 

ANALYSIS OF FUEL ETHANOL TRANSPORTATION ACTIVITY AND POTENTIAL
DISTRIBUTION CONSTRAINTS

Bruce Peterson

Shih-Miao Chin

Sujit Das

Energy and Transportation Science Division

Oak Ridge National Laboratory

Prepared for 

Office of Transportation and Air Quality

U. S. Environmental Protection Agency

March 2009

OAK RIDGE NATIONAL LABORATORY

Oak Ridge, Tennessee 37831

managed and operated by

UT-Battelle, LLC

for the 

U.S. DEPARTMENT OF ENERGY

under contract No. DE-AC05-00OR22725



TABLE OF CONTENTS

1. INTRODUCTION		1

2. SCENARIOS DESCRIPTION	4

3. DATA PREPARATION	5

	3.1 Corn Ethanol Refineries	5

	3.2 Cellulosic Ethanol Refineries	5

	3.3 Ethanol Imports	7

		3.3.1 Ports of Entry	8

		3.3.2 Geographic Information for Ports	8

	3.4 Ethanol Demand	10

	3.5 Gasoline Fuel Terminals	17

	3.6 Ethanol Intermodal Terminals	19

	3.7 Transportation Cost	23

		3.7.1 Rail	23

		3.7.2 Waterways	24

		3.7.3 Truck	25

	3.8 Congestion/Delay	25

		3.8.1 Rail	26

		3.8.2 Waterway	26

		3.8.3 Highway	30

4. GENERATION OF ETHANOL TRANSPORTATION ACTIVITY DATA – 

	MODELING APPROACH	33

	4.1 Origin –Destination Matrix (ODM)	33

	4.2 Traffic Assignment	36

5. RESULTS			40

	5.1 Transportation Activity	40

	5.2 Distribution Cost	51

	5.3 Rolling Stock Requirements	53

	5.4 Distribution Constraint Analysis	55

6. SUMMARY		70

APPENDIX A Location Details and Production Capacity (MGY) of Corn 

		Ethanol Plants for Various Forecast Year	74

APPENDIX B Projected Corn Ethanol Production	82

APPENDIX C Location Details and Production Capacity (MGY) of Cellulosic 

		Ethanol Plants for Various Forecast Years	85

APPENDIX D Projected Cellulosic Ethanol Production 	92



TABLES

Table 1. Annual Ethanol Projections (in billion gallons) under Various
Scenarios	4

Table 2. Projected U.S. Ethanol Imports (billion gallons per year)	8

Table 3. Ports of Entry and Projected Imported Ethanol Volumes (MGY)	9

Table 4. State-Level Annual Ethanol Fuel Demand Projection (in million
gallons)	11

Table 5. Commodity Movements (in thousand tons) by Waterways, Based on
FAF2	27

Table 6. Estimated Future Commodity Flow (in thousand tons) and Growth 

	   Factors for this Study	27

Table 7. Estimated Future Growth Rates for Lock Delay	28

Table 8. Future Growth Factors for 2006 ADT/Lane	32

Table 9. Transportation Activity Summary for the 2022 Scenarios	41

Table 10. A Sample of High-Volume Ethanol Movements by Different Modes 

		for the Three 2022 Scenarios	42

Table 11. Estimated Ethanol Distribution Cost for the Three 2022
Scenarios	52

Table 12. Assumptions Used for Rolling Stock Requirements Estimation	54

Table 13. Estimated Rolling Stock Requirements under 2022 Scenarios	55

Table 14. Estimated Number of Miles Experiencing an Increase in Ethanol
Traffic 

		of Various Categories of Quantity Level under “2012B Scenario”	58

Table 15. Estimated Number of Miles Experiencing an Increase in Ethanol
Traffic 

		of Various Categories of Quantity Level under “2014B Scenario”	59

Table 16. Estimated Number of Miles Experiencing an Increase in Ethanol
Traffic 

		of Various Categories of Quantity Level under “2022B Scenario”	60

Table 17. Estimated Number of Miles Experiencing an Increase in Ethanol
Traffic 

		of Various Categories of Quantity Level under “2022 Controlled”	61



FIGURES

Figure 1. Projected corn ethanol refineries in 2017 and 2022	6

Figure 2. Cellulosic ethanol refineries in 2022	7

Figure 3. Ports of entry for ethanol fuel in 2022	10

Figure 4. VMT distribution at census tract level for Hawaii	12

Figure 5. The Voronoi diagram for the state of Hawaii	13

Figure 6. Ethanol demand at the terminal level for 2010	14

Figure 7. Ethanol demand at the terminal level for 2012	14

Figure 8. Ethanol demand at the terminal level for 2014	15

Figure 9. Ethanol demand at the terminal level for 2017	15

Figure 10. Ethanol demand at the terminal level for 2022	16

Figure 11. Ethanol demand at the terminal level for 2022 AEO Reference
Case	16

Figure 12. Gasoline and/or ethanol terminals	17

Figure 13. Gasoline and/ore ethanol total storage capacity	18

Figure 14. Ethanol rail flows, 2006	21

Figure 15. Distribution of potential ethanol distribution terminals	23

Figure 16. Current train volumes compared to current train capacity	26

Figure 17. Average tow delays for 2007	29

Figure 18. Average tow delays for 2022	29

Figure 19. Average daily traffic per lane trends for rural U.S. 	31

Figure 20. Average daily traffic per land trends for urban U.S.	32

Figure 21. Projected ethanol movements (Ktons) under the 2022 RFS1
Reference 

		scenario	43

Figure 22. Projected ethanol movements (Ktons) under the 2022 AEO
Reference 

		scenario	44

Figure 23. Projected ethanol movements (Ktons) under the 2022 Controlled


		scenario	45

Figure 24. Projected county-level total Kton-mile ethanol movements
under 

		the 2022 RFS1 Reference Scenario	46

Figure 25. Projected county-level total Kton-mile ethanol movements
under 

		the 2022 RFS1 AEO Reference Scenario	47

Figure 26 Projected county-level total Kton-mile ethanol movements under


		the 2022 Controlled Scenario	48

Figure 27. Projected county-level total ethanol loading/unloading
activities

		 (Ktons) under 2022 RFS1 Reference Scenario	49

Figure 28 Projected county-level total ethanol loading/unloading
activities

		 (Ktons) under 2022 AEOference Scenario	50

Figure 29 Projected county-level total ethanol loading/unloading
activities

		 (Ktons) under 2022 Controlled Scenario	51

Figure 30. Projected rail ethanol movements in terms of Ktons and %
increase in 

		base flows for 2012B scenario	62

Figure 31. Projected waterways and highways combined ethanol movements
in 

		terms of Ktons and % increase in  base flows for 2012B scenario	63

Figure 32. Projected rail ethanol movements in terms of Ktons and %
increase in 

		base flows for 2014B scenario	64

Figure 33. Projected waterways and highways combined ethanol movements
in 

		terms of Ktons and % increase in  base flows for 2014B scenario	65

Figure 34. Projected rail ethanol movements in terms of Ktons and %
increase in 

		base flows for 2022 scenario	66

Figure 35. Projected waterways and highways combined ethanol movements
in 

		terms of Ktons and % increase in  base flows for 2022 scenario	67

Figure 36. Projected rail ethanol movements in terms of Ktons and %
increase in 

		base flows for 2022B scenario	68

Figure 37. Projected waterways and highways combined ethanol movements
in 

		terms of Ktons and % increase in  base flows for 2022B scenario	69



Analysis of Fuel Ethanol Transportation Activity and 

Potential Distribution Constraints

1. INTRODUCTION

The Energy Independence and Security Act (EISA) of 2007 mandates that
the Environmental Protection Agency (EPA) establish regulations to
ensure the use of 36 billion gallons of renewable fuel by 2022. EISA
established renewable fuel categories and eligibility requirements,
including setting mandatory greenhouse gas (GHG) reduction thresholds
for the various categories of fuels. Of the total 36 billion gallons of
renewable fuel use required by 2022, EISA requires that 21 billion
gallons be advanced biofuels. This limits the volume of ethanol produced
from grain-based feedstocks that can be used to satisfy the requirement
of the EISA to no more than 15 billion gallons by 2022. EISA further
requires that of the total 36 billion gallons of renewable fuel use
required by 2022, 16 billion gallons be cellulosic biofuel and 1 billion
gallon be biomass-based diesel fuel.

EPA estimated that 34.14 of the 36 billion gallons or renewable fuels
required by 2022 would be ethanol, of which 15 billion gallons would be
corn ethanol, 16 billion gallons would be cellulosic ethanol, and 3.14
billion gallons would be imported ethanol. EPA projected that the
balance of the renewable fuel volume required by EISA in 2022 would be
made up of 0.81 billion gallons of fatty acid methyl ester (FAME)
biodiesel, 0.19 billion gallons of non-co-processed renewable diesel,
and 0.19 billion gallons of co-processed renewable diesel. Ethanol
currently cannot typically be shipped by pipeline due to concerns over
materials compatibility and product contamination. Product contamination
and other concerns currently also has prevented widespread shipment of
FAME-based biodiesel by pipeline. In addition, the location of ethanol
and to a lesser extent FAME biodiesel plants is typically dictated by
the need to be close to the source of the feedstocks used rather than to
fuel demand centers or to take advantage of the existing petroleum fuel
pipeline distribution system. EPA projected that renewable diesel fuel
that is co-processed at petroleum refineries would be distributed
fungibly with petroleum-based diesel fuel and thus be fundamentally
transparent with respect to its handling in the fuel distribution
system. Renewable diesel fuel that is not co-processed at refineries may
need to be transported by rail, barge, and tank truck to the terminals
because the points of production may not be in proximity to the existing
petroleum fuel distribution system. 

Projected biofuels production increases under the EISA will likely spur
demand for transportation services that face capacity constraints. The
extent of infrastructure incompatibilities that may result due to a
shift from the current petroleum pipeline system to rail-based
transportation of ethanol is not fully known. Incongruity between the
geographic locations of production facilities and feedstock supply
networks and the locations of the petroleum distribution system and
consumption demand centers would require adequate infrastructure
development to avoid serious logistical bottlenecks and systemic delays
in biofuels distribution. The country is lacking the infrastructure to
most efficiently transport liquid biofuels that are produced primarily
in the midwest and plains states to population centers in the East and
elsewhere that account for 80% of the U.S. population. It is estimated
that about one-third to one-half of domestic ethanol produced today
travels through Chicago, which is already proving to be a “choke
point.” Several new primary gathering points, unit train loading
facilities, barge receiving and transloading capabilities are either
being built or planned as ethanol demand continues to grow. 

Today, approximately 60 percent of all corn-based ethanol is transported
by the U.S. rail system because the corrosive properties of ethanol
prevent transporting the product by pipeline. Trucks transport 30
percent of the ethanol, and barges transport the remaining 10 percent.
The rail infrastructure for transporting ethanol from biorefineries is
not well developed, and being a major mode of biofuels transportation
requires adequate infrastructure investment for efficient distribution.
A lack of excess capacity increases the sensitivity of the
transportation system to sudden changes in transportation demand and
distribution patterns. Water-borne transportation via barge and ship
faces limitations under the Jones Act, while truck transportation faces
capacity issues, labor shortages, and public opposition to increased
truck traffic on the nation’s highway system. Transportation is
typically the third largest expense to an ethanol producer—after
feedstock and energy. Spatial variation in infrastructure development,
especially refinery location, will influence transportation costs of
renewable fuels. Transportation costs, along with feedstock costs,
production technologies, and economies of scale of refineries influence
biofuels’ competitiveness. It is important to assess how increased
biofuels production levels would change freight flow patterns and
related transportation costs, which are influenced by many factors
including rail accessibility, railcar price differentiation (unit trains
vs. single cars), and network connectivity; rail capacity and
congestion; highway accessibility and congestion; proximity to
waterways; plant location and transportation infrastructure capacity;
and modal competition.

At current levels of shipments, biofuels represent less than 1% of total
railcar ton-miles shipped in the United States.1 It is anticipated that
renewable fuels are unlikely to have more than marginal impact on rail
capacity or congestion. However, the rail system itself is subject to
increasing capacity and congestion constraints that will impact all
commodities shipped by rails, including biofuels. A lack of rail
accessibility at biorefinery sites, the type of rail infrastructure
accessible to biofuels producers, the current state of capacity and
congestion on the U.S. rail network, and logistics and supply-chain
management capabilities at biofuels producers are some of the issues
related to the U.S. rail system that need to be addressed.

A recent study evaluated the impact of ethanol production on grain
transportation under two scenarios, i.e., annual ethanol production
increases to 15 and 20 billion gallons by 2016.2 Transportation impacts
varied under each scenario due in part to modal share differences. It is
estimated that rail and barge demand could decrease if corn exports
decrease, but in the short-term increased ethanol and by-product
shipments could offset decreases in rail grain shipments. However, truck
transportation, the dominant mode of conveying corn to ethanol plants,
would increase by about 104% by 2016 under all scenarios. Similarly, an
earlier study looking at the increased infrastructure needs and
transportation requirements for ethanol production and use of 5 billion
gallons per year (BGY) by 2012 that would result from a “Renewable
Fuels Standard” has found negligible impacts on the transportation
infrastructure. The study found that additional movements by barge,
rail, and truck represent a very small amount of total movements (less
than 1%) and would have no major impact on these transportation sectors.

The objective of the study documented in this report was to conduct an
analysis of ethanol transport by domestic truck, marine, and rail
distribution systems from ethanol refineries to blending terminals using
ORNL’s North American Transportation Infrastructure Network Model
(hereafter referred to as the “Infrastructure Network Model”). As
noted above, the ethanol (the biofuel on which this study focuses) that
EPA projects will be used as a result of the EISA cannot currently be
transported from the point of production to the terminal-level using the
existing petroleum distribution system. As a result, ethanol must be
transported to terminals by truck, marine, and/or rail. As part of the
rulemaking to implement the EISA, EPA must estimate the increased
emissions and costs associated with the transportation of ethanol. To
estimate these impacts, EPA needs transportation activity data for
domestic truck, marine, and rail modes of transportation employed to
convey ethanol from the point of production to the terminal-level on a
county-by-county basis. In addition, EPA must assess the potential
constraints that may exist in the distribution of ethanol under
EISA—hereafter referred to as distribution constraint analysis. An
infrastructure analysis is necessary to identify the potential location
and timing of such constraints and the extent to which the constraints
may represent a barrier to the use of volumes of ethanol that EPA
projects will be used in response to the EISA. Both of these EPA
analytical issues are addressed by ORNL Infrastructure Network Model
based on the projected major inputs such as estimated location of
production facilities, sources of imports, and areas of consumption for
annual volumes of ethanol that EPA estimates will be consumed in
response to the EISA. The following sections at first discuss the major
input data preparation, followed by the estimation procedure and results
obtained.

The projections in this study are a representation of what might happen
given the specific premises, assumptions, and methodologies used. Real
world data were used as a starting point. Such information includes
historic gasoline/ethanol blend sales, current ethanol production, and
actual product terminal locations. However, the actual projections of
the locations of new ethanol production facilities and future ethanol
sales increases are highly dependent on the premises, assumptions, and
methodologies used.

2. SCENARIOS DESCRIPTION

Seven scenarios were considered across five projection years (i.e.,
2010, 2012, 2014, 2017, and 2022) using the EPA assumptions for major
parameters based on EISA, as shown in Table 1. For the detailed
county-by-county transportation activity data, three 2022 scenarios have
been the focus, whereas five of the seven scenarios are considered for
the distribution constraint analyses. Detailed, specific-refinery-level
production-capacity data were provided by EPA for both corn and
cellulosic ethanol. Estimates of ethanol production under the
“Controlled” case reflect the EISA supply requirements, whereas the
“Reference” cases indicate a substantially lower level of supply,
particularly for cellulosic ethanol, in the absence of EISA. The two
2022 reference cases differ with respect to assumptions about imports
and demand. Annual demand estimates were provided by EPA at the state
level and were mapped to specific existing product terminal locations.
Under controlled cases, average ethanol content is assumed to increase
from 8.22% in 2010 to 22.19% in 2022. The average ethanol content under
the 2022 AEO reference case ethanol content is assumed to be 8.96%, and
under the 2022 RFS1 reference case it is assumed to be 4.6%. Similarly,
total ethanol imports data was provided at the level of specific
domestic ports by using the historical import data and potential
state-level import estimates, which were taken into consideration in the
final analysis by allocating these imports to specific product
terminals. 

Table 1. Annual Ethanol Projections (in billion gallons) under Various
Scenarios

	2010	2012	2014	2017	2022	2022

	Controlled	AEO

Reference 	RFS1

Reference

Supply

 -Corn

 -Cellulosic	

11.6

0.10	

12.9

0.5	

14.4

1.75	

15.0

5.5	

15.0

16.039	

12.29

0.25	

6.69

0.0

Imports	0.29	0.18	0.36	1.78	3.14	0.64	0.0

Demand

 -w/o Biodiesel

 -With Biodiesel



11.9	13.62	16.5	22.28	34.14	13.18	6.69

	12.59	14.62	17.51	23.28	35.14	NA	NA



Four scenarios are considered in the distribution constraint analysis.
Additional demand for fatty acid methyl esters (FAME) biodiesel and
non-co-processed renewable diesel has been taken into consideration
under these scenarios by treating the volumes of these fuels as
additional ethanol volume.  The projected biodiesel and non-co-processed
renewable diesel ranged from 3 to 7% of the ethanol demand. The
scenarios considered in the constraint analysis are the controlled cases
with simulated biodiesel/non-co-processed demand for 2012, 2014, and
2022 and the controlled case without simulated
biodiesel/non-co-processed demand for 2022. The remaining years have not
been considered for the analysis since they do not provide any
additional information than the extreme cases already considered for the
analysis. Information on supply in Table 1 is based on ethanol demand
without additional biodiesel/non-co-processed renewable diesel demand,
and so to simulate the additional biodesel/ non-co-processed renewable
diesel demand specific supply sources and imports have been increased
proportionately. The following section discusses in detail the
methodology used in to prepare the EPA data and the other necessary data
used in the analysis.

3. DATA PREPARATION

3.1 Corn Ethanol Refineries

EPA furnished the corn ethanol refinery list and refinery production
outputs for the years 2010, 2012, 2014, 2017, and 2022. For the 2022 AEO
reference case of corn ethanol production outputs, EPA assumed that
necessary small volume of cellulosic ethanol and imports under this case
would be produced and co-located at corn ethanol plants. The RFS1
reference case does not include cellulosic ethanol production.

Based on the information within the corn ethanol refinery list, the
location of each refinery is determined by its latitude and longitude.
The geo-location is estimated by searching for refinery address
information on the internet and geo-coding the address using Google
Earth. GIS-based shape files have been generated for the corn ethanol
refineries. The corn-based ethanol production output information for the
years 2010, 2012, 2014, 2017 and 2022 is then added to the shape
file’s attribute table.

A thematic map of the corn ethanol refineries with their 2017 and
controlled 2022 output (annual facility capacity shown in terms of
million gallons) is depicted in Figure 1 since both years have the same
total maximum corn ethanol supply volume of 15 BGY. As shown in this
figure, the largest share of corn ethanol supply is projected to come
from the Midwest region as it is today. Detailed location information of
corn ethanol plants and maps for other forecast years are shown in
Appendix A and B, respectively.

3.2 Cellulosic Ethanol Refineries

The cellulosic ethanol refinery list and refinery production outputs for
the years 2010, 2012, 2014, 2017, and 2022 were furnished by the EPA.
EPA did not identify locations of the proposed refineries by longitude
and latitude, but by county and state. 

ORNL used a geographic information system (GIS) tool to locate potential
refinery sites within the counties. ORNL assembled street, rail, and
waterway map layers. Earth Science Research Institute’s (ESRI)
StreetMap USA is used for the highway/street network. The US Army Corps
of Engineers’ Inland and Costal Waterway Network dataset prepared by
the Navigation Data Center is used for the waterway network map layer.
The Railway map layer published by the Center for Transportation
Analysis, Oak Ridge National Laboratory is used for the railroad
network. 

Figure 1. Projected corn ethanol refineries in 2017 and 2022

In addition to transportation infrastructure networks for street, rail,
and waterway, the following map layers are also used to help locate the
cellulosic plants. A terrain map layer, i.e., the shaded relief based on
US Geological Survey (USGS) National Elevation Dataset (NED) is used.
This map layer ensures that cellulosic plants will not be located in
areas with terrain not suitable for industrial development. An aerial
imagery map layer, the I-Cubed Nationwide Prime 1-meter or better
resolution imagery for the contiguous United States, is also used under
this task. This map layer enables transportation analysts to visualize
the existing potential locations to build cellulosic plants.

There is no formal algorithm developed to locate new ethanol plants.
Locations are assigned using engineering judgment on a case-by-case
basis. The following basic principles, listed in order of priority, are
employed. 

If the county already has a corn-based ethanol refinery, the cellulosic
refinery is to be located close to the corn-based ethanol refinery.

If the county has more than one corn-based ethanol refinery, the
cellulosic ethanol refinery is to be located close to any one of the
corn-based ethanol refineries.

If the county has access to both railroad and waterway, efforts are made
to locate the cellulosic ethanol refinery close to both rail and
waterway if possible. Otherwise, the cellulosic ethanol refinery is to
be located close to the railroad.

If the county has access to rail only, the cellulosic ethanol refinery
is to be located close to the railroad.

If the county has access to waterway only, the cellulosic ethanol
refinery is to be located close to the waterway.

If the county does not have access to rail or waterway, the cellulosic
refinery is to be located on the outskirts of a small town.

GIS-based shape files have been generated for the cellulosic ethanol
refineries. The cellulosic ethanol production output information for the
years 2010, 2012, 2014, 2017 and 2022 is added to the shape file’s
attribute table. Appendix C shows the detailed location and production
capacity of 181 cellulosic plants considered for the various forecast
years in this analysis, including the location of new cellulosic ethanol
plants close to any specific transportation mode. As shown in this
table, most cellulosic ethanol plants are located in proximity to rail.

A thematic map of the cellulosic ethanol refineries with their 2022
output is depicted in Figure 2, whereas maps of other years are shown in
Appendix D. Unlike the case of corn ethanol plants, the location of
cellulosic ethanol plants is somewhat distributed, with several plants
located in the northeast, southeast, and west coast regions.

