Adaptive Management – 2000 – Part 2

2/6/2001

• Appendix C - Topic Evaluation Summary Forms
• Topic #1 - Snag Levels and Methods
• Topic #5 - Snag Creation in Landscape Areas 2 & 3
• Topic #17 - Down Wood Levels and Methods
• Topic #18 - Snag and Log Creation
• Topic #4 - Road restoration strategy
• Topics #6 and 7 - Substrate and water quality source areas
• Topic # 9 - Refugia objectives and activities
• Topic # 13 - Adjustments to aquatic reserves
• Topic # 21 - Re-evaluation of Aquatic Conservation Strategy Objectives
• Topic #3 - Spatial variability of retention trees
• Topic #8 - Commercial Thinning Regimes
• Topic #11 - Crown Closure
• Topic # 15 - Prescriptions for LA-1
• Topic #: 19 - Complex prescriptions - fire history
• Topic # 14 - Adjustments to landscape area one boundaries
• Topic # 16 - Wetlands
• Topic # 20 - Prescribed Fire

Appendix C - Topic Evaluation Summary Forms

Blue River Landscape Study

Topics are presented in the same order as in the rest of the document, as listed below:

Dead Wood 1. Snag levels and methods 5. Snag creation in landscape areas 2 & 3 17. Down wood levels and methods 18. Dead wood creation in 100% retention areas Aquatic conservation objectives 4. Road restoration strategy 6. Water quality source areas 7. Large wood and coarse sediment source areas 9. Refugia objectives and activities 13. Adjustments to aquatic reserves 21. Aquatic Conservation Strategy Objectives analysis Timber harvest prescriptions 3. Spatial variability & spatial pattern of retention trees 8. Commercial thinning regimes 11. Use of crown closure 15. Canopy cover level in landscape area one 19. Complex prescriptions - fire history Other topics 2. Lands suitable for timber production 14. Adjustments to landscape area one boundaries 16. Wetlands 20. Prescribed fire

Topic #1 - Snag Levels and Methods

Date: 5/24/00

Reviewers: Geary, Mayo, Seitz, Overton, Ford, Swetland.

Compiler: Ruby Seitz

Questions: What snag level should be prescribed and why?

Reason for question: Dead trees are briefly discussed in the BRLP (p.14) but there is a need to review the current prescription process.

Relevant Study Component: Snag management

Adaptation Options:

• Continue to manage for snag habitat based on the 40% population level as a minimum, and vary prescriptions slightly based on Landscape Area.
• Use a method based on a % of overstory retention based on Landscape Area.
• Base snag habitat level on naturally occurring snag habitat levels.
• Base snag habitat level on calculated snags needed to support primary cavity excavator species guild on the Willamette National Forest (4/acre)

Evaluation: Snag habitat management has undergone a high level of change in the past decade, from the initial success of leaving standing green trees and hard snags in logged units to experimentation with various snag creation techniques.

Clearcut logging combined with fire suppression activities for the past several decades have resulted in fewer snags across the Blue River landscape. Continued logging, even with varied levels of overstory retention and longer rotations, will impact snag habitat levels. Snag creation activities are an attempt to mitigate these effects.

Fire suppression has also greatly reduced snag habitat creation across the landscape. The extent is unknown but assumed to be quite high. Prescribed burning may create additional snag habitat to mitigate the loss due to ongoing intensive fire suppression.

At the present time, the minimum snag prescription level on the Blue River Landscape has been at the 40% population level which is the same as the ROD requirement. The Willamette National Forest Plan requires 40% to be met at the subdrainage level with a minimum of 20%. Other standards for snags can be found in FSM 2630.3, which recommends exceeding 40% as a management objective. Region 6 of the Forest Service has given a minimum management objective of 20%.

The 40% population level is based on 4 snags/acre meeting the needs of the primary cavity excavator guild on the Willamette National Forest (WNF Forest Planning Process Paper). This snag level of 4/acre coincides with the level with data from Willamette National Forest ecology plots which indicate an average density of 3.9 hard snags/acre greater than 17 inches dbh (WNF Forest Planning Process Paper). However, this figure appears to be low based on a thorough review of the literature conducted recently by Eugene BLM (Rebecca Thompson, Forester), Willamette Province LSR Assessment, Pam Wright’s Thesis, and more recent data from ~250 ecoplots on the Willamette National Forest which shows an average of 7.9 snags/acre greater than 18" dbh. Snags must be surveyed more intensively than vegetation because they generally occur in a clumped pattern in natural stands.

Snag Ranges in Mature and Old-growth Stands (Willamette LSR Assessment, page 31). Only snags greater than 20" dbh and 16’ tall were counted.

Per Acre Snag Occurrence	Plant Series and 100% Snag Level Ranges
	Pacific silver fir	Western hemlock
Mature Stand	14-29	5-21
Old-Growth Stand	18-43	13-42

Other Options for Snag Habitat Management

Target snag ranges are shown below at the 40% minimum level based on plots taken in mature and old-growth stands (Willamette LSR Assessment, page 31).

Per Acre Snag Prescriptions	Plant Series and 40% Snag Level Ranges
	Pacific silver fir	Western hemlock
Mature Stand	5.6-11.6	2-8.4
Old-Growth Stand	7.2-17.2	5.2-16.8

Brown (1985) calculated snags per acre needed to support maximum populations (100%) of primary excavators on the Willamette National Forest. This resulted in 3.77 snags per acre at low elevations, and 4.12 snags/acre at high elevations.

Recommended Adaptations with Rationale: Snag levels will be prescribed based on a %age of the overstory canopy before harvest by Landscape Area. This will account for varying stand types and diameters. Creation of these snag levels is reasonable based on future expected KV funding and an expectation that some of the snags would be created by post-harvest prescribed fire.

Prescribed Snag Levels by Landscape Area

	Prescribed Overstory Retention Level	Example Trees per Acre Left After Harvest	Additional % Retention for Snag Habitat Based on Overstory Canopy Before Harvest	Example Trees per Acre for Snag Creation	General Guideline
Landscape Area 1	50%	35-50	5%	5	Lowest number of snags
Landscape Area 2	30%	25-35	10%	10	Moderate level of snags
Landscape Area 3	15%	10-20	15%	15	Big pulse of snags

All Landscape Areas. Create high snag levels in Class I and II riparian reserves and the aquatic reserves as previously cut stands recover. Snags should be inventoried and created to meet a goal of four/acre.

There may be special circumstances for all or a portion of a particular unit which may require a snag prescription higher than what the table shows, for example, Survey and Manage or TES species locations which would benefit from snag creation.

With implementation of the Blue River Landscape Plan over time, snag habitat will recover, but this will take several decades. Until that point in time is reached, a somewhat higher level of snag creation is appropriate.

Aspect: The creation of lower numbers of snags on north aspects, and higher numbers on south and west aspects was considered, but recent ecology plot data shows that this phenomenon does not occur in the western Cascades (McCain).

General Guidelines for Snag Habitat Creation during Harvest Activities

Harvest Activity	Stand Health and Snag Creation Level
Regeneration	High disease levels (more inoculation, 1/3 of created snags). Create high numbers of snags.
Commercial Thin at ~40 years	Less disease, less windthrow (low levels of inoculation or none), more competition (girdling). Create fewer snags.
Intermediate Commercial Thinning at ~ 80 years	Intermediate disease levels (moderate inoculation levels, 1/6 of created snags). Create moderate snag levels.

Snag Species Mix: Species mix will lean towards Douglas-fir because these snags last the longest, but hemlocks, true firs and rarely cedar may occasionally be chosen for snag creation if they are abundant in the stand. It is unknown if a fire would create more western hemlock or western red cedar snags, or if these species would be more likely to burn. Some snag creation should include these species.

Snag Sizes: Create snags in size groups using the bigger is better philosophy. It is preferable to create snags in size groups if the remnant stand allows. An example is three size classes of 18-23, 24-28, 29-35" or greater DBH ranges. Snag creation of the older, larger age class should occur in proportion to the percentage of this age class which is removed. For example, if 20% of the largest Douglas-fir age class is logged, approximately 20% of snags created should also come from this age class. Snag creation in the medium and smaller age classes should also approximate the percentage which was removed by logging to a certain point. If the smaller age class contains trees below 12" dbh, there are few cavity nesters which will nest in trees of that size. Those snags, if not charred by fire, would be available as foraging habitat only. Because of this, there should be a tradeoff of the percentage in this smaller age class for more snags in the medium age class.

Smaller sized trees of 14-17" dbh may be chosen for snag creation if no larger trees are available in commercial thinning stands. If there are very few overstory remnant trees, these should not be chosen for snag creation.

