Biomass Byproduct GIS Database
Biomass Byproduct GIS Database: Forest Industry Byproduct Survey and Modeling Results for Northwest Wisconsin August 27, 2003 Richard Barber Northwest Regional Planning Commission
Biomass Byproduct GIS Database: Forest Industry Byproduct Survey and Modeling Results for Northwest Wisconsin August 27, 2003
Completed by: Richard Barber Northwest Regional Planning Commission 1400 South River Street Spooner, Wisconsin 54801 Also Available Online at http://www.nwrpc.com/forestry Funding provided by Northwest Regional Planning Commission, City of Ladysmith, and Wisconsin Focus on Energy Renewable Energy Program
Biomass Byproduct GIS Database
Contents INTRODUCTION .......................................................................................................................................................1 SURVEY DESCRIPTION, ESTIMATION METHODS AND CONCERNS ........................................................2 SURVEY AND ESTIMATE RESULTS ....................................................................................................................5 CONCLUSIONS..........................................................................................................................................................9 REFERENCES ..........................................................................................................................................................10 APPENDIX ................................................................................................................................................................11 COUNTY BYPRODUCT SUMMARY MAPS ...................................................................................................................11 BYPRODUCT PRODUCER LOCATION MAP .................................................................................................................15
i
Biomass Byproduct GIS Database
Introduction Wisconsin’s forest products industry produces large amounts of residues such as bark, sawdust, and solid wood. This project accurately located each producer in a Geographic Information System (GIS) in order to allow accurate delivered cost calculations for a user located in Northwest Wisconsin. Producers who were missed in earlier studies were also included here. While most of this residue is used, few economical markets currently exist. These forest industry byproducts can be used to produce bio-power. Pilot plants are also producing ethanol and chemicals from these materials; large-sale commercial operations should be feasible within the next five to ten years. Production of residue is also highly concentrated with 11 of the 217 producers accounting for a majority of the total output. The smallest producers make up the overwhelming majority of the facilities but account for only a small percentage of the total output.
1
Biomass Byproduct GIS Database
Survey Description, Estimation Methods and Concerns Feedstock cost is typically one of the most important, if not the most critical, factor in determining bio-energy and similar consumptive project locations. Figure 1 below illustrates that, in this example, about every $10 per ton difference in feedstock cost translates into a 10 percent difference in the project’s return on investment. The northwest region of Wisconsin appears to have a relative strength in this area; this study was done to help users locate here and minimize their delivered feedstock cost.
Annual Return on Investment (IRR)
Figure 1 – Example Feedstock Cost Sensitivity Analysis for Bioethanol2 35.0% 30.0% 25.0% 20.0% 15.0% 10.0% 5.0% 0.0% -5.0% $10.00
$20.00
$30.00
$40.00
Feedstock Cost ($/dry ton)
The individual producers were directly surveyed by the Wisconsin DNR in 1999 and 2000 in cooperation with the North Central US Forest Service Experiment Station. Non-respondents were contacted until the DNR felt that they had achieved a complete census, although it was recently determined that some producers were still missed1. Their output was estimated as described below and the results are included with those presented here. Actually, all mill-specific numbers are partially estimated since the US Forest Service is unable to disclose individual responses due to federal regulations. County-level figures, however, are generally from direct reports. In addition, employment estimates were obtained from Harris directory and Reference USA. In most cases, these sources also provided the geographic positions. Where this was not possible, the location of the city given in the address was used. Mill-specific estimates were prepared using linear regression comparing survey summaries of county-level totals for each byproduct and type of mill along with employment estimates for each producer. These initial estimates are shown below as “estimated production”.
2
Biomass Byproduct GIS Database
These estimates were then adjusted so that the total county production matched the reported production. This is shown at right as “reported production”; these are the figures used in the database and the remainder of this report.
Total Dry Tons per Year
Figures 2 and 3 show that the expected outputs obtained through linear regression were generally very close to reported levels for pulp and OSB mills, but there were greater discrepancies for the larger sawmills. This suggests that sawmills below 40 employees or 5000 dry tons can have production accurately estimated with Figure 3 this method, while larger sawmills Reported vs. Expected Production for Sawmills require a direct survey or other 35000 verification. 30000 25000 20000
Reported Production
15000
Estimated Production
10000 5000 0 0
50
100
150
200
250
Employment
Total Dry Tons per Year
Figure 2 Reported vs. Expected Production for Pulp & OSB Mills 180000 160000 140000 120000 100000 80000 60000 40000 20000 0
Reported Production Estimated Production
0
1000
2000
3000
4000
Employment
Figure 4 Impact of Data Generation Method on Total Tons 700000 Dry Tons per Year
600000 500000 400000
Direct Survey
300000
Estimate Only
200000 100000 0 Bark
Coarse
Fine
Byproduct Type
3
The pulp and OSB mills are more predictable regardless of size. This is probably because they are much larger producers and therefore probably based their reported figures on actual numbers, whereas many sawmills do not accurately track their byproducts and could have given less precise reports. In addition, some of the sawmills are conducting other activities, so the reported employment could include people who do not work in the sawmill part of the business. In addition, the overall figures are unlikely to change much in most cases since, on average, the deviation between expected and reported figures is still quite small. Still, additional surveys could be considered for queries that include large differences between estimated and reported production in sawmills. Another significant factor is the added volume from producers who were not included in the original DNR survey. All of these missing producers were sawmills initially identified from the 2000 versions of Harris Directory3 and Reference USA4. The US Forest Service is currently investigating how they were missed. The production of
Biomass Byproduct GIS Database
these mills has been forecast based on the reported employment and the previously discussed regression analysis from survey data. The impact of this adjustment is shown in figure 4. Still, a survey of these producers should be considered where their estimated output makes up a significant part of the desired byproduct for a given location, especially since they are all sawmills. In summary, there are two key factors that could impact the accuracy of results for a site-specific query. First, the variation between estimated and reported production, especially among larger sawmills with more than 40 employees or more than 5000 total dry tons per year. Second, the amount of material originating from mills that were not directly surveyed. In either case, additional verification from the producers should be considered. Still, it is unlikely that this will be an issue since. In general, the average reported production would be very close to the number expected from the regression analysis, thereby lending a high degree of confidence to the result.
