Printer Friendly

Wood chip mill harvest volume and area impacts in North Carolina. (Management).

ABSTRACT

We estimated the impacts of wood chip mills on the volume and area required for timber harvests in North Carolina. Results indicated that new wood chip mills would add to timber processing capacity and increase the area of forest harvested. Wood chip mills were estimated to add about 1,000 acres per year of additional harvest area per 100,000 tons per year of chip mill on average, and be associated with integrated product harvests of a larger area of about 1,700 acres per year per 100,000 tons of chip mill capacity. New wood chip mills would not individually make harvest levels exceed growth levels in the areas within 50 miles of the mills, although some were located in areas where harvests already exceeded growth. The cumulative effects of overlapping chip and other wood mill procurement zones were analyzed with state-level timber projections, which indicated that projected increases in timber harvests and decreases in forest land area would cause softwood and hardwood timber inventories in the state to decl ine.

**********

The contribution of wood chip mills to timber harvests has been an increasingly contentious issue in the South for several years. Environmentalists have called for bans on permits of wood chip plants, citing fears about adverse effects on local communities and on the forest environment. Firms have shifted existing production and added new capacity via wood chip mills, which provide efficient roundwood processing capacity. This paper summarizes some of the findings of a study we conducted in North Carolina regarding potential impacts of wood chip mills on timber harvesting areas. This study was part of a statewide study of the economic and ecological impacts of wood chip mills.

North Carolina forest area, timber volumes, and timber growth and removals increased from 1938 to 1990. Softwood harvests were approximately equal to growth as of 1990; hardwood growth still exceeded removals by about 30 percent in 1990. However, in 1990 more counties had removals that exceeded growth than in 1983, for both softwoods and hardwoods (Cubbage et al. 2000). These increases in timber harvests and concerns about further harvest increases prompted this analysis of wood chip mill impacts.

Schaberg et al. (2000) determined that wood chip processing comprised an increasing amount of the total hardwood pulpwood production ranging from the eastern to western parts of the state. In the two Coastal Plain forest survey units, independent wood chip mills processed from 15 to 16 percent of the total pulpwood production in 1997. In the Piedmont, wood chip production was 49 percent of pulpwood production; in the Mountains, it was 75 percent. Statewide, wood chip mills accounted for 27 percent of pulpwood production. Pulpwood production was only about 40 percent of the state's total timber harvest; the balance was for sawtimber, veneer logs, and other products. Thus, wood chip mills would account for about 12 percent of the total timber production in the state. The 18 wood chip mills in North Carolina used about 88 percent hardwood timber and 12 percent softwood. Schaberg et al. (2000) found statistically significant correlations between increased wood chip mill capacity and timber removal volume in the s tate. Thus, it appears that additional wood volume is being supplied by wood chip mills; they are not just replacing existing concentration yards (Schaberg et al. 2000). Therefore it is appropriate to determine the potential volume and area impacts of wood chip mills.

METHODS

This analysis of the potential impacts of wood chip harvests examined whether wood chip mills would have a measurable impact on increasing the amount of wood volume and the number of acres required to be harvested in the state. The null hypothesis was that wood chip mills had no additional effect on the volume or timber area harvested. This was examined by calculating the number of acres needed to be harvested given the wood chip mill processing capacity and the residual wood volumes from historical timber harvests. These were compared with wood chip mill capacity in the state by forest survey unit.

A range of wood chip mill harvest area impacts was estimated based on the average harvest volumes per acre and the capacity of wood chip mills, and by the share of wood chip mill harvests of total timber product harvests in the state. Cumulative effects of wood chip harvests over time and effects of added harvests on growth and removals also were computed.

To analyze whether wood chip mills would affect forest harvest area, we needed data on: 1) the varying size of wood chip mills and the species mix that they used; 2) estimates of the harvest volumes by product per acre; 3) estimates of the working circles around wood chip mills; and 4) estimates of the residual wood volume that could be harvested on chipped versus non-chipped sites.

If there were enough residual material on existing sites, no new harvest area would be necessary to supply a new wood chip mill. On the other hand, if only new sites were chipped, this would represent the maximum total incremental area that could be attributed to wood chip mills. The area could be greater if wood chips were only a component of other new harvests, but the area attributable just to wood chips would not vary. Intermediate areal impacts would occur to the extent that some proportion of wood chip mill harvest volumes was taken from existing harvests, if this were true, the harvest areas could be reduced in total area, but intensified on any given site, which would result in different environmental impacts.

We analyzed these potential area impacts assuming various levels of clear-cut or partial harvests where the timber would be harvested and processed as wood chips. The amount of timberland needed for harvest was analyzed 1) as a general or model area for different volumes per acre and varying species group mixtures; and 2) in specific locations around 18 identified mills in North Carolina that chip wood and transport it to other locations.

The potential harvest areas were calculated for wood fiber type and for the potential and reported consumption of the chip mills. Potential harvest areas attributed to wood chip mills could come from three sources: 1) the area attributable to increased harvesting intensity on existing harvest sites; 2) the potential amount of incremental harvests that could be attributed totally to new acres; or 3) the area required given partial harvests of new acres of mixed wood chip and sawtimber harvests.

Wood chips probably seldom comprise the brunt of the timber volume removed on any tract; high value products will be sorted out and trucked separately to sawmills or other processing facilities. Thus, our analysis varied the amount of wood volume that would be removed in a timber harvest in order to estimate the different incremental areas of land that would be needed to provide the given wood furnish. Potential harvest volumes were varied from 500 to 6,000 [ft..sup.3]/acre. The average volume harvested in North Carolina in 1990 was 2,350 [ft..sup.3]/acre. These varied assumptions allowed us to approximate the bounds of the areal impact of wood chip harvests in a given location.