Figure 2. Cellulosic ethanol refineries in 2022

3.3 Ethanol Imports

EPA provided estimates of the total amount of ethanol likely to be
imported as a motor fuel from 2010 to 2022. This supply was first
allocated among three U.S. regions, i.e., California/West Coast, Hawaii,
and East/Gulf Coast based on the 2006 Energy Information Administration
(EIA) import locations. The imports, in terms of billion gallons per
year, from Brazil and the Caribbean Basin Initiative (CBI) countries,
are listed in Table 2 along with domestic production estimates. The
relatively small volume of ethanol imports under the AEO 2022 reference
case has been assumed to originate from corn ethanol plants to simplify
the analysis, whereas no imports have been assumed under the RFS1 2022
scenario.

Table 2. Projected U.S. Ethanol Imports (billion gallons per year)



Year	Total Domestic Ethanol	Total

Ethanol	Total

Imports	From Brazil

	From CBI



2009	9.85	10.35	0.50	0.00	0.50

2010	11.65	11.94	0.29	0.00	0.29

2011	12.54	12.70	0.16	0.00	0.16

2012	13.44	13.62	0.18	0.00	0.18

2013	14.75	14.94	0.19	0.00	0.19

2014	16.15	16.51	0.36	0.00	0.36

2015	18.00	18.83	0.83	0.00	0.83

2016	19.25	20.56	1.31	0.18	1.13

2017	20.50	22.28	1.78	0.57	1.21

2018	22.00	24.25	2.25	0.96	1.29

2019	23.50	26.22	2.72	1.33	1.39

2020	25.50	28.20	2.70	1.22	1.48

2021	28.50	31.17	2.67	1.06	1.61

2022	31.00	34.14	3.14	1.34	1.80



3.3.1 Ports of Entry

EPA provided the information about ports of entry. The ports were
identified by reviewing the EIA historical ports of entry for imports of
ethanol and gasoline. After estimating the total ethanol imports demand
by state based on its share of total regional demand, total state level
imports demand was then allocated among various ports within that state
based on the vehicle-miles-traveled. Two ports, i.e., Hampton Roads in
Virginia and Port of Baltimore in Maryland have also been included as
potential ethanol import ports although these currently do not receive
any gasoline imports due to being served by pipelines. The projected
imports, in terms of million gallons per year, at these ports of entry
for years 2010, 2012, 2014, 2017 and 2022 are listed in Table 3.

3.3.2 Geographic Information for Ports

The geographic location information for ports of entry is based on the
U.S. Army Corps of Engineers, Navigation Data Center’s Principal Ports
of the United States database.  A geographic information system (GIS)
based shape file has been created for port of entry locations. The ports
of entry and their ethanol imports in 2022 are displayed in the Figure
3.

Table 3. Ports of Entry and Projected Imported Ethanol Volumes (MGY)

Port Name	2010

	2012

	2014

	2017

	2022



Seattle, WA	13.85	7.36	17.34	64.92	73.26

Vancouver, WA	2.54	1.35	3.18	13.81	15.58

Portland, OR	0.00	0.00	0.00	0.00	13.13

San Francisco, CA	4.23	2.69	5.20	15.70	25.45

Selby, CA	10.06	6.39	12.36	49.87	80.81

Los Angeles, CA	75.01	47.69	92.23	332.84	539.40

Honolulu / Pearl Harbor, HI	2.78	1.84	4.33	15.30	19.36

Belfast / Searsport, ME	0.29	0.19	0.42	2.76	0.21

Portland, ME	2.00	1.33	2.85	15.96	1.20

Boston, MA	8.85	5.88	12.58	74.27	125.65

Providence, RI	1.46	0.97	2.08	12.38	20.94

New Haven, CT	2.40	1.60	3.42	18.92	32.01

Bridgeport, CT	2.67	1.78	3.80	22.04	37.30

New York, NY	16.06	9.72	20.90	115.21	263.38

Newark, NJ	3.95	2.63	5.62	29.24	49.47

Paulsboro / Gloucester, NJ	2.12	1.41	3.02	20.52	34.72

Perth Amboy, NJ	6.29	4.18	8.94	53.22	90.04

Philadelphia, PA	16.09	9.94	21.28	120.58	216.19

Wilmington, NC	12.62	8.38	14.01	81.51	134.48

Charleston, SC	7.26	4.82	10.32	32.99	69.17

Savannah, GA	15.88	7.43	12.53	73.73	173.90

Jacksonville, FL	7.35	4.71	9.39	56.79	93.28

Pt Canaveral, FL	4.43	2.83	5.65	34.50	56.68

Pt Everglade, FL	11.64	7.45	14.86	95.98	157.65

Tampa, FL	8.27	5.29	10.56	65.11	106.95

Port Arthur, TX	1.74	1.02	2.08	10.16	21.45

Houston, TX	26.35	15.45	31.46	177.21	373.97

Brownsville, TX	2.10	1.23	2.51	14.34	30.26

Hampton Roads, VA	12.66	8.41	14.24	83.14	153.85

Baltimore, MD	9.04	6.01	12.85	76.99	130.25



 

Figure 3. Ports of entry for ethanol fuel in 2022

3.4 Ethanol Demand

EPA provided the state-level ethanol demand for the 2010, 2012, 2014,
2017, 2022 controlled and 2022 reference cases (as shown in Table 4).
The data have been disaggregated at the level of existing petroleum
product terminals.

The vehicle-miles-traveled (VMT) at census tract level is estimated
based on the ORNL-developed Transferability Travel Rates, and their
distribution at the census tract level is assumed to be the same for all
scenarios. The U.S. Department of Transportation sponsors and ORNL
conducts a recurring survey of traffic statistics including VMT. The
data are too aggregate in nature to provide any reliable estimate within
small geographic areas such as Census tract and transportation analysis
zone, i.e., from sample households to geographic areas. An alternative
statistical method was developed using the original data in conjunction
with other data sources to transfer estimated travel rates from sample
households geographic areas with similar socioeconomic and demographic
characteristics. Travel estimates using the new method were found to be
second best only to travel parameters estimated with data from
locale-specific household surveys. The state-level ethanol demand for
various future years is shared to the census tract level. A thematic map
of VMT distribution at census tract level for the state of Hawaii is
depicted in Figure 4.

Table 4. State-Level Annual Ethanol Fuel Demand Projections (in million
gallons)

	Controlled Cases	2022 AEO	2022 RFS1

STATE	2010	2012	2014	2017	2022



Alabama	126	281	285	275	509	267	0

Alaska	0	14	27	30	29	16	0

Arizona	293	295	298	343	695	338	257

Arkansas	0	50	142	139	235	36	0

California	1723	1738	2313	3664	5138	1787	1739

Colorado	246	248	303	462	607	274	177

Connecticut	161	162	215	312	437	152	152

Delaware	44	44	45	43	122	42	0

District of Columbia	19	19	19	39	54	19	19

Florida	947	1011	1022	1526	1677	1056	0

Georgia	320	550	556	589	1219	390	0

Hawaii	47	47	63	95	133	46	46

Idaho	74	74	99	158	222	77	0

Illinois	508	513	620	832	1087	501	337

Indiana	365	368	372	398	770	359	157

Iowa	150	152	170	301	422	147	147

Kansas	143	144	145	143	312	140	3

Kentucky	242	244	269	303	673	234	102

Louisiana	131	216	255	356	612	213	0

Maine	44	73	74	114	200	70	0

Maryland	247	288	291	481	755	286	250

Massachusetts	280	282	285	565	793	276	276

Michigan	454	458	564	747	980	429	294

Minnesota	260	263	349	525	737	256	256

Mississippi	197	199	201	197	558	194	0

Missouri	331	334	374	436	934	325	325

Montana	0	5	52	51	60	5	0

Nebraska	88	89	109	178	249	87	87

Nevada	92	93	99	190	257	105	81

New Hampshire	66	66	67	66	188	65	65

New Jersey	391	394	525	784	1100	382	382

New Mexico	122	123	137	259	363	126	126

New York	397	582	827	1079	1391	558	388

North Carolina	0	126	406	418	420	0	0

North Dakota	35	35	46	66	93	32	32

Ohio	534	539	544	522	1256	511	0

Oklahoma	137	239	241	246	696	242	0

Oregon	168	170	226	348	487	170	97

Pennsylvania	223	368	518	624	720	350	123

Rhode Island	46	47	47	94	132	46	0

South Carolina	0	113	234	237	434	0	0

South Dakota	41	41	55	81	114	40	40

Tennessee	329	332	442	681	955	332	332

Texas	1003	1063	1217	1281	2332	1113	0

Utah	0	95	134	145	143	103	0

Vermont	53	53	54	109	153	53	53

Virginia	221	268	408	415	825	219	0

Washington	277	279	282	600	842	293	0

West Virginia	24	87	88	79	124	78	0

Wisconsin	299	302	355	610	855	297	297

Wyoming	43	43	44	42	42	41	41

TOTAL	11943	13620	16514	22281	34142	13177	6697



Figure 4. VMT distribution at census tract level for Hawaii

Petroleum products terminals tend to cluster in the same general
location. There are about 413 clusters of petroleum product terminals,
of which 48 of them had at least 5 petroleum product terminals or more.
The centroids of the existing refined products terminal clusters are
used to develop a Thiessen Polygons or Voronoi Diagram. This Voronoi
diagram provides the terminal cluster service regions. The Voronoi
diagram for the state of Hawaii is depicted in Figure 5.

Figure 5. The Voronoi diagram for the state of Hawaii

By aggregating from the census tract level, the ethanol demand can be
estimated for the terminal cluster service regions (Voronoi diagram
level). The ethanol demands at the terminal cluster service region level
are then shared to individual terminals evenly if there is more than one
terminal in the cluster. Figures 6 thru 11 show projected ethanol demand
at petroleum product terminal levels for six scenarios considered in
this analysis.

Figure 6. Ethanol demand at the terminal level for 2010

Figure 7. Ethanol demand at the terminal level for 2012

Figure 8. Ethanol demand at the terminal level for 2014

Figure 9. Ethanol demand at the terminal level for 2017

Figure 10. Ethanol demand at the terminal level for 2022 controlled case

Figure 11. Ethanol demand at terminal level for 2022 AEO Reference Case

3.5 Gasoline Fuel Terminals

The gasoline fuel terminals database is based on the OPIS/STALSBY
Petroleum Terminal Encyclopedia, 2008 edition, which lists 1,495
petroleum terminals in the United States. Efforts have been made to
isolate terminals that handle gasoline or ethanol for this project.
There are 1056 petroleum terminals in the United States located as 413
clusters that handle gasoline and/or ethanol. These 1056 gasoline and/or
ethanol terminals are depicted in Figure 12.

Figure 12: Gasoline and/or ethanol terminals

It should be noted that the information in the database is not
comprehensive. Most terminals have detailed information on what type of
petroleum products are stored along with their storage tank capacity
information. However, certain terminals only specify their general type,
such as asphalt, crude, jet fuel, lube oil, petrochemicals, natural gas
liquids, and refined products terminals.

A thematic map of total storage capacity at the terminals is presented
in Figure 13. Four terminals in the database have very large total
storage capacities (more than hundreds of million barrels).

Figure 13. Gasoline and/or total storage capacity distribution

For the purpose of this project, a terminal is considered a gasoline
terminal if the terminal in the database is of a refined products type
with no detailed specification on which product it stored. In addition,
a terminal is also considered as a gasoline terminal if the database
specifies that any of the following products are stored at the terminal:

CA Reformulated Blendstock for Oxygenate Blending (gasoline production)
(RBOB)

CA RBOB Regular

RVP Midgrade

RVP Premium

RVP Regular

Sub-Octane

Conventional Regular

Conventional Midgrade

Conventional Premium

Premium RBOB

Refiners Blendstocks

Regular RBOB

RFG Mid Grade

RFG Premium

RFG Regular

Gasoline

Ethanol

Out of these 1056 gasoline terminals, ethanol distribution intermodal
terminals in major metropolitan areas were further examined, as
discussed in the following section. 

3.6 Ethanol Intermodal Terminals

Ethanol must travel from refineries to blending sites. Blending sites
are generally located at multiple terminals in every metropolitan area
and are operated by competing companies that receive petroleum products
from pipeline or by water transport, have tank storage capacity, and
distribute products by truck to retail outlets. Ethanol refineries are
generally located near agricultural production, so the distance between
refineries and blending sites is significant, and the expense of
long-distance truck movements is prohibitive, leaving rail, inland
barge, and marine tanker as modes of choice. Few domestic refineries,
however, are located adjacent to navigable water, so water is available
only as part of a multi-modal trip.

So far, petroleum products pipelines typically cannot batch ethanol with
their other products. Should pipelines become a widespread option, it
would massively rearrange the transportation locations, favoring a Gulf
Coast hub to collect and distribute ethanol.

In principle, each blending site could receive its ethanol directly from
refineries if it had rail access or could receive waterborne imported
ethanol if it had deep berths. At current low levels of consumption,
less than 10% of gasoline, there are several difficulties with this
pattern. First, few products terminals have rail access, and when they
do, it is a single spur that can handle only a handful of cars at a
time. Unit trains can be accommodated only when unit train facilities
are specifically built and when sufficient background storage and
adjacent land is available.

For the foreseeable low level of consumption, an alternate
transportation paradigm is emerging to establish one (or, rarely, two)
dedicated ethanol terminals in a large region which will receive cross
country unit train shipments, and then distribute that ethanol by truck
to each company’s existing products terminals in the region, as shown
in Figure 14 based on 2006 waybill data. The largest ethanol flow has
been 3.2 million tons of alcohol from Houston, Texas to New Orleans,
Louisiana. Those terminals may be hundreds of miles away, but the
dedicated ethanol terminal will often be located adjacent to an existing
concentration of pipeline terminals in a metro area, so truck trips are
more commonly 10 miles long, and may be replaced eventually by dedicated
ethanol pipelines within the petro complex. This pattern will probably
last unless ethanol becomes as common as gasoline, at which time more
products terminals will construct their own unit train facilities, or
the ability to move ethanol by pipeline will be developed.

Several regional dedicated ethanol terminals have been established
within the last few years. They are identified here:

1. Albany, NY. Receives unit trains, distributes ethanol by truck
locally, but primarily by barge for blending in northern NJ (New York
metro area) and the New England coast (Providence and Boston).

2. Northern NJ. Linden and Sewaren (Motiva). Receives unit trains and
seaborne imports, distributes primarily by truck and private pipeline to
local blending sites, and secondarily by barge to the New England coast.


3. Philadelphia, PA. Receives unit trains, and in the future seaborne
imports. Distributes to blending sites mostly on the lower Delaware
River by truck or barge.

4. Baltimore, MD. Receives by unit train or seaborne imports.

5. Houston, TX. At least two sites receive seaborne imports and
originate ethanol rail shipments.

6. Arlington, TX. A new rail receiving facility separate from pipeline
terminals in the Dallas Fort Worth area.

7. Carson, CA (south Los Angeles). A new rail receiving facility,
adjacent to Shell refinery. Truck shipments go to the LA basin, San
Diego, and all of southern California.

8. Sacramento or Stockton, CA. Location unidentified. 2006 rail ethanol
shipments terminated in Sacramento, but references were found to a
Stockton terminal. Truck shipments go to all of northern California and
the Bay Area.

In addition, an examination of the 2006 rail waybills shows significant
ethanol shipments terminating in Columbia, SC, Wilmington, NC,
Kingsport, TN, and Ogden, UT (not associated with any known terminal,
possibly for industrial use). The largest concentrated originations of
rail shipments are at Houston.

There are transload facilities (that do not include storage, sometimes
called “TransFlo” for liquids, and often depend on shipper-supplied
pumps) that transfer ethanol directly from rail cars to trucks for
distribution to blending sites. Although multiple rail cars may be
parked while waiting for trucks to drive up, these transload facilities
do not have unit train capability. The most notable transload facility
is a Norfolk Southern terminal in Alexandria, VA where trucks presumably
make a short drive to the petroleum product terminals (i.e., tank farms)
in Newington and Fairfax. This transfer operation would make no sense if
the terminals had rail facilities. Fairfax is too distant from a rail
line, but Newington is probably achievable. Transload operations also
occur in Richmond, VA. Essentially every city has a transload terminal
where a small number of rail cars may be parked awaiting a direct
transfer to trucks, and small ethanol transfers may be occurring there
now. We expect transload operations to be temporary until more efficient
facilities are constructed as ethanol consumption increases.

A mirror-image collection function can be performed at the beginning of
the trip, collecting ethanol in trucks from refineries not served by
rail, and using on-site storage to load unit trains or barges. The only
two known collection terminals are at Manly, IA (associated with the
Iowa Northern Railways) and at Sauget, IL, for barges. They can also
collect single carloads by rail from refineries without unit train
facilities. However, single- to multiple-car rail shipments do not
require a specific terminal. A railroad can cooperate with a group of
shippers to collect single carloads at any rail yard to form a unit
train provided those shippers have a common destination.

 

Note: Dark blue stars are dedicated ethanol terminals, small purple
stars are petroleum products intermodal terminals, and red markers are
ethanol refinery locations.

Figure 14. Ethanol rail flows, 2006Note that California ethanol
producers import corn from the Midwest by rail, and generally distribute
to California blending sites by truck.

Because ethanol distribution terminals in metro areas are preferentially
located where product distribution terminals are already located, our
analysis of future ethanol terminal operations will not be seriously
limited by assuming they will occur at existing petroleum terminals.
However, the analysis must allow for a single concentrated ethanol rail
destination that accommodates unit trains, followed by truck movements
to possibly distant terminals in the same metro region.

The analysis of transportation options will allow essentially unlimited
throughput at any terminal, in effect assuming that capacity will rise
as demand increases. In cities that have no dedicated ethanol terminal
now, an existing products distribution terminal will be allowed to
handle that function with whatever costs are involved in adding unit
train or barge facilities. The location of the facility is based on the
assumption that dedicated ethanol terminals in the future will be
located adjacent to existing product terminals That will likely be the
terminal with the best apparent access to existing rail lines, however
capability of unit train facilities will be determined for each existing
product terminal based on the access to the existing rail network. Most
likely, as ethanol demand grows, multiple receiving terminals will
appear. The assumption of a concentrated single metro terminal will tend
to overstate slightly the actual amount of truck traffic in the
distribution to blending sites.

Based on the list of existing 1056 petroleum product terminals (as
discussed earlier in Sect. 3.5,) each of them was examined if they
already had rail or waterways access needed for ethanol movement. If
they do not have rail access, the distance for a reasonable route that
would link the fuel terminal by constructing spur line to the nearest
rail line was estimated. A single ethanol distribution terminal was then
selected for each metropolitan area from the 413 cluster of existing
petroleum product terminals (as determined by the Voronoi diagram and
discussed in Sect. 3.4) based on having the minimum access rail
distance. No petroleum product terminal was found to have existing unit
train facilities. Figure 15 shows the distribution of potential ethanol
distribution terminals in major metropolitan areas. These ethanol
distribution terminals are considered as the intermediate ethanol demand
locations, from where ethanol is shipped by trucks to existing petroleum
product terminals. In case, there was only single petroleum product
terminal in a given cluster, ethanol terminal was assumed to be
collocated with the existing product terminal and thereby there was no
final trucking distance in this case. 

 

Figure 15. Distribution of potential ethanol distribution terminals

 Transportation Cost

3.7.1 Rail 

Available rail transportation cost information is based on 2006 waybill
data. Indications are that operating cost varies in the range of 3.8
cents/ton-mile – 17.1 cents/ton-mile depending on the railroad company
used, existence of long-term contracts, and average miles traveled per
trip. In addition, there is a monthly maintenance and leasing cost of
$900 per car13 which when taken into consideration results in total
shipment cost of 5.6 cents/ton-mile – 18.4 cents/ton-mile. Detailed
available ethanol shipment cost data were regressed as a function of
distance, and a cost of 7 cents/ton-mile has been considered in our
analysis assuming an average rail distance of 700 miles based on the
2006 waybill data. In order to include the economies of scale associated
with unit train movements, a 35% reduction in cost, i.e., 4.6
cents/ton-mile has been assumed in that case.

Average rail fuel consumption in 2006 is estimated to be 330
Btu/ton-mile or 0.0023 gal/ton-mile. For the railroad industry in
general, last year fuel outlays were about 22% of its revenues of 53 G$,
compared to 24% for labor. Thereby, fuel outlays in general contribute
to a smaller share of the total rail transportation cost. For a
relatively small share of fuel cost to total rail and waterways
(discussed below under Sect. 3.7.2) transportation cost, future cost
projections for these two modes using the future fuel price trend have
only been used for the determination of various modal distribution costs
(discussed under Sect. 5.2). In the initial development of O-D
(supply-demand) matrix and assignment of specific routes, relative
transportation mode costs are important since the cost differences among
various modes are quite substantial as discussed here.

The future fuel price trend has been assumed to be based on EIA’s
projections for U.S. No.2 Diesel Retail Sales by All Sellers, which
index has also been suggested by the recent ruling of the Surface
Transportation Board (STB) for estimating fuel surcharges to be charged
by railroads. The EIA projections of wholesale price of residential
distillate fuel oil/heating oil, which is quite similar to the proposed
price index, has been used for our analysis. Distillate fuel oil price
is assumed to be $2.00/gallon by 2022 – about the same as today’s
price for all three transportation modes. The EPA diesel fuel sulfur
requirements for locomotive and C1/C2 marine engines to be aligned with
those for highway diesel fuel starting this year may result only in a
small cost difference between these two fuels. It is likely that with
larger economies of scale for diesel fuels used in rail and waterways,
may further help in reducing the fuel cost differentials among the
various modes.  

3.7.2 Waterways

Current ethanol shipment costs by barge in upper Mississippi and
Illinois waterways are estimated to be in the range of 2.35
cents/ton-mile – 2.43 cents/ton-mile based on the Tennessee Valley
Authority barge costing model, having an average hauling distance of
around 1200 miles.13 For our analysis, we assume an average value of 2.4
cents/ton-mile. For barges, average energy consumption is the least,
i.e., 0.0020 gal/ton-mile, close to the estimate for rail but
significantly higher than for trucks. Average energy consumption is
based on the fact that a one standard tow consumes 44 gallons of fuel
per mile with a carrying capacity of 22,500 tons. For future barge
shipment cost projections, we vary only the fuel cost component by using
the EIA distillate fuel cost projections as was done in the case of rail
mode of shipment above.