Snag Creation Implementation: Snags will be cruised approximately four years after completion of logging. This will result in information about total snag numbers, size ranges, decay class, and weakened trees that may soon become future snags. The above snag levels show the final targets after logging and implementation of snag creation. If snags greater than 18" dbh, 40’ height, and decay classes I or II are present after completion of logging, this number will be subtracted from what needs to be created. If the remnant stand shows a high level of dead-topped, broken topped, or heartrot infected trees, numbers of snags to be created may be adjusted if they are at a fairly high level. Each of these weakened trees counts towards _ an inoculated tree, because it is unknown when they will become useful snags.

Future Data Needs to Benefit Snag Habitat Management:

Improved data for naturally occurring snag habitat levels on the Willamette National Forest

Is there an association between levels of conks on live trees and their future lifespan?

Should the type of snag habitat creation be variable depending on the Landscape Area? (see Snag Habitat Implementation Plan)

References:

Brown, E.R. 1985. Management of Wildlife and Fish Habitats in Forests of Western Oregon and Washington, U.S. Department of Agriculture, Forest Service, Pacific Northwest Region.

Mellen, Kim; Ager, Alan. Coarse Woody Debris Model — Version 1.2 — June 1998. U.S. Forest Service, Portland, OR.

Mid-Willamette LSR Assessment, 1998.

Thompson, Rebecca. Draft Snag Management Recommendations, Middle McKenzie Landscape Plan 9/13/99. Eugene District BLM.

Wright, Pam. Down Wood Model, MA Thesis, Oregon State University, Corvallis, Oregon. 1997.

Snag Habitat Implementation Plan

April 13, 2000

At the present time, the minimum prescription level on the Blue River Landscape has been 40% which is the same as the ROD, C-42, which states: "As a minimum, snags are to be retained within the harvest unit at levels sufficient to support species of cavity-nesting birds at 40 percent of potential population levels based on published guidelines and models. The objective is to meet the 40 percent minimum standard throughout the matrix, with per-acre requirements met on average areas no larger than 40 acres." This guideline is the same as that in the Willamette National Forest Plan, FW-122 states "Habitat capability for primary cavity excavators shall be maintained to provide for at least 40% or greater potential populations. Habitat shall be provided and monitored at the subdrainage level." The other guideline which applies is FW-125, "All timber harvest units shall provide snag habitat capable of supporting at least 20% or greater potential populations of cavity-nesting species." Guidelines for the type and size of tree which counts towards a snag is also discussed in FW-129 and 130. This guideline is not being reviewed with this process, however, snags which are less that 18" dbh are counted towards meeting prescribed snag habitat levels after commercial thinning operations or when the size of trees available in the stand does not allow 18" dbh to be met.

Other standards for snags can be found in FSM 2630.3, which recommends exceeding 40% as a management objective. Region 6 of the Forest Service has given a minimum management objective of 20%.

Snag Creation Method Effectiveness: Although snags have been artificially created since the late 1980s, there is a considerable knowledge gap in terms of the effectiveness of the various snag creation methods being used today. Monitoring efforts and documentation are just beginning and there are no large-scale published reports or results available. At this point in time, it appears best to continue to use a variety of methods to best approximate what nature does:

Sawtopping or blasting to simulate wind mortality

Inoculation to initiate stem decay and then long-term disease mortality

Girdling to simulate competition mortality

Prescribed underburns to create fire mortality

Pheromones to cause insect mortality

Plans for 2000 or 2001 are to test snag creation using pheromones which will attract bark beetles to kill the trees. Preliminary information from eastern Oregon shows that tree-baiting with bark beetle pheromones may be less expensive, less hazardous, and a more natural method than topping. Trees killed by bark beetles will be structurally intact and they are immediately inoculated with a variety of microorganisms carried by the beetles. Also, bark beetle pheromones are attractive to a variety of other insects that increase the rate of deterioration and provide a source of food for many other animals (Ross, 1997).

In some cases, pheromones may be used several years after earlier inoculation. Most of the stem heartrots which have been inoculated in Douglas-fir (Phellinus pini, Fomitopsis pinicola, Fomitopsis cajanderi, and Fomitopsis stereum) can all infect and grow in both live and dead trees. Thus, once the pheromones attract enough bark beetles to kill the trees, the heartrot would continue to survive. Until we determine the most effective methods for snag creation, the Blue River Landscape Study and AMA will be an area of creative and intensive experimentation and monitoring.

Amount of each Snag Creation Method: The type of tree mortality in a natural landscape is highly variable and changes over time. Across an entire landscape it seems more possible to consider ranges of mortality by cause in determining the type of snag creation method to choose. The following data was found pertaining to type of tree mortality:

1958 data from dead wood conference: 40% insects, 20% disease, and 5% fire mortality of trees. The remaining 35% mortality is assumed to be wind and competition.

Disease:

Fire: Pulses over time.

Windthrow or uprooting:

Less than one percent of annual tree mortality in stands less than 120 years old (Cline et al. 1980)

Seven percent average (range 3-12 percent) of tree mortality in small, successional Douglas-fir (Graham, 1981)

Suppression or competition: "Major" source of snag creation in unmanaged, even-aged stands resulting from fire or regeneration harvest (Cline et al. 1980).

Insects: Regional data shows high annual fluctuations for insect mortality. Flights are conducted almost annually through a cooperative effort between the Forest Service, National Park Service, Washington Department of Natural Resources and Oregon Department of Forestry. Acres are total numbers of trees of insect mortality by species are available on a GIS map, and were totaled for Oregon and the Blue River Ranger District. Regional entomologists estimate that twice as many trees were probably killed as what was counted. The main use for this data is to show that insect mortality plays a significant role in our ecosystem, and provides a multitude of benefits to wildlife, especially snag-dependent cavity nesting birds. Insect mortality is being suppressed in managed forests because logging and thinning results in a higher proportion of healthy, vigorous trees. Use of insects to create snags should be fully explored as soon as possible.

Snag Development based on Tree Mortality Agent

Natural Snag Creation Agent	Artificial Snag Creation Method	Mortality Timeframe	Decay Process	Type of Snag Created	Expected Snag Use by Primary Cavity Excavators
Disease	Inoculation/heartrot	None or long-term	From the inside out	Soft snag and/or hollow tree. Inoculation also benefits use of future down wood.	Nesting and foraging after at least 5 years (unknown)
Fire	Prescribed underburn	Immediate or later	From the outside in	Hard snag if immediate and tree is only burned to the point of death and not completely charred. Could be soft snag if tree is weakened, allows internal rot to begin and slowly dies.	Foraging could be immediate, and possibly nesting after 10 or more years
Wind	Sawtopping or blasting	Immediate or later depending on number of whorls and branches left	From the outside in	Hard snag	Foraging after 5-10 years
Competition	Girdling	Immediate	From the outside in	Hard snag	Foraging after 5-10 years
Insect attack	Pheromones/bark beetles	Unknown for Westside Cascades but probably immediate	From the outside in	Hard snag	Foraging as soon as beetles congregate and improved over time as secondary insects are attracted
Disease and subsequent insect attack	Inoculation/heartrot and pheromones/bark beetles a few years later	Immediate after pheromone/bark beetle treatment	From the inside out and outside in	Soft snag and future soft log	Nesting and foraging
Disease and subsequent competition mortality	Inoculation/heartrot and girdling a few years later	Immediate after girdling	From the inside out and outside in	Soft snag and future soft log	Nesting and foraging
Disease and wind	Inoculation/heartrot and concurrent topping	Immediate or over time ("time capsule treatment")	From the inside out and outside in	Soft snags and hollow trees, future soft and hollow logs	Nesting and foraging

Each of the above treatment methods may also include artificial cavity creation which may provide immediate nesting habitat. However, this treatment has not been extensively used due to the additional cost. Trees selected for artificial cavity creation should have a minimum of 24" at the site of cavity creation to avoid the risk of tops breaking off during windstorms.

Funding: Snag creation numbers may in some instances depend on funding availability. KV collections have generally paid for snag creation but if there are other, higher priority projects which draw on the KV funding pool combined with a relatively low cost-benefit sale ratio, snag creation may not be entirely funded. This has occurred once so far on the Blue River Ranger District for the North Fork Quartz Timber Sale, due to helicopter yarding, high canopy retention levels and other projects competing for limited KV funds. Snag creation was only funded for half the level requested. At this point in time, snag creation methods used have all required treeclimbing, and bid costs have been similar with girdling at the lower end (~$28-35/tree), and sawtopping, blasting, and inoculation at a slightly higher cost (~$35-40/tree). Pheromone treatments to attract bark beetles and kill trees may offer a future lower cost alternative. With prescribed fire, snag creation would not be an additional cost, but there is uncertainty about the use of fire-killed trees and providing only this type of snag will not provide the full range of ecological benefits.