POTENTIAL RESIDUAL HARVEST VOLUMES AND WOOD CHIP MILL CAPACITY

We calculated the potential residual stand harvest volume available from existing harvest sites, which might be used for wood chips. This was computed based on the number of final harvest acres by survey unit in 1990 for each survey unit, multiplied by the potential wood chip volume per site.

The wood chip volume was computed based on field surveys performed as part of a woody debris study (Hess and Zimmerman 2001). This residual volume estimate was then compared with 1999 wood chip mill capacities in order to estimate whether the residual volumes that could be harvested from sites would be adequate to supply all wood chip mill furnish requirements. The mean residual volume per acre for harvested sites without a wood chip component was 1,288 [ft..sup.3]acre; with a wood chip component it was 834 [ft..sup.3]/acre. The difference was about 450 [ft..sup.3]/acre, or about 18 tons/acre at 25 [ft..sup.3] per ton.

Conversions were made using Timber Mart South (1999) factors as shown in Table 1.

Table 2 shows the calculations by forest survey unit to assess if existing harvest sites could provide adequate material for wood chips, based on 1990 Forest Inventory and Analysis (FIA) data timber removal levels (Johnson 1991a, 1991b, 1990) compared to potential increased harvest levels with wood chip mills. There appears to be more than enough harvest area times the residual volume in the Coastal Plain to provide all the wood chip mill harvest. However, harvesting all the residual volume in the Piedmont and the Mountains would not provide enough volume to feed all the chip mill capacity as of 1997.

Many other factors suggest that some new harvest areas would be needed to furnish wood chip mill harvests. First, wood chip mills comprise about 12 percent of the total timber harvest in the state, so might be expected to require some new harvest area. The added returns from wood chip harvests would make some timber harvest sites profitable that had not been so without chip markets, and this would encourage more harvest area. Our field visits and efforts to sample harvest sites found more sites that did not have wood chip components than those that did. Conversations with foresters and casual observations of harvests in the state indicate that many sites do not have wood chip components. These observations reinforce the premise that all the wood volume needed for chip mills is not being provided from improved utilization on existing sites. A related study of nonindustrial private forest owners showed that economic and market factors would prompt more harvests (Snider and Cubbage 2000). Cointegration statistic al tests indicated that North Carolina timber markets were well integrated; wood chip markets contribute to this status (Aruna and Cubbage 2000).

We also compared these site-specific data with the statewide data available from the 1990 FIA analysis (Brown 1993). These data show that on average 428 million [ft..sup.3] of hardwood growing stock volume was harvested annually from 184,000 acres during the period from 1984 to 1989. This is an average hardwood growing stock volume of 2,320 [ft..sup.3]/acre. About 34 percent found its way into higher value products, and about 31 percent was processed as pulpwood. Roughly 21 percent of total hardwood harvest, about 490 [ft..sup.3]/acre, was not used for products, but was classified as residual growing stock volume or other removals. Acknowledging that some removals represent reserved lands, but knowing also that chip mills can process material too small to be classified as growing stock, our estimate of 450 [ft..sup.3]/acre of incremental volume seems to conform well with historical data.

As Table 2 indicates, there appears to be more than enough harvest area times the residual volume in the Coastal Plain to provide all the wood chip mill harvest. However, removing pine acres from consideration reduces the residual volumes by almost 2 million tons and implies that Coastal chip mill production, which is principally hardwood, would need to utilize virtually all available hardwood acres to meet volume requirements without increasing harvested acres. Harvesting all the residual volume in the Piedmont and the Mountains would not provide enough volume to feed all the chip mill capacity as of 1997. Conclusions about acre effects must be tempered by the observation that some chip mill volume substitutes for volumes previously processed through concentration yards and chipped at pulp mills. The most pertinent factor to consider here would be reductions of wood chipping at pulp mills. While this has certainly occurred, particularly in western North Carolina, these effects were not quantified within the present study. Taken into account, they would reduce our estimates of wood volumes required.

GENERALIZED WOOD CHIP MILL REQUIREMENTS

Chip mills can vary widely in capacity and in the species group mixture consumed. The capacity of a mill is reported in tons per year. The algorithm from Timber Mart-South was employed to convert this tonnage to cubic feet. This harvest volume was calculated under the assumption of total consumption of hardwood and total consumption of softwood for each mill, or mixes of both species.

Using this framework and a given capacity, we calculated generalized tables for the average acres required to annually supply a theoretical wood chip mill for varying volumes per acre and species group mixtures. The volume per acre was increased at a rate of 500 [ft..sup.3]/acre beginning at 500 [ft..sup.3]/acre and ending at 6,000 [ft..sup.3]/acre. The percentage mixture was increased by 10 percent beginning at 10 percent and ending at 100 percent hardwood. This produces a format for considering a chip mill of a given capacity in a general location.

These generalized calculations and tables also could be used to assess the total harvest area impact for harvest where only a portion of the removals were sent to a wood chip facility, or indeed any other wood processing or manufacturing facility. As the base case, we assumed that all the timber harvested at a site was sent to a wood chip mill, not divided among other mills.

EXISTING WOOD CHIP MILLS

We assumed that the identified wood chip mills would procure wood within a 50-mile radius around the center of the mill, which is well within the trucking distance of most logging operations. A smaller radius might prove difficult for procuring enough wood. The procurement circle might be slightly larger or smaller, and certainly would be irregularly shaped. But the 50-mile span provides a reasonable approximation of a mill's primary impact zone. Some of these assumed procurement zones fall outside of the state's boundaries. On one hand, this would indicate that our impact estimates might overestimate removal impacts on inventories if more wood were imported from outside the state. However, our Timber Products Output (TPO) study (Schaberg et al. 2000) indicated that North Carolina is actually a large net exporter of pulpwood, with about 30 percent of the pulpwood harvest in the state actually being exported to another state (25%) or another country (5%). Thus our 50-mile chip mill working circles actually may underestimate timber removal impacts due to more volume exported from the state than imported to the state.

