Assessing the Energy and Greenhouse Gas Emissions Mitigation Effectiveness of Modal Freight Policies
Total transportation accounts for approximately 28 trillion MJ of energy consumption, emits 2 billion tons of Carbon Dioxide equivalents, and costs an estimated $300 billion. My research estimates the total embodied modal freight requirements across the supply chain for each of over 400 sectors using BTS Commodity Flow Survey data and BEA economic input-output tables. Across all sectors, domestic truck and rail are similar in magnitude for embodied freight transportation of goods and services in terms of ton-km. However, they differ significantly in energy consumption, greenhouse gas emissions, and costs per ton-km. Recent pressure to reduce energy consumption and emissions has resulted in exploring how to make more efficient freight mode choices. One solution would be to shift freight transportation away from modes that require more energy and emit more (e.g., truck) to modes with that consume and emit less (e.g., rail and water). My research analyzes various scenarios to determine the overall supply chain resul ts of individual sectors reducing the effects of their truck transportation.
Four scenarios are analyzed: (1) shifting all truck to rail, (2) selecting the top 20% of sectors to shift based on emissions, (3) selecting the top 20% of sectors to shift based on emissions and availability to shift from truck, and (4) increasing truck efficiency. The sectors selected to shift away from truck are back to their 1992 mode share percentages. Increasing truck efficiency by 10% results in similar energy and emissions reductions (approximately 9% for energy and 5% for emissions) as targeting the top 20% sectors when selected based on emissions. However, selecting the top 20% based on availability to shift from truck results in slightly less reductions of energy and emissions. There are no individual sectors for which targeting would significantly decrease the total freight transportation energy and emissions.
From Food-miles to Passenger-miles: Emissions related to consumer grocery travel behavior
For this research we have acquired actual data on food purchases at the individual item transaction level, similar to information individuals receive on a receipt from the grocery store. The amount of money spent at the grocery store is valuable in assessing the supply chain impacts using EIO-LCA. Our data is for roughly one million loyalty card holders in the greater Pittsburgh region, which we can connect to unique households. The goal of this research is to (a) see how the detailed “real data” compares to the CES data above, (b) understand in greater detail the variation in purchase impacts (in dollars, energy, and carbon) of food per household, (c) use demographic data available to estimate the distance traveled to the grocery store and how the energy/carbon impacts of that trip compare to the embedded energy/carbon in the food purchases, and (d) estimate the appropriate “overhead” of driving on a per-food product purchased basis. We expect this study to mark the beginning of a new field of study related to coordinating actual consumer data flows to create dynamic and interactive visualizations. This work will also contribute to research relating impacts from overall and individual product transportation.
Knowing the indirect or ‘embodied’ transportation needed for the manufacturing and distribution of products is helpful when considering transportation costs, impacts, and potential supply chain vulnerabilities. Over the past decades, the movement of goods and the logistics networks supporting them have become faster, more complex, and more important in the US economy. Transportation also represents a major investment of public funds, with over $45 Billion invested in transportation infrastructure by the federal government in 2006 and a comparable amount by state and local agencies. Understanding the preferred methods of transportation, magnitude of commodities shipped, and kinds of commodities shipped can provide insight into the resiliency and future infrastructure improvements necessary to create more reliable and efficient methods of transportation.
We use an input-output based transportation model to estimate the modal freight requirements, direct and indirect ton-km, for 428 sectors of the U.S. economy. Indirect transportation is the transportation that occurs in the supply chain, while direct transportation is the allocated and normalized freight data from the 2002 Commodity Flow Survey. Across all economic sectors, pipeline direct transport is only evident in one aggregate sector (petroleum), but is significant in the supply chain of the remaining 27 aggregate sectors. Domestic truck and rail are similar in magnitude for both direct and indirect transportation of goods and services. International water and air freight transport are also included in the model and approximately double the magnitude of freight transport for each sector, with only a small percentage of commodities being transported by international air.
