To: Ronald M. Pape
Acting Deputy Director for Resource Management
California Department of Forestry and Fire Protection
Coast-Cascade Region Office
135 Ridgeway Avenue / PO Box 670
Santa Rosa, CA 95402

From: Scott Holmen, RPF 2238
VESTRA Resources, Inc.
962 Maraglia Street
Redding, CA 96002

October 12, 1998

RE: Redwood Coast Watersheds Alliance September 11, 1998 letter about THP 1-98-266 MEN and THP 1-97-445 MEN

Dear Acting Deputy Director Pape and CDF:

VESTRA Resources was responsible for the growth and yield analysis presented in Louisiana-Pacific's Sustained Yield Plans. Part of that analysis was the summarization of WHR habitats for the current mix if vegetation found on what are now Mendocino Redwood Company, L.L.C. lands in Mendocino county. I was responsible for designing and implementing the software used to make those WHR summaries. The Redwood Coast Watersheds Alliance appears to have misinterpreted some of the WHR information published in L-P's SYP 95-003 that has lead them to make some erroneous conclusions. I have been asked by MRC to provide you with some background on how this data can be more appropriately interpreted.

In their letter, RCWA states:
"Enclosed please find three pages of statistics taken from Louisiana-Pacific's proposed 'Sustained Yield' Plan (SYP 95-003) 'Wildlife Habitat Relationship' tables on Mendocino coast watersheds. These statistics show the following: 97% of L-P coastal forest lands are in 1-21 inch diameter trees. Of that 97%, 51%is in 11-16 inch trees, 20% in 16-21 inch trees, and 25% in saplings and pole trees. Only 3% of L-P forests contain trees of 24-32 inch diameter. L-P forest land is completely lacking in the 21-24 inch diameter class - the critical link to future old growth.

In Elk Creek watershed (18,000 acres, of which L-P owned 14,000 acres, now belonging to Mendocino Redwood Company), 73% of the L-P/MRC acreage is in 11-16 inch diameter trees, 8.6% in 16-21 inch, and 10% in saplings and pole trees. 8.4% is in 24-32 inch diameter trees. This were the statistics as of 1995.

Neither THP 266 nor THP 445 mentions these publically available statistics in its cumulative effects assessment. Very clearly, any further removal of the oldest trees, and of the 16-21inch trees that would eventually provide the missing diameter class (21-24 inch) and that would, then, eventually provide more old growth, should be disapproved by CDF if 'sustainable logging' has any meaning at all. "

The WHR summary tables which RCWA have based their analysis were included in the SYP to comply with CFPR 1091.4.5(c)(1) which specifies that the SYP shall contain "A summary table of Wildlife Habitat Relationships composition and stand structure types (WHR), or comparable stand structure types within forest vegetation types with their respective acreage. The WHR system, which is herein incorporated by reference, and its uses is described in "A Guide to the California Wildlife Habitat Relationships System", California Department of Fish and Game, March 1988."

The WHR system classifies habitats based upon the following habitat subdivisions: Tree-Dominated, Shrub-Dominated, Herbaceous-Dominated, Aquatic, and Developed. "Terrestrial vegetation habitats are classified using even-structure seral stages" (Garrison, et. al, 1996, Training Manual for the California Wildlife Habitat Relationships System). Tree-dominated habitats are then further classified based upon species composition, average tree size, and canopy density.

One method of computing the average tree size of a forest stand is referred to as quadratic mean diameter at breast height (QMDBH). QMDBH is a statistic used to determine the diameter of a tree of average cross-sectional area (as opposed to a simple arithmetic mean diameter). Most inventory summaries and forest simulation models report QMDBH as part of their standard stand description outputs. QMDBH is the recommended statistic for determining WHR size classes of tree-dominated habitats.

The algorithm used by Louisiana-Pacific in their SYP to predict WHR habitats from inventory data was developed at VESTRA and has been presented to the California Interagency Wildlife Task Force at several of their meetings. This group oversees the WHR system. Details of the methodology can be found on the VESTRA web site, The algorithm is deliberately conservative with regard to prediction of size classes, especially size 6, uneven-structured, multi-layered stands.