The USDA Forest Service Forest Inventory Analysis (FIA) data published in 1990 (Brown 1993) were used to evaluate the availability of wood fiber around existing mills. These data were accessed via the Internet using the USDA Forest Service's Forest Inventory and Analysis Web-Based Data Retrieval System. The latitude and longitude of each of the existing wood chip mills in the state were identified and then a 50-mile radius circle was established around those center coordinates. Plot-level FIA data were summarized for growing stock volume, area of timberland, and growth and removals for each species group within the 50-mile radius of each chip mill analyzed. Figure 1 shows the location of 18 wood chip mills in North Carolina that chipped and transported wood off site, which were used for this study. It also depicts the 50-mile radius of "woodshed" around each of those mills.

According to the FIA, the area of commercial timberland and the number of forested plots varied with the location of the chip mills. The volume of growing stock of each species group and the diversity of the species group mixture were also unique to each location. Therefore, the acres required to supply a chip mill, given the processing capacity and timber species consumed, will depend on where the mill is located, as well as the total timber harvest and wood chip harvest proportion on any given site.

This framework was then applied to the 18 identified wood chip mills in the state. Each chip mill's reported capacity was converted to volume in cubic feet and the respective locations were evaluated for volume per acre and for species group mixtures that would be available. By combining these two characteristics, one can calculate the harvest volume (drain) of timber from the surrounding forests. One can also use the varying volume per acre tables to examine the effect of harvests that included a partial wood chip component. This amount of incremental drain was added to the existing removals in the area for an evaluation of the impact on the growth and removal ratios.

GROWTH AND DRAIN RATIOS

Of the 18 wood chip mills evaluated, 9 were established prior to the collection of the 1990 FIA data for North Carolina and 9 were established after. Thus, the growth and removal ratios published by the USDA Forest Service in 1990 (Brown 1993) included the harvest of the chip mills operating in 1989 or before.

The impact of the nine new mills opened since 1990 depends on the harvest activity prior to the establishment of the chip mill and the added harvest attributable to the new mills. If the growth and removal ratio were close to 1.0, then the additional harvest would push removals above growth. If the removals were already larger than growth, then the addition of the chip mill would expand the imbalance, and if removals were smaller, then the addition could potentially balance the ratio. We also examined the potential effect of overlapping wood chip mills, and in a separate study, projected timber supplies for North Carolina, which account for the effect of all overlapping wood chip mills and other timber harvests in aggregate (Abt et al. 2000b).

RESULTS

ANNUAL HARVEST AREA IMPACTS FOR TYPICAL MILLS

One can calculate various metrics to estimate how much timber volume per acre would be likely to be harvested by a wood chip mill, and then convert that to the number of acres required to feed a mill. The average timber inventory volumes per acre and timber harvest volumes per acre for the state of North Carolina are shown in Table 3.

We assumed that a wood chip mill would harvest timber at average volumes of at least as much as the state average harvest levels in 1990, i.e., about 2,320 [ft..sup.3]/acre. In fact, since it is commonly stated that a wood chip mill would harvest timber at higher intensities than occurred before chip mills, the 1990 average harvest levels per acre would be an effective minimum level of harvest that would occur, and indeed the volume per acre probably would be larger.

If we assumed a slightly greater volume equal to 2,500 [ft..sup.3]/acre, then a chip mill with a 250,000 ton capacity per year, harvesting 100 percent hardwood, would require 2,500 acres per year of incremental harvest areas. Similarly, a mill with a 400,000 ton capacity per year would require 4,000 acres per year of additional hardwood lands. Pine mills would require about the same area; the average volume per acre of pine lands is slightly greater than hardwood lands, but they require more area to equal the same amount of weight per equivalent amount of wood volume.

As part of the general case, three wood chip mill capacities were analyzed: 150,000, 250,000, and 350,000 tons per year (Tables 4 through 6). Each mill was then placed in locations of varying volumes and species group mixtures resulting in a range of average harvest acres per year required from 625 to 8,310, 1,042 to 13,850, and 1,458 to 19,390 acres, respectively. Obviously, the larger the mill, the greater the harvest area required to provide its furnish each year with a given timber harvest volume per acre. Similarly, the greater the harvest volume removed per acre, the smaller the harvest area that would be affected.

ANNUAL HARVEST AREA IMPACTS FOR EXISTING MILLS

For existing wood chip mills, each was analyzed individually for capacity, volume per acre, species group mixture of volume, and species group mixture consumed (Table 7). This resulted in a range of average harvest acres per year required from 340 to 6,768 for all hardwood consumption, 248 to 6,022 for all softwood consumption, and 340 to 6,333 for the current chip mill furnish mixtures consumed.

The smallest area of impact was for a 25 ton per year existing mill, so probably was not relevant. In fact, most modern mills are probably at the higher production levels of the estimates, so could affect harvest areas of up to 4,000 to 5,000 acres per year if they harvested only average volume stands in their region. In reality, they probably harvest stands that have 50 to 100 percent more than the regional average, i.e., mature stands. Thus, the most likely "typical" harvest area affected by new high-capacity chip mills that chip 400,000 tons or so per year would be closer to 3,000 acres per year. Existing lower-capacity mills, which probably chip about 200,000 tons per year, might require clearcut harvests of about 1,500 acres per year given the average wood volumes in a 50-mile circle around the mills. Partial harvests would have larger area impacts, perhaps two or three times as much land area, whether they were selective cuts taking only wood chip material or clearcuts that merchandised out some compone nt of wood chips. However, the incremental area attributed to wood chip harvesting exclusively would be fairly constant.

Therefore, if all chip mills in North Carolina were supplied entirely from dedicated wood chip harvest sites and no substitution effects occurred, the most likely estimate of the number of acres harvested would be 41,000 acres of hardwood harvest to meet chip mill capacity of 4.1 million tons in 1997.

