Determining the Land Use Impact of Two Midwestern Corn Ethanol ...
Determining the Land Use Impact of Two Midwestern Corn Ethanol Plants
Prepared by: Steffen Mueller and Ken Copenhaver Prepared for: Illinois Corn Marketing Board
Date: September 18, 2009
Summary The present study is part of an ongoing effort by the University of Illinois at Chicago Energy Resources Center (UIC-ERC) to consistently evaluate the land use and the global warming impact of corn ethanol production. The study looks at the direct land use changes in the vicinity of two ethanol plants located in Illinois: the Illinois River Energy Center (IRE) ethanol plant located in Rochelle and the Patriot Renewable Fuels (PRF) plant located in Annawan. Both studied ethanol plants have rail access that enables them to deliver to the California market. The study provides an innovative approach to relate direct land use changes determined with high resolution satellite imagery to direct greenhouse gas emissions from the land use changes (while also accounting for different time horizons). Then, the direct land use emissions are added to the remainder of the life cycle greenhouse gas emissions (from the ethanol plant, distribution, denaturing process) to derive a total direct (including direct land use) life cycle global warming impact (GWI) for corn ethanol produced at a specific plant (IRE). Furthermore, data derived from the local, plant-level, land use analysis is compared to state-wide corn production and land use data. Land use surrounding IRE and GWI of its corn ethanol was assessed in previous studies conducted by the University of Illinois at Chicago Energy Resources Center (UIC-ERC).1 However, IRE recently expanded its capacity from 58 million gallons per year (mgpy) to 115 mgpy. The expanded plant started operating in November 2008. This presents also a unique opportunity to provide a longer term assessment of how the land use and the global warming impact change with larger ethanol production volumes produced at the same location. Land use surrounding PRF has not been previously assessed. PRF started operation in September 2008 with a capacity of 100 mgpy. Both IRE and PRF are located in relatively close proximity with a portion of the corn supply area overlapping. The study findings include the following:
The start up of PRF and IRE (expansion) prompted very little conversion of nonagricultural to agricultural land. The land use analysis performed for the corn supply areas showed that less than 1,000 acres were converted in the vicinity of either plant. With respect to land conversion, at a maximum 534 acres of forest (and 105 acres of grassland) were converted to corn around IRE and 609 acres (and 6 acres of grassland) were converted to corn around PRF. This finding supports the notion that corn ethanol plants have a weak influence on direct nonagricultural land conversions.
1
Mueller, S. and K. Copenhaver (2008). “A Bottom-Up Assessment of Land Use Related to Corn Ethanol Production.” Available at http://www.erc.uic.edu/staff/smueller.htm, and Mueller, Steffen and Michelle Wander, Ken Copenhaver (2008). “Global Warming and Land Use Impact of Corn Ethanol Produced at the Illinois River Energy Center”; Report prepared for the Illinois Corn Growers Association. Available at http://www.erc.uic.edu/staff/smueller.htm
2
Using a spreadsheet model (Direct Land Use Mapper “dLUC-Mapper”) developed by the UIC-ERC and land carbon factors from IPCC the greenhouse gas emissions from direct land use changes were assessed. The study finds that the direct land use changes would result in GHG emissions of 1.2 gCO2/MJ for IRE (30 year no discount) and 1.3 gCO2/MJ for PRF (30 year no discount). The respective 100 year, 2% discounted GHG emissions are, as expected, slightly lower.
The GWI from direct emissions of corn ethanol produced at the expanded IRE plant was reassessed as part of the present study. The GWI of the expanded plant is slightly lower at 53.5 compared to 54.8 gCO2e/MJ for the original plant. The difference is a combination of the different energy demand, the different ethanol yield, and the different DDGS production volume.
Adding the 1.2 gCO2e/MJ direct land use emissions to the remaining life cycle emissions of 53.5 gCO2e/MJ, results in total direct (including direct land use) GHG life cycle emissions for IRE of 54.7 gCO2e/MJ.
Despite the start-up of both PRF and IRE (expansion) in the fall of 2008 acres in corn in 2008 went down. During the start-up of the original IRE ethanol plant in 2006 acres in corn in the region increased to levels that could not be attributed to the ethanol plant demand (see original UIC-ERC study). Then, during the start-up of the expanded plant corn acres in the region went actually down (by 37,634 acres between the years 2007-2008). Likewise, corn acres around PRF went also down by 18,020 acres providing further evidence that the ethanol plants have a weak influence on corn rotations.
Also, in IRE’s case, an additional 340,000 acres of grass/pasture/hay land and in PRF’s case an additional 84,000 acres would have been available in 2008 for conversion to corn acres. This did not occur supporting the notion that the studied corn ethanol plants have a weak influence on agricultural land use in general. With intensive pasture and hay management some of these acres would still be available for additional corn production without associated indirect land use change effects.
Furthermore, the study analyzed Illinois-wide corn supply and use data. The analysis shows that corn production increases on constant corn acres were sufficient to support both increasing exports as well as corn for ethanol use.
Consequently, the studied ethanol plants a) did not influence local land use b) did not prompt indirect land use in the State of Illinois, c) did not influence the Illinois corn export balance.
3
1) Land Use Surrounding IRE As part of the original IRE study in 2007, we conducted a survey with growers delivering to the ethanol plant and concluded that an area consisting of a 40 mile circular radius would provide a good representation of the corn supply region for the plant. In support of the present study, IRE provided us with the addresses of all growers delivering directly to the plant, which allowed an even more accurate determination of the corn supply region. Direct grower deliveries account for approximately 60% of all corn delivered to IRE; the rest is sourced through grain elevators. We graphed the grower locations and determined a circular corn draw area that would include a representative amount of growers: We found that a circular corn draw area with a 40 mile radius would include 83% of the grower addresses while increasing the radius to 43 miles (expanding the radius to the Wisconsin border) would capture 92% of the grower addresses. We determined that a 43 mile radius would be a good representation of the corn supply area for the expanded IRE plant. The grower addresses and 43 mile circular area overlaid on top of the Illinois county map is shown in Figure 1.
Addresses of Growers Delivering to IRE 43 Mile Circle
IRE
Figure 1: Growers Delivering to the Expanded IRE Plant
In a next step, we analyzed land use and land use changes within the 43 mile radius corn supply area over the period of 2006, 2007, and 2008. This analysis was performed using the USDA NASS Cropland Data Layers (for crop types) which includes the national land cover dataset (for non-cropland conversions). The software used for this analysis was Erdas Imagine 9.1 developed by Leica Geosystems. While the USDA NASS Cropland Data Layer has been shown to have accurate methods of around 95% for the delineation of corn and soybeans and that dataset is updated every year, the national land cover data set dates back to 2001 and introduces much higher uncertainties for non-agricultural areas. Therefore, the present study analyzes the data
4
layers by a) applying an algorithm that subtracts the unlikely rotation scenario of land in agriculture converting to non-agricultural land and then back to agricultural land, and b) subtracts field fringes and roadway buffers. This approach has been shown by Mueller and Copenhaver (2009) to increase the accuracy of land use change analyses of this type.2 As can be seen in Table 1 the corn draw area includes a total of 3.7 million acres. The area in corn ranges between 1.4 and 1.7 million acres. Land use change from non agricultural lands associated with the introduction of the ethanol plant within the 43 mile radius circle is shown to be minimal (see Table 2). Of the 3.7+ million acres in the study area just over 1,000 acres were found to be converted from non-agricultural categories (grasslands, forest) from 2006 to 2007 and even fewer were converted from 2007 to 2008 after vetting the data for road fringes and misclassification associated with conversions from agriculture to non-agriculture and back to agriculture. This small percentage is well within the classification error of the dataset and could very well be attributed to misclassified pixels, indicating land use change from non-agricultural classes to agriculture did not, in all likelihood, occur despite the introduction of the ethanol plant. Furthermore, based on the USEPA’s definition of agricultural land (cropland, pasture and CRP) in the RFS2, there were over 340,000 acres still available for conversion from pasture/hay to corn for ethanol production in the corn supply area in 2007 and 2008. With intensive pasture and hay management these acres could produce additional corn without associated indirect land use change effects.
