tool, which is available online and can be downloaded for free. It shows the ton miles of commodities moved by distance and also shows the truck and rail mode shares. The truck is in green and the rail is in purple. You can see that, in short distances, truck dominates, and longer distances, rail begins to dominate. You are really looking, in terms of mode shift, you want to isolate commodities that are in the 250 to 1500-mile distance spans because that is where you will get most of the contestable commodities.
There are a number of barriers to achieving mode shift in this category. You need commodities that are low value to weight, so that shippers don't really care about how fast you get them there. In a lot of places, lacking of intermodal infrastructure or circuitous routing, the cost of the drayage at both ends of the trip, other rail costs or spotty service. Just an example again, looking at Connecticut, if you wanted to get freight from the Connecticut coast to the mid-Atlantic coast, you need to route it through Albany, which is really out of the way, or take it on a ferry across Long Island sound. There are no crossings of the Hudson below Albany. Again, this will vary by market of the country. You can use this tool to look at different commodities and look at different distance fans and origins and destinations, if you're interested in looking at freight patterns in your region.
So there are a number of evaluation tools available to look at emissions. Emission factor models from EPA and CARB can look at fleet ages, vehicle types, and certain other strategies. EPA produces a number of resources including the Drayfleet Model, the SmartWay Fleet Performance Model, and the Diesel Emissions Quantifier, that you can use to look at different fleets or different strategies.
U.S. D.O.T. has a Freight Routing and Emissions Analysis Tool that can compare a land-slide and water-side route as well as the Freight Analysis Framework is not specifically for emissions but it is a great data source with freight flows by commodity and mode. I mentioned the Freight Analysis Tool that NREL developed. And finally, for lifecycle emissions, the GREET Fleet Footprint Calculator is a tool available from ANL.
There are also a lot of privately developed tools including carbon footprint calculators that are targeted toward private companies measuring the environmental footprint of their supply chain and logistic practices.
Finally, I just wanted to highlight strategies under control of various parties. National strategies are things the federal government needs to do such as emissions and fuel economy regulations. Regional strategies are things that you need to get together with your neighbors and coordinate on. In the Connecticut example, it wouldn't make much sense just for Connecticut to unilaterally improve its rail infrastructure unless surrounding states did so as well. The state and metro strategies are things that a typical state, air-quality agency, MPO, or Port Authority might be able to implement such as incentive programs for replacing/retrofitting vehicles with cleaner technology, targeted at fleets and individual operators as well as inspection and maintenance programs.
And then carrier-based strategies are things that the private sector can do. These might be especially attractive if they create a net savings in fuel costs. Even if they do, it may require some public sector finance to help cover the upfront cost to help out smaller operators. And there may be ways to pay these back in fuel savings. Finally, just to recap, we are expecting significant decline in criteria pollutants. There are still going be problem areas especially where older trucks are operating around ports, around warehousing distribution hubs, and a lot of opportunities to go in with retrofits and accelerated vehicle retirement to improve emissions in those areas.
Also the growth in freight traffic if it continues as projected, should balance out some of the emissions improvement. In the long run, a lot of uncertainties, in terms of economic structure of production and manufacturing, regulatory issues (is the federal government going to take more action on emissions, on greenhouse gas emissions, on climate change), as well as fuel prices either from a market or policy perspective. Regardless, I think we are going to continue to see challenges in reducing emissions from freight, that we need to address at all scales of policy and decision-making.
That is the end of my presentation. Thank you for listening.
Nicole Coene
Thank you, Chris. I would now like to start off the question-and-answer session with the questions posted online. When we get through those questions, if time allows, we will open up the phone line for questions.
The first question is from Chris. What are the public sector costs you a tribute to the high cost estimate for freight rail improvement?
Chris Porter
That is a good question. Since the railroads are under most generally private entities that manage the rail infrastructure, I think what you might see is if the public sector is looking to accelerate improvement beyond what the private sector is capable of or willing to finance, there might be cost there potentially in terms of either loans or grants for improvements to rail infrastructure. You do see in some cases, states taking control over, for example, short line railroads that provide more local access. So I think the question is: how much does the public sector want to try to accelerate what is being done beyond what the private sector will do itself?