Projecting these estimates into the future, with the increased capacity of chip mills added during the 1990s fully operational, leads to significantly higher area impacts than before chip mills were opened. Our TPO analysis showed that pulpwood harvest levels in North Carolina were substantially higher in the 1990s than in the prior decade. Consistent with this historical data, if we assume that substitution effects are captured by changes prior to 1990, and chip mill expansion after 1990 represents new processing capacity, the area required to support the cumulative gain of the 2.4 million tons of chip mill capacity that has come on line in North Carolina since 1990 would be 24,000 acres chipped per year.

Multi-product harvests would affect more area, although not more due to chip mills alone. Applying the historical average of roughly 1,000 [ft.sup.3]/acre utilized as higher value TPO or fuelwood (Brown 1993), that would leave about 1,500 [ft.sup.3] of pulpwood furnish per acre, and imply an annual harvest impact of 40,000 acres harvested, allowing for 16,000 acres utilized for higher value products and 24,000 acres as pulpwood. Extending these utilization ratios to new chip mill capacity would suggest that each 100,000 tons of additional capacity would result in about 1,700 acres of annual harvest, or that a high capacity 400,000 ton chip mill would induce about 6,700 acres of additional harvest per year, for all products harvested.

CUMULATIVE EFFECTS ON AREA HARVESTED AND ON GROWTH AND REMOVAL RATIOS

The cumulative effects of these annual wood chip harvests would obviously be many times greater than the effects for 1 year. Per the definitions of seral stages used in other components of our chip mill study, one could state that in the first (forbs and shrubs, or seedlings and saplings) seral stage (ages 1 to 10), harvest volumes to provide a chip mill would leave 10 times as much area in the 0 to 10 age class, or as much as 20,000 to 40,000 acres per year. A similar amount would be left in the second youngest seral stage of 11 to 20 years old. For example, a 400,000 ton per year mill would result in about 4,000 acres per year being harvested, and about 40,000 acres in the earliest seral stage of age 1 through 10. The greater harvest intensity per acre would alter residual habitat cover and expose more soil, affecting soil erosion rates and wildlife.

Since hardwoods comprised 88 percent of wood chip mill furnish, the timber harvests would tend to convert increasing areas of North Carolina's hardwood forests into increasingly younger stands. A 400,000 ton chip mill would shift an additional 80,000 to 134,000 acres of hardwood into stands 20 years and younger during its first two decades of operation. Similarly, supplying the 2.4 million ton capacity of the chip mills that have become operational since 1990 could shift an additional 480,000 to 800,000 acres of hardwood into the 20 year or younger age class over a 20 year period. This represents an area of roughly 4 to 6.5 percent of North Carolina's total 1990 hardwood timberland.

The added average volume of annual harvests processed by wood chip mills affects the growth and removal ratios in the assumed 50-mile working circles. These ratios were computed before and after 1989 (Table 8), which was the last year of data for the 1990 FIA report. The wood chip mills established after then have a potential effect on future net annual growth and removals.

Assuming 100 percent hardwood consumption, the total statewide effect of all 18 chip mills harvesting 100 percent hardwood maintains 222 million [ft.sup.3] of annual surplus growth over removals within their assumed 50-mile radius procurement area. If the mills were to utilize only softwood, the net annual growth surplus in the 50-mile radius around mills would be only 30 million [ft.sup.3]

Thus, overall, based on the 1990 FIA data for the sum of all nine wood chip mills opened before 1990, there was more annual timber growth than annual timber removals in the 50-mile radius around each mill. However, of the nine new wood chip mills opened since 1990, four already had either annual softwood or hardwood removal rates (or both) that exceeded annual growth rates in 1990, and these differences would increase after the mill opened. Only one of the mills opened before 1990 was located in an area where annual removals exceeded growth. Thus, new wood chip mills did not cause annual removals in any area to exceed growth, but five of them would exacerbate this difference that already existed.

In practice, wood chip mills have generally located in areas where excess wood supplies are the most plentiful, avoiding those areas where wood supplies were scarcest. Only 6 of 18 chip mills in the state were located in areas where annual removals exceeded annual growth. Thus, while it appears that wood chip mills are greatly increasing the timber harvest around the mill, this is usually in comparison to a relatively low base of removals to begin with. They actually enable forest industry to procure wood in new markets, perhaps as other areas have scarcer supplies.

Even with increased harvests experienced in the 1990s, the state as a whole would generally have sustainable inventory and removal levels in the foreseeable future based on simple growth to removal ratio analyses for the addition of wood chip mills alone. Some areas would have annual removals that exceeded net annual growth. In addition, other processing facilities are increasing their wood requirements as well, and do contribute to the total picture of timber growth and drain.

We examined the effects of multiple chip mills with overlapping wood procurement zones using the Subregional Timber Supply Model (Abt et al. 2000a). SRTS projected timber supply for all private forest lands, including all timber harvests in the state, both for wood chip mills and other wood processing facilities. The projections assumed that timber harvest rates would increase about 1 percent per year, and forest land losses would occur at the -0.35 percent per year level, distributed per historical trends among the five major forest management types in the state. The projections indicated that softwood removals on private forest lands exceeded growth as of 1990, and would continue to do so, decreasing total softwood inventory to 10.3 billion [ft.sup.3]. by 2020, from the 1990 level of 11.2 billion [ft.sup.3]. Hardwood removals would exceed growth by about 2005, decreasing projected inventory slightly from 17.5 billion [ft.sup.3]. in 1990 to 17.3 billion [ft.sup.3]. in 2020. Thus, total projected timber remov als exceeded growth by 178 million [ft.sup.3]. per year by 2020. Timber inventory trends were very sensitive to small changes in both timber harvests or in forest land area (Abt et al. 2000b).

