Wood waste during full ‐ length and cut ‐ to ‐ length harvesting systems in caspian forests

Behjou et al/ International Journal of Advanced and Applied Sciences, 3(5) 2016, Pages: 52‐56 53 1. determining wood waste during bucking, skidding, loading or decking operations in Caspian hardwood sites; 2. Identify how operational factors such as harvesting system, function, tree species, and log size affect damage; 2. Materials and methods Wood waste during logging operations was studied with two harvesting systems on four different Caspian hardwood forest sites. Harvesting functions investigated included bucking, skidding, decking, and loading. Two harvesting systems including chainsaw bucking, skidding with a cable skidder, decking/loading with loader were analyzed (Table 1). The operator's experience in using chainsaws varied from 5 to 16 years (Table 1). The harvest sites were located in the north of Iran and had similar species composition, average slope, and stand conditions (Table 2). The tract size ranged from 46 to 66 ha with average slopes of 20 45%. Average residual stand density for the four tracts was approximately 171-197 trees/ha with an average DBH ranging from 41.4 to 51.2 cm and average merchantable height ranging from 10.5 to 13.1 m, respectively. Selective cuttings were conducted on all sites and harvest intensity was 20.8m3 per hectare. The major species sampled were beech (Fagus orientalis), hornbeam (Carpinus betulus), alder (Alnus subcordata), maple (Acer velutinum), and elm (Ulmus glabra) (Table 2). Table 1: Harvesting machines and operators' experience by site Site 1 Site 2 Site 3 Site 4 Full-length system cut to length system Full-length system cut to length system Bucking Machine Chainsaw Chainsaw Chainsaw Chainsaw Experience 7 11 13 9 Skidding Machine Cable skidder Ranger Cable skidder Ranger Experience 13 14 6 7 Loading Machine Loader Loader Loader Loader Experience 14 11 8 11 Table 2: Conditions for sites, harvests, and machines in the field study Site 1 Site 2 Site 3 Site 4 Harvest system Full-length cut to length Full-length cut to length Harvest method Selective cut Selective cut Selective cut Selective cut Harvest season Winter summer Winter summer Winter summer Winter summer Slope (%) 35-39 20-30 28-35 39-45 Aspect N-NW N-NW NE NW Total sampled log 100


Introduction
*The forests in northern Iran are most often harvested by one of two methods including fulllength and cut to length harvesting systems. On the other side, there have been substantial increases in demand for veneer type products from hardwood species. In most of the cases, logging operations in the region are along with high amount of wood waste and this is one of the problems in logging operations. Williston (1979) found that breakage and skidding damage associated with harvesting operations destroyed almost 6% of the total value of harvested logs. McNeel and Copithorne (1996) stated that species is a factor in defining the amount of breakage expected during harvest. Damage to harvested trees or logs can occur during the felling, bucking, skidding, decking, loading, and hauling functions of the timber harvesting process (Wang et al., 2004). McMorland and Guimier (1984) found that shears cause very many but shallow splits, while saw chains cause less frequent but deeper splits. Greene and McNeel (1989), Faust and Greene (1989) reported log damage by feller-bunchers with both shear and saw heads and found that the damage usually occurred in the first twelve inches of the butt log. Even small improvements in volume recovery could lead to large improvements in financial gains (Boston and Dysart, 2000). Han and Renzie (2005) found that feller-buncher felling results in greater wood volume waste as a results of a thicker saw blade kerf than dose chan saw felling. Wang et al. (2004) found that felling resulted in significantly more log damage when compared to skidding, decking and loading operations. Hall and Han (2006) found that the average stump height of mechanized felling is 5.8cm (17%) lower than manual felling. Unver and Acay (2009) developed a prediction model bye consideration friction vector affecting skidded log, friction surface area of the log and skidding distance parameter. Improper bucking might not damage the log in a physical sense, but it could damage the potential value gained from bucking correctly (Sessions, 1988;Pickens et al., 1992;Zavala, 1995). The implementation of correct log bucking procedures can potentially increase the yield from each tree harvested, which correspondingly can improve profits for the logger and sawmill. Dragging the logs over the ground could present some type of potential damage to the harvested logs (Wang et al. 2004).
Damage to hardwood logs is especially important in the Caspian hardwood forest region, where much of the hardwood timber harvested is sawn into grade lumber or turned into veneer or veneer-based composites. The objectives of this study were to: 1. determining wood waste during bucking, skidding, loading or decking operations in Caspian hardwood sites; 2. Identify how operational factors such as harvesting system, function, tree species, and log size affect damage;