The WHR system was developed to provide a scientifically based methodology for evaluating the quality of wildlife habitats and to allow the comparison of habitats at the landscape level. One of the goals of the system is to " Provide a tool that predicts the potential of habitats to support wildlife and the effects of habitat changes. The habitat classification system was designed to identify and classify wildlife habitats, and it is not a vegetation classification system per se. The habitat classification system is an even-structure or even-aged system. However, uneven-structure stands can be described and analyzed using the CWHR system." (Garrison, et al, 1996, Training Manual for the California Wildlife Habitat Relationships System).

WHR and other vegetation typing systems try to assign useful descriptions to populations of plants that are somehow similar. Since real plant populations are often very complex, typing systems must summarize data into a few simple categories. On a property as large as Mendocino Redwood's there are hundreds of thousands of acres of habitat and millions of individual trees. The WHR system takes all of that individual diversity of soil, slope, rainfall, microclimate, management history, and individual tree variation and lumps it into half-a-dozen sizes and half-a dozen species descriptions. The WHR system simplifies all possible forested habitats in California into a one-page summary table.

A WHR classification of a stand such as DFR4M describes the average species mix and average tree in the population of trees in that habitat type. It does not and cannot describe the range of tree species and sizes found in that habitat type. If a similar system were used to describe the human population of counties in California, the county of San Francisco might be categorized as "female, 18 to 35 years old." This might be a totally accurate description of the statistically average person, but it does not even attempt to comment on the range of variation found throughout the population. Based on that categorization, it would be inappropriate to conclude there are no males older than 35 in San Francisco.

The SYP tables show that 51% of MRC ownership in Elk Creek is in habitats where the average tree diameter is 11-16 inches. That fact tells you nothing about the diameter distribution of trees in those habitats. The average tree size information found in the WHR tables should not be used to describe the distribution of tree diameters found in Elk Creek or anywhere else on the property.

There are two common diameter distributions for tree populations. Single-species, even-aged habitats tend to have "bell-shaped" uniform or Poisson distributions. The other common distribution is a negative exponential, commonly called an "inverse-J-shaped" distribution, with large numbers of small trees and fewer and fewer numbers of larger and larger trees.

The WHR habitat classification system assumes an even-aged or even-structured forest stand. Most of the trees in these populations are similar to the average tree, but some will be much larger and much smaller than the average. Even-structured stands are not common on Mendocino Redwood Company land. Many of their forest stands are not even-aged, and most of the true even-aged stands are not even-structured because of logging and management history. Few of their stands have a "bell-shaped" diameter distribution.

Many forest stands on MRC lands have an uneven-structure with an "inverse-J-shaped" diameter distribution. The inverse-J distribution is commonly found with uneven-aged stands or multi-species even-aged stands. Logging, especially selection logging, often encourages the development of stands with this type of diameter distribution. QMDBH and other average tree diameter statistics are almost meaningless for stands or habitats with this diameter distribution.

The WHR system is the best and most widely used habitat classification system in California, but it is an evolving system. It is currently undergoing a major revision. The problems associated with using even-structured habitat types to classify uneven-aged forest habitats have been presented by myself and others to the coordinator of the WHR program, Barry Garrison, California Department of Fish and Game. We all acknowledge that while it is an imperfect system, it is the best one we have.

We are pleased to see that the information provided in the L-P SYP is being reviewed and closely scrutinized by concerned citizens. After all, disclosure and public review is one of the intended purposes of the SYP. Unfortunately, it is also possible to misinterpret information contained in that document.

Logging of THP 445 will not remove the last large trees in the watershed. There is no 21 to 24 inch size class gap. There are thousands of acres of forest in Elk Creek outside the area in these two THPs which have significant numbers of trees greater than 20 inches in diameter. These trees will continue to grow larger and will help to provide wildlife habitat now and in the future.


Scott Holmen, RPF 2238

VESTRA Resources, Inc.