Prestemon and Abt (2001) updated the SRTS projections with South-wide FIA data for the recent Southern Forest Resource Assessment (SOFRA). These projections reflected the effects of higher assumed forest management intensities, and new FIA data for Tennessee. These more aggressive assumptions indicated that private hardwood timber inventories in North Carolina would be somewhat higher than for the wood chip mill study results, peaking at 18.4 billion [ft.sup.3]. in 2010 and declining to 17.8 billion [ft.sup.3]. in 2025. In addition, the decrease in softwood inventories was reversed in the SOFRA analyses, with inventories increasing from 11.2 billion [ft.sup.3]. in 1995 to 13.7 billion [ft.sup.3]. in 2025, indicating improved sustainability of the state's timber supply in the context of a South-wide timber market.

CONCLUSIONS

This analysis examined the potential impact of wood chip mills on increasing timber harvest area around a wood chip mill. Based on the TPO and ETA data analyzed in this and other parts of the wood chip mill study, we determined that 1) more timber volume was likely to be harvested due to new wood chip mills; and 2) more timber harvest areas were likely to be harvested as well.

Chip mills could take some, but not all, of their wood requirements from existing sites. Findings from our study suggest that chip mill furnish is most likely a combination of more intensive harvest from existing sites, resulting in about 18 tons of additional removals per average acre, and expanded harvest area within the state. As a general rule of thumb, each additional 100,000 tons of chip mill capacity will likely consume 1,000 acres of timber per year, and induce a total clearcut harvest for chips and other products of about 1,700 acres annually. Thus, the 4.1 million tons of wood chip mill capacity would be associated with annual timber harvests on about 40,000 to 70,000 acres per year.

The nine new wood chip mills since 1990 would not individually cause removals to exceed growth in any areas, but would increase those differences around five mills where that situation already existed. We also examined the impacts of overlapping wood chip mills, and found that overlapping procurement circles also would require more wood furnish than is apt to be available from residual material alone. Furthermore, on a state-wide basis, cumulative timber harvests from all sources are apt to cause timber inventories to decline in the future.

Various factors could alter these general guidelines somewhat. If more residual volumes were harvested per acre, less total acres would need to be harvested, as indicated in Tables 4 through 6. Since some mills were already operating in 1990, some residual material used to compute Table 2 was probably already being harvested at that time, which would reduce the number of new acres needed. The fact that only some wood volume would be merchandised for wood chip mills would increase the number of acres affected. On balance, these and other caveats suggest that users of this research should be cautious in applying average impacts indiscriminately to all potential mill impacts. Instead, the estimated removal volumes per acre and the specific mill sizes should be applied to the matrices in Table 4 through 6 to estimate potential impacts for a specific mill size.

These extensive and intensive impacts have different ecological implications for the state's forests. More harvests have the potential to create fragmented forests. Cumulatively, these effects will also result over time in more area in younger hardwood stands, with resulting wildlife habitat effects. Cleaner harvest sites with less residual volume would be likely to expose more bare soil, which is the precursor to soil movement and erosion. While definitive research is lacking in the Southeast, ecologists generally believe that more woody debris on harvest sites provides more cover and habitat for wildlife. Thus, increased harvest intensity could reduce the number of wildlife species on a site, and surely would affect the mix of species on a site. Cumulative impacts of annual chip mill timber would be up to 10 times as much land in the youngest ecological seral stage of shrubs and forbs (seedlings and saplings) of 0 to 10 years old, and 20 times as much including the sapling (11 to 20 year old) age class.

Overall, wood chip mills have both replaced existing wood concentration yards and often opened new processing capacity in areas where timber inventory has been relatively plentiful and previous pulpwood procurement efforts have been small. At least in new pulpwood market areas, wood chip mills prompt additional acres of timber harvest, and influence the distribution of harvests on the landscape. Their biggest impact has been to concentrate pulpwood production increases in the North Carolina Piedmont. Despite their expansion, a large amount of annual hardwood growth remained in comparison to annual removals in the 50-mile radius around 14 of the 18 existing mills. However, other wood processing facilities, coupled with wood chip mills, appear to be increasing annual removals on private forest lands so that they may exceed growth rates.

However, recent southern timber supply projections (Prestemon and Abt 2001) indicate that more intensive management and substitution among states may reduce the adverse impacts on timber inventory levels in North Carolina. In addition, export of wood chips has declined somewhat in the last 2 years due to poor markets for pulp and paper products. In addition, while we did find measurable impacts of wood chip mills in increasing forest areas harvested, these estimates were much lower than commonly presumed or publicized in popular literature opposing wood chip mills. Thus, while significant, the impact of wood chip mills in North Carolina seems to have peaked, at least for the moment.

Like all analyses, those contained here have strengths and limitations. The basic FIA data used to estimate harvest areas for generalized or actual wood chip mills should be as accurate as possible. The new FIA web-based information does provide a good means to tailor analyses to specifically identified areas. The analyses of growth and removals by wood chip mill area should be accordingly robust. The harvest residuals data were obtained from a limited sample of harvested sites, so the estimates of the actual amount of incremental wood that could be obtained for wood chip mills from conventional harvests could be improved. Direct estimates of merchandising from existing harvest sites would be a better approach to estimating the area that might be attributable to wood chip mills.

Other key assumptions for the study should be reasonable. The 50-mile wood chip mill radius might be conservative, but only affected the calculations for the existing wood chip mills' average harvest volumes per acre and the calculations of growth and drain in those "woodsheds." If the radius were larger, the volume per acre would probably be similar, but the growth and drain ratios would be affected less. The harvest volumes per acre will tend to be greater than average, and at least new wood chip mills are likely to be larger rather than average sized. Seldom is all the timber harvest on a site processed at one wood chip mill; high value products are generally sorted out and sent to other sawmills or high-end processors. Thus, more timber harvest area may be required to actually feed a wood chip mill. The purpose of this analysis was to estimate the incremental effect of new wood processing capacity, and those number of acres should be relatively accurate for the given amount of wood fiber required, regardl ess of the type of processing facility.

