Mechanization Potential for Industrial-Scale Fiber and Energy Plantations

Bruce Hartsough, Biological and Agricultural Engineering, University of California, Davis, California

Randall Richter, Simpson Tehama Fiber Farm, Corning, California

ABSTRACT

Current costs for all activities (annual maintenance, harvesting and reestablishment) on an irrigated, short rotation, pulpwood and energy plantation in California were compared with estimated minimum costs for the same activities if high levels of effort were put into mechanization. The plantation, with a total of 4,000 ha, is operating on an eight-year rotation, with 500 ha in each age class. At this scale, move-in costs are insignificant and can be ignored; most of the specialized equipment can be utilized throughout the dry operating season of approximately eight months. Given that the plantation has equal areas in each age class, costs for activities on each age class are incurred each year. The differential cost between current and minimum was derived for each activity, and yearly benefits were calculated. For the California plantation, further mechanization efforts were estimated to provide benefits of up to half a million dollars per year. The majority of this would come from an improved method of delimbing and debarking. Other large gains were projected for a continuous-travel felling machine, lighter chip vans and a better method of handling drip irrigation lines. Minor benefits would accrue from further mechanization of some cultural operations: planting, stump removal and thinning of coppice sprouts.

INTRODUCTION

Several commercial short rotation plantations are now operational in the United States. Many were established within the past six years and have not yet reached the first harvest rotation. Most activities in these plantations - establishment, cultural operations and harvesting - have been or are planned to be carried out with conventional agricultural or forestry equipment. This makes sense for a neophyte industry which is utilizing few pieces of equipment. As the scale of short rotation planting increases, it may be attractive to develop specialized equipment for some of these activities. In this paper, we attempt a first cut at determining where further mechanization efforts have the most potential to improve short rotation activities.

APPROACH

We took the point of view of the owner of a plantation that is operating in a steady-state condition: all the initial establishment has been completed and each age class is represented equally in terms of plantation area. Using "base case" or current costs for each activity, total dollars currently spent on each activity were estimated. Sources of information published over more than a decade were used to derive the cost estimates, but all figures were adjusted to 1993 U.S. dollars. Then we estimated the potential reductions in costs for each activity. Equivalent present worths of infinite series of the yearly reductions can then be used as upper limits on what might be invested in further mechanization or other improvements.

For purposes of illustration, we used Simpson Timber Company's Tehama Fiber Farm, located near Corning, California. The plantation will provide a large part of the hardwood furnish for Simpson Paper Company's pulp mill, and residues will be utilized as fuel by Wheelabrator Shasta Energy Company's freestanding electric power plant. After the last blocks are planted in the next couple of years, the plantation will cover 4,000 ha. Rotation age is expected to be eight years, and yields are assumed to average 20 dry Mg/ha/year. Although planting densities have varied somewhat, 1,500 trees/ha is assumed for this example. The site is dry during the eight warmest months of the year, so drip irrigation is required. We assumed that equipment could travel in the plantation during the eight-month dry period, making it feasible to dedicate equipment for the one plantation. The large scale of the plantation also allowed us to ignore move-in costs. One-way haul distances are 70 km to the pulp mill and energy plant.

CURRENT COSTS

Plantation establishment activities

Before the first planting, Simpson's land was unirrigated pasture. Establishment operations included ripping, installing wells and the underground portion of an irrigation system, disking, installation of above- ground drip lines, and hand-planting of seedlings and/or clones. These activities are more fully described in Jim Rydelius' paper elsewhere in these proceedings. They will be ignored here because these one-time operations are "sunk" and have no bearing on future outcomes for a plantation which is already established.

Annual activities

Information on annual activities was taken from Simpson's experience, some of which was previously reported by Hartsough and Jenkins (1990). Costs of pumping irrigation water constitute the major part of the average annual activity cost (Table 1). Evapotranspiration potential (ET) is 1.07 m/year. Irrigation water is applied at 20 percent of ET during the first year, 30 percent during the second, and 50 percent during the remaining years of the rotation. Nitrogen fertilizer is injected into the drip system and makes up the rest of the cost. Annual application rates average 37 kg N/ha. On Simpson's relatively dry sites, little vegetation control is required after planting. A small amount of pest control is required on the plantation perimeter, but the large plantation area makes this edge application a negligible cost when converted to a per unit area basis.