3.7.3 Truck

The average trucking cost in 2003 is estimated to be about 13.3
cents/ton-mile, based on annual revenues and total ton-mile
movements.16. With the consideration of escalation in diesel fuel price
and since this average cost estimate includes a large number of
single-unit light package deliveries, 2006 trucking cost is assumed to
be 14 cents/ton-mile. It is estimated that 900 trucks with a carrying
capacity of 22,500 tons consume 381 gallons of fuel per mile, or 0.0169
gallons/ton-mile of diesel fuel consumption. As in other transportation
modes, future diesel retail price for trucking in 2022 is assumed to be
the same as today, without explicit consideration of fuel cost component
in total transportation cost. Fortunately in our case, the truck
component in ethanol distribution will be small and inelastic, so
accuracy is not determinative. 

3.8 Congestion/Delay

3.8.1 Rail

Average line haul rail speeds are currently 22 mph. Unlike the other two
major modes of transportation, rail corridors are rarely at widespread
over capacity, since in most cases an optimal balance is achieved
between the level of the capital expenditures for capacity expansions
with the anticipated revenues. As shown in Figure 16, 88 percent of
today’s primary corridor mileage is operating below practical capacity
(Level of Service (LOS) A/B/C), 12 percent is near or at practical
capacity (LOS D/E), and less than 1 percent is operating above capacity
(LOS F). LOS (A/B/C) represents below capacity and volume/capacity in
the range of 0.0-0.7; LOS D at near capacity with a value of 0.7-0.8;
LOS E at capacity with a value of 0.8-1.0 and LOS F above capacity with
a value of greater than 1. It is assumed in the future that rail
management will increase capacity in line with the total traffic on a
route to maintain an optimal balance between congestion costs and
capital costs. Also, there will not be much change occurring in
non-container movements in the future. Manifest traffic has an
additional delay for switching and en route classification, about 1 day
for each 600 miles.

The presumption for rail traffic is that rates cover both the operating
costs and capital costs for the provision of capacity. Therefore, except
for change in fuel price (as discussed earlier in Sect. 3.7.1), rail
costs can be expected not to change significantly in the future, along
with travel speeds. 

 

Figure 16. Current train volumes compared to current train capacity

3.8.2 Waterway

Commodity Flows

For analyzing scenarios that involve future ethanol shipments for the
years 2010, 2012, 2014, 2107 and 2022, future waterway traffic operation
information other than ethanol needs to be forecasted for the
corresponding years. The future waterway traffic operation forecasts are
based on information from the Freight Analysis Framework (FAF).

The FAF integrates data from a variety of sources to estimate commodity
flows by different modes of transportation and related freight
transportation activity among states, regions, and major international
gateways. The Commodity Origin-Destination Database of FAF estimates
tonnage and value of goods shipped by the type of commodity and mode of
transportation among and within 114 areas, as well as to and from seven
international trading regions. The original version, FAF1, provides
estimates for 1998 and forecasts for 2010 and 2020. The new version,
FAF2, provides estimates for 2002 and the most recent year 2007, in
addition to forecasts through 2035 (see Table 5). Future national
freight demands growth factors in reference to 2007, by water mode, are
estimated by interpolating future freight demand information in the FAF2
database and are presented in Table 6. 

Table 5. Commodity Movements (in thousand tons) by Waterways, Based on
FAF2

Commodity

movement	2007	2010	2015	2020	2025	2030	2035

	575,010	613,541	658,618	696,996	749,777	807,244	873,763



Table 6. Estimated Future Commodity Flow (in thousand tons) and Growth
Factors for this Study

	2010	2012	2014	2017	2022

Flow	613,541	631,189	649,344	673,709	717,638

Growth Factor	6.70%	9.77%	12.93%	17.16%	24.80%



3.8.2.2 Estimation of Operation Constraints on National Inland and
Coastal Waterway Networks

In order to archive hydro power generation and deeper navigation
channels, a series of dams have been constructed along national inland
waterways. With the dams is a series of locks constructed to facilitate
marine traffic and regional commerce. These locks and dams currently are
the major bottlenecks for overall marine traffic operations. The average
delay can be as high as thirty hours per tow operation. The Lock
Performance Monitoring System has been used as the base for estimating
the existing and future operation delays.

Lock Performance Monitoring System

Information from United States Army Corps of Engineers’ Lock
Performance Monitoring System (LPMS) is used to describe the current
waterway traffic operations. Information from LPMS is used to formulate
regression models to estimate the future waterway network constraints in
terms of average delay in hours per tow. LPMS collects, edits,
maintains, analyzes, and disseminates data collected at all Corps-owned
and/or Corps operated locks. Data collection began in 1975 and consists
of information about waterway traffic usage and performance of each lock
in the Corps’ national system. Lock statistics for calendar year 2007
are used to estimate the average delays at locks.

Estimate Future Lock Delays

Based on lock statistics for calendar year 2007 information from LPMS,
two regression models (with R2 = 0.7231 and 0.6747 respectively) have
been formulated:

 , and

  .

The average tow delay is the quotient of the total delay and number of
tows. Mathematically, it can be expressed as

  

Instead of estimating the future average tow delay utilizing these
equations directly, a growth factor has been estimated. That is 

 

Using the similar procedure, the future total delay and average delay
growth factors for different target years have been estimated. The
future average delay for a lock is estimated by multiplying average
delay growth factor by target year to average tow delay at 2007. The
existing and 2022 average tow delays for locks in the nation’s
waterway system are depicted in Figures 17 and 18, respectively. Future
total delay and average delay growth factors in reference to 2007 are
presented in Table 7.  

Table 7. Estimated Future Growth Rates for Lock Delay

	2010	2012	2014	2017	2022

Total Delay	105.39%	107.84%	110.35%	113.69%	119.66%

Average Delay	102.09%	103.02%	103.96%	105.18%	107.33%



A histogram analysis of the data indicates that the waterway delay
relative to year 2010 would increase by 0.013 hrs, 0.027 hrs, 0.045 hrs,
and 0.076 hrs in 2012, 2014, 2017, and 2022, respectively. The mode
estimate for the waterway delay distribution in 2010 is estimated to be
1.498 hrs, which would increase to 1.562 hrs by 2022. Lower estimated
delay is due to the fact with an increase in freight demand will
increase the number of tows, and thus not introduce a big change in an
average delay per tow.

 

Figure 17. Average tow delays for 2007

Figure 18. Average tow delays for 2022

3.8.3 Highway 

With the ever-increasing demand for freight and people movement on our
relatively stabilized highway infrastructure, traffic congestion and
delays have become one of the major operation constraints that impede
economic growth and degrade our quality of life. It is important to use
realistic travel speed information with congested traffic in the
scenario analyses for future years. Information presented in the 2007
Urban Mobility Report is used to approximate the travel speeds.

 

The Texas Transportation Institute’s 2007 Urban Mobility Report is an
annual publication which studies urban highway congestion and associated
problems from 437 urban areas. The report presents various measures of
mobility in travel demand, available highway facilities, traffic delays,
excessive fuel consumptions, congestion costs, and effectiveness of
congestion relief counter measures. One of the measures of urban
mobility estimated by the 2007 Urban Mobility Report is area-wide travel
speeds for freeways and arterial streets. 

This project adapts the travel speed estimation procedure used in the
2007 Urban Mobility Report to approximate the travel speeds for
different interstates, other freeways and expressways, and arterials in
different states for both urban and rural areas. The travel speed
estimation procedure is based on the Average Daily Traffic per Lane
(ADT/Lane) statistics. The description of the speed estimation procedure
in the 2007 Urban Mobility Report is incomplete. Therefore, direct
calculation of travel speed based on ADT/Lane statistics is impossible.

However, the 2007 Urban Mobility Report has estimated travel speed
information and Average Daily Traffic per Lane (ADT/Lane) statistics by
freeway and arterial for the top 85 major metropolitan areas. Based on
the data for freeway and arterial from the 85 metropolitan areas, two
linear travel speed models (with R2 = 0.8371 and 0.5284) have been
formulated. Mathematically, they are

 , and

 .

The Average Daily Traffic per Lane (ADT/Lane) statistics for 2006 by
state and by highway functional classification are calculated based on
the vehicle-miles-traveled and highway lane-miles statistics from the
Highway Statistics 2006 published by the Federal Highway Administration
(FHWA), US Department of Transportation (US DOT).

In order to forecast the future Average Daily Traffic per Lane
(ADT/Lane) statistics, vehicle-miles-traveled and highway lane-miles
statistics from 1997 to 2006 have been assembled. The 10-year trends for
national level Average Daily Traffic per Lane (ADT/Lane) for different
freeways and arterials are presented in Figures 19 and 20 for rural and
urban areas, respectively.

Based on the 10-year historical trends, the future Average Daily Traffic
per Lane (ADT/Lane) growth factors can be estimated. The growth factors
for the future analysis years by highway functional classification and
for rural and urban areas are presented in Table 8. 

Figure 19. Average daily traffic per lane trends for rural U.S.

Figure 20. Average daily traffic per lane trends for urban U.S.

Table 8. Future Growth Factors for 2006 ADT/Lane

Year	Rural	Urban

	Interstate	Other

Principal

Arterial	Minor

Arterial	Interstate	Other

Freeways	Other

Principal

Arterial	Minor

Arterial

2010	1.0532	1.0029	1.0085	1.0333	1.0471	1.0054	1.0337

2012	1.0797	1.0043	1.0127	1.0500	1.0706	1.0081	1.0505

2014	1.1063	1.0058	1.0170	1.0666	1.0941	1.0108	1.0674

2017	1.1462	1.0079	1.0234	1.0916	1.1294	1.0148	1.0926

2022	1.2126	1.0115	1.0340	1.1333	1.1882	1.0215	1.1347



By applying these growth factors to the 2006 Average Daily Traffic per
Lane (ADT/Lane) statistics, ADT/Lane statistics for 2010, 2012, 2014,
2017, and 2022 can be estimated. By applying the future ADT/Lane
statistics to the freeway and arterial travel speed models, the travel
speeds for 2010, 2012, 2014, 2017, and 2022 can be approximated. 



4. GENERATION OF ETHANOL TRANSPORTATION ACTIVITY DATA – MODELING
APPROACH

Estimates of ethanol transport activities using the ORNL Transportation
Infrastructure Network Model for three 2022 scenarios (i.e., 2 reference
and 1controlled) were based on the following two steps: 

ODM (origin-destination-mode) matrix

Traffic assignment

4.1 Origin-Destination Matrix (ODM)

At first, an initial origin-destination matrix was developed based on
the following inputs:

(a) A vector of ethanol production by plant location and by base or
projection year. Corn refineries from EPA (Sect. 3.1), cellulosic
refineries as detailed above under Sect. 3.2, including import supplies
as outlined under Sect. 3.3. It is assumed that sufficient storage
capacity exists for a continuous flow over the year.

(b) A vector of ethanol demand by terminal location (where blending
occurs), by year as detailed under Sect. 3.4. It is assumed that some of
these terminals are ethanol distribution terminals, one in each metro
area, the volume of which is exogenously determined as discussed in
Sect. 3.6. Local distribution by truck would then occur from the 413
metro ethanol distribution terminal to blending sites at existing 1056
products terminals (also termed as “tank farms” and as discussed
earlier under Sect.3.5), which have tank storage and currently receive
products from pipeline, ship, or barge, and distribute products by truck
to retail or wholesale customers. Existing product terminals which are
currently not served by either rail or waterways, ethanol movement will
be limited to trucks only.

The goal for each year was to fill in the ODM matrix between refineries
and terminals consistent with these production and supply margins. 

“Location” here means a longitude/latitude that can be examined on a
map. The analysis was done at the county-level resolution, by accounting
for local transportation infrastructure. As discussed earlier, most
refineries have been assumed to be located in proximity to either rail
or waterways mode of transportation.

Procedure: 

1. Attach ethanol facilities to modal networks. 

As part of an examination of ODM routes, refineries and terminals must
be attached to the network in an initial step. Existing locations were
examined to see whether the various modal attachments are already
existent or feasible. By and large, assumptions were made that feasible
connections will be built when they make economic sense, for instance,
that one unit train per week will justify building the tracks and
storage facilities at existing locations to accommodate long trains. The
routing by various modes was then generated to produce a matrix of
distances and costs, which were used in the spatial interaction model to
determine which origins serve which destinations, and their quantities
as discussed below. 

2. A spatial interaction model (SIA) (or gravity model) was applied to
the cost matrix to produce an initial ODM activity matrix. The model is:

  t(o,d,m) = K *O(o)* D(d) * exp[-beta*c(o,d,m)]

where

beta is a distance decay factor. A larger value of beta indicates more
distance sensitive movements, and it is more likely that demand will be
met by nearby suppliers. A smaller beta value produces more cross flows,
that is, more suppliers located in distant parts of the country. It also
incorporates an implicit scale factor for cost units. As it approaches
infinity, the SIA approaches a transportation cost minimization model.
Every choice of beta will result in a different average shipment
distance. Its value was calibrated at 0.85 (a low value characteristic
of high value commodities (like chemicals) that are not very distance
sensitive and which encompass substantial cross-hauling) based on the
average ethanol shipment distance of 700 miles for rail in 2006
(according to the latest 2006 waybill sample) and projected 2010
supply-demand scenario considered in our analysis. Due to lack of data
for other modes, the use of a single calibrated beta value across all
modes resulted in average shipment lengths for 500 miles and 100 miles
for barge and truck mode of transport, respectively. The beta value
remained unchanged for future forecast years due to lack of empirical
data, though as volumes increased and demand became more dispersed,
average shipment distances increased. 

c is the transportation cost from origin (o) to destination (d) by mode
(m). At first the generalized cost of travel between every mode
combination of origin O(i) and destination D(j) is calculated, and from
which the minimum cost per ton among feasible mode choices is used for
c(o,d,m). For the mode specific cost, it is only the relative cost that
matters. As discussed earlier under Sects. 3.7.1 and 3.7.2, the ratio of
relative rates of 2.4 to 7 cents per ton-mile between inland barge and
rail are used, thereby favoring the waterways mode whenever available.
As noted before, truck costs are much higher 14 cents/ton-mile. Many of
the costs that a shipper faces are not included in the carrier’s
charges, such as reliability, inventory costs while a shipment is in
transit, and scale economies, which will often make truck a more
desirable alternative. These are usually modest for bulk commodities
like ethanol, and so those have been ignored here. In addition, we
distinguished between unit train movements and the shipment of
individual tank cars in manifest trains with their en route
classification. Manifest train rates are generally about 40% higher,
which was used in the cost calculation.

K is the calibrated constant, ensuring that total flows match. That is,
the matrix (i.e., t(o,d,m)) that most closely matches the initial SIA
internal distribution while meeting the constraints on individual
supplies and demands for ethanol. In this way, K is actually irrelevant
to the outcome – it is merely a scaling factor for all t(o,d,m)
entries in the OD matrix.

O(o) total amount to be shipped from origin (o),

D(d) total amount demanded at destination (d), and

t is the amount of ethanol to be shipped from (o) to (d) by mode (m)

m is the mode consisting of the following options -- truck, manifest
train (loose cars), unit train, and barge. It is assumed that in most
cases, either one of these modes would be used for ethanol delivery from
refineries to ethanol terminals, from where truck delivery will be used
for the final destination of product terminals. Only in some instances,
e.g., Alaska and Hawaii, initial shipment would be by rail to the port
in the west coast, from where ethanol will be delivered by barges for
the eventual truck delivery to product terminals. In the case of the
Northeast—to be consistent with the existing flow pattern—final
ethanol delivery is assumed to be by barges. Intermodal transfers
between highways and rail, highways and waterways, and waterways to rail
have also been considered in the analysis.

 

The gravity model used here for the generation of ODM is commonly used
for trip distribution in the traditional four-step transportation
forecasting model before mode choice and route assignment are made. This
step matches tripmakers’ origins and destinations to develop a “trip
table” a matrix that displays the number of trips going from each
origin to each destination. It is based on the principle that the
interaction between two locations declines with increasing distance,
time, and cost between them, but is positively associated with the
amount of activity at each location. It uses impedance by travel time,
perhaps stratified by socioeconomic variables, in determining the
probability of trip making. The distance decay factor of 1/distance is
generally a more comprehensive function of generalized cost, which is
not necessarily linear – a negative exponential tends to be the
preferred form as also used in our case illustrated above. The rate of
decline of the interaction, i.e., the impedance factor has to be
empirically measured, and varies by context. Limiting the usefulness of
the gravity model is its aggregate nature.

The resulting matrix is then fitted to the exogenous origin and
destination marginal totals as discussed under items (a) and (b) above.
This results in every cell having some non-zero volume. This is adjusted
by converting the lowest volume cells to zero, provided they were less
than 0.11 M gal/yr and 10% of that origin’s total volume and 10% of
the destination’s total. Instead, those suppliers redistributed their
shipments to customers that already had a significant relationship with
them. On each cycle one destination for each origin and one origin for
each destination were permanently zeroed, and the problem re-run on the
next cycle – thereby forcing the resulting matrix into compliance with
the input origin and destination marginal controls through Iterative
Proportional Fitting (IPF). Costs were initially assumed to be by unit
train, but if insufficient volumes were obtained, costs for unit trains
and water on the next cycle were adjusted upward until manifest train
costs appeared superior. The minimum threshold was 2 unit trains/year
(15 Ktons or 5 Mgal assuming a unit train has 80 tank cars each having
an ethanol capacity of 100 tons or 30,000 gallons) for each possible O-D
pair for the full benefit of unit train pricing. It is further assumed
that at both ethanol supply source (i.e., refineries) and demand
destination (i.e., metro ethanol terminal) meet the minimum threshold of
1 unit train/month (or 12 unit trains/year or 30Mgal) in order to
justify the necessary capital investments for unit train facilities
including sidings. It is important that not only the minimum threshold
between every possible O-D pair on a specific trip be met, but also
every demand and supply centers should also meet the total annual
minimum threshold movements which includes either various supplies
coming to a single demand point or vice versa. The threshold for barge
is assumed to be the same as unit trains, whereas truck is the last
default mode used having an ethanol capacity of 8,000 gallons per truck.

4.2 Traffic Assignment

To determine route-specific flows, the ODM shipment tonnages are routed
over the Intermodal Network containing the 1998 FAF1 base commodity
county-by-county routes by various modes – the only publicly available
detailed data to date. Since commodity flows remain fairly unchanged,
use of the older flow data is inconsequential. The ORNL infrastructure
network model used for the analysis is an extensive and integrated
public domain geographically-based representation of the North American
transportation infrastructure suitable for individual or multimodal
studies. It is a combination of its data, the CTA intermodal network,
and its traffic assignment models to load the network with commodity
flows. The network is composed of independently constructed single-mode
networks for highway, rail, and water, along with a set of intermodal
terminals and a terminal model to connect them. Terminals connect
mode-specific subnetworks, and each terminal is a link in this network
with an impedance that represents the generalized cost of terminal
activities. The result is a unified routable network, with a single node
list, a single link list, and a topology defined by the links’
endpoint nodes, a structure common to most network analysis programs. It
may be easiest to imagine that each modal network occupies a horizontal
plane, while intermodal terminals connecting 2 modes lie between the
planes and are attached above and below by vertical access links.
Traffic generators would typically also be located at a different level
and similarly attached to an arbitrary modal network by vertical access
links. Several types of links are considered, i.e., line haul links,
railroad interlines, terminal links, access links, and shadow links
where by various modes each has an impedance function based on its
characteristics and context. Both links and nodes intentionally share a
common set of attributes, although the attributes of routing impedance
and length and endpoint nodes have meaning only for links. Principal
uses to date have included regional and modal competition effects of
policies and investment in infrastructure (Marine input/output model,
regional routing model, national rail freight infrastructure), optimal
investment strategies for waterways (Ohio River Navigation Investment
Model), simulating the routes of shipments in the Commodity Flow Survey,
routing and scheduling military convoy movements (MOBCON), and risk
assessments of hazardous materials movements.

The three networks, i.e., the Oak Ridge national highway network, the
CTA rail network, and the composite intermodal transportation
network—representing the North American transportation infrastructure
are well documented and available for open distribution.22 The Oak Ridge
national highway network is a database of major highways in the United
States using data from several sources such as USGS national Atlas
digital line graphs, state maps, FHWA highway performance monitoring
system (HPMS), TIGER/line files, and USGS 1:100,000 digital line graphs
etc. The CTA railroad network is a geographically-based representation
of the North American railroad system which is designed foremost to
support analytic transportation studies, such as traffic assignment,
capacity, optimal investment, intermodal terminals, and cost estimation.
The base data for the U.S. portion of the network is the Federal
Railroad Administration’s National Atlas-based strategic rail network.
The intermodal transportation network connects the four individual modal
networks having the modification option in order to estimate different
link impedances. These data are also available in GIS versions easily
importable into commercial software such as TransCAD or ARC/Info. The
other network, i.e., the national waterway network representing the
North American transportation infrastructure is a customized network
with input from the National Waterway GIS Design Committee containing
members from several federal agencies and using data from several
sources. The majority of the inland links are at 1:100,000 scale as is
at least the case with most other networks.

Most software is now written to use the intermodal network to achieve
the maximum flexibility in studying intermodal as well as mode-specific
problems. The core routine is an incremental traffic assignment, which
accepts an origin-destination matrix and finds multiple paths through
the network used by each O-D pair. Traffic is assigned to each of the
paths in order to minimize a generalized cost, or impedance, which
increases as congestion, or the volume/capacity ratio, on each facility
increases. The capacity measure on rail lines is determined by an
approximate function of current traffic (transcribed from 

the Federal Railroad Administration), subjective line quality, and
physical attributes of signal system and number of tracks. 

The ORNL intermodal network is updated on an average cycle of every
three years using press reports and road atlases for updating capacity
and linkage information for rail and highways, respectively. The
intermodal network used for the analysis represents the latest updates
made in 2006. The best test for the adequacy of the assignment is to
compare line volumes produced by the program using the national O-D data
with independent data on observed rail flows, which is available
anecdotally and from the Federal Railroad Administration volume classes.
Major discrepancies have been handled by manual editing of the capacity
measures to enforce better compliance (e.g., limiting Cascade Tunnel to
25 trains/day). For waterways, there are few choices of routes, so the
accuracy of flows is essentially the accuracy of the input O-D matrix.
For trucks, there is little reliable intercity flow data, and the
reasonableness of results depends on mostly anecdotal evidence of actual
routing choices. Therefore, the knowledge and judgment of an analyst for
the reasonableness of routes is essential, absent an extensive data
effort to collect and interpret truck activity by route. 