4
Biomass Byproduct GIS Database
Survey and Estimate Results The unique feature of this project is the GIS database that allows precise, site-specific, demandspecific estimates of supply for any location in the region. All 217 producers in the area are included, and results were cross-checked by calculating expected results as discussed previously. Output can be estimated in a number of formats, typically presented as delivered price and available supply given project parameters. Price will typically vary depending on the supplier, competing uses, and delivery charges. An example summary is provided below in table 1a and 1b for a theoretical ethanol facility in Ladysmith. These numbers are only an example; those desiring an estimate should contact the author directly for project-specific estimates. Example Site-Specific, Project-Specific Summaries Table 1a – Ladysmith Feedstock Availability Summary2 Type of Feedstock Bark Solid Wood Residue Sawdust Timber Total Dry Tons/Day
Average Maximum Daily Dry Tons at Delivered Price/Dry Ton* Undelivered Price/Dry $0-15 $15-20 $20-25 $25-30 $30-35 $35-40 Ton $8 132 351 91 1,010 0 0 $14 0 17 194 266 121 2,564 $14 0 10 193 194 78 2,680 $21 0 0 214 272 1,522 132 378 692 1,743 1,721 5,243
Table 1b – Ladysmith Ethanol Feedstock Purchase Summary at 2200 Tons Per Day 2 Average Purchased Daily Dry Tons at Delivered Price/Ton** Undelivered Price/Dry $0-15 $15-20 $20-25 $25-30 $30-35 $35-40 Ton Bark*** $8 0 0 0 0 0 0 Solid Wood Residue $14 0 8 97 133 60 659 Sawdust $14 0 5 97 97 39 0 Timber $21 0 0 107 136 761 0 Subtotal Dry Tons/Day 0 13 301 366 860 659 Cumulative Dry Tons/Day 0 13 314 680 1,541 2,200 Cumulative Feedstock Cost/Day $0 $247 $7,012 $17,082 $45,047 $65,656 Average Feedstock Cost/Dry Ton $29.84 Average Ethanol Yield Gal./Dry Ton 88.3 Avg. Feedstock Cost $/Gal. Ethanol $0.338 Type of Feedstock
* Numbers shown in table 3a assume 100% availability. ** Purchases shown in table 3b assume using lowest price feedstock first at overall availability of 50% *** No bark residue is utilized since it does not have an attractive composition for use in this example process. Anticipated feedstock yields are given in the appendix.
5
Biomass Byproduct GIS Database
A variety of large-picture supply estimates are also provided. As was shown in earlier surveys, bark makes up the majority of the output; sawmills and pulp mills account for most production; and nearly all of the byproduct is now used. The use is overwhelmingly for fuel, a relatively low-value product. In addition, most of the producers are small sawmills, but the vast majority of the output comes from large facilities. Perhaps of most significance is the concentration of production and its variation for each byproduct. Annual Dry Tons Wood Byproduct Northwest Wisconsin ior uper eS Lak
Total Tons per Year (Bark, Coarse &Fine) 0 - 13,759 13,760 - 48,389 48,390 - 148,409 148,410 - 378,465 U.S. Highways
r rio upe eS Lak
/(2
Bayfield
Douglas
Iron
(/
63
ot es nn Mi
a
Mi c h
Ashland Washburn
/(8
Oneida
(/
94
/(10
Dunn
Pepin
(/53
Clark
(/ T r Buffalo em pe al ea u
Shawano (/10
53
N E S
10
0
10 20 Miles
Source: Northwest RPC.
Menominee
Marathon EauClaire
Forest
Langlade
Taylor Chippewa
(/
Pierce
W
/(45
8
Lincoln
Barron
63
(/12
(/
Rusk
Polk
Vilas
Sawyer Price
St Croix
n
(/51
(/53
Burnett
iga
/(94
LaCrosse
Wood
Portage
Jackson
Waupaca
(/
51
Adams
Monroe Juneau
Waushara
Winneba
An overview of byproduct production distribution is shown at the left. Overall production is concentrated in counties with the largest pulp or OSB mills, but areas with a significant number of sawmills also have significant production. Sawmill concentration is most noticeable in the distribution of coarse residue since this is not a byproduct of pulp mills. Summary maps for the fine, coarse, and bark residues are shown in the appendix along with a larger map of total byproduct output. As shown in figure 6, bark makes up roughly half of the byproduct output and accounts for around 550 thousand dry tons annually in the survey region. Coarse debris, mostly board and log cutoffs, provide 225 thousand dry tons per year while fine residue, or sawdust, makes up another
Marquette Green Lake /(51
260 thousand dry tons annually. Production is also highly concentrated in the largest few producers, all of which are pulp mills except for one OSB mill. This production concentration also explains the output distribution shown on the county-level supply maps; it is a key consideration in project location.