R.Seitz

References

Cline, S.P.; Berg, A.B.; Wight, H.M. Snag characteristics and dynamics in Douglas-fir forests, western Oregon. J. Wildl. Manage. 44(4):773-786; 1980

Graham, R .L.L. Biomass dynamics of dead Douglas-fir and western hemlock boles in mid-elevation forests of the Cascade Range. Corvallis, OR: Oregon State University, 1981. 152p. Dissertation.

Ross, Darrell W. Using Aggregation and Antiaggregation Pheromones of the Douglas-Fir Beetle to Produce Snags for Wildlife Habitat. West. J. Appl. For. 1997. 12(2):52-54.

Topic #5 - Snag Creation in Landscape Areas 2 & 3

Date: 5/24/00

Reviewers: Geary, Mayo, Seitz, Overton, Ford, Swetland.

Compiler: Ruby Seitz

Question: Should snags be created in Landscape Areas 2 and 3?

Reason for question: Snag habitat has been lost due to fire suppression, logging, and salvage for the past several decades. Currently, snags can be created at a contract cost of approximately $40/tree, but there are considerable administrative costs associated with contract implementation which combined with contract costs may reach $60/tree. Because Landscape Areas 2 and 3 will have a 180 and 260 year rotation age, and leave 30% and 15% overstory retention, the question arose whether the relatively costly technique of snag creation should still be used, or perhaps less snag creation should occur.

Relevant Study Component: Snag management

Adaptation Options:

• Create snags within Landscape Areas 2 and 3.
• Create fewer snags within Landscape Areas 2 and 3.
• Do not create snags within Landscape Areas 2 and 3.

Evaluation: The decision for Landscape Areas 2 and 3 should consider the following:

• Need for well distributed snag habitat throughout Landscape Areas 2 and 3
• Stage of forest stand development and naturally occurring snag levels
• Snags remaining after monitoring timber sales with 30% and 15% overstory retention
• Lifespan of snag habitat and frequency of proposed future thinning operations and expected logging systems, which may affect snag retention.
• Potential results of prescribed fire on snag creation for Landscape Areas 2 and 3
• Effects of fire suppression on reduction of snag habitat across the landscape
• Timeframe for acceptable snag levels throughout the rotation, with and without snag creation

More information about forest stand development and naturally occurring snag levels is contained in Topic #1, snag levels and methods.

No snag monitoring of timber sales with 30% and 15% overstory retention has occurred. A cursory review of the Augusta timber sale showed fair numbers, estimated at approximately one/acre, of large diameter snags that were retained after logging and burning, but more data collection is needed to better assess this.

Less intermediate thinnings will result in less of a need for snag creation. As snags are naturally created and become used by wildlife, falling due to the safety hazard to logging operations will be less frequent.

As more experience is gained using prescribed fire under remnant stands with 30% and 15% overstory retention, information will become available about snag levels created due to fire. Historic natural fire in combination with other tree mortality processes would have resulted in a far greater number of snags than currently exists on the landscape.

Medium to large diameter snag levels with a regeneration harvest and one thinning entry throughout rotations of 100, 180, and 260 years, with and without snag creation

Point in Time	Stand Activity	Snag Levels without Snag Creation	Snag Levels with Snag Creation
Year 0	Regeneration harvest	Doesn’t meet prescribed level	Meets prescribed level
Year 40	First thinning entry	Doesn’t meet prescribed level	Meets prescribed level
Year 60	None	Doesn’t meet prescribed level	Meets prescribed level
Year 80	None	Partially meets prescribed level	Meets prescribed level
Year 100	None	Partially meets prescribed level	Meets prescribed level
Year 180	None	Partially meets prescribed level	Meets prescribed level
Year 260	None	Meets prescribed level	Meets prescribed level

Note: Snag levels for Landscape Area 2 are considered moderate (6-10% of original stand), and levels for Landscape Area 3 are considered high (11-15% of overstory retention trees).

Medium to large diameter snag levels with a regeneration harvest and three thinning entries throughout rotations of 100, 180, and 260 years, with and without snag creation

Point in Time	Stand Activity	Snag Levels without Snag Creation	Snag Levels with Snag Creation
Year 0	Regeneration harvest	Doesn’t meet prescribed level	Meets prescribed level
Year 40	First thinning entry	Doesn’t meet prescribed level	Meets prescribed level
Year 60	2nd thinning entry	Doesn’t meet prescribed level	Meets prescribed level
Year 80	3rd thinning entry	Doesn’t meet prescribed level	Meets prescribed level
Year 100	None	Doesn’t meet prescribed level	Meets prescribed level
Year 180	None	Partially meets prescribed level	Meets prescribed level
Year 260	None	Meets prescribed level	Meets prescribed level

Recommended Adaptations with Rationale: Snag creation should continue to be used in Landscape Areas 2 and 3 to avoid unacceptably low snag levels for 80 years after logging. Longer rotations require less snag creation during intermediate thinning entries because overall snag levels will be achieved after the last entry and continue to recover throughout the rest of the rotation. The prescribed number of snags should be created at the time of the initial regeneration harvest. With one thinning entry, snags should be monitored in 40 years and if levels do not meet the prescription, they should be created at that time. With more than one thinning entry, a very low level of snag creation would occur during the intermediate thinning entries to replace those lost during the logging operation. If snag levels drop below half the prescription, snags should be created at that time. During the final thinning entry, snags will be created to achieve the required number originally prescribed during the regeneration harvest.

An example:

A block in Landscape Area 3, 15% retention harvest, and a 260 year rotation on a 180 year old, high elevation stand in Year 2000. Approximately half of the existing snags are located in retention clumps and can be retained by the helicopter logging operation. The safety hazard requires the remainder to be cut. The naturally occurring snag level is interrupted and existing snags felled should be replaced by artificial snag creation methods. If prescribed fire is planned, additional snags may be created which can reduce the need for mechanical (girdling, sawtopping) and biological (inoculation, bark beetle pheromones) snag creation methods. However, it is not preferable that all snags are created by fire. Natural stand development would continue to create additional large snags from the overstory. Since this process has been interrupted, the time of regeneration harvest can be used to create snags. If only one commercial thinning entry is planned for Year 2040, snags can be monitored at that time and created if prescribed levels are not met from both the overstory and understory trees. If additional commercial thinning entries are planned for Years 2060 and/or 2080, it is preferable to not create many additional snags at that time. Only those lost due to the thinning operation should be replaced at this time. During the final thinning entry, snag numbers will again be increased to meet the original prescribed per acre number. Recovery of natural snag development processes would occur during the final 100-180 years of the rotation. If prescribed fire is allowed, additional large snags are likely to develop during this time.

Topic #17 - Down Wood Levels and Methods

Date: 5/24/00

Reviewers: Geary, Mayo, Seitz, Overton, Ford, Swetland.

Compiler: Ruby Seitz

Question: What down wood level should be prescribed and why?

Reason for question: Dead trees are briefly discussed in the BRLP (p.14) but there is a need to review the current prescription process.

Relevant Study Component: Down wood management

Adaptation Options:

• Continue to manage for down wood based a minimum level of 240’ as discussed in the ROD C-40, consideration of site productivity, aspect, and plant associations, and increased levels based on Landscape Area and site-specific needs.
• Use a method based on a % of overstory and Landscape Area.

Evaluation: The Willamette National Forest Plan shows recommendations for down wood based on plant association (FW-212). Since 1996, the Blue River Ranger District has used a stratification of this table to prescribe down wood (see attached Table: Down Woody Material Prescribed Minimum Guidelines). Operationally, down wood linear feet must be converted to trees, which has been done using the attached chart: Tree Dimensions. Down wood levels may be modified based on site-specific recommendations, for example presence of a Survey and Manage species which benefits from down wood. Amounts in the prescription table range from 40 to 320 feet, however, forest level recommendations have been to not leave less than 240 feet as discussed in the ROD. The ROD does leave the option of varying from 240 feet based on site specific conditions. All prescribed down wood has been left standing. Falling is planned to occur four to five years after completion of timber sale units, and only half the prescribed level is felled because future blowdown is assumed. Blowdown present in the unit is monitored before implementation, and the amount is subtracted from what needs to be felled. A study conducted by a student on the Blue River Ranger District showed very high levels of blowdown in harvest units several years after logging, however, none of these units had higher levels of canopy retention which could change results.

Naturally occurring levels of down wood vary enormously and have been found to be significantly greater than prescribed levels (Wright, 1997) (Mid-Willamette LSR Assessment, see attached Table II-10).

