13. Reduced impact logging: does it cost or does it pay? - Wulf Killmann*, Gary Q. Bull, Olaf Schwab** and Reino E. Pulkki***

* Director, Forest Products Division, Food and Agriculture Organization of the United Nations, Via delle Terme di Caracalla 0100 Rome, Italy, Tel: ++(39 6) 5705 3221, Fax: ++(39 6) 5705 5618, E-mail: [email protected]

** 2022, 2424 Main Mall, Faculty of Forestry, University of British Columbia, Vancouver, British Columbia, V6T 1Z4, CANADA, Tel.: 604 822 1553, Fax: 604 822 9106, E-mail: [email protected]

*** Faculty of Forestry and the Forest Environment, Lakehead University, 955 Oliver Road, Thunder Bay, Ontario Canada, P7B-5E1, Tel.: 807-343-8564, Fax: 807-343-8116, E-mail: [email protected]

INTRODUCTION

There has been an increasing interest in reduced impact logging (RIL), particularly in the last decade. The interest has been shaped by a number of developments that include: a political focus on sustainable development at the highest levels, a general consensus about the necessity to manage forests more sustainably and a recognition that better technology is now available to monitor harvesting practices and forest conditions. There is a general desire to reduce negative environmental impacts all the way down to the operational level. It is also recognized that many conventional logging (CL) systems are not going to produce yields of the same volume and/or quality of timber at a sustainable level.

For these reasons, governments are now imposing stricter regulations on forest harvesting and timber markets and demanding responsible harvesting practices. Also, in some markets the demand for certified timber products is growing. Given the political and market demand for change, there is a commensurate desire to further articulate the nature of RIL. Some important questions must be addressed. For example, what have harvesting studies, carried out to date, been telling us and, perhaps even more important, what have these studies not told us? What should researchers do to continue or discontinue the promotion of RIL? Finally does RIL cost or does it pay?

To address these questions the objectives of the initial research briefly described in this paper were to:

• produce a definition and lay down the characteristics of RIL;

• review existing literature to tease out the quantitative information on RIL;

• develop a prototype model for recording data on RIL and CL operations; and

• compile a general cost profile of different logging machines.

DEFINITION

One of the first challenges in addressing the question “Does RIL pay?” is to define what we mean by reduced impact logging. Various authors (Armstrong and Inglis, 2000; Elias, 2000; Van der Hout, 1999; Reid and Rice, 1997; Ruslim et al., 1999; Sist et al., 1998; Sist, 2000; Department of Forests, 1999; Webb, 1997) have proposed definitions or aspects of RIL.^[11]

Summarizing the work of these authors, RIL is defined as:

Intensively planned and carefully controlled implementation of harvesting operations to minimize the impact on forest stands and soils, usually in individual tree selection cutting.

RIL is generally characterized (Table 1) by having stand entries only at a predetermined cutting cycle, which generally should be no shorter than 20 years. No more than one-third of the basal area should be removed at one entry and a pre-harvest operational inventory is recommended strongly. Access roads should be constructed well in advance of harvesting, and climbers should be cut, if required, up to two years before the harvest. Planning should consist of tree marking, location mapping and planned felling direction. This planning should be linked with the layout of a minimal number of extraction trails. Once the logs are removed, they should be placed on landings of minimal size. The logging operations should only be conducted under favourable conditions (e.g. when soils are dry). Forest workers and supervisors should be well trained so they can ensure minimal negative impacts on the site; make maximum utilization of the trees felled; incur minimal damage to residual trees and advanced regeneration; and facilitate rehabilitation of negative impacts that may have occurred on the site. Finally, workers and supervisors should be well qualified to conduct post-harvest assessments.

Table 1. Some major characteristics of RIL techniques

LITERATURE REVIEW, HIGHLIGHTS AND ANALYSIS

Two hundred and sixty-six studies and articles on RIL and/or CL, conducted in closed broad-leaved tropical forests, were reviewed and classified. These publications dated back to 1950. The initial review of 212 articles was conducted in 1997 for the FAO Global Fibre Supply Model project (FAO, 1997). This review was then updated in 2000 when 54 additional articles were added, including some of the German and French literature (FAO, 2001).

