The Big Country Timber Sale was designed as a pilot project to test the feasibility of integrating production and conservation across the landscape.  This is stated in the HCP as a primary objective of the OESF (DNR HCP, IV.81).  Because the DNR is currently in the Restoration Phase of the HCP, silvicultural treatments were designed to promote late-successional forest structure. Four silvicultural treatments were tested: 1) Uniform Thin from below to RD 45-50 or no greater than 40% change from original RD, 2) Variable Density Thin at a fine scale, 3) Variable Density Thin at a coarse scale, and 4) Control (No action).  The first two treatments were based on DNR Procedures.  The third treatment was proposed by Richard Bigley as a more operationally feasible alternative to variable density thinning at a fine scale.             

                A preliminary analysis of proposed silvicultural prescriptions was done for each stand (based on FIU polygons and associated Forest Resource Inventory Summary (FRIS) data) in the Big Country Timber Sale and surrounding Clallam Block region.  The Forest Vegetation Simulator (FVS) growth model was used to project data to the current year and project the original and treated stands to a future point in time.  Calibration of FVS through the use of keyfiles was required to more closely simulate growth to local site conditions.

                After projecting each stand from the inventory base year (1992) to present, each silvicultural prescription was modeled in the Landscape Management System (LMS).  The proposed treatments were specified in DNR Procedure 14-006-080.  The treatments were: No action; Uniform thin to a relative density (RD) of 40 or no greater than 40% change from original RD; Heavy and light variable density thinnings where, for every 10 acres, 1 acre is treated to 0 trees per acre (gap), 1 acre of no touch (skip), 3 acres thinned to 23 to 33 RD, and 5 acres thinned to 33 to 45 RD; Heavy and light variable density thinnings with corresponding ratios of 1:1:4:4; and a regeneration cut.  All thinnings were from below, removing the smallest trees.  Because FVS is a spatially independent growth model that cannot support spatially explicit variable density thinnings, output values were estimated for these treatments by modeling each stand to a single density, then calculating a weighted average (based on the 10 acre target) of the skip, gap, and two thins.

            Outputs obtained from LMS for each stand for each prescription were harvest volumes in board feet per acre, average tree size of the harvested trees, average log size for both 16 and 32 foot log lengths from the harvested trees, and the quadratic mean diameter (QMD) of tail trees.  All volumes reported were scribner board foot volumes to a 4-inch top.  Assumptions used for the log calculations included a 4 inch small end diameter, 1 foot stump height, and 6 inches of trim.  QMD’s (based on trees per acre) was then used to determine the QMD for the stand.  Summary statistics (average, standard deviation, maximum, minimum, median) were also calculated on each output to determine outputs for the entire sale area.             

            The resultant data was used as inputs for various engineering and marketing analyses.  Summarized results are shown in the table below.

Table 5.  A summary of stand data for the planning area.  The values were found by taking the average of all the Research Inventory Units (RIU’s) and applying the various prescriptions to them.  All Volumes are in Scribner bd ft to a 4” top.

	Ave Harvest Vol/Acre (MBF)	Ave Tree Size (bd ft)	Ave Log Size (bd ft)	Tail Tree Sizes (QMD Average)
Uniform Thin	10.6	77	69	20.4
Light Thin	17.4	95	76	20.4
Heavy Thin	23.6	114	88	20.4
Regeneration Cut	38.0	163	117	20.4

 

            Further modeling could be done to analyize the effect of each silvicultural treatment on desired objectives through time.  By defining any objective based on inventory-level measurable criteria, original, treated, and projected inventory can be analyzed to determine how many acres on the landscape meet each objective.  Understanding the limitations of the models used is important.  FVS is a spatially independent growth model, and cannot model many of the characteristics of a spatially explicit operation like variable density thinning.             

            All stands in the planning area have a primary age of approximately 70 years old. The primary species in the area is western hemlock.  Other species present include Douglas fir, red alder, western red cedar, and sitka spruce.               

            Turn weights were calculated using the average log size (32 ft log) and tree length, an average density of 10.9 lbs/bf, and 3.5 logs per turn for cable yarding.               

            Tail trees sizes were based on the largest trees at a 30 by 30 foot spacing (50 trees per acre) including douglas-fir (DF), western hemlock, red cedar, and sitka spruce.  The QMD of DF was determined separately from all other species.  The QMD for other species was reduced by 2” to account for reduced strength properties compared to DF.  The weighted average of DF and other species for all FIU’s in the planning area was found to be 20.4”.