There are separate traffic assignment models for the long-distance and
local components of ethanol transport. The long-distance component of
the model considers receipts of long-distance shipments from refineries
or, in the case of imported ethanol, from seaports at each metro area
assumed to have an ethanol terminal. The movement from ethanol source to
metro receiving terminal could be by rail, water, truck, or any
intermodal combination. If by truck, the long distance trip terminates
directly at the blending site without the use of a dedicated receiving
terminal. After receipt and intermediate storage, the ethanol is
distributed by truck to local blending sites in the same metro area –
considered by the second component of the traffic assignment model.
Those blending sites are existing petroleum products distribution
facilities (sometimes referred to as pipeline terminals or “tank
farms”). There will typically be multiple blending sites operated by
competing distributors. In cities that do not have existing ethanol
distribution terminals, we assume that one will be built adjacent to the
blending site with the best rail or water access. That tank farm may or
may not have a rail siding, and surely cannot presently accommodate unit
trains, but we assume that unit train facilities will be built as demand
in the area increases. Note that all existing ten ethanol receiving
terminals or so can accommodate unit trains today. The distance of local
truck trips is taken as the direct distance between ethanol terminal and
blending site, plus 20% circuity.

The t(o,d,m) OD matrix (as discussed above in Sec. 4.1) describes every
flow from source to consumption. Shipments using the same source and
metro-area receiving terminal are consolidated, and routed over the ORNL
intermodal network. The ethanol sources are attached to the network by
constructing access links from their geographic locations to all the
nearby modes (within 1000 m). Cellulosic plants are located at county
centroids, and use existing county access links to connect to every mode
within that county, regardless of distance. Ports of entry for imported
ethanol are already existing network nodes. Receiving terminals are
similarly attached so that they, too, become network nodes. The
S-network containing lines representing shorelines, US county
boundaries, and North American provinces define county boundaries for
destination and origin ports by extending 3 miles into the water from
shorelines. A minimum impedance path is then found from each record’s
origin node to destination node, allowing any mode or mode combination.
The only connections allowed between modes are existing ethanol and
petroleum products terminals. 

Impedances are chosen to reproduce actual routes used by carriers such
as the data provided by Rail Traffic Atlas by Harry Ladd and Highway
Performance Monitoring System by Federal Highway Administration. For the
best, high volume routes (rural Interstates, rail A-mainlines),
impedances in terms of cost for highway, rail, and water are in the
ratio described above in Sect. 3.7, i.e., 14:7:2.4. This is where the
preponderance of ton-miles will occur. All waterways are treated
equally. Lower class highways and rail lines (surface arterials and
branch lines) receive substantial impedance penalties to shift traffic
to the major routes, in accordance with actual practice. Impedances are
defined only in terms of cost here since they include optimal balance
between congestion and infrastructure cost. As further discussed below
and in Sect. 3.8, congestion/delay may not add significant additional
impedances beyond transportation cost, thereby causing only additional
ton-mile movements existing corridors due to ethanol movements. The
expectation for the future routing of ethanol (or other commodity) is
that over an intermediate time range of 2 to 4 years the assurance of a
steady stream of revenues will finance any needed capacity improvements
along existing routes, and they will continue to be used. In other
words, existing capacities will not constrain future flows. The
interesting question is not whether expansion is economically justified,
but where it will occur, which our analysis attempts to address by
reporting traffic increases by each county and mode.

Unlike rail mode, congestion costs are different on highway and inland
water modes, where capital investment decisions are generally not
optimal for minimizing total system costs. By assuming optimal
investments in rail will continue to be made in the future, congestion
cost is not an issue in this case. Ethanol represents a small proportion
of total water traffic, but it will be affected by the growth in other
traffic. FAF2 projections are for approximately 25% growth up to 2022,
so congestion will rise slightly. Current congestion delays at each lock
on the Upper Mississippi are in the neighborhood of 1.4 hours. This
growth, without capacity improvements, should lead to delays in the
neighborhood of 1.6 hr as discussed earlier in Sect. 3.8.2 above. This
is still a small percentage penalty for the future use of barges
relative to rail. Water carriage costs in the future are likely to
increase due to fuel costs in about the same proportion as rail costs
will, leaving their relative costs nearly equivalent.

Highway congestion is expected to increase, and average speeds decline,
even in rural areas, over the next two decades as discussed above in
Sect. 3.8.3. This will slightly increase truck costs relative to rail.
More importantly, fuel represents a larger proportion of truck than rail
operating costs, where trucks consume about 9 times as much fuel per
ton-mile as discussed earlier under Sect. 3.7.3. This will further
increase the truck price disadvantage relative to rail. By rights, this
growth should have been incorporated in the future year cost
calculations, but the truck proportion of long-distance ethanol movement
is so small that the effect would have been barely perceptible. Truck
dominates the local ethanol distribution from metro-area receiving
terminals to blending sites, but this activity is price inelastic as the
only effective modal competition is from pipeline.

Impedance penalties for using intermodal transfer terminals in other
studies was roughly equivalent to 400 mi of travel, that is, a shipper
or carrier would be willing to accept a route 400 mi longer on a single
mode in order to avoid the costs of an intermodal transfer. This penalty
level resulted in no intermodal ethanol transfers occurring, when we
know that currently New England is served by barge from intermodal
terminals in New Jersey and Albany. To overcome this, intermodal
transfer impedances were reduced to the approximate equivalent of 100
miles. It could be that the costs of intermodal transfers really are
very low for ethanol; it could also be that the marginal costs for water
shipments to New England are low for tank farms that already have well
developed water unloading facilities and no rail facilities. During the
assignment process traffic is accumulated on each link for each O-D pair
whose path traverses it, and is then reported as ton-mi in the
geographic area that contains the link. Traffic is also accumulated at
the county level as the sum of long-distance haul (i.e., from production
refinery/import facilities to metro ethanol terminal) including
intermodal transfers and local truck movements from ethanol metro
terminal to blending terminal as the final ethanol destination point.

5. RESULTS

5.1 Transportation Activity

Transportation activity data by domestic truck, waterways, and rail have
been generated using the ORNL Infrastructure Network model for the three
2022 scenarios, i.e., controlled and AEO and RFS1 reference cases,
summary results of which are shown in Table 9. Amounts (in terms of
Ktons) and Kton-mile movements are listed in this table. The Kton-mile
movements are based on ethanol quantity loaded at the origination point
for each mode of transportation used; thus double-counting occurs from
this perspective.  Since same ethanol quantity may be used several times
depending on the number of times different transportation modes are used
to reach the final destination point, double-counting occurs under those
Kton-mile movement cases . The local Ktons of truck movements are
somewhat less than the final demand quantity that is to be delivered
from dedicated ethanol terminals to blending terminals since a small
amount of ethanol is moved directly from refineries to the blending
terminal. As observed today, major ethanol movements from production
centers to blending terminals will continue to be by rail. Under the
2022 controlled scenario, 72% of total quantity (i.e., in terms of
Ktons) of long-distance ethanol movements are estimated to be by rail.
By ton-mile movement, the estimate is 91% by rail. Unit-train movements
would dominate under the 2022 controlled scenario, to the extent that
manifest train movements become considerably less than under the two
reference scenarios. Ethanol ton-mile movements by waterways are
estimated to be quite low, i.e., about 6.9% of the total under the 2022
controlled scenario. Because of higher costs, the share of highway mode
of transportation is considerably less for the long-distance ethanol
movements. Local truck ton-mile movements would be the least among
various modes considered here due to the small distance to be covered
from the metro ethanol terminal to the blending terminal. Under the 2022
controlled scenario, 

Table 9. Transportation Activity Summary for the 2022 Scenarios



Scenario	

Ethanol Demand(Bgal)	Long- Distance 	Local



Rail (U)	Rail (M)	Barge	Truck	Truck



1	2	1	2	1	2	1	2	1	2

2022 RFS1 Reference	6.69	16800	17218K	1492	1413K	2816	1498K	3131	290K
18431	132K

2022 AEO Reference	13.18	28913	25362K	6182	5288K	4178	2539K	5597	491K
36736	253K

2022 Controlled	34.14	73656	46711K	2481	2705K	11391	3770K	17508	1153K
82260	532K

Note: 1: Ktons; 2: Ktons-mile; (U): Unit train; (M): Manifest train

there would be a 167% increase in total rail ton-mile movements compared
to the 2022 RFS1 scenario. The total volume of domestic freight by rail
in 2005 is estimated to be 1734 billion ton-miles; the ethanol shipments
in the 2022 controlled case increase rail transport about 2.8%.
Similarly, additional stresses from the 2022 controlled scenario on
waterways and highways are estimated to be 0.6% and 0.13%, respectively
of the corresponding 2005 ton-mile movements.

Figures 21 through 23 show Ktons of ethanol movements, including both
long-distance haul and local truck movements, for the three scenarios
considered here. Truck movements in most cases are local, with the
exception of the case where ethanol truck movement is the preferred mode
from a production refinery/import facility directly to the blending
terminal. Because of relatively less Kton movements by waterways and
truck, those movements in some cases overlap and are invisible in the
figures with higher rail Kton movements. Table 10 shows a sample of
high-volume ethanol movements (top three by rail, and one each by
waterways and truck) by different modes for the three 2022 scenarios. As
one would expect, top high-volume shipments are by unit train for all
three scenarios covering long-distance movements. In addition, most
movements are from the Midwest producing regions to high-demand regions,
i.e., northeast, west, and south. Long-distance highway ethanol
movements are mainly limited to within the same state. Because of the
location of dedicated ethanol terminals in proximity to ethanol blending
terminals, local highway mile movements are estimated to be the least in
each case.

Total ton-mile movements for the three 2022 scenarios at the
county-level are shown in Figures 24 thru 26, but not mode-specific
ton-mile movements. As one would expect with no ethanol demand in 21
states under the 2022 RFS1 scenario, no ton-mile movements are estimated
to occur in most southern and southwest states as shown in Figure 24.
The largest ton-mile movements are estimated to occur in San Bernardino
County, CA and quite a few counties in Wyoming. From the county area
perspective, the most intense ton-mile movements is estimated to occur
in the southwest of the United States. Although demand is relatively
less in those regions, larger county size causes total ton-mile
movements to be significantly more intense compared to other regions,
particularly northeast regions with a concentrated level of ethanol
demand. As anticipated, Midwest ethanol producing regions would also
have significant ethanol movements.

Table 10. A Sample of High-Volume Ethanol Movements by Different Modes 

for the Three 2022 Scenarios



Source County	Destination County	Long-Distance

Hwy

(miles)	Unit Train

(miles)	Water-ways

(miles)	Local Hwy

(miles)	Ethanol

Quantity

Shipped

(MGal)

2022 RFS1 Scenario

Floyd, IA	Chittenden, VT	0	1193	147	3	63

Linn, IA	Prince George's, MD	0	999	0	3	51

Hamilton, NE	Butte, CA	0	1607	0	3	48

St.Joseph, IN	Middlesex, CT	0	0	1542	3	19

Green Lake, WI	Fond du Lac, WI	31	0	0	0	30

2022 AEO Scenario

Linn, IA	Orange, FL	0	1437	0	3	86

Putnam, IL	Prince George's, MD	0	870	0	3	85

Macon, IL	Manatee, FL	0	1229	0	3	74

Des Moines, IA	Colbert, AL	0	0	700	3	59

Green Lakes, WI	Fond du Lac, WI	31	0	0	0	33

2022 Controlled Scenario

Bureau, IL	Prince George’s, MD	0	875	0	3	89

Tazewell, IL	Lorain, OH	0	478	0	3	82

St. Joseph, IN	Mercer, NJ	0	771	0	3	80

Hendry, FL	Manatee, FL	0	0	147	3	67

Lane, OR	Lane, OR	35	0	0	0	126



 Figure 21. Projected ethanol movements (Ktons) under the 2022 RFS1
Reference scenario

Figure 22. Projected ethanol movements (Ktons) under the 2022 AEO
Reference scenario

Figure 23. Projected ethanol movements (Ktons) under the 2022 Controlled
scenario

Figure 24. Projected county-level total Kton-mile ethanol movements
under the 2022 RFS1 Reference Scenario

Figure 25. Projected county-level total Kton-mile ethanol movements
under the 2022 AEO Reference Scenario

Figure 26. Projected county-level total Kton-mile ethanol movements
under the 2022 Controlled Scenario

Figure 27. Projected county-level total ethanol loading/unloading
activities (Ktons) under 2022 RFS1 Reference Scenario

 

Figure 29. Projected total ethanol loading/unloading activities (Ktons)
under 2022 Controlled Scenario

level of demand assumed at these sites. Due to the limited level of
estimated intermodal transfers occurring, loading/unloading activities
are not widespread among a large number of counties, as one would
expect. For example, about 4269 Ktons intermodal transfers are estimated
to occur out of total 380,021 Ktons loading/unloading activity under the
controlled 2022 scenario. Southern California and upper Midwest counties
dominate the most activities under all scenarios. The activity
distribution is similar to the distribution of supply centers (i.e.,
Figures 1 and 2) and demand centers (i.e., existing petroleum product
terminals) as shown earlier in Figure 12. Note that the focus of the
ethanol distribution analysis here has been from refineries to the
blending terminals. More activities would have been observed in these
figures with the consideration of final ethanol delivery from blending
terminals to retail outlets.

5.2 Distribution Cost

State-by-state and national average ethanol distribution costs for the
three 2022 scenarios were estimated based on the cost of ton-mile
movements to satisfy the projected demand. Cost of ton-mile movements
were determined by specific mode using the latest available data. The
data sources and methodology used for the cost of various transportation
modes are those discussed earlier under Sect. 3.7. For example, rail
cost was based on the regression analysis of operating cost of various
railroads as reported in the 2006 waybill data, to which monthly
maintenance and leasing costs were added to estimate the total rail cost
in terms of $/ton-mile as a function of distance traveled and fuel cost.
It was further assumed that current diesel fuel cost of around
$2.00/gallon will not be significantly different by the 2022 forecast
year. The unit train distribution cost is assumed to be 35% less than
for the manifest train. In addition, minimum threshold for rail cost has
been assumed to be 1.5 cents/ton-mile. Unlike rail and waterways, the
trucking cost is assumed to be fixed at 14 cents/ton-mile—not a
function of fuel cost as discussed earlier. The sum of mode-specific
costs of ton-mile movements when divided by projected total demand level
provided the composite ethanol distribution cost per gallon both at the
state and national level. Table 11 shows the estimated distribution cost
both in terms of total ($M) and cents/gal at the state-level for the
three 2022 scenarios considered.

It is estimated that the national average ethanol distribution cost
would be in the range of 6.80-9.63 cents/gallon, where the lower end of
the value range reflects the economies of scale obtained due to a
significantly higher level of demand under the 2022 controlled scenario.
At the state level, the distribution cost ranges between 1.2 – 33.2
cents/gallon under the same scenario. It is important to note here that
ethanol distribution cost is purely dictated by the assumed refinery and
ethanol distribution terminal locations. Lower ethanol distribution cost
values are mostly in midwestern states, as most demand is met by local
refineries imports causing thereby a significant lower level of ethanol
transportation. Alaska, Hawaii, some Southwestern states (for example
New Mexico, Utah etc.) and eastern states (e.g., Massachusetts, Vermont
etc.) have higher ethanol distribution costs because of their greater
distances from production locations. New 

Table 11. Estimated Ethanol Distribution Cost for the Three 2022
Scenarios



State	RFS1 2022	AEO 2022	Controlled 2022

	Total ($M)	Rate (¢/gal)	Total ($M) 	Rate (¢/gal)	Total ($M)	Rate
(¢/gal)

AL	-	-	23	8.7	32	6.0

AK	-	-	11	33.8	17	33.2

AZ	7.5	5.7	29	10.7	52	7.8

AR	-	-	9.2	18.6	21	7.6

CA	129	7.8	106	6.7	255	6.3

CO	18	10.1	28	8.3	71	10.1

CT	15	14.1	2.5	4.2	7.5	3.0

DE	-	-	6.1	20.4	6.2	7.1

FL	-	-	64	8.0	45	3.5

GA

-	19	8.0	79	7.1

HI	41	44.5	39	41.2	26	31.9

ID	-	-	2.5	4.7	16	6.6

IL	28	8.5	53	9.6	108	7.1

IN	9.6	6.9	35	9.0	45	5.4

IA	5.0	3.1	5.9	3.2	9.2	1.5

KS	0.2	10.7	13	7.0	24	5.4

KY	8.5	10.6	20	9.7	56	8.3

LA	-	-	36	14.7	12	2.2

ME	-	-	17	10.8	5.7	4.1

MD	28	12.3	32	13.3	52	9.5

MA	33	10.2	31	13.1	102	12.5

MI	32	10.5	61	11.3	144	10.0

MN	24	4.9	23	4.5	71	4.6

MS	-	-	21	9.9	33	5.8

MO	30	8.9	33	8.9	91	7.2

MT	-	-	1.9	19.9	10	13.4

NE	6.5	5.5	7.1	5.5	13	3.8

NV	6.0	9.8	6.3	7.2	12	6.3

NH	10	13.1	4.9	9.7	2.5	1.9

NJ	35	11.8	39	15.3	66	8.5

NM	16	13.8	13	12.6	49	12.4

NY	49	16.5	51	13.0	78	7.3

NC	-	-	-	-	11	3.7

ND	2.1	3.7	2.8	4.8	4.9	3.7

OH	-	-	19.6	6.2	58	5.1

OK	-	-	38	12.4	93	9.7

OR	5.5	6.2	4.6	4.7	3.7	1.2

PA	18	19.9	46	18.4	70	12.2

RI	-	-	1.5	7.5	2.5	2.2

SC	-	-	-	-	7.0	2.1

SD	0.8	2.4	1.2	3.0	30	1.3

TN	22	8.8	39	12.1	94	9.6

TX	-	-	167	12.1	160	6.8

UT	-	-	14	13.3	16	11.9

VT	6.3	10.0	8.3	13.1	27	11.0

VA	2.4	18.7	26	16.3	50	8.0

WA	-	-	23	6.5	42	4.7

WV	-	-	2.7	5.4	6.6	6.4

WI	24	6.3	2.6	6.2	71	5.2

WY	4.9	12.1	7.5	13.3	5.3	12.4

TOTAL	617	9.2	1269	9.6	2335	6.8



Hampshire has a significantly lower distribution cost since it is
cheaper for it to import from neighboring state Massachusetts, instead
of comparatively shorter distance for Massachusetts to import from the
Midwest than it is for New Hampshire. For the two other reference
scenarios, similar discrepancies in the ethanol distribution cost at the
state-level have been observed, however the distribution cost is
significantly higher. For example, in New Hampshire, ethanol
distribution cost is estimated to be 1.9 cents/gallon under the
controlled scenario, compared to 13.1 and 9.7 cents/gallon for AEO and
RFS1 scenarios, respectively. Under 2022 RFS1 scenario, a lack of
ethanol demand in quite a few states causes thereby there no
distribution cost.

Total distribution cost is mainly dictated by the ethanol demand level,
thereby larger demand states such as the California and Texas would have
the most distribution cost. Total annual ethanol distribution cost for
those states has been estimated to be $255M and $160M, respectively,
under the 2022 controlled scenario. Total ethanol distribution cost
expenditure is estimated to be $2,335M under the 2022 controlled
scenario, compared to $1,269M and $617M for the 2022 AEO and RFS1
scenarios, respectively.

5.3 Rolling Stock Requirements 

Rolling stock requirements, i.e., number of rail cars, barges, and
trucks, were estimated for the three 2022 scenarios at the state level
based on that state’s demand. Total estimated ton-mile movements to
satisfy the state level demand when divided by the mode specific speed,
capacity, and total yearly hours of operation (as shown in Table 12)
provided the mode-specific rolling stock requirements. Assumptions for
the number of rail car requirements are based mostly on the assumed
annual number of trips per car. Unit and manifest trains generally make
annually 15-18 trips and 10-12 trips, 

Table 12. Assumptions Used for Rolling Stock Requirements Estimation

Mode	Speed (mph)	Capacity (ton/vehicle)	Operating Time 

(hours/year)

Rail (Unit Train)	15	90	1848

Rail (Manifest)	10	90	1560

Waterways	8	800	6500

Highways	35	27	3500



respectively. We assume the higher value of the range in each case due
to anticipated efficiency in the future ethanol transportation network
Lower number of annual trips in rail cars is mainly due to long times
now used for loading, unloading, and idle time. For example, it is
estimated that loading and idle each may take in the range of 7-10 days
depending on how long it takes to gather all shipments from various
ethanol refineries, particularly in the case of manifest trains. Annual
rail car operating time is estimated based on the difference between
total number of annual operating hours and estimated total rail idle
time, where the latter is dictated by assumed train speed as shown in
the table below.  For example, for unit trains assuming 18 trips/year,
15 mph train speed, and average route distance of 700 miles lead to
about 20 days for a round-trip (i.e., load to load), out of which 4 days
when train will be running. The corresponding numbers for manifest
trains are estimated to be 12 trips/year, 10 mph train speed, 600 miles
of average route distance, 5 days train running out of 30 days of a
total round-trip. It is likely that assumed train speeds are on the
conservative side and applicable only under highly congested and strong
ethanol demand environment. The effect of assumed speed and average rail
route distance is minimum on the estimated annual operating hours, since
highly uncertain total non-operating rail time is the major component of
it. It is assumed that trucks will be operating 10 hours a day for 350
days annually, compared to double the amount of total yearly hours of
operation for waterways modes of transportation.