Recommended Adaptations with Rationale:

Prescribed Down Wood Levels by Landscape Area

	Prescribed Overstory Retention Level	Example Trees per Acre Left After Harvest	Additional % Retention for Down Wood Habitat Based on Overstory Canopy Before Harvest	Example Additional Trees per Acre for Down Wood Habitat Creation (based on 100 TPA)	General Guideline
Landscape Area 1	50%	35-50	1-2%	1-2	Lowest amount of down wood
Landscape Area 2	30%	25-35	3-4%	3-4	Moderate level of down wood
Landscape Area 3	15%	10-20	5%	5	Big pulse of down wood

Include down wood monitoring at time of snag monitoring using 1/10 acre or linear plots after logging and prescribed burning, if applicable, is completed and base recommendations on that result.

Down wood may provide greater usefulness to small and medium-sized mammals if it is hollow or soft which allows easier cavity excavation. Thus, inoculation of standing trees provides not only snag but down wood benefits as well.

Future Data Needs to Benefit Down Wood Management:

Improved data for naturally occurring down wood levels by plant association on the Willamette National Forest

Improved data for expected blowdown rates over time in logged stands with varied levels of overstory canopy retention.

Information sources:

Mellen, Kim; Ager, Alan. Coarse Woody Debris Model — Version 1.2 — June 1998. U.S. Forest Service, Portland, OR.

Mid-Willamette LSR Assessment, 1998.

Thompson, Rebecca. Draft Bear Marten LD Coarse Woody Debris, 11/23/99. Eugene District BLM.

Wright, Pam. Down Wood Model, MA Thesis, Oregon State University, Corvallis, Oregon. 1997.

Topic #18 - Snag and Log Creation

Date: 1/18/00

Reviewers: Geary, Mayo, Seitz, Overton, Ford, Swetland.

Compiler: Ruby Seitz, Karen Geary

Question: Should 100% retention areas have snag and down wood creation?

Reason for question: Frequently the 100% retention areas are designed around sensitive microsites, for example Survey and Manage locations or on unstable ground. These areas are also meant to provide small refugia for species which may require habitat components consisting of snags and down wood. The question is whether logs and snags should be artificially created within these 100% retention areas, which may require killing and falling large trees.

Relevant Study Component: Snag and down wood management

Adaptation Options:

• Create snags and down wood within 100% retention areas.
• Do not create snags and down wood within 100% retention areas.

Evaluation: The goal of retention areas is to keep them as intact as possible in order to provide the highest quality habitat possible while managing the surrounding forest. The location of these relatively small retention areas may vary throughout future rotations.

Factors contributing to the decision of whether or not to create snags and logs within 100% retention areas include the following:

• Size and shape of the 100% retention area
• The reason for the 100% retention area
• Location in block - spatial orientation of existing snags, down wood
• Location in cutting unit - logging systems (corridor, flight path)

Recommended Adaptations with Rationale:

Snags: Overall snag creation guidelines which are specified in the prescription will determine the desired per acre snag level for the entire harvest unit, including the retention areas. For each 100% retention area, the ID team shall discuss and document whether snag creation is allowed. The above four factors will be used to guide this decision. Retention areas less than two acres may have snags created to replace those needing to be falled due to the safety hazard to the logging operation. Retention areas greater than two acres may have snag creation activities if there is a snag deficiency based on the prescription for the entire unit. For example, if the snag prescription for the entire unit is four/acre, and the three acre retention area has only two snags per acre, two additional snags per acre may be created.

If a retention area was placed around a rare species location which requires dead wood, snag creation may be recommended. If existing large snags or down wood is near the retention area, additional dead wood creation may not be required.

Down Wood: Down wood creation in 100% retention areas is not necessary since all existing Class I and II logs within retention areas are prescribed to be maintained. Depending on the location of the areas within the cutting unit it may be prudent to mark the boundaries of the retention area on the ground. Sale contracts include language which prohibits removal of Class I and II down wood which timber sale officers will monitor.

Topic #4 - Road restoration strategy

Date: 1/19/00

Assessment conducted by: John Cissel, Blue River Staff, Blue River Aquatic Team, Fred Swanson, Paul Bennett, Tere Turner

Question: Develop a road restoration strategy to help meet the aquatic conservation strategy objectives and is integrated with the landscape management strategy.

Information source(s):

1. Wemple, B.C., J.A. Jones, and G.E. Grant. 1996. Channel network extension by logging roads in two basins, western Cascades, Oregon. Water Resources Bulletin 32(6): 1195-1207.

2. Process for developing transportation management objectives applied at the watershed scale. 1997. Roseburg District BLM. 19 p.

3. Flanagan, S.A., M.J. Furniss, T.S. Ledwith, S. Thiesen, M. Love, K. Moore, J. Ory. 1998. Methods for inventory and environmental risk assessment of road drainage crossings. USDA Forest Service Technology and Development Program Report. 45 p.

4. Pilot roads analysis. 1999. Willamette National Forest.

5. Blue River road restoration concept paper. 1999. OSU Geosciences class - advanced field methods in geomorphology and landscape ecology.

6. USDA Forest Service. 1999. Roads analysis: informing decisions about managing the National Forest transportation system. USDA Forest Service Washington DC. FS-643.

7. Jones, J.A., F.J. Swanson, B.C. Wemple, and K.U. Snyder. in press. Effects of roads on hydrology, geomorphology, and disturbance patches in stream networks. Conservation Biology.

8. Madej, M.A. in review. Erosion and sediment delivery following removal of forest roads. Earth Surface Processes and Landforms.

9. Wemple, B.C., F.J. Swanson , and J.A. Jones. in review. Effects of forest roads on sediment production and transport, Cascade Range, Oregon. Earth Surface Processes and Landforms.

10.Wemple, B.C. 1998. Investigations of runoff production and

sedimentation on forest roads. Corvallis, OR: Oregon State University. 168

p. Ph.D. dissertation.

Relevant study component: Landscape management strategy - watershed restoration

Background: Attainment of the aquatic conservation strategy objectives in the landscape management plan relies on vegetation management regimes patterned after natural disturbance regimes, various mitigation measures, an aquatic reserve system, and an active watershed restoration program. To date, the vegetation management regimes have been outlined and scheduled into the future with sufficient temporal and spatial detail to support project-level planning, modeling and assessment. However, the road restoration strategy in the landscape management plan contains only general goals and types of projects. There is a need to develop the road restoration strategy in sufficient detail to support project planning and implementation, modeling, assessment, and evaluation of the aquatic conservation strategy objectives.

Road restoration has also become a major public issue and internal concern. Negative effects of roads on a wide variety of watershed and ecological processes have been identified. The existing road system was built based on a large timber sale program, and road maintenance was funded from timber sale receipts. With the sharp decrease in timber sales a huge backlog of unfunded road maintenance needs has arisen. Meanwhile, numerous stocks of anadromous fish species have been identified as threatened species, although none have been listed in the Blue River watershed. National guidance on road analysis has recently been issued in response to these issues.

Recent research in the Blue River watershed and elsewhere has clarified the effects of roads on hydrology, geomorphology, disturbance patches in stream networks, sediment production, and the potential benefits of road restoration. Results from these studies identify aspects of the watershed and road system that should be analyzed to assess risks to aquatic ecosystems.

Evaluation: The road restoration strategy ranks each road by a set of aquatic risk indicators and by ratings of future human use needs. Data to conduct these rankings were assembled from a field inventory and from GIS analyses. Aquatic risk indicators were summed by watershed processes (mass movement risk, fine sediment risk, hydrologic interaction risk), and aggregated into one summary rating for each road. Ten subdrainages were also ranked in terms of aquatic ecosystem risks. A composite aquatic risk rating was formed based on both the individual road and the subdrainage rankings.

Future human use rankings were also summed and categorized based on the degree of need for future access. These ratings determined whether the road should remain on the system or be removed, and for system roads determined whether they should remain open or be closed for future use. Roads slated to remain open were then further screened by their aquatic risk rating. Roads determined to be a high risk to aquatic ecosystems were changed to a closed status. The aquatic ratings were also used to determine the priority for converting roads to non-system or closed status (ensure self-maintenance), and to establish maintenance priorities for open roads.