2
Mueller, S. and K. Copenhaver (2009). “Use of Remote Sensing to Measure Land Use Change from Biofuels Production.” Article submitted to Swords and Ploughshares - Special Issue on Sustainable Biofuels and Human Security - A Publication of the Program in Arms Control, Disarmament, and International Security (ACDIS), March 26, 2009
5
Table 1: Land Use Surrounding IRE 2006 Unvetted 1,391,597
2007 Unvetted 1,694,145
2008 Unvetted 1,656,511
955,428
614,399
666,208
2,347,025
2,308,544
2,322,719
Other Land Uses
1,373,378
1,411,859
1,397,684
Totals:
3,720,403
3,720,403
3,720,403
Corn Soybeans Total Corn and Soybeans
Table 2: Corn Acre Land Use Surrounding IRE 2007 Corn Acres in 2006 Corn
Unvetted 773,811
Vetted ag-nonag-ag 773,885
Vetted 648,689
753,498
753,544
623,530
idle cropland/ fallow
8,042
8,040
5,019
Grasslands
8,737
602
333
Forest
4,126
1,026
734
106,742
106,687
54,352
13,190
7,893
Soybeans
pasture/grass/hay Ag removed Road borders removed
306,528
2008 corn acres in 2007
Unvetted
Vetted ag-nonag-ag
Vetted
Corn
1,010,063
1,010,063
846,408
515,872
515,872
432,697
636
636
518
Grasslands
2,569
164
105
Forest
10,465
622
534
pasture/grass/hay
40,656
40,656
29,091
51,736
16,252
Soybeans idle cropland/ fallow
Ag removed Road borders removed
301,131
As stated above, the expanded plant started operation in November 2008 drawing corn from the 2008 harvest. The plant production levels of the original IRE plant and the expanded plant and the corresponding yields and land demands are shown in Table 3. It is evident that the expanded plant did not prompt additional corn acre conversions. In fact, between the years 2007-2008 corn acres decreased slightly by 37,634. While a relatively large increase of corn acres was observed between 2006-2007 (302,548 acres), this increase is also likely unrelated to the start up of the original ethanol plant in 2006 since the increase far exceeded the ethanol plant’s original corn land demand of 116,777 acres. Taking prevailing yield levels into account, the original and expanded plant require less than 13% of corn acres within the draw area to meet the corn
6
demand. Note that the plant also produces animal feed and that therefore the land demand that should be attributed to corn ethanol production is lower than 13%. Table 3: IRE Land Demand Corn Acres Ethanol Production (gallons) Average ethanol yield per bushel (gal/bu) Required Bushels Yield in Surrounding Counties (bu/acre)3 Required Acres Corn Land Demand (%)
2006 1,391,597
2007 1,694,145
2008 1,656,511
55,820,804 2.73 20,447,181 176 116,177 8.3%
55,820,804 2.73
104,280,111 2.76
20,447,181 186 109,931 6.5%
37,851,220 177 213,849 12.9%
2) Land Use Surrounding PRF In an analysis of growers delivering to PRF, plant personnel determined that 90% of corn receipts are within a 23 mile circle which includes the cities of Princeton, Prophetstown, Geneseo, Kewanee. Multiple reasons may contribute to the fact that the supply area for PRF is smaller than the one for IRE including significantly less urban shares. Also, as can be seen in Figure 2 the corn supply areas for PRF and IRE overlap slightly.
Corn Supply Areas for IRE and PRF
IRE
PRF
Addresses for growers delivering to IRE in 2008 (>95%) City addresses for growers delivering to Patriot in 2008 (>90%)
Figure 2: Growers Delivering to IRE and PRF
With the boundaries of the corn supply area established, we performed the land use analysis for the years 2006, 2007, and 2008. As can be seen in Table 4 the corn supply area includes a total of 1.1 million acres. The area in corn ranges between 441,000 to 3
USDA NASS County Crop Yield Report. www.nass.usda.gov
7
541,000 acres. As was the case for the IRE plant, land use change from non agricultural lands associated with the introduction of the PRF ethanol plant within the 23 mile radius circle is shown to be minimal. Of the 1.1 million acres in the study area just over 600 acres were found to be converted from non-agricultural categories from 2007 to 2008 after vetting the data for road fringes and misclassification associated with conversions from agriculture to non-agriculture and back to agriculture (see Table 5). Furthermore, based on the USEPA’s definition of agricultural land (cropland, pasture and CRP) in the RFS2, there were over 84,000 acres still available for conversion from pasture/hay to corn for ethanol production in the polygon in 2007 and 2008. Again, with intensive pasture and hay management some of these acres would also be available for additional corn production without associated indirect land use change effects. Figure 3 confirms the grass/pasture hay classifications include large contiguous areas suitable for conversion.
PRF and IRE Acres in Pasture/Hay Category 340,421 acres in IRE Corn Supply Area
83,729 acres in PRF Corn Supply Area
2008 Pasture/Hay 2008 Other Classes
Figure 3: Grass/Pasture/Hay Acreage in 2008
The PRF plant started operation in September 2008 drawing corn from the 2008 harvest. Consistent with the findings for the IRE plant, the PRF plant did not prompt additional corn acre conversions. In fact, between the years 2007-2008 corn acres decreased slightly by 18,020 acres. Table 4: Land Use Surrounding PRF Corn
2006 unvetted 441,346
2007 unvetted 558,957
2008 unvetted 540,937
Soybeans
351,510
224,652
248,608
792,856
783,609
789,545
271,777
281,231
275,485
1,064,633
1,064,840
1,065,030
Total Corn and Soybeans Other Land Uses Total Acres
8
Table 5: Corn Acre Land Use Surrounding PRF 2008 corn acres in 2007
Unvetted
Vetted for Ag-nonag-ag
Vetted
Corn
306,294
306,294
255,436
Soybeans
190,902
190,902
159,974
Winter wheat
5,051
5,051
4,142
Other crops
1,111
1,111
985
5
5
5
170
6
6
5,540
700
609
14,434
14,434
10,134
21,236
6,717
Fallow/idle cropland Grassland Forest Pasture/Grass/Hay Ag/Non-Ag/Ag removed Road borders removed
102,222
Due to the smaller supply circle, PRF’s corn land demand to supply the plant as a percentage of overall corn land within the supply circle is much higher than for IRE (34.3% for PRF compared to 13% for IRE).4 This indicates, however, that higher shares of corn acres going to corn ethanol production do not result in increased conversion of non agricultural land. Note that the plant also produces animal feed and that therefore the land demand that should be attributed to corn ethanol production is lower than 34%. Table 6: PRF Land Demand 2006
2007
Corn Acres
540,937
Ethanol Production (gallons)
100,000,000
Average ethanol yield per bushel (gal/bu)
2.8
Required Bushels Yield in Surrounding Counties (bu/acre) Required Acres
2008
35,714,286 174.5
191.5
Corn Land Demand (%)
192.5 185,529 34.30%
3) Correlating Ethanol Plant Corn Supply with Illinois Corn Supply Since, as demonstrated above, ethanol plants have a very weak influence on land use in their vicinity, one could assert that the plants must alter land use elsewhere (i.e. indirectly). To test this effect we analyzed Illinois’ corn production, export, and uses. The ProExporter Network’s Senior Economist Ross Korves provided the data reproduced in Figure 4 and Figure 5. As can be seen, over the last 35 years production of corn nearly doubled from 1.2 to 2.2 billion bushels. In fact production increases are strong enough to support both increases in the “other uses” category, which includes corn delivered to ethanol plants, and corn exports. Furthermore, carryout corn or the amount of corn on hand at the end of the marketing year (inventory) has also been increasing. Feedcorn, on 4
Note that PRF also produces animal feed and that therefore the actual land demand is lower.