Figure 6 Output by Byproduct
Dry Tons per Year
600,000 500,000 400,000 300,000 200,000 100,000 0 Bark
Coarse
Fine
6
Biomass Byproduct GIS Database
Figure 7: Share of Total Production by Output Class More
5000
10000
40000 35000 30000 25000
15000 20000
Figure 8: Share of Number of Producers by Output Class 10000 15000 20000 25000 30000 5000
35000 40000 More
As illustrated by comparing the two graphs at the left, the top 20 percent of producers account for 88 percent of total output. In addition, of the 216 producers in the survey area, the top 11 accounts for the majority of the output. Future survey work and residue-utilizing projects should carefully examine these major suppliers. However, mills producing less than 5000 dry tons could be primarily computermodeled. This is advisable since, as shown at left, they are numerous but have a low portion of total production. In addition, as discussed previously, their output can be accurately modeled.
Sawmills are also the second largest producers behind the pulp mills. 600,000 Veneer and miscellaneous mills 500,000 produce very little as a 400,000 group mainly because Fine there are far more Coarse 300,000 sawmills. Also, note the Bark 200,000 product distribution differences by mill type; 100,000 this variation explains 0 the stark distribution Sawmill Veneer Pulpmill OSB Misc difference between bark and coarse debris. Most pulp mills are in the southern part of the survey area, so the bark and, to a lesser extent, the sawdust, or fine, residues are concentrated in those counties. Dry Tons per Year
Figure 9: Production by Mill Type
7
Biomass Byproduct GIS Database
This survey collected only very basic information about current use, but it clearly 900 shows that virtually the entire 800 production of residue is now 700 used, overwhelmingly for 600 Fine fuel. Although it was not 500 Coarse examined in this study, these 400 Bark fuel markets are mostly 300 internal-use. Excess 200 100 production is sold, mostly 0 during the summer. Not Used Misc Products Fiber Fuel Products Undelivered bark prices are Products generally under $5 per green ton. More efficient users should be able to procure even the internally used portion for an acceptable price. Coarse residue, however, will have a higher market price due to competing uses for fiber products. Most of this, therefore, will be less cost effective for fuel uses, but generally still preferable to virgin feedstock2. Dry Tons per Year
Figure 10: Byproduct Use
Clearly, additional research is needed to more fully explore these competing markets. This should focus on the largest producers with an additional survey of the current buyers of residue to compare consumption and supply proximity.
8
Biomass Byproduct GIS Database
Conclusions Use of this database for project-specific supply estimates is accurate, especially when considering total production and delivery costs. Available supply, however, will need more careful assessment of competing uses. The region surveyed produces a large amount of bark, sawdust, and coarse forest industry residue. This production can be accurately estimated for most producers. This allows inexpensive updates from published producer-specific employment statistics. Larger sawmills, however, will need further examination and verification to accurately assess production levels. In addition, production is concentrated in pulp and OSB mills with the remainder coming primarily from larger sawmills. Nearly the entire production of residue is currently used, mostly for fuel. Coarse residue has the strongest market with most used for fiber. In using this database to estimate site-specific supply and delivered feedstock cost, the following should be carefully considered for additional research: 1. Further investigation of the long-term feedstock costs, availability, and competing uses 2. Verification of production from sawmills producing more than 5000 dry tons per year 3. Careful examination of the largest producers, a few of which will likely make up the majority of the potential supply for a given project. A careful examination of the residue supply is now just beginning. Further research should complete the picture and further improve these critical cost estimates.
9
Biomass Byproduct GIS Database
References Reading, W, et al, “Timber Product Output Database for 2000”, North Central Forest Experiment Station, St. Paul, MN, 2003 1
Barber, R., “Ladysmith Bioethanol Pre-Feasibility Study”, Internal Report, Northwest Regional Planning Commission, Spooner, WI, 6/1/2003 2
Carlsen, F., “Wisconsin Manufacturers Directory 2000”, Harris Infosource, Twinsburg, OH, 1999 3
Reference USA, www.referenceusa.com, InfoUSA Inc., Omaha, NE, 2003 (only used business initially listed 2000 or earlier) 4
10
Biomass Byproduct GIS Database
Appendix County Byproduct Summary Maps
Annual Dry Tons Wood Byproduct Northwest Wisconsin Total Tons per Year (Bark, Coarse & Fine)
r rio upe S ke La
e Lak
0 - 13,759 13,760 - 48,389 48,390 - 148,409 148,410 - 378,465
or eri Sup
(/ 2
Douglas
U.S. Highways
Bayfield Iron
(/63
hi g a
n
so
ta
Ashland /(5 1
in
ne
(/53
Burnett
M
Mi c
Washburn
Vilas
Sawyer Oneida
Price
(/45
Fo
(/ 8
Rusk (/ 8
Polk
Lincoln
Barron
St Croix (/63 (/12
Langlade
Taylor Chippewa (/5 3
/(9 4
Men
Dunn
Marathon Clark
Pierce (/
10
Pepin
Shaw
Eau Claire (/10
E
u ea al pe
W
Buffalo
em Tr
N
(/53
Wood /(94
Portage
Waupaca
(/
Jackson
51
S
10
0
10 20 Miles
La Crosse
Adams Monroe
Source: Northwest RPC.