The generic tables do allow one to compute areal requirements for smaller or larger volume harvests. They also could be used to estimate the potential impact of any new wood processing facility that requires a given amount of timber to be furnished. Again, the general average of 1,000 to 1,700 acres of harvest per 100,000 tons per year of capacity would be a statewide average. Local harvest volumes per acre would increase or decrease this area needed depending on stand volumes, harvest intensities, and product merchandising on any given site.
TABLE 1

Assumptions and algorithms used for analysis of wood chip mill timber
harvest impacts. (a)

Pine Hardwood

5,350 lb. = 1 cord 5,800 lb. = 1 cord
2.675 tons = 1 cord 2.9 tons = 1 cord
2.675 tons = 74 [ft..sup.3] 2.9 tons = 74 [ft..sup.3]
cords/ton = x [ft..sup.3] Cords/ton = x [ft..sup.3]
x = 28 [ft..sup.3]/ton x = 2.5 [ft..sup.3]/ton

Volume harvested per acre

 Procurement circle = 50-mile radius
 [ft..sup.3]/ac. is retrieved from FIA
 Database (total volume/timberland ac.)

Area harvested per mill

 Mill capacity in tons * x =
 capacity in [ft..sup.3]
 Mill capacity in [ft..sup.3]/[ft..sup.3]
 per acre = required harvest acres
 fulfill mill requirements

(a)Source for conversion equations: Timber Mart-South 1999.

TABLE 2

FIA final harvest areas, potential wood chip volumes, and chip mill
capacity.

 1990
 final harvest Residual available 1997 chip capacity

 (ac.) (tons)

Coastal Plain 194,834 3,500,000 1,413,000
Piedmont 72,161 1,300,000 2,060,000
Mountains 27,991 500,000 665,000
State total 294,986 5,300,000 4,138,000

TABLE 3

Timber inventory and harvest volumes per acre in North Carolina (Brown
1990). (a)

 Pine Hardwood All

Harvest volumes

 Harvest volume ([ft..sup.3]) 511,766,000 427,945,000 939,711,000
 Harvest area (ac.)
 Final harvest 144,008 150,978 294,986
 Partial harvest 20,967 31,945 52,912
 Commercial thin 49,717 1,399 51,116
 Total 214,962 184,322 399,014
 Harvest volume/ac. ([ft..sup.3]) 2,381 2,322 2,355

Inventory volumes

 Timber inventory ([ft..sup.3]) 12,530,371 20,211,960 32,742,331
 Total area (ac.) 7,229,474 11,480,907 18,710,381
 Inventory/ac. ([ft..sup.3]) 1,733 1,760 1,750

(a)Pine areas computed as planted pine, natural pine, and 37.5 percent
of oak-pine totals; hardwood areas computed as 62.5 percent of oak-pine
and all of upland and lowland hardwood areas.

TABLE 4

Acres of harvest required annually to furnish a wood chip mill, by
percentage of hardwood and volume per acre for a mill capacity of
150,000 tons/yr.

 Percentage of hardwood
 Volume
 per acre 10 20 30 40 50

([ft.sup.3]) (acres)

 500 8,310 8,220 8,130 8,040 7,950
 1,000 4,155 4,110 4,065 4,020 3,975
 1,500 2,770 2,740 2,710 2,680 2,650
 2,000 2,078 2,055 2,033 2,010 1,988
 2,500 1,662 1,644 1,626 1,608 1,590
 3,000 1,385 1,370 1,355 1,340 1,325
 3,500 1,187 1,174 1,161 1,149 1,136
 4,000 1,039 1,028 1,016 1,005 994
 4,500 923 913 903 893 883
 5,000 831 822 813 804 795
 5,500 755 747 739 731 723
 6,000 693 685 678 670 663

 Percentage of hardwood
 Volume
 per acre 60 70 80 90 100

([ft.sup.3]) (acres)

 500 7,860 7,770 7,680 7,590 7,500
 1,000 3,930 3,885 3,840 3,795 3,750
 1,500 2,620 2,590 2,560 2,530 2,500
 2,000 1,965 1,943 1,920 1,898 1,875
 2,500 1,572 1,554 1,536 1,518 1,500
 3,000 1,310 1,295 1,280 1,265 1,250
 3,500 1,123 1,110 1,097 1,084 1,071
 4,000 983 971 960 949 938
 4,500 873 863 853 843 833
 5,000 786 777 768 759 750
 5,500 715 706 698 690 682
 6,000 655 648 640 633 625

TABLES 5

Acres of harvest required annually to furnish a wood chip mill, by
percentage of hardwood and volume per acre for a mill capacity of
250,000 tons/yr.

 Percentage of hardwood
 Volume
 per acre 10 20 30 40 50

([ft.sup.3]) (acres)

 500 13,850 13,700 13,550 13,400 13,250
 1,000 6,925 6,850 6,775 6,700 6,625
 1,500 4,617 4,567 4,517 4,467 4,417
 2,000 3,463 3,425 3,388 3,350 3,313
 2,500 2,770 2,740 2,710 2,680 2,650
 3,000 2,308 2,283 2,258 2,233 2,208
 3,500 1,979 1,957 1,936 1,914 1,893
 4,000 1,731 1,713 1,694 1,675 1,656
 4,500 1,539 1,522 1,506 1,489 1,472
 5,000 1,385 1,370 1,355 1,340 1,325
 5,500 1,259 1,245 1,232 1,218 1,205
 6,000 1,154 1,142 1,129 1,117 1,104

 Percentage of hardwood
 Volume
 per acre 60 70 80 90 100

([ft.sup.3]) (acres)

 500 13,100 12,950 12,800 12,650 12,500
 1,000 6,550 6,475 6,400 6,325 6,250
 1,500 4,367 4,317 4,267 4,217 4,167
 2,000 3,275 3,238 3,200 3,163 3,125
 2,500 2,620 2,590 2,560 2,530 2,500
 3,000 2,183 2,158 2,133 2,108 2,083
 3,500 1,871 1,850 1,829 1,807 1,786
 4,000 1,638 1,619 1,600 1,581 1,563
 4,500 1,456 1,439 1,422 1,406 1,389
 5,000 1,310 1,295 1,280 1,265 1,250
 5,500 1,191 1,177 1,164 1,150 1,136
 6,000 1,092 1,079 1,067 1,054 1,042

TABLE 6

Acres of harvest required annually to furnish a wood chip mill, by
percentage of hardwood and volume per acre for a mill capacity of
350,000 tons/yr.