9
the other hand, has been decreasing which may be indicative of decreasing livestock but also increased use of distillers grain as feedstock. Furthermore, while corn production increased over the time period corn acres in Illinois remained relatively constant (see Figure 6). In summary, the Illinois corn supply and use data shows that corn production increases on relatively constant corn acres were sufficient to support both increasing exports as well as corn for ethanol use. Consequently, the studied ethanol plants a) did not influence local land use b) did not prompt indirect land use in the State of Illinois, c) did not influence the Illinois corn export balance. Illinois Corn Production and Use (Years 1975-2010) 2,500,000
thousand bushel
2,000,000
1,500,000
1,000,000
500,000
-
1975 1977 1979 1981 1983 1985 1987 1989 1991 1993 1995 1997 1999 2001 2003 2005 2007 2009 Production
CarryOut
FeedUse
OtherUse
NetExports
Figure 4: Illinois Corn Production and Use 1975-2010 Illinois Corn Production and Use (Years 2000-2010) 2,500,000
thousand bushel
2,000,000
1,500,000
1,000,000
500,000
2000
2002
Production
2004
CarryOut
2006
FeedUse
Figure 5: Illinois Corn Production and Use 2000-2010
10
OtherUse
2008
NetExports
2010
Illinois Planted Corn Acres 14,000 12,000
thousand acres
10,000 8,000 6,000 4,000 2,000
09
05
07
20
20
20
01
99
97
95
93
91
03 20
20
19
19
19
19
19
89 19
85
83
81
79
77
87 19
19
19
19
19
19
19
75
0
Figure 6: USDA NASS Statistics on Illinois Planted Corn Acres (Source: www.nass.usda.gov/QuickStats)
These findings are consistent with the overall US export balance. The data compiled by Dr. Darrel Good (U of I Department of Agriculture Economics) and reproduced in Appendix A shows that the US was able to provide a) relatively constant corn exports and b) slightly increasing soy exports on relatively constant principal crop acres. 4) Relating Land use Change to Greenhouse Gas Emissions As detailed above, at a maximum, 534 acres of forest (and 105 acres of grassland) were converted to corn around IRE and 609 acres (and 6 acres of grassland) were converted to corn around PRF during plant start-up. Using a spreadsheet model (Direct Land Use Mapper “dLUC-Mapper”) developed by the UIC-ERC (posted at www.erc.uic.edu/staff/smueller.htm, input screen reproduced in Appendix C) and land carbon factors from IPCC these direct land use conversions would result in GHG emissions of 1.2 gCO2/MJ for IRE (30 year no discount) and 1.3 gCO2/MJ for PRF (30 year no discount). The respective 100 year, 2% discounted GHG emissions are, as expected, slightly lower (see Table 7).5
5
The employed IPCC factors are world average factors for temperate forests. Ideally, emissions factors specific to Midwestern forests should be used but were not available to the author at the time of the study.
11
Table 7: Direct Land Use Emissions Attributable to two Illinois Corn Ethanol Plants IRE
PRF
Ethanol Volume (Billion Gallons) Forest
0.115 -534
0.1 -609
Grassland
-105
-6
30 Year No Discount GHG Emissions (gCO2/MJ)
1.2
1.3
100 Year 2% Discount GHG Emissions (gCO2/MJ)
1.0
1.1
These GHG emissions values which are representative of domestic direct land use effects were derived using a) high resolution remote sensing tools with accuracies exceeding 95%, combined with b) published IPCC emissions factors, and c) an easily verifiable spreadsheet model 5) The Global Warming Impact of Corn Ethanol Produced at IRE The GWI of corn ethanol produced at the expanded IRE plant was assessed by parameterizing Argonne’s GREET model (Version 1.8c) with the energy and production values supplied by IRE listed in Appendix B. Parameterization of GREET was performed using an Interface Macro Tool developed by Life Cycle Associates. The input screen of the Interface Macro Tool is reproduced in Appendix D. The key parameters that were substituted for the GREET default values include the actual ethanol yield (per bushel corn) of the plant, the actual natural gas and electricity demand (and the Illinois electricity grid mix), the actual transportation distances of corn and ethanol, and the blend of denaturant. The GREET current default value for land use change was not included in the analysis. The results are shown in Table 8 and Figure 7. The GWI of the expanded Illinois River corn ethanol plant totals 53.5 gCO2e/MJ for the denatured corn ethanol and 53.2 gCO2e/MJ for the anhydrous corn ethanol. This is close to the original IRE Plant of 54.8 gCO2e/MJ. The direct land use change greenhouse gas emissions for IRE (30 year, no discount), as assessed above, total 1.2 gCO2e/MJ. Adding theses emissions to the remaining life cycle emissions of 53.5 gCO2e/MJ, results in total direct (including direct land use) GHG life cycle emissions for IRE of 54.7 gCO2e/MJ. Table 8: GWI of IRE2 Produced Corn Ethanol Gasoline Ag Phase and Distribution Ethanol Plant Denaturant IRE Total Gasoline
IRE2 Ethanol 23.8 29.4 0.3 92.1
12
The GWI of Corn Ethanol Produced at the Expanded Illinois River Energy Ethanol Plant 100 90 80
gCO2 e /MJ
70 60
Denaturant
50
Ethanol Plant
40
Ag Phase and Distribution
30 20 10 0 Gasoline
IRE2 Ethanol
Figure 7: GWI of the Expanded IRE Plant
13
Appendix A: US Crop Data Provided by U of I Department of Agriculture Economics
U.S. Principal Crop Acres (million acres)
1999-- 329.3 2000-- 328.7 2001-- 324.6 2002-- 327.3 2003-- 325.7 2004-- 322.3 2005-- 317.6 2006-- 315.6 2007-- 320.4 2008-- 324.8 2009– 320.9
US Corn Exports 2.4 B illio n B u s h e ls
2.0 1.6 1.2 0.8 0.4 0.0 79 81 83 85 87 89 91 93 95 97 99 01 03 05 07 09
US Soybean Exports 1.4 1.3 1.2
Billion Bushels
1.1 1.0 0.9 0.8 0.7 0.6 0.5 0.4 0.3 0.2 80
82
84
86
88
90
92
94
96
14
98
00
02
04
06
08
Appendix B: Energy and Production Parameters of the Original and Expanded IRE Plant Unit Plant Performance: Annual total anhydrous ethanol production Annual total denatured ethanol production Description of denaturant used (type)
gallon per year gallon per year Debutanized Natural Gasoline gal/bu