Harvesting activities

Before harvest, the drip lines will be removed; for convenience in the analysis, drip line removal was lumped with the reestablishment activities.

While one short harvesting trial has been conducted by Simpson, the first operational harvesting is scheduled for 1996, so data from other locations were reviewed to help estimate costs. Sources included Arthur and others (1982), Baughman and others (1990), Desrochers (1993), Favreau (1992a), Hartsough (1992), Hartsough and others (1992) and Stokes and Watson (1989). A tricycle feller/buncher and rubber-tired skidder will supply trees to a flail delimber/debarker/chipper at roadside. Delimber/debarker residues will be comminuted with a tub grinder and the processed materials -pulp chips and residue fuel -will be delivered to their respective destinations in chip vans. We assumed that each stump-to-truck activity would handle or process the material from 225 trees in each productive machine hour (PMH), machine utilization was 67 percent for all equipment, and that harvesting labor costs were $15 per scheduled hour including 50 percent loading. Machine cost estimates were based on Brinker and others (1989) and other purchase price estimates. Costs for chipping were calculated separately from those for delimbing/debarking for illustration purposes, even though both activities are to be conducted by one machine.

The resulting costs are shown in Table 2. When adjusted to costs per unit of pulp chips, the estimated stump-to-truck portions are two-thirds of those experienced by James River Corporation at their Lower Columbia River Fiber Farm (Hartsough and others 1992). These estimates appear reasonable as the average tree weight at Simpson is expected to be 107 kg dry versus the 54 kg observed during the James River study.

Reestablishment activities

Simpson is debating whether to remove stumps and replant at the end of each rotation or to rely on coppice regeneration. We assumed a fifty-fifty mix of both methods. Prior to replanting (by hand), stumps will be removed by plowing, the area will be disked, and drip lines will be reinstalled. With coppice regeneration, drip lines will be reinstalled after harvesting. In order to reduce bark percentage in the second rotation, sprouts will be thinned in two stages: with loppers at six months, and by chainsaw when the largest sprouts reach three to four cm in diameter.

Disking, drip line reinstallation and replanting costs were based on extensive experience with initial establishment at Simpson's plantation. Planting costs include $150/ha for seedlings and $82/ha for labor. Costs for removing drip lines and thinning sprouts were estimated from limited trials conducted by Simpson. Costs to remove and reinstall drip lines are estimated at $5 per 100 m, with approximately 3,300 m of lines per ha. Thinning costs include $270/ha for the lopper entry and $300/ha for the chainsaw operation. The cost for stump removal is an educated guess.

Total yearly costs

For a steady-state plantation, the annual activities are carried out on each ha in each year. One-eighth of the area is harvested each year, removing 160 dry Mg per ha; replanting and coppicing are each conducted on one-half of the harvested area. Total yearly costs for each activity for the 4,000-ha plantation are displayed in Figure 1. Although irrigation is the single most expensive activity, combining the activities into the three major groups shows that harvesting is the most costly category (Figure 2). The sum of all costs is approximately 3.3 million dollars per year.

Figure 1. Current yearly costs for a 4,000 ha plantation, by activity.

Figure 2. Current yearly costs for a 4,000 ha plantation, by category.

POTENTIAL FOR IMPROVEMENT

We suspect that there is little further mechanization potential for the annual activities: irrigation, fertilization, pest control and vegetation control. These have evolved over several decades in agricultural operations, and the short rotation forestry conditions do not differ markedly from those on conventional agricultural lands.

In contrast, harvesting equipment has been developed for forest conditions: rough and broken terrain with obstacles such as rocks, large stumps and down logs, and generally for coniferous trees that are larger and less-uniform in size than those produced in short rotation plantations. This indicates possibilities for improvements in harvesting equipment.

It may appear that an improvement in one stump-to-truck activity might have no benefit, because of an imbalance between equipment production rates. However, three or more sets of equipment will be required to harvest 500 ha per year over an eight-month period, and all will be working in close proximity. Therefore the integer balancing problem is less of a concern here than in most conventional forestry operations.