Increased ethanol demand is anticipated to increase the demand for rail
tank cars by a maximum of 41,301 cars under the 2022 controlled scenario
as shown in Table 13. This additional maximum demand for new rail cars
is estimated to be almost double the total new orders placed in 2007.
New orders for rail cars have steadily increased by 12% annually during
2005-2007 period, with a maximum value of 36,292 in 2006. It is
estimated that 60-65% of new rail tank car orders are from the increased
ethanol demand. There continues to be a backlog between the new orders
and deliveries, and the current backlog is estimated to be 25,900 tank
cars. As maximum demand for tank cars under the 2022 controlled scenario
is estimated to be more than double of total 2007 new orders, supply may
be an issue in that case if the high level current backlog continues in
the future. Even under the least constrained 2022 ethanol demand
scenario, i.e., 2022 RFS1,

rail car demand is estimated to be 65% of total new orders. Rail tank
cars are nearly all

Table 13. Estimated Rolling Stock Requirements under 2022 Scenarios

Scenario	Rail Cars	Barges	Trucks

2022 Controlled	41,301	170	1,643

2022 AEO	27,863	122	736

2022 RFS1	15,815	72	403



privately owned, either by leasing companies or shippers. It is likely
that by streamlining the ethanol rail movement in the future, a
significant reduction in delay and thereby total number of new ethanol
rail requirements can be substantially reduced.

It is estimated that total number of barges with each carrying capacity
of 800 tons necessary under the 2022 controlled scenario would be 170,
which is about 4% of total barges used in 2007. Almost 85% of the total
barges used in 2007 were along the Mississippi River system and the Gulf
of Mexico intracoastal waterways. Tank barges used to move petroleum
products and chemicals in the coastal and short-haul (Gulf/South
Atlantic) inter-coastal trades do not face the shortage of double-hulled
vessels as required by the Oil Pollution Act of 1990 (OPA90). OPA90
requires petroleum products and certain petrochemicals to be shipped in
double hulled vessels and a time line to phase out the use of single
hulled vessels. Several single hulled vessels built prior to 1970 were
recently retired from petroleum products service. The double-hulling of
the coastal tank vessel fleets will be virtually complete over the next
three years. The double-hulling process accelerated over the last five
years as non-double-hull vessels built during the 1978-1983 boom period
reached their OPA90 phase-out dates. As of year-end 2007, 73% of total
U.S. flagged tanker barge vessels were double hulled. It is unlikely
that the availability of double-hulled vessels would be an issue to meet
the increased ethanol demand in 2022.

It is estimated that about 1643 maximum more trucks will be on the road
due to increased ethanol demand under the 2022 controlled
scenario—considerably less than other transportation needs, as
trucking will mainly be used for short-distance hauling from dedicated
ethanol terminals to blending terminals. This additional truck demand
will contribute to less than 0.02% of total registered trucks in 2005.

5.4 Distribution Constraint Analysis

As discussed in Section 2, four scenarios were considered for
distribution constraint analysis, out of which three scenarios were
simulated by incorporating additional biodiesel demand in the range of
3-7% of ethanol demand dictated by EISA requirements. The three “with
biodiesel” scenarios were for the years 2012, 2014, and 2022 denoted
here by 2012B, 2014B, and 2022B, respectively. The fourth scenario
considered was for the year 2022 without consideration of additional
biodiesel demand and is denoted as 2022 controlled scenario. To simplify
the analysis, it is assumed here that the location of both supply and
demand of additional biodiesel will be co-located along with original
ethanol locations. For simplicity’s sake, movement of biodiesel was
assumed to follow the same route as ethanol, only with amounts increased
by the additional level of biodiesel demand. This simplified assumption,
thus, explicitly does not consider the strain on feeder lines and other
specific routes that may be used due to different types of feedstocks
and location of production facilities suitable for biodiesel. 

The constraint analysis is based on at first updating the ORNL
intermodal network with latest information on all commodity flows (i.e.,
base), and then determining the stress generated in terms of the
percentage increase in base flows due to ethanol flows under various
scenarios by extent (in terms of miles) of various transportation modal
networks. For railroads where the preponderance of movements will occur,
the network was loaded with current traffic levels from flowing all 2006
Surface Transportation Board Carload waybill sample traffic. The waybill
records provide shipment information about the originating county, the
terminating county, the originating railroad, terminating railroad, and
up to six bridge railroads involved in carrying the shipment; the
interchange junction if the shipment is exchanged between railroads; the
type of service train used to carry the shipment; the car equipment used
to carry the shipment; and the number of carloads per year. For highways
and waterways, nationwide background traffic levels are quite
limited—particularly at the local street level in the former case. So,
a typical level of traffic that a waterway and highway can support is
used, and then future years’ ethanol traffic is shown as a percent of
that “nominal capacity.”

Tables 14 through 17 show the estimated number of miles of rail,
waterway and highway that will experience an increase in traffic given
various ethanol quantities in the four scenarios considered in the
distribution constraint analysis. As discussed above, the increase in
traffic due to ethanol has been estimated based on the reference 2006
flows of all commodities and have been grouped under eight major
categories. The traffic flow increase under the category “5%”
indicates a traffic flow increase in the range of 2-5%, and the number
of miles under the “0%” category indicates along where no ethanol
flows is estimated to occur. Total number of miles under “0%” is the
total number of miles considered in our network for the analysis, which
is significantly higher in the cases of rail and waterways since they
also include Canadian rail and worldwide waterways networks,
respectively. Most number of miles would experience an increase in the
traffic level of less than 10% due to additional ethanol flow. For rail,
most affected miles are where ethanol flow is in the range 10-100
million tons, compared to 0-10 million tons for highways. The effect on
domestic waterways is minimal, less than 2% carrying ethanol volume in
the range of 50-100 million tons. For the 2022 controlled scenario,
10,626 miles of U.S. rail network would have an additional traffic flow
in the range of 0-2% due to transportation of 20-50 million tons loads
of ethanol as shown in Table 17.

As one would expect, the number of miles most affected would be rail,
followed by highways and waterways, respectively. Under the 2022
controlled scenario, 17% rail miles would be affected, compared to 2%
and 0.1% for highways and waterways, respectively. It is likely that the
percentage miles affected value for waterways would be significantly
higher when considered on the basis of U.S.-controlled waterways miles.
An additional biodiesel demand in 2022, wouldn’t cause much change on
the total basis in various ethanol transportation modes, but under
various categories of increased traffic percentages, total number of
affected miles would increase. For example for 2-5% traffic flow
increase, 10,015 miles of rail network would be affected under the 2022B
scenario compared to 9,942 miles for the 2022 controlled scenario. In
prior forecast years, the impacts would be less than in 2022, as one
would expect. For both forecast years, i.e., 2012 and 2014, total rail
miles affected by additional ethanol flows would be similar, 14% and
15%, respectively, compared to 1% for highway miles for both forecast
years. 

It is more important to evaluate the spatial distribution of affected
network miles due to additional ethanol flows (shown in Figures 30-37),
rather than total percentage of affected network miles. For each
scenario, two figures are shown: one for rail and the other one
waterways and highways combined. These figures are very similar in
nature to Figures 21-23, with the major difference being different
colors show different “% increase in base flows” categories, while
the thickness of lines indicates the actual ethanol flow volume as
before. Since end values of various ranges indicating ethanol flow
volumes are the same, thickness of flow lines becomes larger from 2012B
scenario to 2022B scenario with the increased ethanol ton-mile movements
due to higher ethanol demand. The notable features of rail flow (as
shown in Figures 30, 32, 34, and 36) are the increased loadings of the
Overland Corridor from the Corn Belt to the West Coast, and local
loadings of feeder lines within the Corn Belt. Note, however, that
long-distance loadings seldom exceed 10% of base, and that
transportation fees will easily cover the modest increases in
infrastructure required. For the controlled case, ethanol flows are not
that great as imports satisfy a large extent of total ethanol demand,
particularly coastal lying states.

Figures 31, 33, 35, and 37 indicate the spatial distribution of affected
waterways and highways network miles combined for the four scenarios
considered here, since each of these transportation modes contribute
only a small share of total ethanol movements. For waterways, although
congestion levels are currently significant on the upper Mississippi and
Ohio Rivers, ethanol will still be a small proportion of all traffic,
other traffic is not expected to grow, and at worst lock delays will
increase from the current half hour to the hour range, and possibly will
increase not at all. Note that for highways, local trucking from
centralized distribution terminals to petroleum blending terminals is
not considered, and so due to limited long-distance expensive highway
mode of transportation, not much highway ethanol flows are estimated for
the right half of the U.S. regions as shown in these figures. Congestion
will surely increase, but highway congestion is dominated by non-freight
traffic, and long-distance ethanol movements will be very modest. The
exception is local traffic in corn-producing areas, with often a
longer-distance movement of ethanol to transshipment points than the
corn shipments to elevators that they displace. There is a possibility
of consequential impacts on local roads in the vicinity of refineries,
both for incoming feedstock and outgoing ethanol.

Table 14. Estimated Number of Miles Experiencing an Increase in Ethanol
Traffic of Various Categories of Quantity Level under “2012B
Scenario”

RAIL

Traffic	0%	2%	5%	10%	20%	50%	100%	>100%

0Mt	4413	0	0	0	0	0	0	10

0-2Mt	60291	299	316	358	271	320	37	0

2-5Mt	15174	617	311	306	156	123	3	0

5-10Mt	26247	1101	757	773	143	197	0	0

10-20Mt	32226	4377	1216	312	397	20	0	0

20-50Mt	45157	8592	3053	895	48	0	0	0

50-100Mt	11055	5194	2146	610	0	0	0	0

100-500Mt	0	0	0	0	0	0	0	0

Total (miles)	194563	20180	7799	3254	1015	660	40	10

% Rail miles affected = 14%



WATERWAYS

Traffic	0%	2%	5%	10%	20%	50%	100%	>100%

0Mt	2070	0	0	0	0	0	0	5

0-2Mt	0	0	0	0	0	0	0	0

2-5Mt	0	0	0	0	0	0	0	0

5-10Mt	0	0	0	0	0	0	0	0

10-20Mt	0	0	0	0	0	0	0	0

20-50Mt	0	0	0	0	0	0	0	0

50-100Mt	2601879	430	0	0	0	0	0	0

100-500Mt	0	0	0	0	0	0	0	0

Total (miles)	2603949	430	0	0	0	0	0	5



HIGHWAYS

Traffic	0%	2%	5%	10%	20%	50%	100%	>100%

0Mt	3457	0	0	0	0	0	0	1

0-2Mt	295290	2151	746	401	190	38	1	0

2-5Mt	138939	2144	520	45	0	0	0	0

5-10Mt	27397	959	30	2	0	0	0	0

10-20Mt	10031	32	0	0	0	0	0	0

20-50Mt	45189	193	0	0	0	0	0	0

50-100Mt	109	0	0	0	0	0	0	0

100-500Mt	0	0	0	0	0	0	0	0

Total (miles)	520412	5479	1296	448	190	38	1	1

% Highway miles affected = 1%





Table 15. Estimated Number of Miles Experiencing an Increase in Ethanol
Traffic at Various Categories of Quantity Level under  “2014B
Scenario”

RAIL

Traffic	0%	2%	5%	10%	20%	50%	100%	>100%

0Mt	4415	0	0	0	0	0	0	8

0-2Mt	60442	110	290	448	278	286	35	3

2-5Mt	15095	549	486	155	265	135	5	0

5-10Mt	26087	1076	1360	343	181	165	7	0

10-20Mt	32306	3960	1381	524	304	74	0	0

20-50Mt	44677	8849	3110	1057	53	0	0	0

50-100Mt	10522	5536	2099	848	0	0	0	0

100-500Mt	0	0	0	0	0	0	0	0

Total (miles)	193544	20080	8726	3375	1081	660	47	11

% Rail miles affected = 15%



WATERWAYS

Traffic	0%	2%	5%	10%	20%	50%	100%	>100%

0Mt	2071	0	0	0	0	0	0	5

0-2Mt	0	0	0	0	0	0	0	0

2-5Mt	0	0	0	0	0	0	0	0

5-10Mt	0	0	0	0	0	0	0	0

10-20Mt	0	0	0	0	0	0	0	0

20-50Mt	0	0	0	0	0	0	0	0

50-100Mt	2601808	501	0	0	0	0	0	0

100-500Mt	0	0	0	0	0	0	0	0

Total (miles)	2603879	501	0	0	0	0	0	5



HIGHWAYS

Traffic	0%	2%	5%	10%	20%	50%	100%	>100%

0Mt	3457	0	0	0	0	0	0	1

0-2Mt	295004	1964	937	624	276	12	1	0

2-5Mt	138914	1954	705	75	0	0	0	0

5-10Mt	27458	895	34	2	0	0	0	0

10-20Mt	10027	36	0	0	0	0	0	0

20-50Mt	45282	100	0	0	0	0	0	0

50-100Mt	109	0	0	0	0	0	0	0

100-500Mt	0	0	0	0	0	0	0	0

 Total (miles)	520251	4949	1676	701	276	12	1	1

% Highway miles affected = 1%



Table 16. Estimated Number of Miles Experiencing an Increase in Ethanol
Traffic of Various Categories of Quantity Level  under  “2022B
Scenario”

RAIL

Traffic	0%	2%	5%	10%	20%	50%	100%	>100%

0Mt	4408	0	0	0	0	0	0	15

0-2Mt	60165	174	228	315	497	411	85	17

2-5Mt	14752	301	680	319	379	202	56	1

5-10Mt	25620	1238	675	662	794	64	165	0

10-20Mt	32042	2455	2195	1385	273	199	0	0

20-50Mt	40815	10575	3548	2419	388	0	0	0

50-100Mt	10603	3933	2689	877	902	0	0	0

100-500Mt	0	0	0	0	0	0	0	0

Total (miles)	188405	18676	10015	5977	3233	876	306	33

% Rail miles affected = 17%



WATERWAYS

Traffic	0%	2%	5%	10%	20%	50%	100%	>100%

0Mt	2067	0	0	0	0	0	0	8

0-2Mt	0	0	0	0	0	0	0	0

2-5Mt	0	0	0	0	0	0	0	0

5-10Mt	0	0	0	0	0	0	0	0

10-20Mt	0	0	0	0	0	0	0	0

20-50Mt	0	0	0	0	0	0	0	0

50-100Mt	2600715	1593	0	0	0	0	0	0

100-500Mt	0	0	0	0	0	0	0	0

Total (miles)	2602782	1593	0	0	0	0	0	8



HIGHWAYS

Traffic	0%	2%	5%	10%	20%	50%	100%	>100%

0Mt	3458	0	0	0	0	0	0	0

0-2Mt	294018	2154	952	961	412	288	34	0

2-5Mt	138163	2074	529	458	418	6	0	0

5-10Mt	27017	1156	164	51	0	0	0	0

10-20Mt	9984	64	15	0	0	0	0	0

20-50Mt	44931	451	0	0	0	0	0	0

50-100Mt	109	0	0	0	0	0	0	0

100-500Mt	0	0	0	0	0	0	0	0

Total (miles)	517680	5899	1660	1470	830	294	34	0

% Highway miles affected = 2%





Table 17. Estimated Number of Miles Experiencing an Increase in Ethanol
Traffic of Various Categories of Quantity Level under the “2022
Controlled Scenario”

RAIL

Traffic	0%	2%	5%	10%	20%	50%	100%	>100%

0Mt	4408	0	0	0	0	0	0	15

0-2Mt	60179	160	240	302	584	325	85	16

2-5Mt	14752	391	592	330	414	155	56	1

5-10Mt	25620	1241	673	663	797	60	165	0

10-20Mt	32042	2614	2366	1098	277	152	0	0

20-50Mt	40955	10626	3366	2411	387	0	0	0

50-100Mt	10603	3950	2705	844	902	0	0	0

100-500Mt	0	0	0	0	0	0	0	0

Total (miles)	188559	18982	9942	5648	3361	692	306	32

% Rail miles affected = 17%



WATERWAYS

Traffic	0%	2%	5%	10%	20%	50%	100%	>100%

0Mt	2067	0	0	0	0	0	0	8

0-2Mt	0	0	0	0	0	0	0	0

2-5Mt	0	0	0	0	0	0	0	0

5-10Mt	0	0	0	0	0	0	0	0

10-20Mt	0	0	0	0	0	0	0	0

20-50Mt	0	0	0	0	0	0	0	0

50-100Mt	2600793	1516	0	0	0	0	0	0

100-500Mt	0	0	0	0	0	0	0	0

Total (miles)	2602860	1516	0	0	0	0	0	8



HIGHWAYS

Traffic	0%	2%	5%	10%	20%	50%	100%	>100%

0Mt	3458	0	0	0	0	0	0	0

0-2Mt	294037	2181	999	908	371	288	34	0

2-5Mt	138223	2022	541	447	409	6	0	0

5-10Mt	27028	1147	170	44	0	0	0	0

10-20Mt	9984	77	2	0	0	0	0	0

20-50Mt	44965	417	0	0	0	0	0	0

50-100Mt	109	0	0	0	0	0	0	0

100-500Mt	0	0	0	0	0	0	0	0

Total (miles)	517804	5844	1712	1399	780	294	34	0

% Highway miles affected = 2%

 

Figure 30. Projected rail ethanol movements in terms of Ktons and %
increase in base flows for 2012B scenario.

Figure 31. Projected waterways and highways combined ethanol movements
in terms of Ktons and % increase in base flows for 2012B scenario.

Figure 32. Projected rail ethanol movements in terms of Ktons and %
increase in base flows for 2014B scenario.

Figure 33.  Projected waterways and highways combined ethanol movements
in terms of Ktons and % increase in base flows 

for 2014B scenario.

Figure 34. Projected rail ethanol movements in terms of Ktons and %
increase in base flows for 2022 scenario.

Figure 35. Projected waterways and highways combined ethanol movements
in terms of Ktons and % increase

in base flows for 2022 scenario.

Figure 36. Projected rail ethanol movements in terms of Ktons and %
increase in base flows for 2022B scenario

Figure 37.  Projected waterways and highways combined ethanol movements
in terms of Ktons and % increase in base flows 

for 2022B scenario.

6. Summary

The projected need for 34.14 billion gallons of ethanol by 2022 under
the EISA compared to the 2007 production level of 6.5 billion gallons
will likely spur demand for transportation services that face capacity
constraints. Incongruity between the geographic locations of production
facilities and feedstock supply networks and the locations of the
petroleum distribution system and consumption demand centers would
require adequate infrastructure development to avoid serious logistical
bottlenecks and systemic delays in biofuels distribution. This study
provides an analysis of ethanol transport by domestic rail, waterways,
and truck distribution systems from ethanol refineries to blending
terminals using ORNL’s Infrastructure Network model. The scope of this
study was limited to the intermediate portion of the entire ethanol
supply chain from the ethanol producer/point of importation to the
terminal where ethanol is blended with gasoline., Distribution issues
related to transportation of feedstocks for ethnaol production and of
ethanol from blending terminals to retail gas outlets was not considered
in this study. Although the focus of this study has been on ethanol,
several scenarios have been considered that include simulating
additional demand for biodiesel and non-co-processed renewable diesel
fuel, in the range of 3-7% of total ethanol demand.

Two scenario sets have been considered to facilitate EPA’s to
evaluation of the adequacy of the domestic biofuel transportation system
to deliver the volumes of biofuels projected to be used in response to
the requirements of the EISA. The first set of scenarios considered
three 2022 scenarios (i.e. two reference cases having a demand level of
13.2 billion gallons and 6.7 billion gallons denoted by AEO and RFS1,
respectively, and one based on the EISA of 34.14 billion gallons
ethanol), focusing on the transportation activity analysis of ethanol
from the point of production to the blending terminal-level on a
county-by-county basis. This analysis is to support the EPA analysis
activity related to the estimation of emissions associated with the
transportation of ethanol. The constraint analysis (the focus of the
second set of scenario analyses) was conducted to evaluate potential
constraints in the biofuel distribution system which may limit the
ability of the system to cope with the projected increase in biofuel
volumes under the EISA. Four scenarios have been considered here, where
three scenarios represent three forecast years under EISA, i.e., 2012,
2014, and 2022 and for each forecast year total biofuels demand includes
also biodiesel/non-co-processed renewable diesel fuel demand. Total
biofuel demand assumed for these three forecast years are 14.6, 17.5,
and 35.1 billion gallons, respectively. The fourth scenario is exactly
the same as the 34.14 billion gallons of ethanol volume projected to be
used in 2022 under EISA—one of the scenarios considered under the
first set of scenarios.

The analysis of fuel ethanol transportation and activity and potential
distribution constrains using the ORNL Infrastructure Network model are
based on the data provided both by EPA and available commercial sources.
EPA provided major data inputs by each forecast year including the
estimated location of ethanol production facilities, sources of ethanol
imports, and state-level consumption for the annual volumes of ethanol
that EPA estimates will be consumed in response to the EISA. EPA also
provided the projected volumes of biodiesel and non-co-processed
renewable diesel fuel that would be used under the EISA. To simplify the
analysis, biodiesel and non-co-processed diesel fuel volumes were
assumed to originate from the ethanol production facilities and follow
projected ethanol use patterns. Several geographic information systems
(GIS) were used for the geo-location of refineries needed by the ORNL
model. EPA provided state-level ethanol imports were geo-coded at the
specific port level and ethanol demand were mapped to the existing
petroleum product terminals using the vehicle-miles-traveled data at the
census tract level. Commercial data source was used for geo-coding the
1056 existing U.S. petroleum product terminals. It was assumed that the
concept of existing 10+ ethanol distribution terminals in metro areas
will continue, most likely occurring close to existing petroleum
terminals as found today that would accommodate unit trains, followed by
truck movements to possibly distant in the same metro region. A single
ethanol distribution terminal was selected for each metropolitan area
from the 413 cluster of existing petroleum product terminals based on
the minimum access rail distance. Hence, all three modes of
transportation, i.e., rail, waterways, and highways were considered for
the initial ethanol distribution from refineries to the ethanol
distribution terminal, followed by local trucking to the final blending
terminal. 

Estimation of transportation activity is based on the development of an
initial origin-destination matrix using the spatial interaction model,
followed by the actual ethanol route selection using the traffic
assignment process. Relative transportation cost (where waterways the
most preferred mode of transportation), total supply and demand level to
be met at various locations, and the distance between dictated the
linkage between various supply demand margins points. The traffic
assignment is based on the latest updated and validated ORNL intermodal
network and its traffic assignment models for both long-distance and
local ethanol transport, where the network is composed of independently
constructed single-mode networks of highways, rail, and water, along
with a set of intermodal terminals and a terminal mode to connect them.
Ethanol traffic assignment on the network is based on the minimum
impedance principle, i.e., relative transportation costs by various
transportation modes in addition to the reproduction of data available
of actual routes used by various carriers. The expectation for the
future routing of ethanol along existing routes presumes that the
assurance of a steady stream of revenues will finance any needed
capacity movements and thereby existing capacities will not constrain
future flows. Impedance penalties for using intermodal transfer
terminals was assumed to be equivalent to 400 miles of additional
travel, and in some cases lower where intermodal facilities provide the
most optimal transportation mode.