Average ethanol yield per bushel (anhydrous) Plant Energy Systems: Annual total natural gas consumption HHV Annual total electricity consumption Natural Gas (HHV) per unit Anhydrous Ethanol Production Natural Gas (HHV) per unit Denatured Ethanol Production Natural Gas (LHV) per unit Anhydrous Ethanol Production Natural Gas (LHV) per unit Denatured Ethanol Production Electricity per unit Anhydrous Ethanol Production Electricity per unit Denatured Ethanol Production By-Products: Annual total DDGS production Annual avg DDGS moisture Annual total WDG(S) production Annual avg WDG(S) moisture Annual total S production - as product sold Annual avg S moisture Transportation Logistics: Corn by truck Corn by rail DDGS shipments by truck
% % %
DDGS shipments by rail
%
DDGS shipments by ship
%
WDGS shipments by truck Ethanol shipments by truck Ethanol shipments by rail Ethanol shipments by barge Avg ethanol distance transported by truck (per trip - one way) Avg ethanol distance transported by rail (per trip one way) Avg ethanol distance transport from terminal to retail outlet (per trip one way) DDGS (11% Moisture) DDGS equivalent of WDGS (corrected to 11% Moisture) DDGS equivalent of WDGS Total DDGS equivalent Denaturant by %
lbs/gal tons
IRE1
IRE2 55,820,804 57,812,280 DNG 2.73
104,280,111 106,709,420 DNG 2.755
Btu kWh Btu/gal
1,671,765,900,000 39,898,320 29,949
Btu/gal
28,917
28,071
26,981
25,907
26,051
25,289
0.71 0.69
0.692 0.675
kWh/gal kWh/gal tons % tons % tons %
3,000,521,873,185 72,057,134 28,757
153,213 11 13,488 30 5,036 60
281,939 11 7,340 60 7,990 60 100 0 11.1
% % % % mi
100 0 Backhaul Shipment Backhaul Shipment Backhaul Shipment 100 98 2 0 80
mi
1,000
918
mi
10
10
lbs/gal lbs/gal
15
5.49
1.82 87.08 100 96.21 3.79 0 80
5.41 3,743 0.072 5.48 2.3%
Appendix C: Direct GHG Emissions Spreadsheet Model Direct GHG Emissions Surrounding Corn Ethanol Plants Author: Steffen Mueller, PhD Copyright: University of Illinois at Chicago Yellow Cells Allow Customization for Different Ethanol Plants Ethanol Plants: IRE Plant Ethanol Volume (Billion Gallons)
PRF 0.115
0.1
-534
-609
-105
-6
Land Conversion: Forest Land Conversion: Grassland 0.115
Ethanol Volume
Billion Gallons
Production Years (Foregone C-Sequestration Period) Select Duration:
List of Possible Durations: 20 30 100
30 Years
Analytical Horizon (Amortization, Above Ground Carbon Release Horizon) Select Duration
List of Possible Durations:
30 Years
30 50 100 Recovered Carbon (Restoration of original vegetation after biofuels program)
0% Percent
Input Recovered Carbon:
for 0% use 0.0001 to avoid division by 0
Error Message: Carbon Recovery Time Insufficient Emissions Discounting Method
0.00% Percent
Input Discount Rate: Carbon Flux Duration Above Ground Biomass Below Ground Biomass
Starting in Year 1 5
0 0
Foregone Net Primary Production
30
0
Recovered Carbon
0.01
30
Undiscounted Emissions Factors
Region Forest Grassland
Area Cover Change (acre)
Area Cover Change (ha) -534 -105
Above Ground tC/ha -216 -43
Net Primary Below Ground Production tC/ha tC/ha-yr 57.0 96.0 6.3 7.0 236.0 4.2
Net Present Value of Emissions Factors
Region Forest Grassland
Recovered Carbon
Net Emissions
Recover ed Recovered Carbon Carbon Undiscounte Discoun Below Ground Net Primary Production ted Forest tCO2/ha d Above Ground tC/ha tC/ha tC/ha-yr NPV tC/ha 57.0 96.00 187.47 340.5 0.0 0.0 1250 7.0 235.99 127.18 370.2 0.0 0.0 1359
Forest Total Emissions tCO2e Annualized Emissions tCO2e Annualized Emissions gCO2e/gal Annualized Emissions gCO2e/MJ
Indirect Land Use Change (gCO2e/MJ):
tCO2e Grassland tCO2e Forest+Grassl. tCO2e 270,138 57,752 327,890 9,006 1,925 10,931 78 17 95 1.0 0.2 1.2
1.2 16
Appendix D: GREET Interface Macro Tool for IRE Corn Ethanol Corn Cultivation General Target year
2010
Corn farming energy (Btu/bu)
7,800
Bailing energy (Btu/bu)
0
Total ag. energy input (Btu/bu)
7,800
Corn yield (Bu/acre)
196.0
Stover Yield (Tons/acre)
0.00
Bu per ton stover
19600000.0
Farming Energy Shares Diesel
100.0%
Gasoline
0.0%
Natural G as
0.0%
LPG
0.0%
Electricity
0.0%
Electricity Generation Mix for Farming Electricity mix
IL SERC
Fuel shares if electricity mix is "User Defined": Residual oil Natural gas
1.8% 10.0%
Coal
57.3%
Nuclear
25.2%
Biomass
1.9%
Other (renewables)
3.8% International
Chemical Inputs (Fertilizer, Herbicide, Insecticide) N input (g/bu)
368.0
Production Share Ocean Tanker (mi)
Ammonia
70.7%
60.0%
Urea
21.1%
60.0%
5,200
3,000
Ammonium nitrate P2 O5 input (g/bu)
8.2%
60.0%
3,700
193.6
60.0%
4,200
K2 O input (g/bu)
300.5
60.0%
3,900
1,419.4
0.0%
Herbicide input (g/bu)
8.36
60.0%
4,000
Insecticide input (g/bu)
0.70
60.0%
4,000
CaCO3 input (g/bu)
Soil N Emissions Baseline N content of above and below ground biomass (g/bu) N removed in stover
141.6 0.0
Above and below ground N available for N2O loss N in N 2 O as % of N in fertilizer and biomass
141.6 1.3%
Land Use Change and Sustainability Land-use change CO2 emissions (g/bu)
0.0
17
Corn and Stover Transport Corn Field to Stack Medium duty truck Corn Stack to Ethanol Plant
Distance (mi)
Share
0.1
100.0%
Distance (mi)
Share
Barge
350
0.0%
Rail
400
0.0%
Heavy duty truck
40
100.0%
Corn Stover Transport Heavy duty truck
Distance (mi)
Share
0
100.0%
Dry Mill Ethanol Plant (Anhydrous EtOH Basis) Thermal energy for NG, coal, biomass (Btu/gal EtOH), LHV Ethanol yield (gal/bu) Dry Mill Ethanol Plant Thermal Energy Shares: Natural gas Coal Biomass
25,907 2.76
100.0% 0.0% 0.0%
Biomass type shares Farmed Trees Herbaceous biomass Corn stover
0.0% 0.0% 100.0%
Electricity Inputs Imported grid electricity (kW h/gal EtOH)
0.69
Biomass-based electricity export (kWh/gal EtOH)
0.00
Electricity Generation Efficiency
32.1%
Displaced electricity mix for co-product electricity
G REET Selected
DGS co-product displacement ratios DG S yield (wet lbs/gal)
5.480
DG S displacing feed corn (lb/lb co-prod.)
0.992
DG S displacing soybean meal (lb/lb co-prod.)
0.306
DG S displacing N in Urea (lb/lb co-prod.)