Juneau
11
Waushara Marquette Green Lake
W
Biomass Byproduct GIS Database
Annual Dry Tons - Bark Northwest Wisconsin ke La
or eri Sup
U.S. Highways Tons per Year 0 - 11,710 11,710 - 63,590 63,590 - 97,840 97,840 - 256,620
or eri Sup e Lak
/( 2
Bayfield
Douglas
Iron
(/6 3
n
ta
so
(/5 1
ne in
Washburn
Vilas
Sawyer Oneida
Price
/(45
Fore
(/ 8
Rusk (/
h i ga
Ashland (/53
Burnett
M
Mi c
8
Polk
Lincoln
Barron
St Croix (/6 3 (/ 12
/(5 3 (/94
Langlade
Taylor Chippewa
Meno
Dunn
Marathon Clark
Pierce (/
10
Pepin
Shaw
Eau Claire /( 10
E
(/5 3
u ea al pe
W
Buffalo
em Tr
N
Wood /(9 4
Portage
Waupaca
(/
Jackson
51
S
10
0
10 20 Miles
La Crosse
Adams Monroe
Source: Northwest RPC.
Juneau
12
Waushara Marquette Green Lake
W
Biomass Byproduct GIS Database
Annual Dry Tons - Coarse Northwest Wisconsin U.S. Highways Tons per Year 0 - 4,010 ior r 4,010 - 9,960 e up eS k 9,960 - 17,980 a L 17,980 - 28,230
r rio upe S ke La
/( 2
Bayfield
Douglas
Iron
(/63
Mi c
hi ga
n
ta
Ashland (/5 1
ne
so
(/5 3
M
in
Burnett
Washburn
Sawyer Oneida
Price
/(4 5
Fo
(/ 8
Rusk (/
Vilas
8
Polk
Lincoln
Barron
St Croix (/63 (/12
/(53 (/94
Langlade
Taylor Chippewa
Me
Dunn
Marathon Clark
Pierce (/10
Pepin
Sha
Eau Claire /(10
E S
10
0
(/53
u ea al pe
W
Buffalo
em Tr
N
Wood /(94
(/
Jackson
Adams Monroe
Source: Northwest RPC.
Juneau
13
Waupac
51
10 20 Miles La Crosse
Portage
Waushara Marquette Green Lake
W
Biomass Byproduct GIS Database
Annual Dry Tons - Fine Northwest Wisconsin r rio upe S ke La
r rio upe S e Lak
U.S. Highways Tons per Year 0 - 6,470 6,470 - 13,140 13,140 - 25,310 25,310 - 98,570
/( 2
Bayfield
Douglas
Iron
(/63
Mi c
hi ga
n
ta
Ashland (/5 1
ne
so
(/5 3
M
in
Burnett
Washburn
Sawyer Oneida
Price
/(4 5
Fo
(/ 8
Rusk (/
Vilas
8
Polk
Lincoln
Barron
St Croix (/63 (/12
/(53 (/94
Langlade
Taylor Chippewa
Me
Dunn
Marathon Clark
Pierce (/10
Pepin
Sha
Eau Claire /(10
E S
10
0
(/53
u ea al pe
W
Buffalo
em Tr
N
Wood /(94
(/
Jackson
Adams Monroe
Source: Northwest RPC.
Juneau
14
Waupac
51
10 20 Miles La Crosse
Portage
Waushara Marquette Green Lake
W
Biomass Byproduct GIS Database
Byproduct Producer Location Map
Annual Dry Tons - Wood Byproducts by Mill* Northwest Wisconsin *NOTE: Mill Locations are approximations only.
S ke La
ior er p u
#
# #
ke La
# ##
#
(/ 2
Douglas
#
Bayfield# #
ta
r rio
#
#
#
#
Iron
#
o s
e
(/
n
in
##
##
####
#
#
#
#
#
#
#
#
Vilas #
Sawyer
#
#
/(5 1
##
#
Washburn
Burnett
an
#
# Ashland
##
#
hig
#
#
#
Mic
#
(/6 3
#
M
pe Su
#
##
53
Tons per Year (Bark, Coarse & Fine) < 1,000 # 1,000 - 36,000 36,000 - 82,000 # # 82,000 - 163,000 Railroads U.S. Highways
#
# Price
#
#
Oneida #
#
#
###
#
(/ 8
#
Polk
#
Rusk#
#
(/
94
#
#
#
## #
Lincoln
# #
#
#
/(5 3
Taylor #
#
#
#
#
#
#
#
#
#
#
###
#
#
(/10
#
Pepin #
Eau Claire
#
Meno
##
## # Marathon # #
# #
##
#
#
#
#
##
Clark
#
#
#
#
#
#
#
Pierce
Langlade
#
# #
#
Chippewa
Dunn
#
#
#
#
#
#
#
#
#
##
#
St Croix (/63 (/12
(/ 8
#
#
#
Barron
#
#
## #
Fore
##
# #
(/
45
Shawa
#
#
/( 10
# #
E
Buffalo
#
##
(/5 3 #
u ea al pe
W
em Tr
##
N
##
##
#
#
#
#
/(9 4 ##
Jackson
## # ## #
# Wood
Portage
Waupaca
#
(/
51
#
S
10
0
#
10 20 Miles
Waushara #
Source: Northwest RPC.