The majority of the studies were published in the last decade (Figure 1). However, as many of the graphs in this paper indicate, the statistics were not presented according to a standardized system. In the 1970s, the number of publications on CL increased rapidly, whereas RIL was covered in only a few publications prior to 1980.

Once the articles were compiled and analysed, the statistics were organized into a database (Figure 2). The data extracted are summarized in the figures presented in this paper in order to give an indication of some general trends. It should be noted, however, that the data found in the various reports are not adjusted to account for inflation.

Figure 1. Number of forest harvesting publications by decade

Figure 2. Literature summary of cost-related characteristics of RIL and CL

Logging intensity

Figure 3 is a summary of 130 observations (n=37 for RIL and n=93 for CL). The median^[12] volume harvested was 8 m³/ha lower, with RIL at 37 m³/ha and CL at 45 m³/ha. In the majority of the RIL studies the logging intensity was <60 m³/ha, while for CL there was a significant portion >60 m³/ha.

Figure 3. Logging intensity in terms of volume per hectare removed

In Figure 4, 101 observations (45 RIL and 56 CL) showed no significant difference in the number of trees harvested per hectare between the two forms of logging. The median number of trees harvested was 8 trees/ha for each cutting cycle. Taken together with the observation in Figure 3, we can conclude that CL involves extracting larger trees.

Figure 4. Logging intensity in terms of trees per hectare removed

Costs

Figure 5 summarizes the findings on planning costs per cubic metre. In 10 observations the median cost of RIL was US$ 0.28/m³ higher than that of CL. In another two observations the planning costs were reported as costs per hectare. The range of US$ 5.06 to US$ 50.00/m³ indicates serious shortcomings in the statistics on planning at this time.

It should also be noted that inventory costs were also woefully lacking and no statistical summary could be generated from the literature reviewed. Future studies must record and report planning and pre-harvest operational inventory costs in more detail.

In a total of 10 observations, the median felling cost of RIL was US$ 0.56/m³ higher than that of CL. This represents a cost that is 48 percent higher.

Figure 5. Summary of planning costs with RIL and CL

Figure 6. Felling costs in RIL and CL

One reason for the significant difference between the two types of logging is that RIL, on average, extracts smaller trees, and therefore felling costs per cubic metre are higher. Time studies on felling in RIL and CL operations showed that the time required to fell each tree is significantly longer in RIL operations. This additional time is used for felling preparations, such as determining the appropriate felling direction, and careful directional felling to prevent the stem from splitting and damaging advanced regeneration.

In a review of 11 articles (Figure 7) no appreciable difference in skidding costs between the two forms of logging was found. We did postulate that skidding costs might be lower with RIL, but these observations do not support this hypothesis. The average size of the trees is one of the factors that affects skidding costs per cubic metre. Therefore, the harvesting of larger trees in CL operations would result in lower skidding costs per cubic metre compared with RIL. Another factor may be that with fewer skid trails, the overall extraction distance would be longer.

Figure 7. Skidding costs in RIL and CL

In these studies, no standard methodology was used to determine inventory, planning, operational and total costs. Therefore the total cost comparison (see Figure 8), which for 23 observations shows an increase in costs of US$ 8.50/m³, is not comparable with the previous cost graphs. The total costs in these studies were not broken down by activity. Therefore, it was impossible to assess which logging phases had been included.

The comparison indicates an increase in total costs (often not explicitly defined in the articles) of more than 43 percent. This highlights the need for a statistical framework that would make it possible to use field researchers’ and practitioners’ data more widely if they were presented according to a standard format.

Most studies that state that the total costs would be higher for RIL mention at the same time that the lower level of damage to the residual stand in RIL operations is likely to compensate for some of the additional costs in the future.