0.022
% co-products used for new feed markets
0.0%
Methane reduction from cattle fed with DG S instead of corn
-3,381
Transport & Distribution Distance (mi)
Share
Barge
520.0
0.0%
Pipeline
600.0
0.0%
Rail
918.0
3.8%
Heavy duty truck
80.0
96.2%
10.0
100.0%
Ethanol Plant to Bulk Terminal
Bulk Terminal to Refueling Station Heavy duty truck to fuel station
Fuel Combustion Vehicle methane emissions (g/mi) Vehicle nitrous oxide emissions (g/mi) Vehicle energy use (Btu/mi)
0.015 0.012 4908
18
Prepared by: Steffen Mueller and Ken Copenhaver Prepared for: Illinois Corn Marketing Board
Date: September 18, 2009
Summary The present study is part of an ongoing effort by the University of Illinois at Chicago Energy Resources Center (UIC-ERC) to consistently evaluate the land use and the global warming impact of corn ethanol production. The study looks at the direct land use changes in the vicinity of two ethanol plants located in Illinois: the Illinois River Energy Center (IRE) ethanol plant located in Rochelle and the Patriot Renewable Fuels (PRF) plant located in Annawan. Both studied ethanol plants have rail access that enables them to deliver to the California market. The study provides an innovative approach to relate direct land use changes determined with high resolution satellite imagery to direct greenhouse gas emissions from the land use changes (while also accounting for different time horizons). Then, the direct land use emissions are added to the remainder of the life cycle greenhouse gas emissions (from the ethanol plant, distribution, denaturing process) to derive a total direct (including direct land use) life cycle global warming impact (GWI) for corn ethanol produced at a specific plant (IRE). Furthermore, data derived from the local, plant-level, land use analysis is compared to state-wide corn production and land use data. Land use surrounding IRE and GWI of its corn ethanol was assessed in previous studies conducted by the University of Illinois at Chicago Energy Resources Center (UIC-ERC).1 However, IRE recently expanded its capacity from 58 million gallons per year (mgpy) to 115 mgpy. The expanded plant started operating in November 2008. This presents also a unique opportunity to provide a longer term assessment of how the land use and the global warming impact change with larger ethanol production volumes produced at the same location. Land use surrounding PRF has not been previously assessed. PRF started operation in September 2008 with a capacity of 100 mgpy. Both IRE and PRF are located in relatively close proximity with a portion of the corn supply area overlapping. The study findings include the following:
The start up of PRF and IRE (expansion) prompted very little conversion of nonagricultural to agricultural land. The land use analysis performed for the corn supply areas showed that less than 1,000 acres were converted in the vicinity of either plant. With respect to land conversion, at a maximum 534 acres of forest (and 105 acres of grassland) were converted to corn around IRE and 609 acres (and 6 acres of grassland) were converted to corn around PRF. This finding supports the notion that corn ethanol plants have a weak influence on direct nonagricultural land conversions.
1
Mueller, S. and K. Copenhaver (2008). “A Bottom-Up Assessment of Land Use Related to Corn Ethanol Production.” Available at http://www.erc.uic.edu/staff/smueller.htm, and Mueller, Steffen and Michelle Wander, Ken Copenhaver (2008). “Global Warming and Land Use Impact of Corn Ethanol Produced at the Illinois River Energy Center”; Report prepared for the Illinois Corn Growers Association. Available at http://www.erc.uic.edu/staff/smueller.htm
2
Using a spreadsheet model (Direct Land Use Mapper “dLUC-Mapper”) developed by the UIC-ERC and land carbon factors from IPCC the greenhouse gas emissions from direct land use changes were assessed. The study finds that the direct land use changes would result in GHG emissions of 1.2 gCO2/MJ for IRE (30 year no discount) and 1.3 gCO2/MJ for PRF (30 year no discount). The respective 100 year, 2% discounted GHG emissions are, as expected, slightly lower.
The GWI from direct emissions of corn ethanol produced at the expanded IRE plant was reassessed as part of the present study. The GWI of the expanded plant is slightly lower at 53.5 compared to 54.8 gCO2e/MJ for the original plant. The difference is a combination of the different energy demand, the different ethanol yield, and the different DDGS production volume.
Adding the 1.2 gCO2e/MJ direct land use emissions to the remaining life cycle emissions of 53.5 gCO2e/MJ, results in total direct (including direct land use) GHG life cycle emissions for IRE of 54.7 gCO2e/MJ.
Despite the start-up of both PRF and IRE (expansion) in the fall of 2008 acres in corn in 2008 went down. During the start-up of the original IRE ethanol plant in 2006 acres in corn in the region increased to levels that could not be attributed to the ethanol plant demand (see original UIC-ERC study). Then, during the start-up of the expanded plant corn acres in the region went actually down (by 37,634 acres between the years 2007-2008). Likewise, corn acres around PRF went also down by 18,020 acres providing further evidence that the ethanol plants have a weak influence on corn rotations.
Also, in IRE’s case, an additional 340,000 acres of grass/pasture/hay land and in PRF’s case an additional 84,000 acres would have been available in 2008 for conversion to corn acres. This did not occur supporting the notion that the studied corn ethanol plants have a weak influence on agricultural land use in general. With intensive pasture and hay management some of these acres would still be available for additional corn production without associated indirect land use change effects.
Furthermore, the study analyzed Illinois-wide corn supply and use data. The analysis shows that corn production increases on constant corn acres were sufficient to support both increasing exports as well as corn for ethanol use.
Consequently, the studied ethanol plants a) did not influence local land use b) did not prompt indirect land use in the State of Illinois, c) did not influence the Illinois corn export balance.
3
1) Land Use Surrounding IRE As part of the original IRE study in 2007, we conducted a survey with growers delivering to the ethanol plant and concluded that an area consisting of a 40 mile circular radius would provide a good representation of the corn supply region for the plant. In support of the present study, IRE provided us with the addresses of all growers delivering directly to the plant, which allowed an even more accurate determination of the corn supply region. Direct grower deliveries account for approximately 60% of all corn delivered to IRE; the rest is sourced through grain elevators. We graphed the grower locations and determined a circular corn draw area that would include a representative amount of growers: We found that a circular corn draw area with a 40 mile radius would include 83% of the grower addresses while increasing the radius to 43 miles (expanding the radius to the Wisconsin border) would capture 92% of the grower addresses. We determined that a 43 mile radius would be a good representation of the corn supply area for the expanded IRE plant. The grower addresses and 43 mile circular area overlaid on top of the Illinois county map is shown in Figure 1.
Addresses of Growers Delivering to IRE 43 Mile Circle
IRE
Figure 1: Growers Delivering to the Expanded IRE Plant
In a next step, we analyzed land use and land use changes within the 43 mile radius corn supply area over the period of 2006, 2007, and 2008. This analysis was performed using the USDA NASS Cropland Data Layers (for crop types) which includes the national land cover dataset (for non-cropland conversions). The software used for this analysis was Erdas Imagine 9.1 developed by Leica Geosystems. While the USDA NASS Cropland Data Layer has been shown to have accurate methods of around 95% for the delineation of corn and soybeans and that dataset is updated every year, the national land cover data set dates back to 2001 and introduces much higher uncertainties for non-agricultural areas. Therefore, the present study analyzes the data
4
layers by a) applying an algorithm that subtracts the unlikely rotation scenario of land in agriculture converting to non-agricultural land and then back to agricultural land, and b) subtracts field fringes and roadway buffers. This approach has been shown by Mueller and Copenhaver (2009) to increase the accuracy of land use change analyses of this type.2 As can be seen in Table 1 the corn draw area includes a total of 3.7 million acres. The area in corn ranges between 1.4 and 1.7 million acres. Land use change from non agricultural lands associated with the introduction of the ethanol plant within the 43 mile radius circle is shown to be minimal (see Table 2). Of the 3.7+ million acres in the study area just over 1,000 acres were found to be converted from non-agricultural categories (grasslands, forest) from 2006 to 2007 and even fewer were converted from 2007 to 2008 after vetting the data for road fringes and misclassification associated with conversions from agriculture to non-agriculture and back to agriculture. This small percentage is well within the classification error of the dataset and could very well be attributed to misclassified pixels, indicating land use change from non-agricultural classes to agriculture did not, in all likelihood, occur despite the introduction of the ethanol plant. Furthermore, based on the USEPA’s definition of agricultural land (cropland, pasture and CRP) in the RFS2, there were over 340,000 acres still available for conversion from pasture/hay to corn for ethanol production in the corn supply area in 2007 and 2008. With intensive pasture and hay management these acres could produce additional corn without associated indirect land use change effects.