Results of the analysis are displayed as a series of maps depicting road and subdrainage rankings and restoration priorities, individual road and subdrainage ranking spreadsheets, and analysis process documentation. Results will be displayed on the Blue River Landscape Study web site as well as hard copy. The following outline provides more detail:

Blue River Road Assessment

I. SITUATION

Land allocations

• H.J. Andrews Experimental Forest
• Central Cascades Adaptive Management Area

Landscape management plan

• vegetation management modeled after historical fire regimes
• spatially- and temporally-specific timber harvest schedule
• alternative approach to aquatic conservation strategy objectives

Road data

• spatial data confusing, standards unclear
• attribute data old and unmanaged

II. OBJECTIVES

Develop an integrated road restoration strategy

- establish road restoration priorities

- establish future maintenance levels and priorities

Apply concepts and results from recent research

Identify future research and monitoring needs

Clean up road GIS layer

III. ANALYSIS PROCESS

Aquatic risks

Road ranking

risk factors

- field measurements

- GIS-derived data

watershed process rankings

- mass movements

- fine sediments

- hydrologic effects

Subdrainage ranking

Restoration priority

Human uses

Public

Private

Management

Research

IV. ROAD RANKINGS - RISK FACTORS

Field measurements

- adjacency to fish-bearing streams

- culvert fill height

- sideslope steepness

- sustained steep road gradients

- road prism stability

GIS-derived data

- road age

- steep, shallow soils

- slope position

- rain-on-snow susceptibility

- stream/road crossing density

- fine sediment soils

V. ROAD RANKINGS - WATERSHED PROCESSES

Mass movement

risk of initiation = sideslope steepness + stability (x2) + slope position + road age + shallow soils + rain-on-snow susceptibility

transport risk = sideslope steepness + slope position + crossing density

magnitude of effect = fish (x2)+ culvert fill depth + crossing density

Fine sediment

risk of initiation = sideslope steepness (x2)+ road gradient (x2) + culvert fill depth + crossing density + road age + fine sediment soils

transport risk = sideslope steepness + slope position + crossing density

magnitude of effect = fish (x2)+ culvert fill depth + crossing density

Hydrologic effects

risk of initiation = sideslope steepness + slope position + shallow steep soils + rain-on-snow susceptibility

transport risk = road age + road gradient + crossing density + sideslope steepness

magnitude of effect = fish (x2)+ culvert fill depth + crossing density

VI. SUBDRAINAGE RANKING

Ranking factors

- rain-on-snow susceptibility

- terrain

- road density

- stream/road crossing density

- aquatic habitat

- special status

VII. RESTORATION PRIORITIES

Road rankings summed for each road and assigned high/moderate/low

Subdrainage rankings summed for each road and assigned high/moderate/low

Restoration priority assigned to each road by following table:

Subdrainage Priority	Road Priority
	Low	Moderate	High
Low	Low	Low	Moderate
Moderate	Low	Moderate	High
High	Moderate	High	High

VIII. HUMAN USES

Uses ranked

• private
• mining
• cultural
• recreation
• research
• timber
• silviculture
• fire

Maintenance levels

• scores summed for each road and assigned off-system/level 1/level 2
• if any individual use rated maximum score, road assigned to level 2

Recommendations: The draft road restoration strategy should be reviewed and refined and then incorporated into the landscape management strategy, the evaluation of the aquatic conservation strategy objectives, and future monitoring, modeling and research. The strategy identifies road restoration priorities for individual roads and by subdrainage.

Topics 6 and 7 - Substrate and water quality source areas

Aquatic Conservation Strategy Update for the

Blue River Landscape Strategy: Connectivity of Hill Slope and Fluvial Processes to Potential Habitat Locations

Topic #: 6,7,21

Date: 8/23/00

Reviewers: D Kretzing, J.Phillips, R.Rivera, J.Cissel

Question: Can specific locations of existing or potential high value aquatic habitats be identified, and linked to source areas within the landscape that have the potential to provide resources necessary for the formation and maintenance of these habitats?

Information Sources:

"General Landscape Theory of Organized Complexity" - Benda, Miller, Sias, Dunne, Reeves, 1999

"A Disturbance Based Ecosystem Approach to Maintaining and Restoring Freshwater Habitats of Evolutionarily Significant Units of Anadromous Salmonids in the Pacific Northwest" - Reeves, Benda, Burnett, Bisson, Sedell, 1995

"Understanding the Role of Sediment Waves and Channel Conditions Over Time and Space" — Lisle, 1997

"Comments on Historical Variation and Desired Condition as Tools for Terrestrial Landscape Analysis" — Millar, 1997

"Multiscale Thermal Refugia and Stream Habitat Associations of Chinook Salmon in Northeastern Oregon" — Torgersen, Price, Li, McIntosh, 1999

"Discussion of Substrate Source Area selection process" - John Phillips

Relevant Study Component: Spatial pattern of retention trees and distribution of refugia

Background:

The Blue River Landscape Strategy employs temporal and spatial patterns of vegetation treatments over long rotation lengths to restore forest structures and disturbance patterns to a condition more closely resembling structures and patterns resulting from natural disturbance agents such as wildfire. In the first iteration of the landscape strategy (4/18/97), the default riparian reserve prescriptions of the Northwest Forest Plan were replaced with a network of large headwater aquatic reserves and corridor aquatic reserves along fish bearing streams. Intermittent and non-fish bearing perennial streams were not included in the reserve system. The objective of this system was to meet the objectives of the Aquatic Conservation Strategy on a landscape basis, over time. The initial concept was that long rotations and higher retention levels would limit the extent of disturbance in any decade. Then, a strategy of retention tree placement that emphasized continuity between riparian processes and adjacent upland habitats combined with an alternative reserve system would provide an equivalent level of protection to watershed and riparian processes as the original default riparian reserve prescriptions in the Northwest Forest Plan.

This first iteration of the landscape strategy coarsely evaluated the relative aquatic habitat value or potential of the various fish bearing streams, specifying reserve widths of one to two site potential tree heights based on valley form characteristics. Generally, constrained valleys received one site-potential tree height reserve prescriptions, while unconstrained valley types received two site potential tree height reserve prescriptions. In addition, aquatic reserves where management disturbance was eliminated were situated in headwater areas upstream from the corridor reserves in locations that are least likely to experience natural fire disturbance, and in areas adjacent to major stream confluences.

Experience gained during planning efforts in the watershed on several projects intended to implement the landscape strategy suggests that adjustments to the strategy could improve our ability to meet the Aquatic Conservation Strategy objectives in the Northwest Forest Plan. Specifically, adjustments were made to the basic strategy of aquatic reserves to offer additional protection to streams, to maintain the integrity of these habitats. These adjustments were initially made on a project and stream basis, rather than a landscape basis, consistent with expectations in the original landscape plan. In later projects, linkages between stream segments and the source areas for cold water, large wood, sediment, and nutrients were explored on a drainage basis. This second iteration of the landscape plan is an attempt to apply some of what has been learned to date from projects that have already been implemented.

Evaluation:

A. High Potential Stream Reaches

Streams within the watershed were further evaluated to identify which reaches have the highest aquatic habitat value or potential value. The majority of the streams in the Blue River Watershed have high channel gradients and flow through narrow restricted valleys with little meaningful floodplain area. These streams possess a great deal of power and readily transport large wood and coarse gravel and cobble sized sediment downstream on an annual basis. Most of these streams have little opportunity to store these materials, or to use them to develop complex channels.

Areas with the highest potential to actually store large wood and coarse sediment were identified by evaluation of valley shape. This was retained from the original landscape design element that identified unconstrained storage reaches and constrained transport reaches. Then, stream segments that flow through terrain with valley gradients approximately 3 percent or less were identified by evaluation of valley gradient on topographic maps. This process allowed us to focus more tightly on those reaches that had both unconstrained valley forms, and relatively gentle valley gradients. Relatively lower stream power in these reaches would allow corresponding increases in the ability to store and use large wood and coarse sediment.

After these high potential stream reaches were identified and mapped, the surrounding landscape was evaluated and source areas for substrate materials, nutrients, and abundant cool water were located.

B. Substrate Source Areas

Coarse, gravel and cobble size sediment, and large wood are the primary substrate materials that these streams can use to develop complex channel structure and resulting aquatic habitats. These materials are contributed to streams throughout the landscape in relatively low concentration as trees fall into streams and occasional bank scour occurs. However, there are two mechanisms that lead to larger, concentrated pulses of these materials into streams, often as a complete package of both coarse sediment and large wood. The first mechanism is slope failure that results in a torrent of material that is delivered to the stream from adjacent slopes. The second mechanism is toe slope failure of active earth flows, where adjacent streams have undermined them. Both of these processes can and do occur naturally in the Blue River Watershed, and play an important role in maintaining stream channel complexity and resulting aquatic habitats. An important characteristic of these mechanisms is that they have the potential to provide coarse sediment and large wood simultaneously. Consequently, management to retain large wood on areas susceptible to slope failure and on areas situated on earth flow terrain will insure that large wood will be input to the system along with sediments of varying size class. Groupings of blocks with these characteristics that are located adjacent to, or upstream of the high potential reaches were identified as Substrate Source Areas. Selections were made based on GIS queries of slope, soil, and drainage characteristics, experience gained during the floods and associated slope failures of 1996 and 1997, and other local experience. A detailed description of this process has been prepared by John Phillips (attached).