Nicole Coene
Thank you, Chris. Another question for you. What you think about the ongoing shift from direct rail service, to truck service, only sometimes shifting intermodally on the rail network, sometimes door-to-door by truck?
Chris Porter
The issue of intermodal, we certainly have been seeing a lot more of that in recent years, in terms of containers, trucks on flat cars, and that sort of thing. So I certainly think there is more potential in that area. Part of it will depend on the density of the freight movements and if you have enough freight density to support the infrastructure to make that happen. And in areas where you don't have a lot of freight density, it may be hard to develop the intermodal infrastructure. But certainly, we are hoping to see more creative applications of shifting from one mode to another and using lower-cost methods.
Nicole Coene
Thank you. Another question for you, Chris. In evaluating the effectiveness and benefits of rail technology, did you look at idle reduction technology usable with conventional locomotives, like AESS, APUs, shorepower, and yard air?
Chris Porter
That is a good comment for locomotives. Iowa reduction technology is another strategy that should be considered. And I think I neglected to put that on the list. I think that is going to be most effective in areas where you have switching yards. Probably with some of the switching locomotives it will be a lot more effective than with the longer distance movements. But yes, that has been shown to be a fairly cost-effective strategy.
Nicole Coene
Again, thank you, Chris. This question has already been partially addressed by Frank Gallivan but I would like to give the other speakers and opportunity to answer. How, if at all, do your modal outputs present uncertainty in the values?
Mikhail Chester
So uncertainty can result from a variety of different dimensions in the analyses. In my presentation, the first one I showed results that were per ton kilometer for example. When I think about uncertainty, I think about it, in terms of the underlying data that is used, that would be parameter uncertainty. But then as you saw in some of the latter figures, I had scaled things up to what would happen in the future? So that gets us into, for example, scenario uncertainty. Then regarding the first parameter uncertainty, we tend to do a robust uncertainty assessment that considers all sorts of information about the underlying data that we use. It includes the geographical representativeness of the underlying data. So for example, whether or not an emissions factor we are using is for, say Arizona, or if it was developed for another state. If so, we ask: what does that mean? We think about the temporal uncertainty of the data we are using. So generally we are not using data that was developed yesterday. We are using data that were developed maybe a year ago or five years ago or sometimes 10 years ago. And then we ask questions about whether or not the data that we are using are reasonable for what has happened today. Or whether or not there has been innovation that we need to take into account. So, we thinking about the underlying parameter uncertainty in a variety of different ways. We have a structured way of doing developing uncertainty assessments, which ultimately gives us insight into where we focus our efforts to improve data quality. So that gives us the sense of whether or not in our revisions to our lifecycle assessment, we should focus on one factor in our model versus another. And what we tend to do is focus on the pieces that have the highest uncertainty first and the greatest effect on the overall results. We ask: what can we do better? Maybe that is sometimes going out and collecting the data ourselves or doing another round of modeling of the underlying data to improve that.
Nicole Coene
Thank you, Mikael. Another question for you. How do you account for the proportion of infrastructure, roadways, etc., related emissions describable to freight carrying trucks versus on road vehicles?
Mikhail Chester
Terrific question. I glossed over this and did not mention this at all actually. I will tell you how we do it first and then the reasoning for that. The way we do it is, we assign all of the construction impacts of road to cars in the model results that you saw, and then we assign all of the rehabilitation impacts to trucks. The reason for this is because -- damage to roadways is a fourth power relationship per axle road. So essentially the damage done to roads is almost entirely heavy duty vehicles. So the rehabilitation of roads essentially is allocated to trucks whereas roadways are designed capacity wise based largely on passenger cars, which are the bulk of the volume in many places. So that is the reasoning behind that. In the Fisonia 2007 results that I showed, the very first figure that I showed, done in 2007, they had actually allocated impacts of roadways based on the the economic evaluation of the roads between passenger and freight. That leads to something like 60/40 split, between passenger and freight. So that was initially how we thought about it maybe 10 years ago. We have moved more toward allocation mechanisms that include actual measures of damage based on the physical principles of roadway maintenance and rehabilitation.