2
Mueller, S. and K. Copenhaver (2009). “Use of Remote Sensing to Measure Land Use Change from Biofuels Production.” Article submitted to Swords and Ploughshares - Special Issue on Sustainable Biofuels and Human Security - A Publication of the Program in Arms Control, Disarmament, and International Security (ACDIS), March 26, 2009
5
Table 1: Land Use Surrounding IRE 2006 Unvetted 1,391,597
2007 Unvetted 1,694,145
2008 Unvetted 1,656,511
955,428
614,399
666,208
2,347,025
2,308,544
2,322,719
Other Land Uses
1,373,378
1,411,859
1,397,684
Totals:
3,720,403
3,720,403
3,720,403
Corn Soybeans Total Corn and Soybeans
Table 2: Corn Acre Land Use Surrounding IRE 2007 Corn Acres in 2006 Corn
Unvetted 773,811
Vetted ag-nonag-ag 773,885
Vetted 648,689
753,498
753,544
623,530
idle cropland/ fallow
8,042
8,040
5,019
Grasslands
8,737
602
333
Forest
4,126
1,026
734
106,742
106,687
54,352
13,190
7,893
Soybeans
pasture/grass/hay Ag removed Road borders removed
306,528
2008 corn acres in 2007
Unvetted
Vetted ag-nonag-ag
Vetted
Corn
1,010,063
1,010,063
846,408
515,872
515,872
432,697
636
636
518
Grasslands
2,569
164
105
Forest
10,465
622
534
pasture/grass/hay
40,656
40,656
29,091
51,736
16,252
Soybeans idle cropland/ fallow
Ag removed Road borders removed
301,131
As stated above, the expanded plant started operation in November 2008 drawing corn from the 2008 harvest. The plant production levels of the original IRE plant and the expanded plant and the corresponding yields and land demands are shown in Table 3. It is evident that the expanded plant did not prompt additional corn acre conversions. In fact, between the years 2007-2008 corn acres decreased slightly by 37,634. While a relatively large increase of corn acres was observed between 2006-2007 (302,548 acres), this increase is also likely unrelated to the start up of the original ethanol plant in 2006 since the increase far exceeded the ethanol plant’s original corn land demand of 116,777 acres. Taking prevailing yield levels into account, the original and expanded plant require less than 13% of corn acres within the draw area to meet the corn
6
demand. Note that the plant also produces animal feed and that therefore the land demand that should be attributed to corn ethanol production is lower than 13%. Table 3: IRE Land Demand Corn Acres Ethanol Production (gallons) Average ethanol yield per bushel (gal/bu) Required Bushels Yield in Surrounding Counties (bu/acre)3 Required Acres Corn Land Demand (%)
2006 1,391,597
2007 1,694,145
2008 1,656,511
55,820,804 2.73 20,447,181 176 116,177 8.3%
55,820,804 2.73
104,280,111 2.76
20,447,181 186 109,931 6.5%
37,851,220 177 213,849 12.9%
2) Land Use Surrounding PRF In an analysis of growers delivering to PRF, plant personnel determined that 90% of corn receipts are within a 23 mile circle which includes the cities of Princeton, Prophetstown, Geneseo, Kewanee. Multiple reasons may contribute to the fact that the supply area for PRF is smaller than the one for IRE including significantly less urban shares. Also, as can be seen in Figure 2 the corn supply areas for PRF and IRE overlap slightly.
Corn Supply Areas for IRE and PRF
IRE
PRF
Addresses for growers delivering to IRE in 2008 (>95%) City addresses for growers delivering to Patriot in 2008 (>90%)
Figure 2: Growers Delivering to IRE and PRF
With the boundaries of the corn supply area established, we performed the land use analysis for the years 2006, 2007, and 2008. As can be seen in Table 4 the corn supply area includes a total of 1.1 million acres. The area in corn ranges between 441,000 to 3
USDA NASS County Crop Yield Report. www.nass.usda.gov
7
541,000 acres. As was the case for the IRE plant, land use change from non agricultural lands associated with the introduction of the PRF ethanol plant within the 23 mile radius circle is shown to be minimal. Of the 1.1 million acres in the study area just over 600 acres were found to be converted from non-agricultural categories from 2007 to 2008 after vetting the data for road fringes and misclassification associated with conversions from agriculture to non-agriculture and back to agriculture (see Table 5). Furthermore, based on the USEPA’s definition of agricultural land (cropland, pasture and CRP) in the RFS2, there were over 84,000 acres still available for conversion from pasture/hay to corn for ethanol production in the polygon in 2007 and 2008. Again, with intensive pasture and hay management some of these acres would also be available for additional corn production without associated indirect land use change effects. Figure 3 confirms the grass/pasture hay classifications include large contiguous areas suitable for conversion.
PRF and IRE Acres in Pasture/Hay Category 340,421 acres in IRE Corn Supply Area
83,729 acres in PRF Corn Supply Area
2008 Pasture/Hay 2008 Other Classes
Figure 3: Grass/Pasture/Hay Acreage in 2008
The PRF plant started operation in September 2008 drawing corn from the 2008 harvest. Consistent with the findings for the IRE plant, the PRF plant did not prompt additional corn acre conversions. In fact, between the years 2007-2008 corn acres decreased slightly by 18,020 acres. Table 4: Land Use Surrounding PRF Corn
2006 unvetted 441,346
2007 unvetted 558,957
2008 unvetted 540,937
Soybeans
351,510
224,652
248,608
792,856
783,609
789,545
271,777
281,231
275,485
1,064,633
1,064,840
1,065,030
Total Corn and Soybeans Other Land Uses Total Acres
8
Table 5: Corn Acre Land Use Surrounding PRF 2008 corn acres in 2007
Unvetted
Vetted for Ag-nonag-ag
Vetted
Corn
306,294
306,294
255,436
Soybeans
190,902
190,902
159,974
Winter wheat
5,051
5,051
4,142
Other crops
1,111
1,111
985
5
5
5
170
6
6
5,540
700
609
14,434
14,434
10,134
21,236
6,717
Fallow/idle cropland Grassland Forest Pasture/Grass/Hay Ag/Non-Ag/Ag removed Road borders removed
102,222
Due to the smaller supply circle, PRF’s corn land demand to supply the plant as a percentage of overall corn land within the supply circle is much higher than for IRE (34.3% for PRF compared to 13% for IRE).4 This indicates, however, that higher shares of corn acres going to corn ethanol production do not result in increased conversion of non agricultural land. Note that the plant also produces animal feed and that therefore the land demand that should be attributed to corn ethanol production is lower than 34%. Table 6: PRF Land Demand 2006
2007
Corn Acres
540,937
Ethanol Production (gallons)
100,000,000
Average ethanol yield per bushel (gal/bu)
2.8
Required Bushels Yield in Surrounding Counties (bu/acre) Required Acres
2008
35,714,286 174.5
191.5
Corn Land Demand (%)
192.5 185,529 34.30%
3) Correlating Ethanol Plant Corn Supply with Illinois Corn Supply Since, as demonstrated above, ethanol plants have a very weak influence on land use in their vicinity, one could assert that the plants must alter land use elsewhere (i.e. indirectly). To test this effect we analyzed Illinois’ corn production, export, and uses. The ProExporter Network’s Senior Economist Ross Korves provided the data reproduced in Figure 4 and Figure 5. As can be seen, over the last 35 years production of corn nearly doubled from 1.2 to 2.2 billion bushels. In fact production increases are strong enough to support both increases in the “other uses” category, which includes corn delivered to ethanol plants, and corn exports. Furthermore, carryout corn or the amount of corn on hand at the end of the marketing year (inventory) has also been increasing. Feedcorn, on 4
Note that PRF also produces animal feed and that therefore the actual land demand is lower.