In felling, reduction in cost is possible by developing a continuous-travel feller/buncher, similar to those proposed by Golob (1986), prototyped by Hyd-Mech for the Bioenergy Program of the National Research Council of Canada (Curtin and Barnett 1986), and tested by Stokes and others (1986). Effective derivatives of the Hyd-Mech FB-7 or FB-12 would eliminate the stop-and-go, forward-and-back travel pattern inherent to conventional feller/bunchers. Although limited studies show that feller/bunchers can be highly productive in short rotation plantations (Stokes and McDonald 1993), it is difficult to imagine a conventional machine competing with a continuous-travel machine over the long term. Impressive results with continuous- travel machines for harvesting willow in Sweden support this concept (Culshaw 1993) harvesting productivity is higher than for traditional forestry operations even though both the willow volume per ha and average stem size are much less than in traditional plantation clearcuts. Based on results for the FB-7 reported by Stokes and others (1986) and Stoke's unpublished data on the FB-12, we estimated that current felling and bunching costs might be reduced by 40 percent. This assumes a purchase price of $300,000 for the machine, and an average travel speed of 2 km per hour.

Skidders were designed for rugged terrain, but potential reductions in skidding costs via modifications of skidders seem minimal. Existing skidders can be fully utilized by carrying relatively large turns of small trees on the flat agricultural sites.

Flail delimbing and debarking is widely considered the bottleneck in operations where clean chips are being produced (e.g. Thompson and Jackson 1991). Many studies indicate that either chipping production rates must be reduced to obtain low bark contents, or high bark levels are accepted in order to maintain higher production rates. Bark contents between one and six percent have been reported (Favreau 1992b, Franklin 1992, Sauder 1989, Stokes and Watson 1989, Thompson and Jackson 1991); lower production rates will have to be accepted to meet tolerances of one percent, as specified by some pulp mills.

Chain costs represent the largest single component of flail delimbing/debarking costs. Stokes and Watson (1989) estimated chain costs to be $0.9 to $1.8 per dry Mg of conifer chips, or 20 to 28 percent of total flailing costs, assuming 25 PMH per set of chains. Empirical studies have shown these early estimates to be too optimistic. For conifer chips, Carte (1991) and Sauder (1989) pegged costs at $2.0 to $2.6 per dry Mg. Reported chain costs for hardwood have been higher: $3.8 (Sauder 1989), $4.6 (Hartsough and others 1992) and $6 per dry Mg of chips (Kaiser 1994). This may be due to inherent differences between bark of conifer and hardwoods, differences that might be exploited to design a more efficient debarking method for hardwoods.

Compared with processes such as chipping, flail delimbing/debarking is inefficient from an energy standpoint. Chipping requires on the order of 10 to 20 MJ per dry Mg of wood (Rodgers 1948). Based on fuel consumption figures reported by Stokes and Watson (1989), flails use 40 to 70 MJ per dry Mg of residues. An additional 40 MJ per dry Mg is necessary to comminute the residues with a tub grinder (Arthur and others 1982).

Moving incrementally from the chain flail, delimbing/debarking for short rotation hardwoods can be improved by 1) increasing production rate so the delimber/debarker does not limit chipper and tub grinder production and 2) reducing the chain cost and 3) using a more-efficient concept. The first is easy: build a bigger, more powerful unit, possibly with longer drums or a third drum. Examination of the costs of forestry equipment within a single class, such as skidders, shows that the purchase cost per horsepower declines as machine power increases. Assuming flail throughput rate would be proportional to power, the delimbing/debarking cost per unit material would decline if power was increased. Finding a concept that is more efficient and has lower cost than a chain flail is more problematic. While a potentially-successful concept is not immediately apparent, we assumed that chain costs could be halved, and improved efficiency would reduce other costs by one-sixth. The combined maximum potential reduction from current levels would then be approximately 45 percent.

Chipping is a mature science; we assumed no improvement in chipping capability. Chipping costs and residue grinding costs, however, may be reduced by an increase in delimbing/debarking rate, since the bark content limit constrains the production of existing equipment. We have assumed a potential reduction in chipping and grinding costs of one-sixth.

Hauling is also mature, and yet recent reductions in log trailer weights (Stuart 1993) may indicate a potential for similar reductions in chip van weights. We assume a potential increase in net load weights, to 13 dry Mg from the current 12 Mg. Loading times would also be reduced by the increase in delimbing/debarking, chipping and grinding rates, and unloading times might also be reduced, possibly by up to five minutes per load. In sum, these two factors might reduce hauling costs by up to 14 percent, but hauling reductions are considered less-likely than others.