 Percentage of hardwood
 Volume
 per acre 10 20 30 40 50

([ft..sup.3]) (acres)

 500 19,390 19,180 18,970 18,760 18,550
 1,000 9,695 9,590 9,485 9,380 9,275
 1,500 6,393 6,393 6,323 6,253 6,183
 2,000 4,848 4,795 4,743 4,690 4,638
 2,500 3,878 3,836 3,794 3,752 3,710
 3,000 3,232 3,197 3,162 3,127 3,092
 3,500 2,770 2,740 2,710 2,680 2,650
 4,000 2,424 2,398 2,371 2,345 2,319
 4,500 2,154 2,131 2,108 2,084 2,061
 5,000 1,939 1,918 1,897 1,876 1,855
 5,500 1,763 1,744 1,725 1,705 1,686
 6,000 1,616 1,598 1,581 1,563 1,546

 Percentage of hardwood
 Volume
 per acre 60 70 80 90 100

([ft..sup.3]) (acres)

 500 18,340 18,130 17,920 17,710 17,500
 1,000 9,170 9,065 8,960 8,855 8,750
 1,500 6,113 6,043 5,973 5,903 5,833
 2,000 4,585 4,533 4,480 4,428 4,375
 2,500 3,668 3,626 3,584 3,542 3,500
 3,000 3,057 3,022 2,987 2,952 2,917
 3,500 2,620 2,590 2,560 2,530 2,500
 4,000 2,293 2,266 2,240 2,214 2,188
 4,500 2,038 2,014 1,991 1,968 1,944
 5,000 1,834 1,813 1,792 1,771 1,750
 5,500 1,667 1,648 1,629 1,610 1,591
 6,000 1,528 1,511 1,493 1,476 1,458

Table 7

Acres of harvest required annually to furnish actual wood chip mills in
North Carolina by capacity, species group, and species group mixture
consumed.

 Mill requirements
Chip
mill Capacity Hardwood Softwood Current % hardwood

 (tons) ([ft..sub.3]) (%)

 A 25,000 625,000 700,000 625,000 100
 B 200,000 5,000,000 5,600,000 5,300,000 50
 C 362,234 9,055,850 10,142,552 9,055,850 100
 D 350,000 8,750,000 9,800,000 8,855,000 90
 E 160,000 4,000,000 4,480,000 4,096,000 80
 F 300,000 7,500,000 8,400,000 7,725,000 75
 G 325,000 8,125,000 9,100,000 8,470,000 65
 H 300,000 7,500,000 8,400,000 7,500,000 100
 I 240,000 6,000,000 6,720,000 6,288,000 60
 J 350,000 8,750,000 9,800,000 9,012,500 75
 K 175,000 4,375,000 4,900,000 4,427,500 90
 L 400,000 10,000,000 11,200,000 10,000,000 100
 M 150,000 3,750,000 4,200,000 3,975,000 50
 N 100,000 2,500,000 2,800,000 2,650,000 50
 0 300,000 7,500,000 8,400,000 7,770,000 70
 P 100,000 2,500,000 2,800,000 2,560,000 80
 Q 200,000 5,000,000 5,600,000 5,060,000 90
 R 100,000 2,500,000 2,800,000 2,500,000 100

 Acres required/yr.
Chip
mill Hardwood Softwood Current



 A 340 248 340
 B 2,872 2,500 2,686
 C 6,068 5,839 6,068
 D 5,273 4,661 5,212
 E 2,421 1,732 2,284
 F 5,004 3,831 4,711
 G 5,667 4,757 5,348
 H 3,317 2,591 3,317
 I 3,034 2,511 2,825
 J 6,768 5,028 6,333
 K 2,322 1,711 2,261
 L 6,380 6,022 6,380
 M 2,346 2,018 2,182
 N 1,437 1,396 1,417
 0 4,157 2,446 3,644
 P 1,441 1,018 1,356
 Q 4,096 3,246 4,011
 R 1,690 1,296 1,690

TABLE 8

Comparisons of growth and removals before and after wood chip mill
openings after 1989 in North Carolina.

 Hardwood Softwood

 Removals
Chip Year
mill opened Growth Before After Growth

 (million[ft..sup.3])

 E 1970 64.7 49.7 Same 95.1
 K 1978 66.2 46.8 Same 88.1
 M 1981 81.2 58.7 Same 64
 I 1985 106 72.8 Same 51.9
 O 1985 5.7 2.9 Same 23.8
 B 1986 43.3 25.8 Same 34.6
 P 1986 50 19.5 Same 13.8
 N 1986 41.9 31.9 Same 55.1
 G 1988 81.1 52.3 Same 60.4
 F 1990 87.8 69.4 76.9 90.1
 C 1990 76.1 76.8 85.8 131.1
 Q 1990 33.7 18.9 23.9 61.7
 L 1990 87.1 99.8 109.8 121.6
 J 1991 83.4 66.5 75.2 105.3
 A 1991 108.4 61.4 62.0 47.3
 D 1995 105.7 106.1 114.8 106.4
 R 1996 82.7 98.6 101.1 84.5
 H 1998 96.5 61.6 69.1 53.3

Total 1,301.5 1,019.5 1,079.0 1,288.1

 Softwood

 Removals
Chip
mill Before After

 (million[ft..sup.3])

 E 83.4 Same
 K 93.7 Same
 M 60.4 Same
 I 50.9 Same
 O 5.7 Same
 B 25.6 Same
 P 5.6 Same
 N 40.8 Same
 G 59 Same
 F 90.8 99.2
 C 131 141.1
 Q 48 53.6
 L 126.9 138.1
 J 90.8 100.6
 A 25.4 26.1
 D 92.3 102.1
 R 121.8 124.6
 H 39.2 47.6

Total 1,191.3 1,258.1


LITERATURE CITED

Abt, R.C., F.W. Cubbage, and G. Pacheco. 2000a. Southern forest resource assessment using the Subregional Timber Supply (SRTS) model. Forest Prod. J. 50(4):25-33.