9
the other hand, has been decreasing which may be indicative of decreasing livestock but also increased use of distillers grain as feedstock. Furthermore, while corn production increased over the time period corn acres in Illinois remained relatively constant (see Figure 6). In summary, the Illinois corn supply and use data shows that corn production increases on relatively constant corn acres were sufficient to support both increasing exports as well as corn for ethanol use. Consequently, the studied ethanol plants a) did not influence local land use b) did not prompt indirect land use in the State of Illinois, c) did not influence the Illinois corn export balance. Illinois Corn Production and Use (Years 1975-2010) 2,500,000
thousand bushel
2,000,000
1,500,000
1,000,000
500,000
-
1975 1977 1979 1981 1983 1985 1987 1989 1991 1993 1995 1997 1999 2001 2003 2005 2007 2009 Production
CarryOut
FeedUse
OtherUse
NetExports
Figure 4: Illinois Corn Production and Use 1975-2010 Illinois Corn Production and Use (Years 2000-2010) 2,500,000
thousand bushel
2,000,000
1,500,000
1,000,000
500,000
2000
2002
Production
2004
CarryOut
2006
FeedUse
Figure 5: Illinois Corn Production and Use 2000-2010
10
OtherUse
2008
NetExports
2010
Illinois Planted Corn Acres 14,000 12,000
thousand acres
10,000 8,000 6,000 4,000 2,000
09
05
07
20
20
20
01
99
97
95
93
91
03 20
20
19
19
19
19
19
89 19
85
83
81
79
77
87 19
19
19
19
19
19
19
75
0
Figure 6: USDA NASS Statistics on Illinois Planted Corn Acres (Source: www.nass.usda.gov/QuickStats)
These findings are consistent with the overall US export balance. The data compiled by Dr. Darrel Good (U of I Department of Agriculture Economics) and reproduced in Appendix A shows that the US was able to provide a) relatively constant corn exports and b) slightly increasing soy exports on relatively constant principal crop acres. 4) Relating Land use Change to Greenhouse Gas Emissions As detailed above, at a maximum, 534 acres of forest (and 105 acres of grassland) were converted to corn around IRE and 609 acres (and 6 acres of grassland) were converted to corn around PRF during plant start-up. Using a spreadsheet model (Direct Land Use Mapper “dLUC-Mapper”) developed by the UIC-ERC (posted at www.erc.uic.edu/staff/smueller.htm, input screen reproduced in Appendix C) and land carbon factors from IPCC these direct land use conversions would result in GHG emissions of 1.2 gCO2/MJ for IRE (30 year no discount) and 1.3 gCO2/MJ for PRF (30 year no discount). The respective 100 year, 2% discounted GHG emissions are, as expected, slightly lower (see Table 7).5
5
The employed IPCC factors are world average factors for temperate forests. Ideally, emissions factors specific to Midwestern forests should be used but were not available to the author at the time of the study.
11
Table 7: Direct Land Use Emissions Attributable to two Illinois Corn Ethanol Plants IRE
PRF
Ethanol Volume (Billion Gallons) Forest
0.115 -534
0.1 -609
Grassland
-105
-6
30 Year No Discount GHG Emissions (gCO2/MJ)
1.2
1.3
100 Year 2% Discount GHG Emissions (gCO2/MJ)
1.0
1.1
These GHG emissions values which are representative of domestic direct land use effects were derived using a) high resolution remote sensing tools with accuracies exceeding 95%, combined with b) published IPCC emissions factors, and c) an easily verifiable spreadsheet model 5) The Global Warming Impact of Corn Ethanol Produced at IRE The GWI of corn ethanol produced at the expanded IRE plant was assessed by parameterizing Argonne’s GREET model (Version 1.8c) with the energy and production values supplied by IRE listed in Appendix B. Parameterization of GREET was performed using an Interface Macro Tool developed by Life Cycle Associates. The input screen of the Interface Macro Tool is reproduced in Appendix D. The key parameters that were substituted for the GREET default values include the actual ethanol yield (per bushel corn) of the plant, the actual natural gas and electricity demand (and the Illinois electricity grid mix), the actual transportation distances of corn and ethanol, and the blend of denaturant. The GREET current default value for land use change was not included in the analysis. The results are shown in Table 8 and Figure 7. The GWI of the expanded Illinois River corn ethanol plant totals 53.5 gCO2e/MJ for the denatured corn ethanol and 53.2 gCO2e/MJ for the anhydrous corn ethanol. This is close to the original IRE Plant of 54.8 gCO2e/MJ. The direct land use change greenhouse gas emissions for IRE (30 year, no discount), as assessed above, total 1.2 gCO2e/MJ. Adding theses emissions to the remaining life cycle emissions of 53.5 gCO2e/MJ, results in total direct (including direct land use) GHG life cycle emissions for IRE of 54.7 gCO2e/MJ. Table 8: GWI of IRE2 Produced Corn Ethanol Gasoline Ag Phase and Distribution Ethanol Plant Denaturant IRE Total Gasoline
IRE2 Ethanol 23.8 29.4 0.3 92.1
12
The GWI of Corn Ethanol Produced at the Expanded Illinois River Energy Ethanol Plant 100 90 80
gCO2 e /MJ
70 60
Denaturant
50
Ethanol Plant
40
Ag Phase and Distribution
30 20 10 0 Gasoline
IRE2 Ethanol
Figure 7: GWI of the Expanded IRE Plant
13
Appendix A: US Crop Data Provided by U of I Department of Agriculture Economics
U.S. Principal Crop Acres (million acres)
1999-- 329.3 2000-- 328.7 2001-- 324.6 2002-- 327.3 2003-- 325.7 2004-- 322.3 2005-- 317.6 2006-- 315.6 2007-- 320.4 2008-- 324.8 2009– 320.9
US Corn Exports 2.4 B illio n B u s h e ls
2.0 1.6 1.2 0.8 0.4 0.0 79 81 83 85 87 89 91 93 95 97 99 01 03 05 07 09
US Soybean Exports 1.4 1.3 1.2
Billion Bushels
1.1 1.0 0.9 0.8 0.7 0.6 0.5 0.4 0.3 0.2 80
82
84
86
88
90
92
94
96
14
98
00
02
04
06
08
Appendix B: Energy and Production Parameters of the Original and Expanded IRE Plant Unit Plant Performance: Annual total anhydrous ethanol production Annual total denatured ethanol production Description of denaturant used (type)
gallon per year gallon per year Debutanized Natural Gasoline gal/bu