Another factor is that RIL is fairly new and it is likely that costs will be reduced as logging crews become more experienced.

Figure 8. Total costs of RIL and CL logging systems

Damage

Another issue of concern in the literature is damage to site and stands. Here we found significant differences between the two approaches to forest harvesting.

In 75 observations (Figure 9) we found that RIL had 41 percent less residual stand damage, when compared to 49 percent for CL systems.

Other studies (Figure 10), which measured stand damage in a different manner (a total of seven observations), found that median rates of damage for RIL and CL were 124 and 131 trees per hectare, respectively.

The data used to generate Figure 10 are not clear; no indication is given of tree size, proportion of tree species of commercial interest and the share of residual stand volume or basal area damaged. Again, this supports the need for a standardized approach to reporting logging impacts.

Figure 9. Stand damage (percent of residuals)

Figure 10. Stand damage (stems/ha)

In 15 observations (Figure 11) the damage to trees per trees felled was 56 percent less in RIL operations when compared to CL operations. In RIL operations the first step of directional felling is to determine the general felling direction according to the layout of the skid trails. Within this frame the felling direction is adjusted in order to minimize damage to the residual stand by felling into the section with the lowest density of trees of commercial interest.

Figure 11. Stand damage (trees/trees felled)

Figure 12 was generated using 39 observations. There was a significant decline in skid trail damage with the adoption of RIL. The area covered by skid trails in RIL operations is almost 50 percent less than in CL. One of the most important characteristics of RIL is the layout and planning of skid trails prior to felling. The combination of marked skid trails and maps showing the location of each tree is very effective in minimizing the area affected by skidders when compared with the large area covered with trails when each stem is approached in a random manner as is done in CL operations.

Figure 12. Site damage (skid trails)

Despite only eight observations (Figure 13) of road damage, the trend was as expected with a much lower level of damage (41 percent) when using RIL techniques. As mentioned earlier, the planning of access routes prior to harvesting is a very effective tool to optimize the construction of roads and trails.

Figure 13. Compartment area damaged by roads

Figure 14, with 58 observations, indicates that site damage at RIL sites was 57 percent less than at CL sites. This is a very important finding. One of the most important problems with CL practices is that the high impact on the residual stand and soil severely limits the ability of the site to regenerate completely within the proposed cutting cycles. It is possible to reduce the impact of timber harvesting drastically by implementing RIL (Figure 14). In tropical forests, the highest level of nutrient loss due to leaching was detected in areas where the mineral soil was exposed. With RIL it is possible to restrict these severe damages to a relatively small area so that nutrient losses can be minimized while advanced regeneration is maintained to enable the site to recover completely within one cutting cycle.

Figure 14. Site damage (total area)

Utilization, lost timber and canopy opening

Figure 15, with 21 observations, indicates that utilization rates^[13] are better with RIL; however, many more studies are needed. The training of workers and supervisors for RIL techniques usually includes some instructions on efficient bucking. In CL operations most stems were topped before the first branch, although there might be another valuable log above the first branch.

In 33 observations (Figure 16), with the adoption of RIL the volume of lost timber^[14] was 60 percent lower than in CL operations. The maps used in RIL operations clearly indicate the position of each log so that the skidder operator can proceed systematically. In CL operations, it is often the case that logs are not found by the skidder operator.

The median for lost timber in RIL operations may be lower than reported here. This is because in many studies a loss was reported for CL operations but none was given for RIL operations. This could be interpreted as zero loss of timber; however, since this had not been stated explicitly, it was not considered in the analysis.

Figure 15. Utilization of felled timber

Figure 16. Lost timber (percent of volume removed)

In 25 observations (Figure 17) with RIL the canopy opening was 36 percent smaller and again this was expected. In CL operations most trees are felled in their leaning direction so that the felling gaps rarely overlap. In RIL operations several trees are felled into the same gap, whenever possible, thereby maintaining a high level of crown cover.