Nicole Coene
Thank you again. We have a couple of questions related to the use of the tool and who was using it. I wanted to give the speakers an opportunity to discuss that more and what organizations are testing the tool.
John Davies
This is John here. There have been a couple of components, in terms of the process for engaging a practitioner's perspective. As we were developing the tool, we had a sort of informal advisory group that comprised representatives from both MPOs and NDOT. We got a lot of really good input from that group, especially from NYSDOT and Caltrans. We also had a fairly accelerated pilot testing phase for the tool before releasing it. In addition to the advisory group, which include Caltrans, NYSDOT, and Pike's Peak Regional Council, we had offers to pilot test the tool from a couple of additional MPOs, including NCTCOG and SACOG. The most active pilot tester was NCTCOG. They evaluated system level emissions associated with their LRTC.
There are many different ways the tool can be used. You can use it in the simplest way at a region level of state level just for inventory purposes. So you can go from just having inventory of tailpipe emissions from transportation sources. This is sort of the foundation for developing an infrastructure component. This is the kind of strategy that public agencies can have a lot of control over. If you take it a step further, you could use the tool, for instance, to evaluate the greenhouse gas reduction associated with various green construction and maintenance strategies. You could also use it to evaluate some decisions that would be tied to planning. For instance, you could evaluate various scenarios of transportation investment. So if there is a question of how your entire system performs, you could use the carbon estimator to provide an estimate of various greenhouse gases emissions associated with different levels of infrastructure investment. And you can use MOVES to come up with the corresponding vehicle operations estimates. And when you bring those two pieces together, you can compare the overall system performance of the impact of various investments at the system level. FHWA's position has typically been that greenhouse gas analysis is most meaningful at the system-level but if you have an interest in evaluating a project's emission, the tool can be scaled down to evaluate those as well. Another way you can use the tool would be to evaluate various alternatives for NEPA analysis. That would serve as a foundation for providing construction estimates that are very uncommon part of the NEPA work and the state of practice there. I don't know. Jeff and Frank, do you have anything to add?
Jeff Houk
This is Jeff. I will just add that the tool has only been out for a few months. So we are still sort of in outreach mode here. But we expect interest will grow, especially if CEQ finalizes the climate change guidance and we to start looking at greenhouse gas impacts on more projects. Abby asked a question in the chat pod that I will address. She asked if the construction dollar costs factor into the 10 month payback period for the case study. This is just an emissions payback exercise, not a cost payback exercise. But you do bring up an interesting concept, that is, that when we reviewed climate action plans from different states, many of them have a transportation component where the climate action plan will say: we are going to build more transit or HOV lanes or bike paths. The plans calculate the benefits of those strategies but none of the plans account for the emissions required to build the stuff in the first place. The context of greenhouse gas emissions, where the construction emissions impact the atmosphere just as much as the vehicle emissions, it is really important to consider those construction impacts. So if you're looking at two different strategies that can reduce greenhouse gas emissions by 1000 tons per year and one of them has a 10 month payback period and another has a ten-year payback period, that can be useful information to the people that are deciding which strategies to fund.
Nicole Coene
Thank you, Jeff. It looks like most of the questions have been answered at this point. You gentlemen want to expand upon anything that you answered in the chat earlier?
If not, I will go ahead and open the phone lines for questions.
Operator
If you would like to ask a telephone question at this time, please press *1 on your telephone keypad. We will pause for a moment to compile the Q&A roster.
Again, if you would like to ask a question, press *1 on your telephone keypad.
There are no telephone questions at this time. I will turn the call back over to Ms. Coene.
Nicole Coene
Thank you. That concludes the seminar for today. Thanks for attending. The recorded version will be available within the next few weeks on the Talking Freight website. The next seminar will be held on May 20th and the topic is TBD. Registration is not yet available, but I will send a notice out through the freight planning listserv announcing the topic and the availability of registration. I encourage you to join the freight planning listserv if you have not already done so.
Operator
This concludes today's conference call. You may now disconnect.