Average ethanol yield per bushel (anhydrous) Plant Energy Systems: Annual total natural gas consumption HHV Annual total electricity consumption Natural Gas (HHV) per unit Anhydrous Ethanol Production Natural Gas (HHV) per unit Denatured Ethanol Production Natural Gas (LHV) per unit Anhydrous Ethanol Production Natural Gas (LHV) per unit Denatured Ethanol Production Electricity per unit Anhydrous Ethanol Production Electricity per unit Denatured Ethanol Production By-Products: Annual total DDGS production Annual avg DDGS moisture Annual total WDG(S) production Annual avg WDG(S) moisture Annual total S production - as product sold Annual avg S moisture Transportation Logistics: Corn by truck Corn by rail DDGS shipments by truck
% % %
DDGS shipments by rail
%
DDGS shipments by ship
%
WDGS shipments by truck Ethanol shipments by truck Ethanol shipments by rail Ethanol shipments by barge Avg ethanol distance transported by truck (per trip - one way) Avg ethanol distance transported by rail (per trip one way) Avg ethanol distance transport from terminal to retail outlet (per trip one way) DDGS (11% Moisture) DDGS equivalent of WDGS (corrected to 11% Moisture) DDGS equivalent of WDGS Total DDGS equivalent Denaturant by %
lbs/gal tons
IRE1
IRE2 55,820,804 57,812,280 DNG 2.73
104,280,111 106,709,420 DNG 2.755
Btu kWh Btu/gal
1,671,765,900,000 39,898,320 29,949
Btu/gal
28,917
28,071
26,981
25,907
26,051
25,289
0.71 0.69
0.692 0.675
kWh/gal kWh/gal tons % tons % tons %
3,000,521,873,185 72,057,134 28,757
153,213 11 13,488 30 5,036 60
281,939 11 7,340 60 7,990 60 100 0 11.1
% % % % mi
100 0 Backhaul Shipment Backhaul Shipment Backhaul Shipment 100 98 2 0 80
mi
1,000
918
mi
10
10
lbs/gal lbs/gal
15
5.49
1.82 87.08 100 96.21 3.79 0 80
5.41 3,743 0.072 5.48 2.3%
Appendix C: Direct GHG Emissions Spreadsheet Model Direct GHG Emissions Surrounding Corn Ethanol Plants Author: Steffen Mueller, PhD Copyright: University of Illinois at Chicago Yellow Cells Allow Customization for Different Ethanol Plants Ethanol Plants: IRE Plant Ethanol Volume (Billion Gallons)
PRF 0.115
0.1
-534
-609
-105
-6
Land Conversion: Forest Land Conversion: Grassland 0.115
Ethanol Volume
Billion Gallons
Production Years (Foregone C-Sequestration Period) Select Duration:
List of Possible Durations: 20 30 100
30 Years
Analytical Horizon (Amortization, Above Ground Carbon Release Horizon) Select Duration
List of Possible Durations:
30 Years
30 50 100 Recovered Carbon (Restoration of original vegetation after biofuels program)
0% Percent
Input Recovered Carbon:
for 0% use 0.0001 to avoid division by 0
Error Message: Carbon Recovery Time Insufficient Emissions Discounting Method
0.00% Percent
Input Discount Rate: Carbon Flux Duration Above Ground Biomass Below Ground Biomass
Starting in Year 1 5
0 0
Foregone Net Primary Production
30
0
Recovered Carbon
0.01
30
Undiscounted Emissions Factors
Region Forest Grassland
Area Cover Change (acre)
Area Cover Change (ha) -534 -105
Above Ground tC/ha -216 -43
Net Primary Below Ground Production tC/ha tC/ha-yr 57.0 96.0 6.3 7.0 236.0 4.2
Net Present Value of Emissions Factors
Region Forest Grassland
Recovered Carbon
Net Emissions
Recover ed Recovered Carbon Carbon Undiscounte Discoun Below Ground Net Primary Production ted Forest tCO2/ha d Above Ground tC/ha tC/ha tC/ha-yr NPV tC/ha 57.0 96.00 187.47 340.5 0.0 0.0 1250 7.0 235.99 127.18 370.2 0.0 0.0 1359
Forest Total Emissions tCO2e Annualized Emissions tCO2e Annualized Emissions gCO2e/gal Annualized Emissions gCO2e/MJ
Indirect Land Use Change (gCO2e/MJ):
tCO2e Grassland tCO2e Forest+Grassl. tCO2e 270,138 57,752 327,890 9,006 1,925 10,931 78 17 95 1.0 0.2 1.2
1.2 16
Appendix D: GREET Interface Macro Tool for IRE Corn Ethanol Corn Cultivation General Target year
2010
Corn farming energy (Btu/bu)
7,800
Bailing energy (Btu/bu)
0
Total ag. energy input (Btu/bu)
7,800
Corn yield (Bu/acre)
196.0
Stover Yield (Tons/acre)
0.00
Bu per ton stover
19600000.0
Farming Energy Shares Diesel
100.0%
Gasoline
0.0%
Natural G as
0.0%
LPG
0.0%
Electricity
0.0%
Electricity Generation Mix for Farming Electricity mix
IL SERC
Fuel shares if electricity mix is "User Defined": Residual oil Natural gas
1.8% 10.0%
Coal
57.3%
Nuclear
25.2%
Biomass
1.9%
Other (renewables)
3.8% International
Chemical Inputs (Fertilizer, Herbicide, Insecticide) N input (g/bu)
368.0
Production Share Ocean Tanker (mi)
Ammonia
70.7%
60.0%
Urea
21.1%
60.0%
5,200
3,000
Ammonium nitrate P2 O5 input (g/bu)
8.2%
60.0%
3,700
193.6
60.0%
4,200
K2 O input (g/bu)
300.5
60.0%
3,900
1,419.4
0.0%
Herbicide input (g/bu)
8.36
60.0%
4,000
Insecticide input (g/bu)
0.70
60.0%
4,000
CaCO3 input (g/bu)
Soil N Emissions Baseline N content of above and below ground biomass (g/bu) N removed in stover
141.6 0.0
Above and below ground N available for N2O loss N in N 2 O as % of N in fertilizer and biomass
141.6 1.3%
Land Use Change and Sustainability Land-use change CO2 emissions (g/bu)
0.0
17
Corn and Stover Transport Corn Field to Stack Medium duty truck Corn Stack to Ethanol Plant
Distance (mi)
Share
0.1
100.0%
Distance (mi)
Share
Barge
350
0.0%
Rail
400
0.0%
Heavy duty truck
40
100.0%
Corn Stover Transport Heavy duty truck
Distance (mi)
Share
0
100.0%
Dry Mill Ethanol Plant (Anhydrous EtOH Basis) Thermal energy for NG, coal, biomass (Btu/gal EtOH), LHV Ethanol yield (gal/bu) Dry Mill Ethanol Plant Thermal Energy Shares: Natural gas Coal Biomass
25,907 2.76
100.0% 0.0% 0.0%
Biomass type shares Farmed Trees Herbaceous biomass Corn stover
0.0% 0.0% 100.0%
Electricity Inputs Imported grid electricity (kW h/gal EtOH)
0.69
Biomass-based electricity export (kWh/gal EtOH)
0.00
Electricity Generation Efficiency
32.1%
Displaced electricity mix for co-product electricity
G REET Selected
DGS co-product displacement ratios DG S yield (wet lbs/gal)
5.480
DG S displacing feed corn (lb/lb co-prod.)
0.992
DG S displacing soybean meal (lb/lb co-prod.)
0.306
DG S displacing N in Urea (lb/lb co-prod.)
0.022
% co-products used for new feed markets
0.0%
Methane reduction from cattle fed with DG S instead of corn
-3,381
Transport & Distribution Distance (mi)
Share
Barge
520.0
0.0%
Pipeline
600.0
0.0%
Rail
918.0
3.8%
Heavy duty truck
80.0
96.2%
10.0
100.0%
Ethanol Plant to Bulk Terminal
Bulk Terminal to Refueling Station Heavy duty truck to fuel station
Fuel Combustion Vehicle methane emissions (g/mi) Vehicle nitrous oxide emissions (g/mi) Vehicle energy use (Btu/mi)
0.015 0.012 4908
18