C. Water Quality Source Areas

Criteria used to identify landscape blocks that provide nutrient inputs to the system focused on blocks with substantial wetland habitats. Most wetlands in the Blue River landscape are hardwood dominated, and provide substantial amounts of leaf litter to these streams. Blocks that met this criteria were identified visually by comparing a block map with a GIS overlay of wetland areas in the watershed.

Criteria used to identify landscape blocks that provide substantial quantities of cool water included those blocks that have a relatively high contribution to base flows during the summer period. These blocks were identified by examining a GIS generated map of the watershed that rated the landscape into high, medium, and low potential, based on characteristics of aspect, elevation, precipitation, and soil depth. The presence of wetlands also indicates an ability to provide substantial flows of cool water, so blocks that possessed either or both of these traits were selected as candidate source areas for cool water.

Groupings of blocks with the characteristics needed to provide nutrients and abundant cool water that are located upstream of the low gradient reaches were identified as Water Quality Source Areas, and management prescription elements were developed. These source areas were mapped in relation to the locations of the low gradient stream reaches, and those areas that could be linked to downstream low gradient reaches were identified. Prescriptions were developed for these areas to maintain and restore their ability to produce the materials needed downstream to facilitate the development and retention of desired habitat features in the low gradient reaches.

Recommendation:

A. Prescription elements to be applied in Substrate source areas include:

Fifty percent retention of evenly spaced mature trees will be applied within Blocks where the soils are shallow, coarse texture and occupy slopes greater than seventy percent. Leave tree retention along streams within sediment sources areas will be those designated for that Landscape Area. Retention trees should not be allocated from elsewhere in the Block, but in addition to the green tree retention a described for that Landscape Area.

Active earthflows are identified and dropped from the timber base. Within identified quaternary earthflow terrain or glacial deposits adjacent to perennial streams there will be, generally, a one site tree height no harvest buffer, and on Class IV streams a one half to one full tree height no harvest buffer in order to maintain a supply of large wood to the stream system while maintaining maximum stability and retention of fine sediments. Dependent of operability, slope and topographic characteristics no harvest buffers may vary in width (longer and shorter). Where possible the entire toe of the earthflow should be deferred from harvest inorder to maintain toe stability and to retain the large wood source. Retention trees should not be allocated from elsewhere in the Block, but in addition to the green tree retention as described for the Landscape Area.

B. Prescription elements to be applied in Water Quality source areas include:

No road construction, ground skidding, or other activity with the potential to affect surface and subsurface water flow should be permitted within two site-potential tree height of wetlands, unless site specific analysis indicates that surface and subsurface flows will not be affected.

All perennial streams with substantial flows will have a one site-potential tree height buffer where at least 70% canopy cover will be retained. On streams flowing east to west, the entire buffer will be situated on the south side of the stream. On streams flowing north to south, the buffer will extend for one half site-potential tree height on each side of the stream.

Silvicultural treatments such as pre-commercial thinning, fertilization, and commercial thinning should be evaluated and utilized to accelerate development of large wood, shade, and late successional stand structure in existing managed stands, adjacent to perennial streams with substantial flows.

Use of ground based yarding equipment and road construction should not be permitted within one site potential tree height of wetlands, and use of this equipment or construction of new roads within an additional site potential tree height should only occur if site specific evaluation indicates that alteration of subsurface water patterns will not occur.

Blue River Landscape Study

Substrate Source Areas (SSA) Mapping Rational

Prepared by

John C Phillips

Soil Scientist

Blue River/McKenzie Ranger Districts

DRAFT

7/21/00

Introduction

The Blue River Landscape Strategy (April, 1997) emphasizes the spatial patterning of vegetation over long rotations to restore forest structures and disturbance patterns similar to those left by naturally occurring wildfires. Management activities are scheduled for sub-watersheds to mimic the temporal and spatial affect of stand replacing fires; recognizing wild fire burn in discrete areas affecting the aquatic habitat of that area or sub-watershed while the remainder of the watershed remained intact. That iteration of the landscape plan addressed landform stability and large wood recruitment in accordance with the Willamette National Forest LMP and the Northwest Forest Plan. Essentially actively unstable areas are removed from the timber base, and potentially unstable areas were deferred from treatment or heavier retention was prescribed without recognizing the role slope failures play in the disturbance ecology of the watershed. Large wood recruitment for aquatic habitats was met with various riparian management strategies described in the study.

The Aquatic Conservation Strategy Objective #5 Maintain or restore the sediment regime under which the aquatic ecosystems evolved. Elements of the sediment regime include the timing, volume, rate and character of sediment input, storage, and transport was addressed at the watershed scale through meeting the above criteria while implementing the landscape strategy of retention and long rotations. This approach however did not; 1) identify high risk areas of slope instability relative to aquatic habitat establishment, succession and maintenance, 2) adequately illustrate the natural temporal, spatial and magnitude of variability of substrate (sediment and large wood) transport mechanisms, nor did it 3) describe the nature of the sediment (fine or coarse textured) and its affect on water quality or aquatic habitat.

Background

Landscapes and watersheds are dynamic systems composed of biotic (vegetation, wildlife, etc) and abiotic (geology, soils, water, climate, etc.) components that are periodically subjected to disturbance under the influence of climate. It is the stochastic nature of climate acting upon the topography, geology, soils, vegetation and hydrology within a watershed that created the topography of today. Throughout the Holocene wildfires occurred naturally as a result of drought, and floods and slope failures during periods of unusually high precipitation or rain on snow events. Fires and floods create conditions of instability, they transport and process sediment and organic matter (stream substrate materials) and re-establish early succession upland and aquatic habitats. Wildfires and flooding and the associated effects are common to the Blue River watershed and we know that the terrestrial and aquatic species evolved and persisted within this natural disturbance regime.

In general in the western cascades, fire and flood come infrequently following decades to hundreds of years of few or low magnitude disturbance events. Stand replacement fire and infrequent flooding can cause severe disturbance over a spatially discrete area in a very short period of time, and are generally followed by a long periods of recovery.

The present sediment regime differs from the historic regime by having a transportation system that is a chronic low amplitude source of fine sediment throughout the watershed, and act as sources and sinks of substrate materials during flood events. Road systems appear to have increased the magnitude and frequency of peak flows, debris slides and debris flows relative to fully forested and fire disturbed areas (Jones, J. A. et. al., Sept., 1999) and have interrupted the cascade of events that allow substrate transport into riparian areas and stream systems. During the 1964-65 and 1996 storms triggered nearly 100 mass movement events. Of these approximately 80% of slope failures were initiated along roads and in clear-cut units (Swanson and Dyrness, 1975 and Swanson, unpublished data). The majority of slope failures (approx. 60%) in the Blue River Watershed in 1996/1997 storms originated from roads. Therefore; based on the above data between 20 to 40 landslides occur within the watershed as a result of natural geologic processes, independent of land management activities, during 50 and 100-year storm events. Research has found debris flows were 14 times more frequent per unit area of roads than on undisturbed forest in the Andrews Forest and Blue River watershed (Synder, 2000). Road failures above managed slopes with young trees often resulted in soils striped form hills sides with little or no opportunity for rapid recovery (personal observation). By maintaining mature trees on managed slopes that naturally are susceptible to slope failures in the form of debris slide or earthflow toe slope failures the processes, quantities and quality of substrate is maintained. In addition, the fire return intervals have been interrupted over the last hundred years and influenced the rate of large wood recruitment and substrate transport from upslope positions.

Wildfire fire: The return intervals for high severity stand replacement fires in the Blue River watershed range from 100 to greater than 300 years. Landforms that burn relatively infrequently experience high severity fire due to the heavier fuel loads that accumulate over time and the subsequent longer residence time. High severity stand replacement fires can remove the majority of standing live vegetation, and consume all of the vegetative cover and leaf litter on the soil surface. Often soils become hydrophobic following intense fire.

The combination of vegetation and leaf litter removal along with hydrophobic conditions can lead to severe erosion of hill slopes in the form of sheet, rill and gully erosion. These conditions can persist for several years following the fire event. Typically, sand, silt and clay sized particles are selectively eroded leaving the coarser sediment (gravel and larger) behind. Sand and clay sized sediments can adversely affect water quality and fish habitat by reducing water clarity and imbedding spawning gravels. Dependent on the soil texture, slope gradients and slope topography; rates of soil erosion can exceed 50 tons per acre following fires of high severity and persist for several years. Rates of soil erosion slowly recover to background levels (<0.1 ton per acre per year) as vegetation becomes established and effective ground cover is restored.