15
Wi
Biomass Byproduct GIS Database: Forest Industry Byproduct Survey and Modeling Results for Northwest Wisconsin August 27, 2003
Completed by: Richard Barber Northwest Regional Planning Commission 1400 South River Street Spooner, Wisconsin 54801 Also Available Online at http://www.nwrpc.com/forestry Funding provided by Northwest Regional Planning Commission, City of Ladysmith, and Wisconsin Focus on Energy Renewable Energy Program
Biomass Byproduct GIS Database
Contents INTRODUCTION .......................................................................................................................................................1 SURVEY DESCRIPTION, ESTIMATION METHODS AND CONCERNS ........................................................2 SURVEY AND ESTIMATE RESULTS ....................................................................................................................5 CONCLUSIONS..........................................................................................................................................................9 REFERENCES ..........................................................................................................................................................10 APPENDIX ................................................................................................................................................................11 COUNTY BYPRODUCT SUMMARY MAPS ...................................................................................................................11 BYPRODUCT PRODUCER LOCATION MAP .................................................................................................................15
i
Biomass Byproduct GIS Database
Introduction Wisconsin’s forest products industry produces large amounts of residues such as bark, sawdust, and solid wood. This project accurately located each producer in a Geographic Information System (GIS) in order to allow accurate delivered cost calculations for a user located in Northwest Wisconsin. Producers who were missed in earlier studies were also included here. While most of this residue is used, few economical markets currently exist. These forest industry byproducts can be used to produce bio-power. Pilot plants are also producing ethanol and chemicals from these materials; large-sale commercial operations should be feasible within the next five to ten years. Production of residue is also highly concentrated with 11 of the 217 producers accounting for a majority of the total output. The smallest producers make up the overwhelming majority of the facilities but account for only a small percentage of the total output.
1
Biomass Byproduct GIS Database
Survey Description, Estimation Methods and Concerns Feedstock cost is typically one of the most important, if not the most critical, factor in determining bio-energy and similar consumptive project locations. Figure 1 below illustrates that, in this example, about every $10 per ton difference in feedstock cost translates into a 10 percent difference in the project’s return on investment. The northwest region of Wisconsin appears to have a relative strength in this area; this study was done to help users locate here and minimize their delivered feedstock cost.
Annual Return on Investment (IRR)
Figure 1 – Example Feedstock Cost Sensitivity Analysis for Bioethanol2 35.0% 30.0% 25.0% 20.0% 15.0% 10.0% 5.0% 0.0% -5.0% $10.00
$20.00
$30.00
$40.00
Feedstock Cost ($/dry ton)
The individual producers were directly surveyed by the Wisconsin DNR in 1999 and 2000 in cooperation with the North Central US Forest Service Experiment Station. Non-respondents were contacted until the DNR felt that they had achieved a complete census, although it was recently determined that some producers were still missed1. Their output was estimated as described below and the results are included with those presented here. Actually, all mill-specific numbers are partially estimated since the US Forest Service is unable to disclose individual responses due to federal regulations. County-level figures, however, are generally from direct reports. In addition, employment estimates were obtained from Harris directory and Reference USA. In most cases, these sources also provided the geographic positions. Where this was not possible, the location of the city given in the address was used. Mill-specific estimates were prepared using linear regression comparing survey summaries of county-level totals for each byproduct and type of mill along with employment estimates for each producer. These initial estimates are shown below as “estimated production”.
2
Biomass Byproduct GIS Database
These estimates were then adjusted so that the total county production matched the reported production. This is shown at right as “reported production”; these are the figures used in the database and the remainder of this report.
Total Dry Tons per Year
Figures 2 and 3 show that the expected outputs obtained through linear regression were generally very close to reported levels for pulp and OSB mills, but there were greater discrepancies for the larger sawmills. This suggests that sawmills below 40 employees or 5000 dry tons can have production accurately estimated with Figure 3 this method, while larger sawmills Reported vs. Expected Production for Sawmills require a direct survey or other 35000 verification. 30000 25000 20000
Reported Production
15000
Estimated Production
10000 5000 0 0
50
100
150
200
250
Employment
Total Dry Tons per Year
Figure 2 Reported vs. Expected Production for Pulp & OSB Mills 180000 160000 140000 120000 100000 80000 60000 40000 20000 0
Reported Production Estimated Production
0
1000
2000
3000
4000
Employment
Figure 4 Impact of Data Generation Method on Total Tons 700000 Dry Tons per Year
600000 500000 400000
Direct Survey
300000
Estimate Only
200000 100000 0 Bark
Coarse
Fine
Byproduct Type
3
The pulp and OSB mills are more predictable regardless of size. This is probably because they are much larger producers and therefore probably based their reported figures on actual numbers, whereas many sawmills do not accurately track their byproducts and could have given less precise reports. In addition, some of the sawmills are conducting other activities, so the reported employment could include people who do not work in the sawmill part of the business. In addition, the overall figures are unlikely to change much in most cases since, on average, the deviation between expected and reported figures is still quite small. Still, additional surveys could be considered for queries that include large differences between estimated and reported production in sawmills. Another significant factor is the added volume from producers who were not included in the original DNR survey. All of these missing producers were sawmills initially identified from the 2000 versions of Harris Directory3 and Reference USA4. The US Forest Service is currently investigating how they were missed. The production of
Biomass Byproduct GIS Database
these mills has been forecast based on the reported employment and the previously discussed regression analysis from survey data. The impact of this adjustment is shown in figure 4. Still, a survey of these producers should be considered where their estimated output makes up a significant part of the desired byproduct for a given location, especially since they are all sawmills. In summary, there are two key factors that could impact the accuracy of results for a site-specific query. First, the variation between estimated and reported production, especially among larger sawmills with more than 40 employees or more than 5000 total dry tons per year. Second, the amount of material originating from mills that were not directly surveyed. In either case, additional verification from the producers should be considered. Still, it is unlikely that this will be an issue since. In general, the average reported production would be very close to the number expected from the regression analysis, thereby lending a high degree of confidence to the result.