Figure 17. Canopy opening (percent)

COST DATA BY MACHINE TYPE

Another project was conducted to find standardized cost data that permit a comparison of data found in field trials with data assigned to various pieces of logging equipment by manufacturers. With cost information, it is possible to detect significant variances and better explain variability, particularly from region to region or from country to country.

In total, 231 pieces of ground-based logging equipment were reviewed with regard to machine costs, and all results were stored in a database. Figure 18 illustrates the types of cost and production data for a feller-buncher.

PROTOTYPE MODEL

The information presented in this paper clearly demonstrates a weakness in the existing economic data on CL and RIL. To address this issue we developed a RIL statistical prototype to collect, assemble and disseminate information for researchers and practitioners.

Figures 19-21 demonstrate just a small portion of the Visual Basic programme. As indicated on the front page, the structure of the information system is based on our findings in existing literature, but no logging-cost model by region is available at the moment. This is a feature that can be added easily once the system is adapted for the web environment, which will be administered by a group of editors considered experts in their field or in their region.

Figure 20 shows the mechanism for data entry in the following categories:

1. General planning
2. Felling
3. In-stand operations
4. Extraction
5. Roadside operations
6. Costs and impacts

Figure 18. Example of standard ground-based machine costs

Figure 19. RIL statistical framework prototype

Figure 20. Logging method and system description form

The statistics that are to be compiled for general planning and compartment information include: general background information, logging compartment information, planning levels and road access.

In Figure 21, for example, the logging compartment information includes: stand data, terrain data, weather and silvicultural system data.

Figure 21. Logging compartment information data required

CONCLUDING REMARKS

RIL does cost more, but not as much as one might expect. It is important to bear in mind that the cost data presented represent a short-term analysis and they do not reflect all ‘costs’ such as, for example, the costs that might be expected when harvesting in the next logging cycle. Also, experience in RIL is new and higher costs can be expected. Once logging crews become more familiar with RIL it is expected that costs will be reduced.

In the longer term, and from ecological, social and economic points of view, it does pay to secure a more sustainable higher-valued timber supply with the application of RIL techniques. Many of the figures presented in this paper on site and stand damage, utilization rates and size of canopy opening support this assertion. In this context, it is also clear that RIL would support much more effectively the objectives of financial institutions, such as the World Bank, in their emphasis on poverty alleviation than other forms of logging. Additionally, the market will require widespread implementation of RIL if the forests are to be certified.

There are serious data deficiencies with respect to planning and inventory, and there is a serious lack of standards in data collection. We conclude that the existing data should be integrated with the proposed statistical framework to the greatest extent possible. In addition, new data on cable and aerial logging systems should also be collected and added to the existing statistical framework.

It would also be very helpful to build a logging cost model in order to make the information useful to practitioners, researchers, equipment manufacturers and modellers.

Finally, in order to make the necessary advances, it would be necessary to move to a web-based data compilation and information dissemination system. Input and participation of other researchers is required to develop a more reliable harvest information system.

REFERENCES

Armstrong, S. & Inglis, C.J. 2000. RIL for real: introducing reduced impact logging techniques into a commercial forestry operation in Guyana. International Forestry Review, 2(1): 17-23.

Elias. 1999. Introducing a manual on reduced impact timber harvesting in the Indonesian selective cutting and planting system. ITTO Tropical Forestry Update 9, (3): 26+30.

FAO. 1997. Literature synthesis on logging impacts in moist tropical forests, by R. Pulkki. Global Fibre Supply Study Working Paper Series #No. 6. Food and Agriculture Organization of the United Nations, Rome.

FAO. 2001. Literature review on logging impacts in moist tropical forests, by O. Schwab, R. Pulkki & G.Q. Bull. Global Fibre Supply Model Working Paper Series No. 7. Food and Agriculture Organization of the United Nations, Rome.

Reid, J.W. & Rice, R.E. 1997. Assessing natural forest management as a tool for tropical forest conservation. Ambio, 26(6): 382-386.