We have been fairly conservative in our choices of harvesting systems to evaluate. Some methods not considered here that may have benefit include:

• a feller/chipper, with separation of pulp and fuel components carried out at the pulp mill,
• chipping whole trees at roadside, with separation at the mill,
• a feller/loader for whole trees; trees would then be processed at the mill in a drum debarker.

Data on use of a large reel system to install and retrieve drip lines on agricultural lands (Coates 1985) was used to estimate costs for an improved system for handling drip lines. We assumed a crew of four, working with a tractor, could reel in lines at a rate of two ha/PMH; the same crew could reinstall lines on 3.5 ha/PMH. Utilization was assumed to be 80 percent and labor costs were assumed to be $10 per SMH, including loading. This system would reduce current costs by about half.

We assumed that stump removal costs, uncertain at present, might be reduced by about 25 percent via further development of equipment.

Based on data from other areas (Christopherson 1989, Culshaw 1993), we estimate that existing continuous-furrow mechanical tree planters could cover up to one ha an hour, resulting in costs in the range of $50 to $80/ha. Intermittent-furrow planters are more expensive, on the order of $90/ha (Miles and others 1985). Hand planting is used by Simpson to insure seedlings are placed in the zones wetted by the drip emitters. It is feasible to design equipment to sense wet spots; assuming this feature could be added to a continuous planter at negligible extra cost, the maximum potential reduction from the current cost is about $30 per ha, or 40 percent, excluding seedling costs which are assumed to remain constant.

An improved motor-manual scheme might reduce the costs of thinning coppice sprouts; we have assumed a potential reduction of one-eighth. The major challenge here is preventing damage to the crop sprouts while thinning.

The potential yearly cost reductions by activity for the 4,000 ha plantation are displayed in Figure 3. Potential cost reductions for a 4,000 ha plantation, by activity.

Because all of the gains in chipping and grinding are due to improvemen ts in delimbing/de barking, the three have been combined in Figure 4, Potential cost reductions for a 4,000 ha plantation, by source of improvement

which shows reductions by source of improvement. Of a total potential reduction of half a million dollars per year, over half is due to improvements in delimbing/debarking. Felling and hauling developments each account for over ten percent of the reductions, and improved drip line handling for almost ten percent. Relatively small gains can be made in the three cultural operations: thinning, planting and stump removal. This indicates that development work should primarily be focused on the two harvesting activities: delimbing/debarking and felling.

How much should be spent on developing improved equipment and methods? No more than the expected present worth of the benefits, which is the yearly benefit times the probability of success of the development work, divided by the risk-free, inflation-free discount rate:

CONCLUSIONS

For the example 4,000 ha pulp/fiber plantation, potential mechanization benefits total approximately half a million dollars per year. The majority is due to delimbing/debarking improvements. Other substantial amounts are available in felling, hauling and drip line handling. Minor amounts can be gained in cultural operations.