4
Biomass Byproduct GIS Database
Survey and Estimate Results The unique feature of this project is the GIS database that allows precise, site-specific, demandspecific estimates of supply for any location in the region. All 217 producers in the area are included, and results were cross-checked by calculating expected results as discussed previously. Output can be estimated in a number of formats, typically presented as delivered price and available supply given project parameters. Price will typically vary depending on the supplier, competing uses, and delivery charges. An example summary is provided below in table 1a and 1b for a theoretical ethanol facility in Ladysmith. These numbers are only an example; those desiring an estimate should contact the author directly for project-specific estimates. Example Site-Specific, Project-Specific Summaries Table 1a – Ladysmith Feedstock Availability Summary2 Type of Feedstock Bark Solid Wood Residue Sawdust Timber Total Dry Tons/Day
Average Maximum Daily Dry Tons at Delivered Price/Dry Ton* Undelivered Price/Dry $0-15 $15-20 $20-25 $25-30 $30-35 $35-40 Ton $8 132 351 91 1,010 0 0 $14 0 17 194 266 121 2,564 $14 0 10 193 194 78 2,680 $21 0 0 214 272 1,522 132 378 692 1,743 1,721 5,243
Table 1b – Ladysmith Ethanol Feedstock Purchase Summary at 2200 Tons Per Day 2 Average Purchased Daily Dry Tons at Delivered Price/Ton** Undelivered Price/Dry $0-15 $15-20 $20-25 $25-30 $30-35 $35-40 Ton Bark*** $8 0 0 0 0 0 0 Solid Wood Residue $14 0 8 97 133 60 659 Sawdust $14 0 5 97 97 39 0 Timber $21 0 0 107 136 761 0 Subtotal Dry Tons/Day 0 13 301 366 860 659 Cumulative Dry Tons/Day 0 13 314 680 1,541 2,200 Cumulative Feedstock Cost/Day $0 $247 $7,012 $17,082 $45,047 $65,656 Average Feedstock Cost/Dry Ton $29.84 Average Ethanol Yield Gal./Dry Ton 88.3 Avg. Feedstock Cost $/Gal. Ethanol $0.338 Type of Feedstock
* Numbers shown in table 3a assume 100% availability. ** Purchases shown in table 3b assume using lowest price feedstock first at overall availability of 50% *** No bark residue is utilized since it does not have an attractive composition for use in this example process. Anticipated feedstock yields are given in the appendix.
5
Biomass Byproduct GIS Database
A variety of large-picture supply estimates are also provided. As was shown in earlier surveys, bark makes up the majority of the output; sawmills and pulp mills account for most production; and nearly all of the byproduct is now used. The use is overwhelmingly for fuel, a relatively low-value product. In addition, most of the producers are small sawmills, but the vast majority of the output comes from large facilities. Perhaps of most significance is the concentration of production and its variation for each byproduct. Annual Dry Tons Wood Byproduct Northwest Wisconsin ior uper eS Lak
Total Tons per Year (Bark, Coarse &Fine) 0 - 13,759 13,760 - 48,389 48,390 - 148,409 148,410 - 378,465 U.S. Highways
r rio upe eS Lak
/(2
Bayfield
Douglas
Iron
(/
63
ot es nn Mi
a
Mi c h
Ashland Washburn
/(8
Oneida
(/
94
/(10
Dunn
Pepin
(/53
Clark
(/ T r Buffalo em pe al ea u
Shawano (/10
53
N E S
10
0
10 20 Miles
Source: Northwest RPC.
Menominee
Marathon EauClaire
Forest
Langlade
Taylor Chippewa
(/
Pierce
W
/(45
8
Lincoln
Barron
63
(/12
(/
Rusk
Polk
Vilas
Sawyer Price
St Croix
n
(/51
(/53
Burnett
iga
/(94
LaCrosse
Wood
Portage
Jackson
Waupaca
(/
51
Adams
Monroe Juneau
Waushara
Winneba
An overview of byproduct production distribution is shown at the left. Overall production is concentrated in counties with the largest pulp or OSB mills, but areas with a significant number of sawmills also have significant production. Sawmill concentration is most noticeable in the distribution of coarse residue since this is not a byproduct of pulp mills. Summary maps for the fine, coarse, and bark residues are shown in the appendix along with a larger map of total byproduct output. As shown in figure 6, bark makes up roughly half of the byproduct output and accounts for around 550 thousand dry tons annually in the survey region. Coarse debris, mostly board and log cutoffs, provide 225 thousand dry tons per year while fine residue, or sawdust, makes up another
Marquette Green Lake /(51
260 thousand dry tons annually. Production is also highly concentrated in the largest few producers, all of which are pulp mills except for one OSB mill. This production concentration also explains the output distribution shown on the county-level supply maps; it is a key consideration in project location.