Materials and methods
Wood waste during logging operations was studied with two harvesting systems on four different Caspian hardwood forest sites. Harvesting functions investigated included bucking, skidding, decking, and loading. Two harvesting systems including chainsaw bucking, skidding with a cable skidder, decking/loading with loader were analyzed (Table 1). The operator's experience in using chainsaws varied from 5 to 16 years ( Table 1).
The harvest sites were located in the north of Iran and had similar species composition, average slope, and stand conditions ( Table 2). The tract size ranged from 46 to 66 ha with average slopes of 20 -45%. Average residual stand density for the four tracts was approximately 171-197 trees/ha with an average DBH ranging from 41.4 to 51.2 cm and average merchantable height ranging from 10.5 to 13.1 m, respectively. Selective cuttings were conducted on all sites and harvest intensity was 20.8m 3 per hectare. The major species sampled were beech (Fagus orientalis), hornbeam (Carpinus betulus), alder (Alnus subcordata), maple (Acer velutinum), and elm (Ulmus glabra) ( Table 2).  In addition to site and stand conditions, variables measured for each sawlog in the field included bottom and top diameter, length, tree species, damage type, damage location, damage dimensions, log grades, and comments for each individual log. Bottom and top diameters were measured in centimeters and the length of each saw-log was recorded in meters.
A total of 1000 hardwood logs from each of the four harvesting sites were sampled for measurable damage sustained during the harvesting operations. Of the 1000 observations per harvesting site, 250 observations were made for each of the four harvesting functions: bucking, skidding, decking, and loading. Sampling was done one function at a time. Damage was evaluated before and after each function to eliminate double counting of any damage caused by previous function. Bottom diameters of observed logs were between 73.4 and 96.9cm while top diameter varied from 21.1 cm to 31.3 cm. The length of observed logs ranged from 4.3 to 7.9 m and the log volume ranged from 1.01 to 1.45 m 3 .