Transportation activity analysis of the three 2022 scenarios indicate a
significant increase in the level of ton-mile activity under the
controlled scenario compared to two other reference scenarios. As
observed today, major ethanol shipments from production centers to
blending terminals will continue to be by rail – 91% in terms of
ton-mile movements under the controlled scenario. Unit-train movements
would dominate under the controlled scenario to the extent that manifest
train movements become considerably lower than those under two reference
scenarios. Under the controlled scenario, there would also be an
increase of 167% in total rail movements compared to the 2022 RFS1
scenario. However, the additional stress on U.S. transportation network
is estimated to be minimal even under the controlled (i.e. EISA)
scenario. The percentage increases in ton-mile movements by rail,
waterways, and highways are estimated to be 2.8%, 0.6%, and 0.13%,
respectively, compared to the corresponding 2005 total domestic flows by
various modes. Most high-volume ethanol movements are estimated to occur
from the Midwest producing regions to high-demand regions, i.e.,
northeast, west, and south. At the county-level, larger county sizes,
particularly southwest quadrant of the United States indicate a
significantly higher level of ton-mile movements because of larger
distance covered. Concentrations of loading/unloading activities are
estimated to be in counties where ethanol supply centers (Midwest) and
blending terminals coincide with the locations of existing petroleum
product terminals.

State-by-state ethanol distribution cost shows the effect of economies
of scale, as demand level increases considerably in 2022 from RFS1
scenario to controlled scenario. It is estimated that the national
average ethanol distribution cost under the controlled scenario is 6.8
cents/gallon, compared to 9.2 cents/gallon under the RFS1 scenario. At
the state level, the distribution cost ranges between 1.2-33.2
cents/gallon under the 2022 controlled scenario. Lower ethanol
distribution cost values are mostly in midwestern states as most demand
is met by local refineries causing thereby a significantly lower level
of ethanol transportation. Some of the southwestern states and eastern
states have higher ethanol distribution cost as most ethanol
distribution occurs by the long-distance unit rail mode of
transportation. Assumptions made regarding locations of refineries and
ethanol distribution terminals are key factors to these distribution
cost estimates. Total ethanol distribution cost expenditure is estimated
to be $2,335M under the controlled scenario, compared to $1,269M and
$617M for the 2022 AEO and RFS1 scenarios, respectively. States with
comparatively a higher level of ethanol demand as well as having fewer
local refineries such as California and Texas indicate the most annual
ethanol distribution cost expenditures.  

Rolling stock requirements, i.e., number of rail cars, barges, and
trucks were estimated for the first set of three 2022 scenario
considered under the ethanol transportation activity analysis at the
state level based on its demand. Increased ethanol demand is anticipated
to increase the demand for rail tank cars by 41,301 under the 2022
controlled scenario, compared to 15,815 under the 2022 RFS1 scenario.
This maximum demand for rail tank cars is estimated to be about double
the total new orders placed in 2007. The increase in number of barges is
considerably less, about 4% total barges used in 2007 under the 2022
controlled scenario. It is unlikely that the availability of
double-hulled vessels under OPA90 would be an issue to meet the
increased ethanol demand, since the double-hulling process has
accelerated over the last five years and as of year-end 2007, 73% of
total U.S. flagged tanker barge vessels were double-hulled. Since
trucking is limited to short-distance hauling from dedicated ethanol
terminals to blending terminals, the additional truck demand is
estimated to contribute to less than 0.02% of total registered trucks in
2005.

The distribution constraint analysis considered for four scenarios
encompassing three different forecast years indicate the most network
miles affected compared to 2006 base commodity flows would be the rail
as it’d be used as the major mode of ethanol transportation. The
percentage miles affected would be 17% for the 2022 scenarios, compared
to 2-3% lower in 2012 and 2014 scenarios. The effect would be
considerably less in the case of highways, i.e., 1-2%, range depending
on the scenario as trucking is used mainly for the ethanol
transportation from central ethanol distribution terminals to existing
petroleum product terminals. The notable features of rail flow are the
increased loadings of the Overland Corridor from the Corn Belt to the
West Coast, and local loadings of feeder lines within the Corn Belt. For
waterways, although congestion levels are currently significant on the
upper Mississippi and Ohio Rivers, ethanol will still be a small
proportion of all traffic, other traffic is not expected to grow, and at
worst lock delays will increase from the current half hour to the hour
range, and possibly will increase not at all. Congestion will surely
increase, but highway congestion is dominated by non-freight traffic,
and long-distance ethanol movements will be very modest. The difference
in additional stress on the U.S. transportation network due to biodiesel
demand consideration was minimal (as observed between the 2022
controlled vs. 2022B cases) since the difference in demand between them
was only 4%. It is also likely that the impacts would be more on feeder
lines and different supply chain network representing biodiesel if
considered in the analysis.

In summary, significant future ethanol demand would have minimal impacts
on transportation infrastructure overall.  In addition, the simplified
analysis of biodiesel/non-co-processed renewable fuel transportation
indicates that the additional burden on the distribution system from the
increased volumes of these fuels (relatively small compared to ethanol)
under the EISA also would have minimal impacts. However, there will be
spatial impacts due to a considerable increase in rail traffic from
refineries to ethanol distribution terminals requiring a significant
level of investment. Assumed locations of both refineries and ethanol
distribution terminals would impact significantly not only the projected
ethanol flows but also distribution cost. Feedstock availability and the
relative cost of transporting ethanol vs. feedstock are some of the
major parameters that need to be considered for estimating refinery
locations. Similarly, intermodal ethanol terminal locations may be
different from existing petroleum-based intermodal terminals, depending
on ethanol supply locations and on the extent of availability of
pipelines (not considered in this study) as one of the modes of future
ethanol transportation.  Further more, any change in the current trend
of the citing of ethanol distribution terminals close to existing
petroleum product terminals into the future will affect the future
ethanol flows and thereby transportation infrastructure requirements.  

APPENDIX A

Location Details and Production Capacity (MGY) of Corn Ethanol Plants
for Various Forecast Years

Location Details and Production Capacity (MGY) of Corn Ethanol Plants
for Various Forecast Years

ID	FEEDSTOCK	LONGI-TUDE	LATI-TUDE	CITY	COUNTY	Vol

2010	Vol

2012	Vol

2014	Vol

2017

2022	

Vol

2022

Ref. AEO	

Vol

2022

Ref. RFS1

1	Abengoa Bioenergy Corporation	-97.53	37.79	Colwich	Sedgwick 	25.00
25.00	25.00	25.00	25.00	25

2	Abengoa Bioenergy Corporation	-101.39	37.18	Hugoton	Stevens 



88.00

0

3	Abengoa Bioenergy Corporation	-103.37	34.17	Portales	Roosevelt 	30.00
30.00	30.00	30.00	30.00	30

4	Abengoa Bioenergy Corporation	-98.87	41.02	Ravenna	Buffalo 	95.00
95.00	95.00	95.00	95.00	0

5	Abengoa Bioenergy Corporation	-97.54	40.87	York	York 	55.00	55.00
55.00	55.00	55.00	55

6	Aberdeen Energy, LLC	-98.78	45.43	Mina	Edmunds 	100.00	100.00	100.00
100.00	100.00	0

7	Absolute Energy, LLC	-92.92	43.38	St. Ansgar	Mitchell 	100.00	100.00
100.00	100.00	100.00	0

8	ACE Ethanol	-90.96	44.96	Stanley	Chippewa 	39.60	39.60	39.60	39.60
39.60	39.6

9	Adkins Energy, LLC	-89.80	42.36	Lena	Stephenson 	42.50	42.50	42.50
42.50	42.50	42.5

10	Advanced BioEnergy (ABE)	-97.60	40.61	Fairmont	Fillmore 	110.00
110.00	110.00	110.00	110.00	0

11	Ag Processing Inc. (AGP)	-98.34	40.59	Hastings	Adams 	52.00	52.00
52.00	52.00	52.00	52

12	Agri-Energy, LLC	-96.22	43.64	Luverne	Rock 	21.00	21.00	21.00	21.00
21.00	21

13	Al-Corn Clean Fuel	-93.00	44.04	Claremont	Dodge 	50.00	50.00	50.00
50.00	50.00	50

14	AltraBiofuels Coshocton Ethanol LLC.	-81.86	40.21	Coshocton	Coshocton
	60.00	60.00	60.00	60.00	60.00	0

15	AltraBiofuels Indiana, LLC.	-86.80	39.54	Cloverdale	Putnam 	92.00
92.00	92.00	92.00	92.00	0

16	AltraBiofuels Nebraska, LLC.	-97.69	40.30	Carleton	Thayer 	113.00
113.00	113.00	113.00	113.00	0

17	AltraBiofuels Phoenix Bio Industries, LLC.	-119.43	36.36	Goshen
Tulare 	31.50	31.50	31.50	31.50	31.50	31.5

18	Amaizing Energy	-95.40	41.98	Denison	Crawford 	60.00	60.00	60.00
60.00	60.00	60

19	Amaizing Energy Atlantic, LLC (formerly Cassco & EK-SEN)	-95.03	41.42
Atlantic	Cass 

	100.00	100.00

0

20	Archer Daniels Midland (ADM)	-91.69	41.93	Cedar Rapids	Linn 	250.00
250.00	250.00	250.00	250.00	250

21	Archer Daniels Midland (ADM)	-91.69	41.93	Cedar Rapids	Linn 	275.00
275.00	275.00	275.00	275.00	0

22	Archer Daniels Midland (ADM)	-90.21	41.82	Clinton	Clinton 	190.00
190.00	190.00	190.00	190.00	190

23	Archer Daniels Midland (ADM)	-97.29	41.42	Columbus	Platte 	275.00
275.00	275.00	275.00	275.00	0

24	Archer Daniels Midland (ADM)	-97.29	41.42	Columbus	Platte 	95.00
95.00	95.00	95.00	95.00	95

25	Archer Daniels Midland (ADM)	-88.90	39.86	Decatur	Macon 	290.00
290.00	290.00	290.00	290.00	290

26	Archer Daniels Midland (ADM)	-95.79	44.46	Marshall	Lyon 	40.00	40.00
40.00	40.00	40.00	40

27	Archer Daniels Midland (ADM)	-89.61	40.68	Peoria	Peoria 	210.00
210.00	210.00	210.00	210.00	210

28	Archer Daniels Midland (ADM)	-97.89	48.91	Walhalla	Pembina 	25.00
25.00	25.00	25.00	25.00	25

29	Arkalon Energy	-100.92	37.04	Liberal	Seward 	110.00	110.00	110.00
110.00	110.00	0

30	Aventine Aurora West	-98.04	40.86	Aurora	Hamilton 

	113.00	113.00

0

31	Aventine Renewable Energy - Mt. Vernon, LLC.	-87.85	37.92	Mount
Vernon	Posey 

	113.00	113.00

0

32	Aventine Renewable Energy, Inc.	-89.66	40.56	Pekin	Tazewell 	56.50
56.50	56.50	56.50	56.50	56.5

33	Aventine Renewable Energy, Inc. (formerly Williams Energy)	-89.66
40.56	Pekin	Tazewell 	100.00	100.00	100.00	100.00	100.00	100

34	Badger State Ethanol, LLC	-89.66	42.60	Monroe	Monroe 	52.00	52.00
52.00	52.00	52.00	52

35	Big River Resources, LLC	-91.23	40.84	West Burlington	Des Moines 
92.00	92.00	92.00	92.00	92.00	52

36	Big River Resources, LLC - Galva	-90.03	41.16	Galva	Henry 

	100.00	100.00

0

37	Blue Flint Ethanol LLC	-101.19	47.46	Underwood	McLean 	50.00	50.00
50.00	50.00	50.00	50

38	Bonanza BioEnergy	-100.83	37.95	Garden City	Finney 	55.00	55.00	55.00
55.00	55.00	0

39	Bridgeport Ethanol, LLC.	-103.14	41.71	Bridgeport	Morrill 

	45.00	45.00

0

40	Buffalo Lake Energy, LLC	-94.49	43.67	Fairmont	Martin 	115.00	115.00
115.00	115.00	115.00	0

41	Bushmills Ethanol	-94.79	45.13	Atwater	Kandiyohi 	49.00	49.00	49.00
49.00	49.00	49

42	Calgren Renewable Fuels	-119.30	36.00	Pixley	Tulare 	55.00	55.00
55.00	55.00	55.00	0

43	Cardinal Ethanol	-84.86	40.19	Harrisville / Union City	Randolph 

100.00	100.00	100.00	100.00	0

44	Cargill Inc.	-92.65	41.14	Eddyville	Wapello 	35.00	35.00	35.00	35.00
35.00	35

45	Cargill, Inc.	-96.10	41.54	Blair	Washington 	85.00	85.00	85.00	85.00
85.00	85

46	Cascade Grain Products	-123.16	46.17	Clatskanie	Columbia 	113.00
113.00	113.00	113.00	113.00	0

47	Castlerock Renewable Fuels, LLC.	-90.07	43.99	Necedah	Juneau 	50.00
50.00	50.00	50.00	50.00	0

48	Center Ethanol Co.	-90.18	38.60	Sauget	St. Clair 	54.00	54.00	54.00
54.00	54.00	0

49	Central Indiana Ethanol, LLC	-85.71	40.56	Marion	Grant 	48.00	48.00
48.00	48.00	48.00	48

50	Central Minnesota Ethanol Co-op (CMEC)	-94.34	46.02	Little Falls
Morrison 	21.50	21.50	21.50	21.50	21.50	21.5

51	Chief Ethanol Fuels Inc.	-98.34	40.58	Hastings	Adams 	62.00	62.00
62.00	62.00	62.00	62

52	Chippewa Valley Ethanol Co., LLC (CVEC)	-95.61	45.31	Benson	Swift 
46.00	46.00	46.00	86.00	46.00	46

53	Cilion	-120.92	37.56	Keyes	Stanislaus 	55.00	55.00	55.00	55.00	55.00
0

54	Commonwealth Agri-Energy, LLC	-87.41	36.82	Hopkinsville	Christian 
35.00	35.00	35.00	35.00	35.00	35

55	Corn LP	-93.92	42.73	Goldfield	Wright 	50.00	50.00	50.00	50.00	50.00
50

56	Corn Plus, LLP	-94.16	43.76	Winnebago	Faribault 	46.00	46.00	46.00
46.00	46.00	46

57	Cornhusker Energy Lexington (CEL)	-99.73	40.77	Lexington	Dawson 
40.00	40.00	40.00	150.00	40.00	40

58	Dakota Ethanol, LLC	-96.96	43.98	Wentworth	Lake 	48.00	48.00	48.00
48.00	48.00	48

59	Dean CEG LLC	-99.18	35.36	Burns Flat	Washita 	2.00	2.00	2.00	2.00
2.00	2

60	Didion Milling / Grand River Distribution	-89.10	43.54	Cambria /
Courtland	Columbia 	50.00	50.00	50.00	50.00	50.00	0

61	Diversified Energy Company (DENCO), LLC	-95.91	45.57	Morris	Stevens 
25.00	25.00	25.00	25.00	25.00	25

62	E Caruso Ethanol (part of Goodland Energy Center)	-101.71	39.37
Goodland	Sherman 	20.00	20.00	20.00	20.00	20.00	0

63	E Energy Adams	-96.55	40.49	Adams	Adams 	50.00	50.00	50.00	50.00
50.00	0

64	East Kansas Agri-Energy, LLC (EKAE)	-95.24	38.27	Garnett	Anderson 
35.00	35.00	35.00	35.00	35.00	35

65	Elkhorn Valley Ethanol	-97.39	42.06	Norfolk	Madison 	40.00	40.00
40.00	40.00	40.00	0

66	ESE Alcohol Inc.	-101.41	38.48	Leoti	Wichita 	1.50	1.50	1.50	1.50
1.50	1.5

67	Ethanol Grain Processors, LLC (EGP)	-89.15	36.29	Obion / Rives	Obion 

100.00	100.00	100.00

0

68	First United Ethanol LLC (FUEL)	-84.16	31.17	Camilla	Mitchell 

100.00	100.00	100.00	100.00	0

69	Front Range Energy, LLC	-104.91	40.46	Windsor	Weld 	48.00	48.00	48.00
48.00	48.00	48

70	Glacial Lakes Energy, LLC (GLE)	-97.11	44.88	Watertown	Codington 
100.00	100.00	100.00	100.00	100.00	50

71	Global Ethanol	-94.09	43.38	Lakota	Kossuth 	100.00	100.00	100.00
200.00	100.00	100

72	Global Ethanol / Midwest Grain Processors	-83.85	41.82	Riga	Lenawee 
57.00	57.00	114.00	114.00	57.00	57

73	Golden Cheese Company of CA	-117.58	33.89	Corona	Riverside 	5.00	5.00
5.00	5.00	5.00	5

74	Golden Grain Energy LLC	-93.20	43.16	Mason City	Cerro Gordo 	120.00
120.00	150.00	150.00	120.00	120

75	Golden Triangle Energy, LLC	-95.37	40.20	Craig	Holt 	20.00	20.00
20.00	20.00	20.00	20

76	Grain Processing Corp	-91.06	41.40	Muscatine	Muscatine 	20.00	20.00
20.00	20.00	20.00	20

77	Granite Falls Energy, LLC (GFE)	-95.52	44.77	Granite Falls	Chippewa 
50.00	50.00	50.00	50.00	50.00	50

78	Greater Ohio Ethanol (GOE)	-84.11	40.74	Lima	Allen 	55.00	55.00	55.00
55.00	55.00	0

79	Green Plains Renewable Energy, Inc. (GPRE)	-95.40	40.76	Shenandoah
Page 	50.00	50.00	50.00	50.00	50.00	0

80	Green Plains Renewable Energy, Inc. (GPRE) (formerly Superior
Ethanol)	-94.95	43.43	Superior	Dickinson 	50.00	50.00	50.00	50.00	50.00
0