Mortality of the overstory can result in susceptibility of slope failure 10-20 years following the event. Removal of the live root network reduces evapotransporation demands and subsequently increases in hydrostatic pore pressure. Due to the soil texture, depth and topography in the Blue River watershed slope failures are the dominant means of substrate transport.

In general, vegetative communities and soils exposed to relatively frequent fires (< 150 years) experience a lower magnitude effect than fire with infrequent return intervals. Relatively frequent fires remove much of the ground and aerial fuels that can build up over time and avoid large accumulations fuels. The lower severity leaves the majority of the mature viable trees, and retains much of the litter and ground vegetation. Rarely do fires of low and moderately severity cause widespread erosion or hydrophobic soils. The species composition in stands subjected to low severity fires are generally dominated by fire resistant species (e.g. Douglas Fir) that are less susceptible to mortality during low and moderate severity fire. Slope instability in general will not increase due to the remaining live root network.

Debris Slope Failures: In the Western Cascades and the Blue River Watershed, slope failures generally occur episodically one to two decades following a high severity fire during rain on snow or during intense rainfall sufficient to produce flooding. Slope failures are naturally occurring events that can and do occur independent of fire, but are most often associated. Storm or rain fall events of sufficient intensity to cause slope failures occur with a return interval of 5 to 10 years; however, severe flood events have return intervals of 50 to 100 years or greater. The frequency and magnitude relationship and the effects on slope stability follow a similar frequency/magnitude reverse correlation as with fire. Landforms that experience relatively frequent slope failures often result in relatively low magnitude of effects when compared to infrequent slope failure regimes and the subsequent high magnitude effects (Binda et al, October, 1999). Examples include the extremely steep slopes and shallow soils within Tidbits, Cook and Quentin sub-drainages compared to the toe or front of deep fine textured slow moving earthflows in Mann Creek sub-drainage. The steep slopes in Cook and Quentin Creeks have a relatively frequent (100-1000 years) incidence of slope failures in the form of rock fall shoots, debris slides and torrents. These failures are relatively small (100’s to 10,000’s cubic yards) and generally transport substrate material into the stream system. Substrate generated from these landforms are dominated by coarse fragment of gravel to boulder sized material and can be beneficial to stream and aquatic habitats. Slope failures with a return frequency of greater than 10,000 years are capable of producing massive amounts of geologic materials. As was the case with the quaternary earthflow that blocked the Blue River access to the McKenzie River at Forest Road 15, diverting it into the present channel through the community of Blue River (Blue River Watershed Analysis).

Current models of rapidly moving debris torrents and landslides on unmanaged slopes describe the effects of slope failure as long linear features that have a great deal of complexity (Jones, et al DRAFT). Torrents travel down streams leaving large wood, logjams and coarse sediments deposited along its path leaving a new template for establishment and succession to occur (HJA cite and J C Phillips; personal observation). Torrents can travel distances ranging from hundreds of feet to a few miles miles. In steep mountainous terrain with narrow valley bottoms large wood and sediment are typically deposited along fan margins or within the stream itself where as in terrain with wide valley floors sediments and large wood are stored in fans and along toe slopes (Benda et al, Oct., 1999). Eventually these stored sediments are incorporated into the stream system through stream meander and bank erosion. Slope failures on landforms void of mature trees or on slopes with young stands generally result in little beneficial additions of substrate to the streams that benefit aquatic species.

Landforms that produce rapidly moving landslides are often the most sensitive to land management activities such as timber harvest and road construction. These landforms are also important sources of substrate necessary for maintaining aquatic habitat, fish bearing reaches, and for properly functioning aquatic and riparian ecosystems and habitats. The sediment generated during these slope failures is gravel and larger sized materials that often benefit the aquatic habitat as long as large wood and boulder material is present to retain these substrates.

Earthflow terrain, on the other hand, moves very slowly (a few to tens of centimeters per year) resulting in an almost continual instability at the toe and margins of the flow. The movement is largely governed by climate and yearly fluctuations in rainfall. Land use effects on earthflows and associated streamside slide are less well known and potentially less significant than effects on the debris slide initiated disturbance events (Nakamura, et. al., 1999). Earthflows are generally dominated by deep relatively fine textured soils or soft bedrock and provide a chronic source of fine sediment by erosion of the toe or front of the flow at the stream interface. At the toe the slow movement of the earthflow constricts the stream channel causing an increase stream flow velocities and erosive power, which undercuts the toe of the earthflow and deflects stream flows into the opposite bank causing calving of sediment and large wood into the stream channel. The result is relatively frequent (yearly) calving of small amounts sediments (10’s cubic yards) and large wood directly into the stream. This process of movement and failure is chronic and is largely unaffected by land management activities except at the toe or front of the flow. The upper portions of Mann and Wolf Creeks are dominated by Quaternary earthflow terrain of which some slope elements are currently active. The fine textured soils and geologic deposits (i.e. high is silts and clays) associated with earthflows can have an adverse effect on water quality and fish habitat by increased turbidity, reduced visibility, and filling of gravel beds with fine textured material.

Selection Process

This iteration of the Landscape Study sought to identify geographic areas and topographic elements prone to natural slope failures as Substrate Source Areas. By identifying areas that are potentially unstable, that are capable of producing slope failures that transport substrate (coarse sediment and large wood) into response stream reaches that are capable of retaining sediment and providing fish habitat; we can prescribe road treatments, silvicultural treatments and green tree retention levels in locations that maintain or enhance the natural sediment regime and meet the Aquatic Conservation Strategy Objectives.

This is not to say that the entire watershed can’t and doesn’t produce sediment naturally; but it is those areas that have a highest potential for sediment and large wood transport that can influence fish habitat and water quality that are of most the interest. Although it is recognized that these processes are a natural part of a dynamic system, slope failures can pose a risk to aquatic and riparian habitats, listed fish species and municipal water supplies.

Coarse Substrate Source Areas

The evaluation and selection of substrate source areas (SSAs) focused on the soil, slopes and stream densities as indicators of slope instability and the landform’s capability to produce rapid down slope movements of coarse sediment (> sand size) and large woody material that has the potential for incorporation into the stream system along response reaches. A Geographic Information System (GIS) query initially used the landscape blocks that were dominated by slopes greater than 70% with coarse soils less than 3 feet deep (see SSA draft Map; T. Turner; November, 1999). Further refinement of SSAs was performed using slope gradient mapping, stream densities, and site-specific knowledge of the location of slope failures during the 1996/97 floods. Slope, stream density and soil/slope stability maps of the 10-3 watershed above the reservoir provided by J. Lloyd; October, 1999, Slope Percent Map of Blue River Watershed; T. Turner, August. 1999, Slope Stability Analysis within Landscape Blocks; T. Turner, August, 1999. and stream gradient maps D. Kretzing, November, 1999, and Fish Population Mapping of the Blue River Watershed, Ramon Rivera, November, 1999 were also used to refine and define important SSAs.

Selection Criteria

Coarse substrate source areas occur on very steep slopes (>70%) with shallow coarse textured soils. Selection of substrate source areas was based on the data provided by the GIS queries, first hand knowledge of the watershed, the proximity of the Landscape Blocks for sediment source area to fish bearing streams, high stream densities or high topographic relief, and proximity to response reaches (low gradient streams). Blocks of SSAs were selected to encompass definable geographical areas with the greatest potential for naturally occurring slope failure. Clusters of Blocks were selected by on the criteria below.

Treatment Recommendations:

• Fifty percent retention of evenly spaced mature trees within Blocks designated as Substrate Source Areas.
• Leave tree retention along streams within substrate sources areas will be those designated for that Landscape Area.
• Retention trees should not be allocated from elsewhere in the Block, but in addition to the green tree retention a described for that Landscape Area.

Fine Sediment Sources Areas

Areas capable of producing fine sediment were identified initially by GIS query of the Soil Resource Inventory units composed of deep-seated earthflow terrain or glacial deposits that are dominated by silt and clay size particles, or where known Quaternary Earthflow terrain exists. Green tree retention criteria are limited to the toe, margins and interface of glacial deposits and earthflow terrain at the interface with stream channels. Clusters of Blocks were selected by on the criteria below.

Selection Criteria

Blocks dominated by glacial deposits or earthflow terrain that are adjacent to streams or in areas of relatively high stream density, and/or adjacent to or above response reaches with known populations of fish. Blocks were selected in definable geographical areas that are composed of or dominated by earthflow and glacial deposits that are capable of transporting fine sediment or large wood to the stream system.