REFERENCES

1. Arthur, J.F., R.A. Kepner, J.B. Dobie, G.E. Miller and P.S. Parsons. 1982. Tub grinder performance with crop and forest residues. Transactions of the ASAE 25(6):1488-1494.
2. Baughman, R.K., B.J. Stokes and W.F. Watson. 1990. Utilizing residues from in-woods flail processing. In: Stokes, B.J. (ed.) Harvesting small trees and forest residues. IEA/BA Task VI Activity 3 Workshop, Copenhagen, Denmark, May 28. USDA Forest Service Southern Forest Experiment Station, Auburn, Alabama. pp.21-30.
3. Brinker, R.W., D. Miller, B.J. Stokes and B.L. Lanford. 1989. Machine rates for selected forest harvesting machines. Circular 296, Alabama Agricultural Experiment Station, Auburn University, Alabama.
4. Carte, I.C. 1991. Maintaining chain flail delimber-debarker chain. Technical Release 91-R-56, American Pulpwood Association, Washington, DC.
5. Christopherson, N.S. 1989. Mechanization of fast-growing forests: worldwide progress. Paper 89-7604. ASAE International Winter Meeting, New Orleans, Louisiana, December 12-15.
6. Coates, W. 1985. Comparison of equipment for the installation and retrieval of drip irrigation laterals. Transactions of the ASAE 28(4):1173-1177.
7. Culshaw, D. 1993. Study tour report. In: Culshaw, D. (ed.) Status of short rotation forestry mechanization worldwide. IEA/BA Task IX Activity 1 Workshop and Study Tour, Sweden - March 2-4. ETSU, Harwell Laboratory, Oxford, UK. pp.62-75.
8. Curtin, D.T. and P.E. Barnett. 1986. Development of forest harvesting technology: application in short rotation intensive culture (SRIC) woody biomass. Technical Note B58, Division of Land and Economic Resources, Tennessee Valley Authority, Knoxville.
9. Desrochers, L. 1993. Recovery of residues from delimbing and delimbing-debarking using tub grinders. Technical Note TN-209, FERIC, Pointe Claire, Quebec.
10. Favreau, J. 1992a. In-woods chipping: a comparative cost analysis. Technical Report TR-105, FERIC, Pointe Claire, Quebec.
11. Favreau, J. 1992b. Peterson-Pacific DDC 5000 delimber-debarker-chipper: new observations. Field Note No. Processing-29, FERIC, Pointe Claire, Quebec.
12. Franklin, G.S. 1992. Model 23 flail chiparvestor delimber-debarker-chipper: productivity and chip quality in hardwood. Technical Note TN-187, FERIC, Pointe Claire, Quebec.
13. Golob, T.B. 1986. Analysis of short rotation forestry operations. NRCC No. 26014, National Research Council of Canada, Ottawa, Ontario.
14. Hartsough, B. R. 1992. Product/harvesting options for agroforestry plantations in the San Joaquin Valley, California. Transactions of the ASAE 35(6):1987-1993.
15. Hartsough, B.R. and B.M. Jenkins. 1990. Energy and cost accounting for large-scale eucalyptus plantations. California Committee in the Relation of Energy to Agriculture Report No. 89-04. University of California, Davis.
16. Hartsough, B. R., B. J. Stokes, and C. Kaiser. 1992. Short-rotation poplar: a harvesting trial. Forest Products Journal 42(10):59-64.
17. Kaiser, C. 1994. Personal communication. Superintendent, James River Lower Columbia River Fiber Farm, Clatskanie, Oregon.
18. Miles, J.A., J.E. Burk and R.J. Merala. 1985. Engineering field evaluation of intermittent tree planters. Transactions of the ASAE 28(4):1077-1081.
19. Rodgers, H.W. 1948. The wood chipper. Paper Industry 30(6):883-888.
20. Sauder, E.A. 1989. Satellite chipping frozen small-diameter timber using a chain flail delimber/debarker. Paper 89-7594. ASAE International Winter Meeting, New Orleans, Louisiana, December 12-15.
21. Stokes, B. and T. McDonald. 1993. Status of short rotation forestry in the USA. In: Culshaw, D. (ed.) Status of short rotation forestry mechanization worldwide. IEA/BA Task IX Activity 1 Workshop and Study Tour, Sweden - March 2-4. ETSU, Harwell Laboratory, Oxford, UK. pp. 21-44.
22. Stokes, B.J. and W.F. Watson. 1989. Field evaluation of in-woods flails in the southern United States. In: Hudson, J.B. and A. Twaddle (eds.) Integrated Harvesting Systems Workshop. IEA/BA Task VI Activity 2 Meeting, New Orleans, Louisiana, May 30-31. Aberdeen University Department of Forestry, Aberdeen, Scotland. pp. 99-111.
23. Stokes, B.J., D.J. Frederick and D.T. Curtin. 1986. Field trials of a short-rotation biomass feller buncher and selected harvesting systems. Biomass 11:185-204.
24. Stuart, W.B. 1993. (Oral presentation on advances in logging trailer design.) Presented at ASAE International Winter Meeting, Chicago, Illinois, December 14-17.
25. Thompson, M.A. and L. Jackson. 1991. Peterson stationary chain flail evaluation. Technical Release 91-R- 87, American Pulpwood Association, Washington, DC.

File posted on March 5, 1996; Date Modified: February 21, 1999