81	Hawkeye Menlo, LLC	-94.38	41.52	Menlo	Guthrie 

	115.00	115.00

0

82	Hawkeye Renewables, LLC	-92.03	42.64	Fairbank	Buchannan 	115.00
115.00	115.00	115.00	115.00	115

83	Hawkeye Renewables, LLC	-93.29	42.51	Iowa Falls	Hardin 	115.00	115.00
115.00	115.00	115.00	115

84	Hawkeye Shell Rock, LLC	-92.63	42.74	Shell Rock	Butler 

	115.00	115.00

0

85	Heartland Corn Products	-94.34	44.54	Winthrop	Sibley 	99.00	99.00
99.00	99.00	99.00	99

86	Heartland Grain Fuels, LP	-98.54	45.46	Aberdeen	Brown 	55.30	55.30
55.30	55.30	55.30	10

87	Heartland Grain Fuels, LP	-98.25	44.37	Huron	Beadle 	32.00	32.00
32.00	32.00	32.00	32

88	Heron Lake BioEnergy, LLC	-95.30	43.81	Heron Lake	Jackson 	50.00
50.00	50.00	50.00	50.00	0

89	Holt County Ethanol	-98.66	42.45	O’Neill	Holt 



100.00

0

90	Homeland Energy Solutions	-92.14	43.07	Lawler	Chickasaw 

	100.00	100.00

0

91	Husker Ag, LLC	-97.77	42.32	Plainview	Pierce 	26.00	66.00	66.00	66.00
66.00	26

92	Idaho Ethanol Processing, LLC (former J.R. Simplot plant)	-116.73
43.67	Caldwell	Canyon 	5.00	5.00	5.00	5.00	5.00	3

93	Illinois River Energy LLC (IRE)	-89.07	41.91	Rochelle	Ogle 	50.00
50.00	115.00	115.00	50.00	50

94	Indiana Bio-Energy, LLC	-85.17	40.76	Bluffton	Wells 

100.00	100.00	100.00	100.00	0

95	Iroquois BioEnergy Company, LLC (IBEC)	-87.06	40.93	Rensselaer	Jasper
	40.00	40.00	40.00	40.00	40.00	40

96	KAAPA Ethanol, LLC	-98.95	40.51	Minden	Kearney 	60.00	60.00	60.00
60.00	60.00	60

97	Kansas Ethanol, LLC	-98.22	38.35	Lyons	Rice 	55.00	55.00	55.00	55.00
55.00	0

98	Land O’ Lakes / Melrose Dairy Proteins	-94.80	45.67	Melrose	Stearns
	2.60	2.60	2.60	2.60	2.60	2.6

99	Levelland/Hockley County Ethanol LLC	-102.27	33.59	Levelland	Hockley 
40.00	40.00	40.00	40.00	40.00	0

100	LifeLine Foods / ICM	-94.85	39.74	St. Joseph	Buchanan 	45.00	45.00
45.00	45.00	45.00	0

101	Lincolnland Agri-Energy	-87.63	39.01	Palestine / Robinson	Crawford 
48.00	48.00	48.00	48.00	48.00	48

102	Lincolnway Energy LLC	-93.51	42.02	Nevada	Story 	50.00	50.00	50.00
50.00	50.00	50

103	Little Sioux Corn Processors	-95.76	42.82	Marcus	Cherokee 	100.00
100.00	100.00	100.00	100.00	88

104	Marquis Energy LLC.	-89.34	41.25	Hennepin	Putnam 	100.00	100.00
100.00	100.00	100.00	0

105	Marysville Ethanol	-82.49	42.88	Marysville	St. Clair 	50.00	50.00
50.00	50.00	50.00	0

106	Merrick & Company (Coors Brewery)	-105.20	39.76	Golden	Jefferson 
3.00	3.00	3.00	3.00	3.00	3

107	MGP Ingredients, Inc.	-95.13	39.56	Atchison	Atchison 	25.00	25.00
25.00	25.00	25.00	25

108	MGP Ingredients, Inc.	-89.67	40.56	Pekin	Tazewell 	90.00	90.00	90.00
90.00	90.00	90

109	Mid-Missouri Energy (MME)	-93.45	39.20	Malta Bend	Saline 	45.00
45.00	45.00	45.00	45.00	45

110	Midwest Renewable Energy, LLC (MRE)	-101.09	41.16	Sutherland	Lincoln
	20.00	20.00	20.00	20.00	20.00	20

111	Minnesota Energy	-94.62	44.74	Buffalo Lake	Renville 	18.00	18.00
18.00	18.00	18.00	18

112	Nebraska Energy	-98.04	40.86	Aurora	Hamilton 	50.00	50.00	50.00
50.00	50.00	50

113	NEDAK Ethanol	-98.98	42.53	Atkinson	Holt 	44.00	44.00	44.00	44.00
44.00	0

114	Nesika Energy	-97.82	39.79	Scandia	Republic 	10.00	10.00	10.00	10.00
10.00	0

115	New Energy Corp.	-86.29	41.65	South Bend	St. Joseph 	102.00	102.00
102.00	102.00	102.00	102

116	Nexsun Ethanol	-101.36	37.58	Ulysses	Grant 

	40.00	40.00

0

117	North Country Ethanol (NCE)	-96.74	45.88	Rosholt	Roberts 	25.00
25.00	25.00	25.00	25.00	25

118	Northeast Biofuels, LP (NEB)	-76.37	43.30	Volney	Oswego 	114.00
114.00	114.00	114.00	114.00	0

119	Northwest Renewable, LLC.	-122.98	46.14	Longview	Cowlitz 

55.00	55.00	55.00

0

120	One Earth Energy, LLC	-88.38	40.46	Gibson City	Ford 

	100.00	100.00

0

121	Orion Ethanol (formerly Gateway Ethanol)	-98.72	37.60	Pratt	Pratt 
55.00	55.00	55.00	55.00	55.00	0

122	Otter Tail Ag Enterprises (OTAE)	-96.09	46.30	Fergus Falls	Otter
Tail 	55.00	55.00	55.00	55.00	55.00	0

123	Pacific Ethanol - Columbia	-119.77	45.81	Boardman	Morrow 	40.00
40.00	40.00	40.00	40.00	0

124	Pacific Ethanol - Madera	-119.97	36.92	Madera	Madera 	40.00	40.00
40.00	40.00	40.00	40

125	Pacific Ethanol - Magic Valley	-113.81	42.52	Burley	Cassia 	50.00
50.00	50.00	50.00	50.00	0

126	Pacific Ethanol - Stockton	-121.34	37.94	Stockton	San Joaquin 

50.00	50.00	50.00	50.00	0

127	Panda Ethanol - Hereford	-102.40	34.82	Hereford	Deaf Smith 

115.00	115.00	115.00	115.00	0

128	Parallel Products	-117.54	34.10	Rancho Cucamonga	San Bernardino 
4.00	4.00	4.00	4.00	4.00	4

129	Parallel Products	-85.77	38.22	Shively / Louisville	Jefferson 	5.40
5.40	5.40	5.40	5.40	5.4

130	Patriot Renewable Fuels	-89.90	41.40	Annawan / Mineral	Henry 	100.00
100.00	100.00	100.00	100.00	0

131	Penford Products Co.	-91.67	41.97	Cedar Rapids	Linn 

45.00	45.00	45.00	45.00	0

132	Permeate Refining	-91.25	42.34	Hopkinton	Delaware 	1.50	1.50	1.50
1.50	1.50	1.5

133	Pinal Energy, LLC	-112.03	33.05	Maricopa	Pinal 	50.00	50.00	50.00
50.00	50.00	50

134	Pine Lake Corn Processors, LLC	-93.06	42.46	Steamboat Rock	Hardin 
20.00	20.00	20.00	20.00	20.00	20

135	Pioneer Trail Energy	-98.61	40.82	Wood River	Hall 	115.00	115.00
115.00	115.00	115.00	0

136	Platinum Ethanol , LLC. (formerly Willmark)	-95.34	42.28	Arthur	Ida 

110.00	110.00	110.00	110.00	0

137	Plymouth Energy, LLC	-96.25	42.72	Merrill	Plymouth 	50.00	50.00
50.00	50.00	50.00	0

138	POET Biorefining - Alexandria	-85.65	40.29	Alexandria	Madison 	65.00
65.00	65.00	65.00	65.00	0

139	POET Biorefining - Ashton	-95.81	43.27	Ashton	Osceola 	55.00	55.00
55.00	55.00	55.00	55

140	POET Biorefining - Big Stone	-96.55	45.30	Big Stone City	Grant 
75.00	75.00	75.00	75.00	75.00	75

141	POET Biorefining - Bingham Lake	-95.06	43.98	Bingham Lake	Cottonwood
	33.00	33.00	33.00	33.00	33.00	33

142	POET Biorefining - Caro	-83.41	43.47	Caro	Tuscola 	52.00	52.00	52.00
52.00	52.00	52

143	POET Biorefining - Chancellor	-96.96	43.37	Chancellor	Turner 	100.00
100.00	100.00	100.00	100.00	50

144	POET Biorefining - Coon Rapids	-94.63	41.86	Coon Rapids	Carroll 
54.00	54.00	54.00	54.00	54.00	54

145	POET Biorefining - Corning	-94.80	40.97	Corning	Adams 	60.00	60.00
60.00	60.00	60.00	60

146	POET Biorefining - Emmetsburg	-94.64	43.10	Emmetsburg	Palo Alto 
56.00	56.00	56.00	93.75	56.00	56

147	POET Biorefining - Fostoria	-83.39	41.17	Fostoria	Seneca 

65.00	65.00	65.00

0

148	POET Biorefining - Glenville East	-93.31	43.63	Glenville / Albert
Lea	Freeborn 	45.00	45.00	45.00	45.00	45.00	45

149	POET Biorefining - Glenville West	-93.31	43.63	Glenville / Albert
Lea	Freeborn 

65.00	65.00	65.00

0

150	POET Biorefining - Gowrie	-94.29	42.33	Gowrie	Webster 	62.00	62.00
62.00	62.00	62.00	62

151	POET Biorefining - Groton	-98.14	45.46	Groton	Brown 	53.00	53.00
53.00	53.00	53.00	53

152	POET Biorefining - Hamlontown	-93.39	43.29	Hanlontown	Worth 	45.00
45.00	45.00	45.00	45.00	45

153	POET Biorefining - Hudson	-96.48	43.10	Hudson	Lincoln 	55.00	55.00
55.00	55.00	55.00	55

154	POET Biorefining - Jewell	-93.66	42.33	Jewell	Hamilton 	62.00	62.00
62.00	62.00	62.00	62

155	POET Biorefining - Laddonia	-91.65	39.25	Laddonia	Audrain 	50.00
50.00	50.00	50.00	50.00	50

156	POET Biorefining - Lake Crystal	-94.27	44.14	Lake Crystal	Blue Earth
	56.00	56.00	56.00	56.00	56.00	56

157	POET Biorefining - Leipsic	-83.99	41.10	Leipsic	Putnam 	65.00	65.00
65.00	65.00	65.00	0

158	POET Biorefining - Macon	-92.38	39.75	Macon	Macon 	42.00	42.00	42.00
42.00	42.00	42

159	POET Biorefining - Marion	-83.16	40.62	Marion	Marion 

65.00	65.00	65.00	65.00	0

160	POET Biorefining - Mitchell	-98.10	43.80	Mitchell	Davison 	60.00
60.00	60.00	60.00	60.00	60

161	POET Biorefining - North Manchester	-85.80	40.94	North Manchester
Wabash 

65.00	65.00	65.00

0

162	POET Biorefining - Portland	-85.02	40.42	Portland	Jay 	65.00	65.00
65.00	65.00	65.00	0

163	POET Biorefining - Preston	-92.09	43.69	Preston	Fillmore 	42.00
42.00	42.00	42.00	42.00	42

164	POET Research Center	-97.72	43.15	Scotland	Bon Homme 	9.00	9.00	9.00
9.00	9.00	9

165	Prairie-Horizon Agri-Energy, LLC (PHAE)	-99.33	39.76	Phillipsburg
Phillips 	40.00	40.00	40.00	40.00	40.00	40

166	Quad-County Corn Processors	-95.42	42.47	Galva	Ida 	27.00	27.00
27.00	27.00	27.00	27

167	Red Trail Energy, LLC	-102.31	46.88	Richardton	Stark 	50.00	50.00
50.00	50.00	50.00	50

168	Redfield Energy	-98.71	44.91	Redfield	Spink 	50.00	50.00	50.00	50.00
50.00	50

169	Reeve Agri-Energy	-100.88	37.95	Garden City	Finney 	12.00	12.00
12.00	12.00	12.00	12

170	Renew Energy	-88.79	43.03	Jefferson Junction	Jefferson 	130.00
130.00	130.00	130.00	130.00	0

171	Renova Energy	-113.74	42.55	Heyburn	Minidoka 



20.00

0

172	Show Me Ethanol, LLC	-93.49	39.36	Carrollton	Carroll 



60.00

0

173	Siouxland Energy & Livestock Coop (SELC)	-96.23	43.09	Sioux Center
Sioux 	57.00	57.00	57.00	57.00	57.00	22

174	Siouxland Ethanol, LLC	-96.59	42.45	Jackson	Dakota 	50.00	50.00
50.00	50.00	50.00	50

175	Southwest Iowa Renewable Energy (SIRE)	-95.83	41.16	Council Bluffs
Pottawattamie 

	110.00	110.00

0

176	Standard Ethanol - Horizon	-100.16	40.29	Cambridge	Furnas 	44.00
44.00	44.00	44.00	44.00	0

177	Standard Ethanol - Wheatland	-101.54	40.85	Madrid	Perkins 	44.00
44.00	44.00	44.00	44.00	0

178	Sterling Ethanol, LLC	-103.19	40.64	Sterling	Logan 	42.00	42.00
42.00	42.00	42.00	42

179	Sun Energy	-102.39	37.22	Walsh	Baca 	3.00	3.00	3.00	3.00	3.00	3

180	SW Energy, LLC.	-100.62	40.20	McCook	Red Willow 	0.10	0.10	0.10	0.10
0.10	0.1

181	Tate & Lyle	-94.31	42.50	Fort Dodge	Webster 

	115.00	115.00

0

182	Tate & Lyle	-84.32	35.74	Loudon	Loudon 	126.00	126.00	126.00	126.00
126.00	66

183	The Andersons Albion Ethanol, LLC	-84.79	42.26	Albion	Calhoun 	55.00
55.00	55.00	55.00	55.00	55

184	The Andersons Clymers Ethanol LLC	-86.44	40.72	Clymers	Cass 	110.00
110.00	110.00	110.00	110.00	110

185	The Andersons Marathon Ethanol, LLC	-84.59	40.08	Greenville	Darke 
110.00	110.00	110.00	110.00	110.00	0

186	Theraldson Ethanol	-97.26	46.90	Casselton	Cass 

	100.00	100.00

0

187	Trenton Agri-Products, LLC. (TAP)	-101.01	40.17	Trenton	Hitchcock 
40.00	40.00	40.00	40.00	40.00	40

188	United Ethanol, LLC	-88.92	42.76	Milton	Rock 	50.00	50.00	50.00
50.00	50.00	50

189	United Wisconsin Grain Producers, LLC. (UWGP)	-89.07	43.59	Friesland
Coumbia 	60.00	60.00	80.00	80.00	60.00	50

190	Utica Energy, LLC	-88.65	43.98	Oshkosh / Utica	Winnebago 	52.00
52.00	52.00	52.00	52.00	52

191	Verasun - Albert City (formerly US Bio)	-94.95	42.78	Albert City
Buena Vista 	110.00	110.00	110.00	110.00	110.00	110

192	Verasun - Albion (formerly ASAlliances)	-97.99	41.68	Albion	Boone 
110.00	110.00	110.00	110.00	110.00	0

193	VeraSun - Aurora	-96.69	44.32	Aurora	Brookings 	120.00	120.00	120.00
120.00	120.00	120

194	Verasun - Bloomingburg (formerly ASAlliances)	-83.39	39.60
Bloomingburg	Fayette 	110.00	110.00	110.00	110.00	110.00	0

195	Verasun - Central City (formerly US Bio)	-97.96	41.11	Central City
Merrick 	100.00	100.00	100.00	100.00	100.00	100

196	VeraSun - Charles City	-92.74	43.10	Charles City	Floyd 	110.00
110.00	110.00	110.00	110.00	110

197	Verasun - Dyersville (formerly US Bio)	-91.15	42.49	Dyersville
Dubuque 

110.00	110.00	110.00	110.00	0

198	VeraSun - Fort Dodge	-94.31	42.51	Fort Dodge	Webster 	110.00	110.00
110.00	110.00	110.00	110

199	Verasun - Hankinson (formerly US Bio)	-96.89	46.08	Hankinson
Richland 	110.00	110.00	110.00	110.00	110.00	0

200	VeraSun - Hartley	-95.51	43.18	Hartley	O’Brien 	110.00	110.00
110.00	110.00	110.00	0

201	Verasun - Janesville (formerly US Bio)	-93.71	44.11	Janesville
Waseca 

	110.00	110.00

0

202	Verasun - Linden (formerly ASAlliances)	-86.89	40.21	Linden
Montgomery 	110.00	110.00	110.00	110.00	110.00	0

203	Verasun - Marion (formerly US Bio)	-97.26	43.43	Marion	Turner 
110.00	110.00	110.00	110.00	110.00	0

204	Verasun - Ord (formerly US Bio)	-98.92	41.58	Ord	Valley 	50.00	50.00
50.00	50.00	50.00	50

205	VeraSun - Welcome	-94.65	43.66	Welcome	Martin 

110.00	110.00	110.00	110.00	0

206	Verasun - Woodbury (formerly US Bio)	-85.07	42.77	Lake Odessa	Ionia 
50.00	50.00	50.00	50.00	50.00	50

207	Western New York Energy, LLC	-78.30	43.19	Shelby	Orleans 	50.00
50.00	50.00	50.00	50.00	0

208	Western Plains Energy, LLC	-100.85	39.23	Campus / Oakley	Gove 	45.00
45.00	45.00	45.00	45.00	45

209	Western Wisconsin Energy, LLC (WWE)	-91.98	45.05	Boyceville	Dunn 
45.00	45.00	45.00	45.00	45.00	45

210	White Energy	-102.52	34.87	Hereford	Deaf Smith 	100.00	100.00	100.00
100.00	100.00	0

211	White Energy (formerly Plainview BioEnergy, LLC.)	-101.66	34.18
Plainview	Hale 	100.00	100.00	100.00	100.00	100.00	0

212	White Energy (formerly USEP)	-98.86	38.89	Russell	Russell 	50.00
50.00	50.00	50.00	50.00	50

213	Wind Gap Farms (Anheuser/Miller Brewery)	-84.18	31.35	Baconton
Mitchell 	0.40	0.40	0.40	0.40	0.40	0.4

214	Wyoming Ethanol	-104.35	42.13	Torrington	Goshen 	9.00	9.00	9.00	9.00
9.00	9

215	Xethanol	-92.09	41.89	Blairstown	Benton 	5.00	5.00	5.00	5.00	5.00	5

216	Yuma Ethanol	-102.68	40.25	Yuma	Yuma 	50.00	50.00	50.00	50.00	50.00
0

TOTAL



	11595	12890	14438	14994	12540	6697





APPENDIX B

Projected Corn Ethanol Production 

Figure B-1. Corn Ethanol Production in 2010

Figure B-2. Corn Ethanol Production in 2012

Figure B-3. Corn Ethanol Production in 2014

Figure B-4. Corn Ethanol Production AEO Reference Case in 2022



APPENDIX C

Location Details and Production Capacity (MGY) of Cellulosic Ethanol
Plants for Various Forecast Years

Appendix C

Location details and production capacity (MGY) of cellulosic ethanol
plants for various forecast years

ID	FIPS	FEEDSTOCK	LONGI-TUDE	LATI-TUDE	COUNTY/

STATE	Transport

Mode	Vol

2010	Vol

2012	Vol

2014	Vol

2017	

Vol

2022

1	17189	Corn Stover with some Forest Residue and MSW	-89.41	38.35
WASHINGTON IL	R



	84.77

2	17159	Corn Stover with some Forest Residue and MSW	-88.09	38.71
RICHLAND IL	R



	80.96

3	17029	Corn Stover with some Forest Residue and MSW	-88.22	39.52	COLES
IL	R



	77.24

4	17019	Corn Stover with some Forest Residue and MSW	-88.20	40.14
CHAMPAIGN IL	R



	88.63

5	17075	Corn Stover with some Forest Residue and MSW	-87.82	40.75
IROQUOIS IL	R



	80.39

6	17063	Corn Stover with some Forest Residue and MSW	-88.42	41.29	GRUNDY
IL	RW



	76.57

7	17039	Corn Stover with some Forest Residue and MSW	-88.90	40.17	DEWITT
IL	R



	99.83

8	17173	Corn Stover with some Forest Residue and MSW	-88.81	39.39	SHELBY
IL	R



	67.69

9	17135	Corn Stover with some Forest Residue and MSW	-89.48	39.23
MONTGOMERY IL	R



	77.62

10	17129	Corn Stover with some Forest Residue and MSW	-89.80	40.03
MENARD IL	R



	99.11

11	17137	Corn Stover with some Forest Residue and MSW	-90.20	39.72
MORGAN IL	R



	67.11

12	17179	Corn Stover with some Forest Residue and MSW	-89.51	40.51
TAZEWELL IL	RW



107.42	107.42

13	17011	Corn Stover with some Forest Residue and MSW	-89.53	41.40
BUREAU IL	R



	129.66

14	17095	Corn Stover with some Forest Residue and MSW	-90.21	40.93	KNOX
IL	R



	88.82

15	17043	Corn Stover with some Forest Residue and MSW	-88.09	41.85
DUPAGE IL	R

	109.44	127.65	127.65

16	17141	Corn Stover with some Forest Residue and MSW	-89.32	42.04	OGLE
IL	R



94.58	94.58

17	17015	Corn Stover with some Forest Residue and MSW	-89.93	42.07
CARROLL IL	R



	77.43

18	18179	Corn Stover with some Forest Residue and MSW	-85.22	40.73	WELLS
IN	R



77.14	77.14

19	18095	Corn Stover with some Forest Residue and MSW	-85.72	40.16
MADISON IN	R



78.05	78.05

20	18079	Corn Stover with some Forest Residue and MSW	-85.63	39.00
JENNINGS IN	R



	94.39

21	18027	Corn Stover with some Forest Residue and MSW	-87.07	38.70
DAVIESS IN	R



	93.09

22	18109	Corn Stover with some Forest Residue and MSW	-86.45	39.48
MORGAN IN	R



100.12	100.12

23	18023	Corn Stover with some Forest Residue and MSW	-86.48	40.30
CLINTON IN	R



	80.43

24	18049	Corn Stover with some Forest Residue and MSW	-86.26	41.05
FULTON IN	R



	73.56

25	18161	Corn Stover with some Forest Residue and MSW	-84.92	39.63	UNION
IN	R



	82.29

26	18033	Corn Stover with some Forest Residue and MSW	-85.00	41.40
DEKALB IN	R



	91.46

27	18121	Corn Stover with some Forest Residue and MSW	-87.21	39.77	PARKE
IN	R



	91.79

28	18007	Corn Stover with some Forest Residue and MSW	-87.31	40.61
BENTON IN	R



	91.62

29	18073	Corn Stover with some Forest Residue and MSW	-87.12	41.02
JASPER IN	R



	81.53

30	18163	Corn Stover with some Forest Residue and MSW	-87.59	38.03
VANDERBURGH IN	RW



	73.94

31	20109	Corn Stover with some Forest Residue and MSW	-101.15	38.92
LOGAN KS	R



	74.88

32	20189	Corn Stover with some Forest Residue and MSW	-101.31	37.19
STEVENS KS	R

30.00	74.42	86.80	86.80

33	20173	Corn Stover with some Forest Residue and MSW	-97.46	37.68
SEDGWICK KS	R



	70.57

34	20131	Corn Stover with some Forest Residue and MSW	-96.01	39.78
NEMAHA KS	R



	78.02

35	19061	Corn Stover with some Forest Residue and MSW	-90.88	42.47
DUBUQUE IA	R



70.43	70.43

36	19011	Corn Stover with some Forest Residue and MSW	-92.07	42.08
BENTON IA	R



	68.91

37	19019	Corn Stover with some Forest Residue and MSW	-91.84	42.47
BUCHANAN IA	R



83.34	83.34

38	19123	Corn Stover with some Forest Residue and MSW	-92.64	41.34
MAHASKA IA	R



	80.29

39	19075	Corn Stover with some Forest Residue and MSW	-92.79	42.40
GRUNDY IA	NT



	82.64

40	19037	Corn Stover with some Forest Residue and MSW	-92.32	43.06
CHICKASAW IA	R



82.21	82.21

41	19033	Corn Stover with some Forest Residue and MSW	-93.26	43.08	CERRO
GORDO IA	R



79.37	79.37

42	19069	Corn Stover with some Forest Residue and MSW	-93.26	42.73
FRANKLIN IA	R



	79.54

43	19169	Corn Stover with some Forest Residue and MSW	-93.47	42.04	STORY
IA	R



89.27	89.27

44	19175	Corn Stover with some Forest Residue and MSW	-94.24	41.03	UNION
IA	R



	75.54

45	19077	Corn Stover with some Forest Residue and MSW	-94.50	41.68
GUTHRIE IA	R



85.35	85.35

46	19187	Corn Stover with some Forest Residue and MSW	-94.18	42.43
WEBSTER IA	R



85.83	85.83

47	19155	Corn Stover with some Forest Residue and MSW	-95.54	41.34
POTTAWATTAMIE IA	R



84.39	84.39

48	19093	Corn Stover with some Forest Residue and MSW	-95.51	42.39	IDA
IA	R



87.72	87.72

49	19021	Corn Stover with some Forest Residue and MSW	-95.15	42.74	BUENA
VISTA IA	R



83.63	83.63

50	19147	Corn Stover with some Forest Residue and MSW	-94.68	43.08	PALO
ALTO IA	R

31.00	64.58	75.33	75.33

51	19141	Corn Stover with some Forest Residue and MSW	-95.62	43.08
O’BRIEN IA	R



80.45	80.45

52	19167	Corn Stover with some Forest Residue and MSW	-96.18	43.08	SIOUX
IA	R



72.10	72.10

53	19139	Corn Stover with some Forest Residue and MSW	-91.11	41.48
MUSCATINE IA	RW