The sector with the largest embodied freight transportation in consumption is final petroleum products followed by government services, construction, and food products. Overall, pipeline contributes 6% to the total profile of freight movement per sector, air transport is generally under 1%, and rail and truck transportation are the most dominant modes (10% each) for domestic freight transportation for the average sector. International water is the largest mode (50%) even compared to domestic modes, and international air contributes less than 1%.
This model estimates the ton-km required for each sector in the US economy which is important in understanding freight movement, modal resiliency, the relative dependence on transportation and transportation choices for various products and services. Relating this estimated freight movement to energy requirements and greenhouse gas emissions may lead to more effective policies in reducing the freight transportation emissions.
Advances in information and communication technology have provided consumers the option of shopping on-line instead of driving to a traditional retail store for many products. E-commerce has now grown from less than 1% of retail sales in 2000 to 3% in 2008. The alternative retail channels have some distinct differences with regard to environmental costs, including overstock inventory, physical store space, and consumer transport in traditional retail stores and individual packaging and last mile delivery for e-commerce. We build on prior comparative research and conduct a streamlined Life Cycle Assessment (LCA) to quantify variations in energy use and carbon dioxide (CO2) emissions for the alternative systems using data received from the e-commerce industry for an electronic product. This report reviews our assumptions and analysis and provides conclusions and offer recommendations to decrease logistics LCA uncertainties.
The major differences between the traditional retail model and the e-commerce model are the transportation from the warehouses to the retail store or the distribution center, data center energy usage, individual vs. bulk packaging, and the transportation, from the store or distribution center to the consumer, often called the “last mile” of delivery. These differences vary in energy usage and intensity. The objective of our work was to assess which network uses less energy and produces less greenhouse gas carbon dioxide (CO2) equivalent emissions.
Our results confirm prior findings that e-commerce delivery uses less primary energy and produces less CO2 emissions than traditional retailing. Considering retail and e-commerce logistics differences, the three largest contributors were customer transport, packaging, and last mile delivery. Customer transport encompassed approximately 65% of the traditional retail primary energy expenditures and CO2 equivalent emissions on average. For e-commerce, packaging and last mile delivery were responsible for approximately 22% and 32% of the e-commerce energy usage, respectively. Overall, e-commerce had 35% lower energy consumption and CO2 emissions compared to traditional retail using calculated mean values.
There was significant uncertainty and variability in many of the numbers used in the analysis, particularly in terms of customer transport to the retail store (i.e., fuel economy, trip length, purposes per trip, etc). We used Monte Carlo simulations and scenario analysis to estimate that e-commerce being the less energy-consumptive option approximately 70 % of the time with average delivery logistics and 50% with air-only delivery logistics for e-commerce. To make the LCA transportation model more robust, actual data from a traditional retail business and more detailed information on consumer shopping behaviors are necessary.
Identification of Total Coliform, E. coli, and Enterococci and Variability in the Asnebumskit Watershed Through Data Collection and Modeling (UMass--Amherst)
This project aims to determine spatial and temporal trends in bacteria loading along the Asnebumskit Brook that are a concern due to previously high fecal coliform counts in the area. Water quality testing was done on multiple sites along the Asnebumskit Brook for total coliform, E. coli, and Enterococci. The watershed is comprised of primarily residential homes where high bacteria counts are not expected. A unique model was created and implemented to assess the bacteria loading in the watershed, specifically to target what kind of influence rainfall and runoff have on the bacteria loading of the Asnebumskit Brook. The model is based on a plug flow of the contaminants as they travel in runoff over land, through the storm-sewer system, and in the brook itself with relation to temperature, distance, and time. When compared to the data collected, the model aids in evaluating the possible sources of high bacteria counts within the Asnebumskit Brook, including sediments, domestic animals, wildlife, and storm-sewers. Additionally, ribotyping, an in-depth analysis of the E. coli collected as compared to a known database of E. coli sources, was done to further determine the sources of the E. coli within the Asnebumskit Brook. These findings can be used to implement rules and regulations on land use adjacent to important water sources if it is determined that one or more of these non-point sources are significant contributors to poor water quality.