Ruslim, Y., Hinrichs, A. & Ulbricht, R. 1999. Technical guideline for reduced impact tractor logging. SFMP Document No. 10a. Indonesian-German Technical Cooperation. Ministry of Forestry and Estate Crops in Cooperation with Deutsche Gesellschaft für Technische Zusammenarbeit (GTZ).

Sist, P. 2000. Reduced impact logging in the tropics: Objectives, principles and impacts. International Forestry Review, 2(1): 3-10.

Sist, P., Dykstra, D. & Fimbel, R. 1998. Reduced impact logging guidelines for lowland and hill dipterocarp forests in Indonesia. CIFOR Occasional Paper No. 15. Jakarta, Center for International Forestry Research. Jakarta.

Van der Hout, P. 1999. Reduced impact logging in the tropical rain forest of Guyana. Dissertation. University Utrecht, Utrecht.

Department of Forests. 1999. Vanuatu reduced impact logging guidelines. Vanuatu, Department of Forests, Vanuatu.

Webb, E.L. 1997. Canopy removal and residual stand damage during controlled selective logging in lowland swamp forest of northeast Costa Rica. Forest Ecology and Management, 95: 117-129.

APPENDIX 1

Definitions of reduced impact logging

Armstrong, S. & Inglis, C.J. 2000. RIL for real: introducing reduced impact logging techniques into a commercial forestry operation in Guyana. International Forestry Review, 2(1): 17-23.

RIL should at least imply a systematic approach to harvesting, specifically improved pre-harvest planning on the basis of appropriate and accurate information.

Elias. 1999. Introducing a manual on reduced impact timber harvesting in the Indonesian selective cutting and planting system. ITTO Tropical Forestry Update, 9(3): 26+30.

Reduced impact timber harvesting includes the following:

• Forest surveys prior to harvesting to generate data required for planning of the harvesting operations;

• A tree location and topographical map as a guide for felling and skidding;

• Climber cutting prior to felling;

• Regular training and adequate supervision;

• Routine briefing on procedures and techniques;

• Adoption of a premium wage system consisting of a base wage and premiums depending on quality and quantity of production as well as terrain difficulty.

Van der Hout, P. 1999. Reduced impact logging in the tropical rain forest of Guyana. Dissertation. University Utrecht. 1999.

The term ‘reduced impact logging’ (RIL) surfaced around the mid 1990s (Pinard et al., 1995), but the concept is also referred to as ‘low impact logging’ (Blate, 1997; Holmes et al., 1999), ‘planned’ (as opposed to ‘unplanned’) logging (Johns et al., 1996; Barreto et al., 1998), ‘environmentally sound harvesting’ (Hendrison, 1990). There is a need to clarify the substance covered by these terms, because we may be comparing apples and oranges. The adjective ‘reduced’ hints at a comparison with another logging method, which is obviously the current, local practice. The current practice may cover a broad range of methods, varying from ‘hit-and-miss, unplanned’ logging to ‘standard practice’ logging (Van der Hout and Van Leersum, 1998). The place a current practice may take on this scale depends largely on the scale and level of capitalization of the operation. Several elements are common to most RIL systems including the following (ITTO 1990):

• Pre-harvest inventory and mapping;

• Pre-harvest planning of roads and skid trails;

• Pre-harvest climber cutting;

• Directional felling;

• Optimum recovery of utilisable timber;

• Winching of logs to planned skid trails.

Reid, J.W. & Rice, R.E. 1997. Assessing natural forest management as a tool for tropical forest conservation. Ambio, 26(6): 382-386.

Natural forest management is controlled and regulated harvesting, combined with silvicultural and protective measures, to sustain or increase the commercial value of stands, all relying on natural regeneration of native species.