	82.58

54	19057	Corn Stover with some Forest Residue and MSW	-91.18	40.92	DES
MOINES IA	R



87.22	87.22

55	19145	Corn Stover with some Forest Residue and MSW	-95.15	40.74	PAGE
IA	R



80.88	80.88

56	21233	Corn Stover with some Forest Residue and MSW	-87.68	37.52
WEBSTER KY	R



	69.54

57	26163	Corn Stover with some Forest Residue and MSW	-83.29	42.29	WAYNE
MI	RW



84.68	84.68

58	26025	Corn Stover with some Forest Residue and MSW	-85.01	42.25
CALHOUN MI	R



108.67	108.67

59	26157	Corn Stover with some Forest Residue and MSW	-83.42	43.46
TUSCOLA MI	R



105.33	105.33

60	26067	Corn Stover with some Forest Residue and MSW	-85.07	42.95	IONIA
MI	R

	100.62	117.37	117.37

61	26159	Corn Stover with some Forest Residue and MSW	-86.02	42.25	VAN
BUREN MI	R



	89.12

62	27133	Corn Stover with some Forest Residue and MSW	-96.25	43.67	ROCK
MN	R



	72.56

63	27091	Corn Stover with some Forest Residue and MSW	-94.55	43.67
MARTIN MN	R



94.64	94.64

64	27043	Corn Stover with some Forest Residue and MSW	-93.95	43.67
FARIBAULT MN	R



	87.69

65	27039	Corn Stover with some Forest Residue and MSW	-92.86	44.02	DODGE
MN	R



85.90	85.90

66	27083	Corn Stover with some Forest Residue and MSW	-95.84	44.41	LYON
MN	R



	83.94

67	27143	Corn Stover with some Forest Residue and MSW	-94.23	44.58
SIBLEY MN	R



	101.73

68	27037	Corn Stover with some Forest Residue and MSW	-93.07	44.67
DAKOTA MN	R



113.56	113.56

69	27145	Corn Stover with some Forest Residue and MSW	-94.61	45.55
STEARNS MN	R



	67.81

70	27023	Corn Stover with some Forest Residue and MSW	-95.57	45.02
CHIPPEWA MN	R



	91.66

71	27149	Corn Stover with some Forest Residue and MSW	-96.00	45.59
STEVENS MN	R



76.01	76.01

72	29007	Corn Stover with some Forest Residue and MSW	-91.84	39.22
AUDRAIN MO	R



86.23	86.23

73	29041	Corn Stover with some Forest Residue and MSW	-92.96	39.51
CHARITON MO	R



	73.73

74	29177	Corn Stover with some Forest Residue and MSW	-93.99	39.35	RAY
MO	RW



99.91	99.91

75	29075	Corn Stover with some Forest Residue and MSW	-94.41	40.21
GENTRY MO	R



	95.44

76	29143	Corn Stover with some Forest Residue and MSW	-89.65	36.59	NEW
MADRID MO	R



	84.19

77	29045	Corn Stover with some Forest Residue and MSW	-91.74	40.41	CLARK
MO	R



	89.40

78	29189	Corn Stover with some Forest Residue and MSW	-90.44	38.64	ST.
LOUIS MO	RW



	124.77

79	31179	Corn Stover with some Forest Residue and MSW	-97.12	42.21	WAYNE
NE	R



	95.77

80	31089	Corn Stover with some Forest Residue and MSW	-98.78	42.46	HOLT
NE	R



	91.48

81	31155	Corn Stover with some Forest Residue and MSW	-96.64	41.23
SAUNDERS NE	R



	99.87

82	31011	Corn Stover with some Forest Residue and MSW	-98.07	41.71	BOONE
NE	R



98.35	98.35

83	31109	Corn Stover with some Forest Residue and MSW	-96.69	40.78
LANCASTER NE	R



	73.93

84	31129	Corn Stover with some Forest Residue and MSW	-98.05	40.18
NUCKOLLS NE	R



	75.78

85	31185	Corn Stover with some Forest Residue and MSW	-97.60	40.87	YORK
NE	R



93.51	93.51

86	31041	Corn Stover with some Forest Residue and MSW	-99.73	41.39
CUSTER NE	R



	84.18

87	31111	Corn Stover with some Forest Residue and MSW	-100.75	41.05
LINCOLN NE	R



	81.45

88	31087	Corn Stover with some Forest Residue and MSW	-101.04	40.18
HITCHCOCK NE	R



	83.41

89	31083	Corn Stover with some Forest Residue and MSW	-99.40	40.18
HARLAN NE	R



	78.47

90	39133	Corn Stover with some Forest Residue and MSW	-81.20	41.17
PORTAGE OH	R



97.84	97.84

91	39139	Corn Stover with some Forest Residue and MSW	-82.54	40.77
RICHLAND OH	R



	83.33

92	39049	Corn Stover with some Forest Residue and MSW	-83.01	39.97
FRANKLIN OH	R



	77.48

93	39027	Corn Stover with some Forest Residue and MSW	-83.81	39.41
CLINTON OH	R



99.83	99.83

94	39091	Corn Stover with some Forest Residue and MSW	-83.77	40.39	LOGAN
OH	R



	74.83

95	39173	Corn Stover with some Forest Residue and MSW	-83.62	41.36	WOOD
OH	R



	84.96

96	39011	Corn Stover with some Forest Residue and MSW	-84.22	40.56
AUGLAIZE OH	R



	79.87

97	46037	Corn Stover with some Forest Residue and MSW	-97.61	45.37	DAY
SD	R



	84.73

98	46077	Corn Stover with some Forest Residue and MSW	-97.49	44.37
KINGSBURY SD	R



	98.02

99	46079	Corn Stover with some Forest Residue and MSW	-97.13	44.02	LAKE
SD	R



83.21	83.21

100	46125	Corn Stover with some Forest Residue and MSW	-97.15	43.31
TURNER SD	R



88.50	88.50

101	46045	Corn Stover with some Forest Residue and MSW	-99.22	45.42
EDMUNDS SD	R



	79.94

102	55055	Corn Stover with some Forest Residue and MSW	-88.78	43.02
JEFFERSON WI	R



94.09	94.09

103	55025	Corn Stover with some Forest Residue and MSW	-89.42	43.07	DANE
WI	R



	75.76

104	55043	Corn Stover with some Forest Residue and MSW	-90.71	42.87
GRANT WI	RW



	68.08

105	55077	Corn Stover with some Forest Residue and MSW	-89.40	43.82
MARQUETTE WI	R



	65.48

106	55015	Corn Stover with some Forest Residue and MSW	-88.22	44.08
CALUMET WI	R



	91.42

107	55141	Corn Stover with some Forest Residue and MSW	-90.04	44.46	WOOD
WI	R



	59.11

108	55035	Corn Stover with some Forest Residue and MSW	-91.29	44.73	EAU
CLAIRE WI	R



	64.59

109	55033	Corn Stover with some Forest Residue and MSW	-91.90	44.95	DUNN
WI	R



	62.85

110	40035	Corn Stover with some Forest Residue and MSW	-95.21	36.76
CRAIG OK	R



	130.00

111	40101	Corn Stover with some Forest Residue and MSW	-95.38	35.62
MUSKOGEE OK	R



	118.00

112	40063	Corn Stover with some Forest Residue and MSW	-96.25	35.05
HUGHES OK	R



	91.00

113	40113	Corn Stover with some Forest Residue and MSW	-96.40	36.63
OSAGE OK	NW



	116.00

114	40081	Corn Stover with some Forest Residue and MSW	-96.88	35.70
LINCOLN OK	R



	120.00

115	40073	Corn Stover with some Forest Residue and MSW	-97.94	35.95
KINGFISHER OK	R



	110.00

116	40051	Corn Stover with some Forest Residue and MSW	-97.88	35.02
GRADY OK	R



	108.00

117	40027	Corn Stover with some Forest Residue and MSW	-97.33	35.20
CLEVELAND OK	R





	118	54033	Corn Stover with some Forest Residue and MSW	-80.38	39.28
HARRISON WV	R



	149.00

119	06007	Forestry and MSW	-121.60	39.67	BUTTE CA	R

12.00	80.92	94.39	94.39

120	06077	Forestry and MSW	-121.27	37.93	SAN JOAQUIN CA	RW

	103.05	120.21	120.21

121	06093	Forestry and MSW	-122.54	41.59	SISKIYOU CA	R



	102.20

122	06059	MSW	-117.76	33.70	ORANGE CA	R	22.00	22.00	114.02	133.20	133.00

123	41039	Forestry and MSW	-122.85	43.94	LANE OR	R



	125.91

124	41071	Forestry and MSW	-123.31	45.23	YAMHILL OR	RW	16.00	43.00
101.25	118.11	118.11

125	53067	Forestry and MSW	-122.83	46.92	THURSTON WA	R



	97.11

126	53007	Forestry and MSW	-120.62	47.87	CHELAN WA	RW



	78.04

127	30089	Forestry and MSW	-115.13	47.68	SANDERS MT	R



	91.54

128	23021	Forestry and MSW	-69.29	45.84	PISCATAQUIS ME	R



	91.11

129	23019	Forestry and MSW	-68.65	45.40	PENOBSCOT ME	R



	99.88

130	33003	Forestry and MSW	-71.20	43.87	CARROLL NH	R



	136.10

131	33003	Switchgrass NE (NH, CT &MA)	-71.20	43.87	CARROLL NH	R



	34.92

132	05061	Forestry and MSW	-93.99	34.09	HOWARD AR	R



	96.80

133	05147	Forestry and MSW	-91.24	35.18	WOODRUFF AR	RW



	102.09

134	28043	Forestry and MSW	-89.80	33.77	GRENADA MS	R



	106.98

135	22035	Forestry and MSW	-91.24	32.73	EAST CARROLL LA	RW

36.00	88.25	102.93	102.93

136	22013	Forestry and MSW	-93.06	32.35	BIENVILLE LA	R



	115.21

137	22053	Forestry and MSW	-92.81	30.27	JEFFERSON DAVIS LA	R

50.00	74.59	87.01	87.01

138	22033	Forestry and MSW	-91.10	30.54	EAST BATON ROUGE LA	RW



105.54	105.54

139	28035	Forestry and MSW	-89.26	31.19	FORREST MS	R



	106.78

140	01063	Forestry and MSW	-87.96	32.86	GREENE AL	RW



	108.33

141	01103	Forestry and MSW	-86.85	34.45	MORGAN AL	R



	95.98

142	01053	Forestry and MSW	-87.16	31.13	ESCAMBIA AL	R



	112.00

143	01113	Forestry and MSW	-85.19	32.29	RUSSELL AL	RW



	100.52

144	13131	Forestry and MSW	-84.23	30.87	GRADY GA	R



	130.00

145	13283	Forestry and MSW	-82.57	32.40	TREUTLEN GA	RW	43.00	43.00	84.07
98.06	98.06

146	13245	Forestry and MSW	-82.07	33.36	RICHMOND GA	R

50.00	86.74	101.18	101.18

147	45015	Forestry and MSW	-79.95	33.20	BERKELEY SC	RW



104.58	104.58

148	45083	Forestry and MSW	-81.99	34.93	SPARTANBURG SC	R



	107.99

149	37067	Forestry and MSW	-80.26	36.13	FORSYTH NC	R

	89.09	103.92	103.92

150	37051	Forestry and MSW	-78.83	35.05	CUMBERLAND NC	RW



	109.58

151	37117	Forestry and MSW	-77.11	35.84	MARTIN NC	RW



	101.80

152	51083	Forestry and MSW	-78.94	36.77	HALIFAX VA	R



	97.57

153	51149	Forestry and MSW	-77.23	37.18	PRINCE GEORGE VA	RW

	85.19	99.38	99.38

154	47123	Forestry and MSW	-84.25	35.44	MONROE TN	R

37.00	82.85	96.64	96.64

155	48005	Forestry and MSW	-94.61	31.26	ANGELINA TX	R



	114.10

156	13127	Forestry and MSW	-81.55	31.24	GLYNN GA	RW



	107.63

157	05025	Forestry and MSW	-92.19	33.90	CLEVELAND AR	R



	98.91

158	01121	Forestry and MSW	-86.17	33.38	TALLADEGA AL	R

	98.40	114.78	114.78

159	48201	Forestry and MSW	-95.39	29.86	HARRIS TX	RW



79.92	79.92

160	12051	Bagasse	-81.16	26.55	HENDRY FL	NW

37.00	77.16	90.00	90.00

161	12099	Bagasse	-80.47	26.65	PALM BEACH FL	R



	100.00

162	12099	Bagasse	-80.47	26.65	PALM BEACH FL	R



	100.00

163	12099	Bagasse	-80.47	26.65	PALM BEACH FL	R



	100.00

164	48215	Bagasse	-98.18	26.40	HIDALGO TX	R



	100.00

165	48061	Bagasse	-97.53	26.13	CAMERON TX	RW



	100.00

166	48489	Sorghum	-97.68	26.47	WILLACY TX	R



	100.00

167	48113	MSW-Dallas, TX	-96.78	32.77	DALLAS TX	R

	44.58	52.20	52.00

168	48029	MSW-San Antonio, TX	-98.52	29.45	BEXAR TX	R



16.20	16.00

169	48453	MSW-Austin, TX	-97.78	30.33	TRAVIS TX	R



14.00	14.00

170	22055	Bagasse	-92.06	30.21	LAFAYETTE LA	NW



	100.00

171	22047	Bagasse	-91.35	30.26	IBERVILLE LA	RW



	90.00

172	22057	Bagasse	-90.44	29.58	LAFOURCHE LA	RW



	50.00

173	22097	Bagasse	-92.01	30.60	ST. LANDRY LA	R



	100.00

174	22077	Bagasse	-91.60	30.71	POINTE COUPEE LA	RW



	100.00

175	22009	Bagasse	-92.00	31.08	AVOYELLES LA	R



	100.00

176	22003	Bagasse	-92.83	30.65	ALLEN LA	R

	42.87	50.00	50.00

177	36119	MSW-New York, NY	-73.75	41.17	WESTCHESTER NY	R	16.00	63.00
61.73	72.00	72.00

178	32003	MSW-Las Vegas, NV	-115.02	36.21	CLARK NV	R



17.20	17.00

179	12011	MSW-Miami, FL	-80.49	26.15	BROWARD FL	RW

	26.58	31.20	31.00

180	08001	MSW-Denver, CO	-104.34	39.88	ADAMS CO	R	3.00	10.00	24.00	28.00
28.00

181	42091	MSW-Phillidelphia, PA	-75.37	40.21	MONTGOMERY PA	R

36.00	36.01	42.00	42.00

TOTAL

100	500	1750	5500	16039



Note: R= Rail; RW= Rail and Water; NT= Near Town; and NW= Near Water



APPENDIX D

Projected Cellulosic Ethanol Production 



Figure D-1. Cellulosic Ethanol Production in 2010

Figure D-2. Cellulosic Ethanol Production in 2012

Figure D-3. Cellulosic Ethanol Production in 2014

Figure D-4. Cellulosic Ethanol Production in 2017

 Boudreaux, T. (2007). Ethanol and Biodiesel: What You Need to Know.
Hart Energy Publishing LLP. McLean, VA. 

 U.S. Department of Agriculture (USDA) (2007). “Ethanol Transportation
Backgrounder: Expansion of U.S. Corn-based Ethanol from the Agricultural
Perspective,” Agricultural Marketing Service, Sept.

 Downstream Alternatives, Inc. (2002). “Transportation and
Infrastructure Requirements for a Renewable Fuels Standard,” Phase III
Project Deliverable Report prepared for Oak Ridge National Laboratory,
Oak Ridge, TN, Aug. 20.

 ARCGIS 9, Data, Maps and StreetMap USA, DVD, Redland Ca, 2006.

 Data downloaded from   HYPERLINK
"http://www.iwr.usace.army.mil/NDC/data/datanwn.htm on Febuary 10" 
http://www.iwr.usace.army.mil/NDC/data/datanwn.htm on February 10 ,
2008.

 More information can be found at http://ned.usgs.gov/

 More information can be found at http://www.i3.com/.

 More information on this map layer can be found at
http://www1.arcwebservices.com/v2006/content/publisher/service_summary.d
o?name=ArcWeb%3AI3.Imagery_Tiles.World&service=MapImage&publisher=i-cube
d&show=true.

 The information can be found at   HYPERLINK
"http://www.ndc.iwr.usace.army.mil/data/datappor.htm" 
http://www.ndc.iwr.usace.army.mil/data/datappor.htm .

 Hu, P., T. Reuscher, and R.L. Schmoyer, Transferring 2001 National
Household Travel Survey, Oak Ridge, TN 2007. Document can be found at
http://fmip.ornl.gov/nhts/TransferabilityReport.pdf.

 The Voronoi Diagram for the terminal centroids is the partition of the
US states which all points within a partition(i) are closer to the
terminal i than any other terminals. 

 More information can be found at http://opisnet.com/directories/pte.asp

 Dager, C. (2008). Tennessee Valley Authority, Knoxville, TN, Personal
communication with Sujit Das, Oak Ridge National Laboratory, Oak Ridge,
TN, Sept. 22.

 Davis, S. C., Diegel, S. W. and Boundy, R. G. (2008). “Transportation
Energy Data Book: Edition 27, Oak Ridge National Laboratory, Oak Ridge,
TN.

 Eno Foundation, Transportation in America, 20th Edition, p. 32 and 40.

 Energy Information Administration (EIA) 2008. “Annual Energy Outlook
2008,” supplemental tables 92-117,   HYPERLINK
"http://www.eia.doe.gov/oiaf/aeo/supplement/sup_ogc.xls"  Petroleum,
Natural Gas, Coal, Macroeconomic, Petroleum, and Import , U.S.
Department of Energy, Washington, DC. Accessed at the website:
http://www.eia.doe.gov/oiaf/aeo/supplement/sup_ogc.xls

 American Association of Railroads (AAR) (2004). Accessed from  
HYPERLINK
"http://findarticles.com/p/articles/mi_m1215/is_12_205/ai_n8967508" 
http://findarticles.com/p/articles/mi_m1215/is_12_205/ai_n8967508  on
1/26/09.

 Cambridge Systematics (2007). “National Rail Freight Infrastructure
Capacity and Investment Study: Final Report,” prepared for Association
of American Railroads, Cambridge, MA, Sept.

 Freight Analysis Framework (FAF) Version 2.2, User Guide, Federal
Highway Administration, 2006.

 Fleming, Marilyn V.; Wood, Donna E.; Goodwin, Robert J. Lock
Performance Monitoring System User’s Manual for Data Analysis, Corps
Of Engineers Fort Belvoir VA Water Resources Support Center, 1985

 David Schrank, and Lomax, T. (2007) The 2007 Urban Mobility Report,
Texas Transportation Institute, The Texas A&M University System. Copy of
the report is available at   HYPERLINK
"http://tti.tamu.edu/documents/mobility_report_2007_wappx.pdf" 
http://tti.tamu.edu/documents/mobility_report_2007_wappx.pdf 

Highway Statistics Annual Publications, Federal Highway Administration
(FHWA), US Department of Transportation (US DOT). Tables from Highway
Statistics series are available through   HYPERLINK
"http://www.fhwa.dot.gov/policy/ohpi/hss/hsspubs.cfm" 
http://www.fhwa.dot.gov/policy/ohpi/hss/hsspubs.cfm .

 Levinson, D. and Kumar, A. 1995. A Multi-modal Trip Distribution
Model,” Transportation Research Record #1466, pp. 124-131.

 “CTA Transportation Networks,” can be accessed at
http://cta.ornl.gov/transnet/

Southworth, F. (2005). “A Multimodal Regional Routing and Multi-port
Analysis Model, Prepared for the Institute for Water Resources, United
States Army Corps of Engineers, Fort Belvoir, Alexandria VA 22315-3868,
Nov. 

 Cambridge Systematics, Inc. (2007). “National Rail Freight
Infrastructure Capacity and Investment Study,” prepared for
Association of American Railroads, Cambridge, MA, Sept. 

U.S. Department of Transportation (2008). “Pocket Guide to
Transportation 2008,” Bureau of Transportation Statistics, Research &
Innovative Technology Administration, Washington, DC, Feb.

 Dager, C. (2008). Tennessee Valley Authority, Knoxville, TN. Personal
communication with Sujit Das, Oak Ridge National Laboratory, Oak Ridge,
TN, Sept. 22.

Hadenfeldt, K. (2009). GATX Corporation, Chicago, IL. Personal
communication with Bruce Peterson and Sujit Das of Oak Ridge National
Laboratory, Oak Ridge, TN on 1/26/09.

 Tom Simpson (2009). Railway Supply Institute, Washington, DC. Personal
communication with Sujit Das, Oak Ridge National Laboratory, Oak Ridge,
TN, Jan. 5.

 U.S. Army Corps of Engineers (2009). “Waterborne Transportation Lines
of the United States: Calendar Year 2007 – Volume 1: National
Summaries,” Alexandria, VA. Accessed from the website:   HYPERLINK
"http://www.iwr.usace.army.mil/ndc/publicatons.htm" 
http://www.iwr.usace.army.mil/ndc/publicatons.htm  on 1/5/09.

 U.S. Army Corps of Engineers (2009). “The U.S. Waterway System Facts:
Fact Card 2008 – Volume 1: National Summaries,” Alexandria, VA.
Accessed from the website:   HYPERLINK
"http://www.iwr.usace.army.mil/ndc/factcard/fc07/factcard.pdf" 
http://www.iwr.usace.army.mil/ndc/factcard/fc07/factcard.pdf  on 1/5/09.

 An additional overall motor fuel volume will need to be trucked to
retail facilities to satisfy demand due to the lower energy density of
ethanol relative to gasoline.

 EPA assumed that co-processed renewable diesel would be distributed
fungibly with petroleum-based diesel fuel from petroleum refineries,
whereas biodiesel and non-processed renewable diesel fuel would not be
able to take advantage of the established petroleum fuel distribution
system.  

 PAGE   

  PAGE  1 

Ethanol Transport Activity & Potential Distribution Constraints	Oak
Ridge National Laboratory

Ethanol Transport Activity & Potential Distribution Constraints		Oak
Ridge National Laboratory

 PAGE   48 

 PAGE   

  PAGE  61 

 PAGE   85 

Ethanol Transport Activity & Potential Distribution Constraints		Oak
Ridge National Laboratory

  PAGE  70 

 PAGE   

 PAGE   89 