Figure 6 Output by Byproduct
Dry Tons per Year
600,000 500,000 400,000 300,000 200,000 100,000 0 Bark
Coarse
Fine
6
Biomass Byproduct GIS Database
Figure 7: Share of Total Production by Output Class More
5000
10000
40000 35000 30000 25000
15000 20000
Figure 8: Share of Number of Producers by Output Class 10000 15000 20000 25000 30000 5000
35000 40000 More
As illustrated by comparing the two graphs at the left, the top 20 percent of producers account for 88 percent of total output. In addition, of the 216 producers in the survey area, the top 11 accounts for the majority of the output. Future survey work and residue-utilizing projects should carefully examine these major suppliers. However, mills producing less than 5000 dry tons could be primarily computermodeled. This is advisable since, as shown at left, they are numerous but have a low portion of total production. In addition, as discussed previously, their output can be accurately modeled.
Sawmills are also the second largest producers behind the pulp mills. 600,000 Veneer and miscellaneous mills 500,000 produce very little as a 400,000 group mainly because Fine there are far more Coarse 300,000 sawmills. Also, note the Bark 200,000 product distribution differences by mill type; 100,000 this variation explains 0 the stark distribution Sawmill Veneer Pulpmill OSB Misc difference between bark and coarse debris. Most pulp mills are in the southern part of the survey area, so the bark and, to a lesser extent, the sawdust, or fine, residues are concentrated in those counties. Dry Tons per Year
Figure 9: Production by Mill Type
7
Biomass Byproduct GIS Database
This survey collected only very basic information about current use, but it clearly 900 shows that virtually the entire 800 production of residue is now 700 used, overwhelmingly for 600 Fine fuel. Although it was not 500 Coarse examined in this study, these 400 Bark fuel markets are mostly 300 internal-use. Excess 200 100 production is sold, mostly 0 during the summer. Not Used Misc Products Fiber Fuel Products Undelivered bark prices are Products generally under $5 per green ton. More efficient users should be able to procure even the internally used portion for an acceptable price. Coarse residue, however, will have a higher market price due to competing uses for fiber products. Most of this, therefore, will be less cost effective for fuel uses, but generally still preferable to virgin feedstock2. Dry Tons per Year
Figure 10: Byproduct Use
Clearly, additional research is needed to more fully explore these competing markets. This should focus on the largest producers with an additional survey of the current buyers of residue to compare consumption and supply proximity.
8
Biomass Byproduct GIS Database
Conclusions Use of this database for project-specific supply estimates is accurate, especially when considering total production and delivery costs. Available supply, however, will need more careful assessment of competing uses. The region surveyed produces a large amount of bark, sawdust, and coarse forest industry residue. This production can be accurately estimated for most producers. This allows inexpensive updates from published producer-specific employment statistics. Larger sawmills, however, will need further examination and verification to accurately assess production levels. In addition, production is concentrated in pulp and OSB mills with the remainder coming primarily from larger sawmills. Nearly the entire production of residue is currently used, mostly for fuel. Coarse residue has the strongest market with most used for fiber. In using this database to estimate site-specific supply and delivered feedstock cost, the following should be carefully considered for additional research: 1. Further investigation of the long-term feedstock costs, availability, and competing uses 2. Verification of production from sawmills producing more than 5000 dry tons per year 3. Careful examination of the largest producers, a few of which will likely make up the majority of the potential supply for a given project. A careful examination of the residue supply is now just beginning. Further research should complete the picture and further improve these critical cost estimates.
9
Biomass Byproduct GIS Database
References Reading, W, et al, “Timber Product Output Database for 2000”, North Central Forest Experiment Station, St. Paul, MN, 2003 1
Barber, R., “Ladysmith Bioethanol Pre-Feasibility Study”, Internal Report, Northwest Regional Planning Commission, Spooner, WI, 6/1/2003 2
Carlsen, F., “Wisconsin Manufacturers Directory 2000”, Harris Infosource, Twinsburg, OH, 1999 3
Reference USA, www.referenceusa.com, InfoUSA Inc., Omaha, NE, 2003 (only used business initially listed 2000 or earlier) 4
10
Biomass Byproduct GIS Database
Appendix County Byproduct Summary Maps
Annual Dry Tons Wood Byproduct Northwest Wisconsin Total Tons per Year (Bark, Coarse & Fine)
r rio upe S ke La
e Lak
0 - 13,759 13,760 - 48,389 48,390 - 148,409 148,410 - 378,465
or eri Sup
(/ 2
Douglas
U.S. Highways
Bayfield Iron
(/63
hi g a
n
so
ta
Ashland /(5 1
in
ne
(/53
Burnett
M
Mi c
Washburn
Vilas
Sawyer Oneida
Price
(/45
Fo
(/ 8
Rusk (/ 8
Polk
Lincoln
Barron
St Croix (/63 (/12
Langlade
Taylor Chippewa (/5 3
/(9 4
Men
Dunn
Marathon Clark
Pierce (/
10
Pepin
Shaw
Eau Claire (/10
E
u ea al pe
W
Buffalo
em Tr
N
(/53
Wood /(94
Portage
Waupaca
(/
Jackson
51
S
10
0
10 20 Miles
La Crosse
Adams Monroe
Source: Northwest RPC.