Ruslim, Y., Hinrichs, A. & Ulbricht, R. 1999. Technical guideline for reduced impact tractor logging. SFMP Document No. 10a. Indonesian-German Technical Cooperation. Ministry of Forestry and Estate Crops in Cooperation with Deutsche Gesellschaft für Technische Zusammenarbeit (GTZ).

Reduced Impact Logging aims to:

• reduce damage to the residual stand and soil (compaction and erosion);

• leave the environment in good condition;

• improve utilisation of timber potential and reduce waste; and

• reduce rehabilitation costs.

RIL includes the following planning and harvesting stages:

• pre-harvest inventory and topographic survey;

• contour and tree location map plotting;

• skid trail and log landing planning on the map;

• skid trail and log landing marking in the field;

• RIL using the marked trails and directional felling;

• skidding using “winching”;

• closing up;

• block inspection, quality control and reporting; and

• payment based on quality of work.

Sist, P., Dykstra, D. & Fimbel, R. 1998. Reduced impact logging guidelines for lowland and hill dipterocarp forests in Indonesia. Occasional Paper No. 15. Bogor, Center for International Forestry Research.

Reduced impact logging aims to reduce soil disturbance, impacts on wildlife, and damage to residual trees. RIL can be characterized through the following activities:

• stock survey (harvestable trees, potential crop trees, protected trees species, trees for non-timber forest products, important wildlife resource trees);

• climber cutting (all climbers >2 cm dbh that are attached to the canopy of harvestable trees);

• topography assessment;

• protected areas (unworkable areas, sacred areas, conservation areas, stream buffer zones);

• road, landing and skid trail planning;

• final pre-logging RIL survey (marking of skid trails, adjusting and marking felling directions, excluding trees from harvest where the extraction intensity would exceed 8 trees/ha);

• tactical maps and written plans;

• directional felling;

• skid trail marking and opening;

• rehabilitation of skid trails (cross drains, removing temporary stream crossing structures);

• road closure (removal of temporary bridges and culverts, cross drains); and

• other post-harvesting operations (controlled access to the permanent forest estate; proper maintenance of road surfaces, ditches, cross drains and stream crossings; clean landings and temporary camps).

Sist, P. 2000. Reduced impact logging in the tropics: objectives, principles and impacts. International Forestry Review, 2(1): 3-10.

Reduced Impact Logging (RIL) is also called Low Impact Logging (LIL) or Low Impact Harvesting (LIH). It includes the following principles:

• planning of logging operations at the annual coupe scale;

• pre-harvest forest inventory (100% timber inventory);

• planning of felling;

• planning of secondary roads, landings and skidding trails;

• supervision of logging operations;

• planning of post-logging operations; and

• detailed tactical logging map 1:2 000.

RIL is not only a technique to reduce the damage to the residual stand; it is also a procedure to optimize resource utilization through forest inventory and planning of harvesting.

Vanuatu Department of Forests. 1999. Vanuatu reduced impact logging guidelines. Vanuatu, Department of Forests.

The Vanuatu Reduced Impact Logging Guidelines were designed to reduce the impact of forest harvesting on soil and residual trees in comparison to damage levels incurred during conventional tractor logging in natural forests in Vanuatu. The main objective of RIL is to protect the regeneration and advance growth trees (potential crop trees) required for the next harvesting cycle and to minimize harvesting costs and optimize utilizable log volume. These objectives can be achieved through:

• appropriate pre-operational planning;

• careful implementation of acceptable harvesting practices and silvicultural prescriptions; and

• post-operational restoration and maintenance.

Webb, E.L. 1997. Canopy removal and residual stand damage during controlled selective logging in lowland swamp forest of northeast Costa Rica. Forest Ecology and Management 95: 117-129.

Selective logging is a harvesting system that produces disturbances similar to natural tree-fall gaps. Under optimal conditions selective logging does not significantly change forest structure, but stimulates natural regeneration and growth with the formation of gaps. A selective logging operation that utilizes methods (e.g. directional felling, planned extraction) to reduce damage to the residual stand is termed a controlled selection system.