Results
There were 94 logs that had wood damage and 451 logs that had bark damage, which accounted for 2.35% and 11.27% of the total 4000 observed logs, respectively (Table 3). Damage volume was a volumetric measurement of the damage area and was calculated based on the length, width, and depth of the damaged area at the greatest point. The grade of each log was determined based on the log's dimensions and the visual defects of the log (Carpenter et al., 1989;Hanks et al., 1980). Saw-log damage was analyzed in terms of volume losses. Volume loss was computed based on the damage dimensions. Pre-and post-damage log grades were given to all logs based on their dimensions, superficial condition, and damage severity. Damage volume was deducted from a log with slight damage while both volume deduction and degradation were considered for a severely damaged log. A log pricing system was developed based on current saw-log market prices from the local hardwood lumber industry and Tehran (Capital city of Iran) timber market reports (Emanuel and Rhodes, 2002). A monetary volume was assigned to the log based on its species, grade, and current hardwood saw-log market prices. This system then assigned a postdamage volume loss and monetary value to each log based on the amount of damage the log had sustained. Each log then had a pre-damage and a post-damage monetary value associated with it. Log size was computed based on its volume and grouped into five different categories of 1.0, 1.1, 1.2, 1.3, and 1.4 m 3 . In order to compare the log damage, percentage of volume loss were defined and computed based on the ratios of damage volume over log volume, respectively.
The data was analyzed by using Statistical Analysis Systems (SPSS). The general linear model (GLM) procedure, which is a type of analysis of variance (ANOVA), was used to determine if significant differences existed among harvesting systems, harvesting functions, tree species, and log sizes (Wang et al., 2004). The GLM can be expressed as follows (Eq. 1): V ijklm = μ + HS i + HF j + SP k + LS l + HS i * SP k + HF j * SP k + HS i * LS l+ HF j * LS l+ SP k *LS + ε ijklm i= 1, 2 j= 1,2,3,4 k= 1,2,…,7 l = 1,2,…,5 m= 1,2,…,n where V ijklm represents the mth observation of the volume loss for a log; μ is the mean of response variable; HS i is the effect of ith harvesting system; HF j is the jth effect of harvesting function; SP k is the kth effect of species, LS l is the lth effect of log size; ε ijklm is an error component for all uncontrolled variability; and n is the number of observations within each treatment. The interactions among harvesting systems, functions, tree species, and log sizes were also considered in the model. The response variables were tested with Tukey's Test at the 5-percent level to determine if there was any significant difference of damage volume loss among harvesting systems, functions, tree species, and log sizes.
Gouge was the most common damage type found in the field study. However, split and slab caused the most volume loss to logs. Split caused 6.6% of volume loss while slab resulted in 6.2% volume loss (Table 4). The percentage of volume loss per cubic meter was significantly different between the two studied harvesting systems. Full-length systems caused a loss equal to 15.6% of a log's volume while cut-tolength systems lost 35.54% of volume (Table 5). Bucking damage caused 12.2% volume loss of the logs, which differed significantly from the damage caused during skidding, decking, and loading operations. However, log damage was not significantly different among skidding, decking, and loading operations ( Table 5).
The percentage of damaged logs by species ranged from 0.7% for hornbeam to 2.4% for maple (Table 5). Damage to beech resulted in 1.5% loss of the volume, also damage to elm resulted in 1.1% volume loss (Table 5). Volume loss of damaged logs was not significantly different between alder and maple but was significant among these two species (alder and maple) with other species (beech, elm, and hornbeam) ( Table 5).
The percentage of volume loss varied as the log size decreased from 0.5% to 4.5% based on different length of logs (Table 5).

Discussion
The majority of the volume loss was caused during the bucking function when using a cut-tolength harvesting system. The damage caused by bucking operations in cut to length harvesting system overshadows damage from all other parts of the timber harvesting process. Results suggest that logs skidding, decking and loading have very little impact on damage levels (Table 5). Except for alder and maple all other major hardwood species in this region consisting of beech, hornbeam, and elm showed similar volume loss (Table 5). Alder and maple presented a significant difference in volume loss compared to the other three species observed. One reason for this could be the mechanical properties of Alder and maple. It has been suggested in past studies that more brittle species (such as alder and maple) have a greater susceptibility to damage than other species (such as beech, elm, and hornbeam) showing more resistance. Another reason for differences among species could be the time of harvest. Although, in this study all observations were made during the winter, spring and summer.
Another species-specific point to consider is the hardwood market. As demand for certain species rises and falls, more or less of that species may be harvested. Also, a high demand may generate greater care in harvesting certain species, whereas a low demand may be have the opposite effect.
Damage to saw-logs is occurring during harvesting operations, but to a lesser degree than might be perceived by the wood products industry. Certainly, there is room for improvement of motormanual felling operations. This could be accomplished by formal chainsaw felling instruction, and by paying more attention to value and grade, rather than volume alone. Damage to logs is likely to vary from site to site, and maybe even within a single site. These variations may be driven by geographic factors not considered in this study, which focused on north of Iran.
The information gained from the field study is important in creating new guidelines or training to help minimize hardwood log damage occurring during the timber harvesting process. It would be beneficial to determine how log damage relates to skid trail density, road capacity, multiple landings and their locations, and length of skid. Along with these factors, terrain slope could be analyzed more closely, especially its correlation with the occurrence of switchbacks that may damage logs.
Implementing new and better guidelines or training requirements for fellers could create monetary gains for loggers while better utilizing the forest resources by not damaging valuable hardwood logs in Caspian forests.