Juneau
11
Waushara Marquette Green Lake
W
Biomass Byproduct GIS Database
Annual Dry Tons - Bark Northwest Wisconsin ke La
or eri Sup
U.S. Highways Tons per Year 0 - 11,710 11,710 - 63,590 63,590 - 97,840 97,840 - 256,620
or eri Sup e Lak
/( 2
Bayfield
Douglas
Iron
(/6 3
n
ta
so
(/5 1
ne in
Washburn
Vilas
Sawyer Oneida
Price
/(45
Fore
(/ 8
Rusk (/
h i ga
Ashland (/53
Burnett
M
Mi c
8
Polk
Lincoln
Barron
St Croix (/6 3 (/ 12
/(5 3 (/94
Langlade
Taylor Chippewa
Meno
Dunn
Marathon Clark
Pierce (/
10
Pepin
Shaw
Eau Claire /( 10
E
(/5 3
u ea al pe
W
Buffalo
em Tr
N
Wood /(9 4
Portage
Waupaca
(/
Jackson
51
S
10
0
10 20 Miles
La Crosse
Adams Monroe
Source: Northwest RPC.
Juneau
12
Waushara Marquette Green Lake
W
Biomass Byproduct GIS Database
Annual Dry Tons - Coarse Northwest Wisconsin U.S. Highways Tons per Year 0 - 4,010 ior r 4,010 - 9,960 e up eS k 9,960 - 17,980 a L 17,980 - 28,230
r rio upe S ke La
/( 2
Bayfield
Douglas
Iron
(/63
Mi c
hi ga
n
ta
Ashland (/5 1
ne
so
(/5 3
M
in
Burnett
Washburn
Sawyer Oneida
Price
/(4 5
Fo
(/ 8
Rusk (/
Vilas
8
Polk
Lincoln
Barron
St Croix (/63 (/12
/(53 (/94
Langlade
Taylor Chippewa
Me
Dunn
Marathon Clark
Pierce (/10
Pepin
Sha
Eau Claire /(10
E S
10
0
(/53
u ea al pe
W
Buffalo
em Tr
N
Wood /(94
(/
Jackson
Adams Monroe
Source: Northwest RPC.
Juneau
13
Waupac
51
10 20 Miles La Crosse
Portage
Waushara Marquette Green Lake
W
Biomass Byproduct GIS Database
Annual Dry Tons - Fine Northwest Wisconsin r rio upe S ke La
r rio upe S e Lak
U.S. Highways Tons per Year 0 - 6,470 6,470 - 13,140 13,140 - 25,310 25,310 - 98,570
/( 2
Bayfield
Douglas
Iron
(/63
Mi c
hi ga
n
ta
Ashland (/5 1
ne
so
(/5 3
M
in
Burnett
Washburn
Sawyer Oneida
Price
/(4 5
Fo
(/ 8
Rusk (/
Vilas
8
Polk
Lincoln
Barron
St Croix (/63 (/12
/(53 (/94
Langlade
Taylor Chippewa
Me
Dunn
Marathon Clark
Pierce (/10
Pepin
Sha
Eau Claire /(10
E S
10
0
(/53
u ea al pe
W
Buffalo
em Tr
N
Wood /(94
(/
Jackson
Adams Monroe
Source: Northwest RPC.
Juneau
14
Waupac
51
10 20 Miles La Crosse
Portage
Waushara Marquette Green Lake
W
Biomass Byproduct GIS Database
Byproduct Producer Location Map
Annual Dry Tons - Wood Byproducts by Mill* Northwest Wisconsin *NOTE: Mill Locations are approximations only.
S ke La
ior er p u
#
# #
ke La
# ##
#
(/ 2
Douglas
#
Bayfield# #
ta
r rio
#
#
#
#
Iron
#
o s
e
(/
n
in
##
##
####
#
#
#
#
#
#
#
#
Vilas #
Sawyer
#
#
/(5 1
##
#
Washburn
Burnett
an
#
# Ashland
##
#
hig
#
#
#
Mic
#
(/6 3
#
M
pe Su
#
##
53
Tons per Year (Bark, Coarse & Fine) < 1,000 # 1,000 - 36,000 36,000 - 82,000 # # 82,000 - 163,000 Railroads U.S. Highways
#
# Price
#
#
Oneida #
#
#
###
#
(/ 8
#
Polk
#
Rusk#
#
(/
94
#
#
#
## #
Lincoln
# #
#
#
/(5 3
Taylor #
#
#
#
#
#
#
#
#
#
#
###
#
#
(/10
#
Pepin #
Eau Claire
#
Meno
##
## # Marathon # #
# #
##
#
#
#
#
##
Clark
#
#
#
#
#
#
#
Pierce
Langlade
#
# #
#
Chippewa
Dunn
#
#
#
#
#
#
#
#
#
##
#
St Croix (/63 (/12
(/ 8
#
#
#
Barron
#
#
## #
Fore
##
# #
(/
45
Shawa
#
#
/( 10
# #
E
Buffalo
#
##
(/5 3 #
u ea al pe
W
em Tr
##
N
##
##
#
#
#
#
/(9 4 ##
Jackson
## # ## #
# Wood
Portage
Waupaca
#
(/
51
#
S
10
0
#
10 20 Miles
Waushara #
Source: Northwest RPC.
15
Wi