_____, R. Schaberg, and G. Hess. 2000b. Forest resource trends and projections for North Carolina. Working Pap. No.4. In: Economic and Ecologic Impacts Associated With Wood Chip Production in North Carolina. Southern Center for Sustainable Forests: North Carolina State Univ. College of Forest Resources, Duke Univ. School of the Environment, and North Carolina Dept. of Natural Resources Div. of Forest Resources. 13 pp. + append. July 31. http://www.env.duke.edu/scsf/.

Aruna, P.B. and F.W. Cubbage. 2000. Regional economic analyses of the forest products and tourism sectors in North Carolina. Working Pap. No. 9. In: Economic and Ecologic Impacts Associated With Wood Chip Production in North Carolina. Southern Center for Sustainable Forests: North Carolina State Univ. College of Forest Resources, Duke Univ. School of the Environment, and North Carolina Dept. of Natural Resources Div. of Forest Resources. 102 pp. + append. July 31. http://www.env.duke.edu/scsf/.

Brown, M.J. 1993. North Carolina's forests, 1990. Resource Bull. SE-142. USDA Forest Serv., Southeastern Forest Expt. Sta., Asheville, NC. 101 pp.

Cubbage, F.W., W.S. Burleson, and J.D. Dodrill. 2000. North Carolina's forests: 1938 to 1990. Working Pap. No. 1. In: Economic and Ecologic Impacts Associated With Wood Chip Production in North Carolina. Southern Center for Sustainable Forests: North Carolina State Univ. College of Forest Resources, Duke Univ. School of the Environment, and North Carolina Dept. of Natural Resources Div. of Forest Resources. 26 pp. + append. July 31. http://www.env.duke.edu/scsf/.

Hess, GR, and D. Zimmerman. 2001. The effect of satellite chip mills on post-harvest woody debris. Southern J. of Appl. Forestry 25(4):173-177.

Johnson, T.G. 1990. Forest Statistics for the Southern Coastal Plain of North Carolina, 1990. Resource Bull. SE-111, USDA Forest Serv., Southeastern Forest Expt. Sta., Asheville, NC.

_____. 1991a. Forest Statistics for the Mountains of North Carolina, 1990. Resource Bull. SE-118. USDA Forest Serv., Southeastern Forest Expt. Sta., Asheville, NC.

_____. 1991b. Forest Statistics for North Carolina, 1990. Resource Bull. SE-120. USDA Forest Serv., Southeastern Forest Expt. Sta., Asheville, NC.

Prestemon, J.P. and R.C. Abt. 2001. Timber products supply and demand. In: Southern Forest Resource Assessment, Draft Report. http://www.srs.fs.fed.us/sustain/report/index.htm.

Schaberg, R., F. Cubbage, and D. Richter. 2000. Timber products output trends in North Carolina. Working Pap. No. 2. In: Economic and Ecologic Impacts Associated With Wood Chip Production in North Carolina. Southern Center for Sustainable Forests: North Carolina State Univ. College of Forest Resources, Duke Univ. School of the Environment, and North Carolina Dept. of Natural Resources Div. of Forest Resources. 34 pp. + append. July 31. http://www.env.duke.edu/scsf/.

Snider, A. and F. Cubbage. 2000. Nonindustrial private forests: An analysis of changes in potential returns as a result of shifts in demand. Working Pap. No. 8. In: Economic and Ecologic Impacts Associated With Wood Chip Production in North Carolina. Southern Center for Sustainable Forests: North Carolina State Univ. College of Forest Resources, Duke Univ. School of the Environment, and North Carolina Dept. of Natural Resources Div. of Forest Resources. 58 pp. + append. July 31. http://www.env.duke.edu/scsf/.

Timber Mart-South. 1999. Southern timber volume conversion factors. Warnell School of Forest Resources, Athens, GA.

The authors are, respectively, Graduate Research Assistant, Professor, Research Assistant Professor, and Professor, Dept. of Forestry, 3120 Jordan Hall, Box 8008, North Carolina State Univ., Raleigh, NC 27695-8008. This paper was received for publication in June 2001. Reprint No. 9327.

FREDERICK W. CUBBAGE *

* Forest Products Society Member.

[C]Forest Products Society 2002. Forest Prod. J. 52(11/12):29-37.
COPYRIGHT 2002 Forest Products Society
No portion of this article can be reproduced without the express written permission from the copyright holder.
Copyright 2002 Gale, Cengage Learning. All rights reserved.

Article Details
Printer friendly Cite/link Email Feedback
Author:Dodrill, John D.; Cubbage, Frederick W.; Schaberg, Rex H.; Abt, Robert C.
Publication:Forest Products Journal
Geographic Code:1U5NC
Date:Nov 1, 2002
Words:8841
Previous Article:Advanced engineered wood composites: innovative technologies for Maine's forest products industry.
Next Article:Movement of boron and fluoride from rod formations into Douglas-fir heartwood. (Solid Wood Products).
Topics:

Terms of use | Copyright © 2017 Farlex, Inc. | Feedback | For webmasters