Treatment Recommendations

• Active earthflows are identified and dropped from the timber base.
• Within identified Landscape Blocks where quaternary earthflow terrain or glacial deposits occur adjacent to perennial and intermittent streams there will be a one-site tree height no harvest buffer.
• Dependent of operability, slope and topographic characteristics the no harvest buffers may vary in width (longer). Where possible the entire toe of the earthflow should be deferred from harvest in order to maintain the source of large wood.
• Retention trees should not be allocated from elsewhere in the Block, but in addition to the green tree retention as described for the Landscape Area.

Bibliography

Benda, Lee, D. Miller, J. Sias, T. Dunne, and G. Reeves. October, 1999. General Landscape Theory of Organized Complexity

Jones, J. A. Hydrologic Processes and Peak Discharge Response to Forest Harvest, Regrowth and Roads in Ten Small Experimental Basins, Western Cascades, Oregon. In Press, Water Resources Research; draft 4/3/00.

Nakamura, F., F. J. Swanson, and S. Wondzell. Disturbance Regimes of Stream and Riparian Systems, a disturbance-cascade perspective. Hydrological Processes 00, 1-12 (2000).

Jones, J. A., F. J. Swanson, B. C. Wemple and K. U. Snyder. Effects of Roads on Hydrology, Geomorphology, and Disturbance patches in Stream Network. Conservation Biology, pages 76-85; September 6, 1999.

Snyder, K. 1999. Patterns of debris flows in streams of the H. J. Andrews Experimental Forest. M. S. Thesis. Department of Forest Science, Oregon State University, Corvallis.

Swanson, F. J., Dyrness, C. T. 1975. Impact of clear-cutting and road construction on soil erosion by landslides in the western Cascade Range, Oregon. Geology 3(7): 393-396.

USDA Forest Service, 1996. Blue River Watershed Analysis.

Topic # 9 - Refugia objectives and activities

Date:

2/29/00

Reviewers:

Blue River Landscape Team, J. Cissel compiler

Question:

What are the objectives for the 40-year refugia (e.g., Cook/Quentin), and what activities are compatible with those objectives?

Information source(s):

1. Operational experience - activities proposed in the Cook/Quentin area have raised questions about objectives and allowable activities

Relevant study component:

Landscape management plan - refugia

Background:

The original Blue River landscape management plan (4/17/97) included the concept of temporally-limited (40 years) refugia for late-successional and aquatic species in portions of the managed landscape, but did not identify which uses should take place in these refugia with two exceptions: regeneration harvesting should not occur and road closures should occur. These refugia were intended to alternate on a 40-year cycle to other portions of the Blue River watershed. Since then commercial thinning and other activities have been proposed within designated refugia. Greater clarity about the purpose of the refugia is needed to evaluate proposed activities.

Evaluation:

The Landscape Team spent portions of two meetings discussing the purpose of the refugia, the level of specificity desirable in a landscape management plan, and an overall management strategy for temporally-limited refugia.

Recommendations:

Spatial extent - Cook/Quentin subwatersheds alternate with Tidbit/Mann/Wolf subwatersheds.

Temporal extent - designated refugia switch on a 40-year cycle, starting with 1995-2035.

Purpose - to maintain or enhance late-successional and aquatic habitat so that populations using this habitat are able to recolonize other subwatersheds where disturbance is occurring.

Functions - reproduction, growth, dispersal, and recruitment of late-successional and aquatic species.

Management strategy - 1. coordinate the scheduling of management activities so necessary projects can be conducted prior to road closure to reduce future management disturbance, including noise; 2. provide a focus and priority for road and watershed restoration activities; 3. delay regeneration harvests until after refugia time-period.

Consistent activities - 1. desirable activities - those that maintain or enhance late-successional and aquatic habitat; 2. allowable activities - those that are neutral to late-successional and aquatic habitat; 3. not allowable activities - those that degrade late-successional and aquatic habitat

Topic # 13 - Adjustments to aquatic reserves

Date:

4/10/00

Reviewers:

Aquatic subteam (Dave Kretzing, John Phillips, Ray Rivera, John Cissel) evaluated and made recommendation to full landscape team

Question:

Are the aquatic reserves in the right places? Should any adjustments be made to reserve locations?

Information source(s):

Operational experience

Relevant study component:

Landscape area mapping and related analysis

Background:

Aquatic reserves were established based on criteria identified in the initial landscape management plan (4/17/97). Concerns about the location of a couple of the small-watershed reserves have arisen during project planning. Two concerns have surfaced: 1) possibly inadequate consideration of wetlands; and 2) the current amount of late-successional habitat within the small-watershed reserves.

Evaluation:

The aquatic subteam compiled data for each of the small-watershed reserves. Each reserve was individually ranked in terms of:
- stream density

- percentage of mature and old forests

- amount and percentage of reserve with high or moderate summer baseflow rankings

- amount and percentage of reserve with steep, shallow soils

- amount and percentage of reserve with earthflow soils

- amount and percentage of reserve with wetland soils

The subteam then reviewed the data and discussed the desirability of either changing the criteria or modifying the boundaries of individual small-watershed reserves based on the new data.

Recommendations:

The subteam recommended, with full team concurrence, that the aquatic reserve criteria remain unaltered, but that the location of one small-watershed reserve located at the confluence of Cook Creek and Blue River be moved slightly upstream to include more late-successional habitat. All other reserve locations were suitable as currently mapped.

Topic # 21 - Re-evaluation of Aquatic Conservation Strategy Objectives

Date: 2/6/01

Reviewers: Dave Kretzing, John Phillips, Ray Rivera, John Cissel

Question: Do we have sufficient new information to warrant updating the Aquatic Conservation strategy Objectives analysis in the landscape management plan?

Information source(s):

New information to be incorporated in the revised analysis of aquatic conservation strategy objectives includes:

1. Water quality and substrate source areas and related management guidelines

2. An integrated road restoration strategy

3. Landscape simulations projecting future conditions so that a more quantitative assessment could be made

4. Fire history information for riparian and lower slope areas allowing comparison of planned disturbance rates to historical disturbance rates

Relevant study component: the evaluation of the Aquatic Conservation strategy Objectives

Background:

Development of the water quality and substrate source areas, and of the road restoration strategy, provides a more complete and coherent approach to meeting the aquatic conservation strategy objectives. This approach consists of the following components:

1. A less intense timber management regime patterned after historical fire regimes - this results in lower timber harvest frequencies and intensities as compared to Matrix land management in the NFP.

2. A stream corridor reserve system applied to fish-bearing streams - this less extensive riparian reserve network (as compared to the NFP) allows for implementation of timber harvest at spatial scales and patterns more similar to historical fires.

3. A small-watershed reserve system consisting of 200-600 acre blocks distributed across the watershed - these reserves are intended to meet multiple objectives including maintenance of watershed processes, aquatic habitats, and provision of interior late-successional habitat.

4. Large wood, coarse sediment and water quality source area management - areas most likely to provide these materials to key stream reaches are mapped and specific prescriptive elements are provided to ensure continued delivery of these materials to streams.

5. Riparian and lower slope prescriptions - specific prescriptive elements are included to ensure retention of large trees and hardwoods in riparian and lower slope areas.

6. Road restoration strategy - all roads in the watershed have been evaluated for risks to the aquatic ecosystem, and restoration priorities have been established and integrated with the overall landscape management plan.

7. Watershed restoration - a variety of other restoration activities are being implemented including addition of large wood to stream channels, encouraging growth of large conifers near streams, and removal of human-placed migration barriers.

8. Timber harvest scheduling - we've scheduled timber harvest over the watershed to act more like a pulse disturbance & less like a press disturbance.

This strategy needs to be incorporated into the evaluation of the aquatic conservation strategy objectives (ACSOs).

Evaluation:

In addition to the new components of the landscape management plan, more detailed projections of future landscape conditions are now available (see Cissel et al. 1999 and supporting data). And an evaluation of general fire history in riparian and lower slope areas has been completed (attached). These two sources provide additional information that could be included in a re-evaluation of the aquatic conservation strategy objectives.

Recommendations:

Complete re-evaluation of the ACSOs using these new sources of information.

Blue River Landscape Study - Aquatic Conservation Strategy Objectives Evaluation
Riparian and lower slope historical fire occurrence

2/6/01 - John Cissel

I. Purpose

The purpose of this assessment is to provide a general interpretation of historical fire frequency and severity in riparian and lower slope areas in the Blue River watershed to support evaluation of the aquatic conservation strategy objectives. Several of these objectives refer to maintaining or restoring historical landscape patterns or watershed processes (e.g., #1, #5). These interpretations provide an important context to evaluate the significance and effect of potential future disturbance rates and severities in the landscape management plan.

II. Background

More is known of the fire history in the Blue River watershed than any other watershed in the Pacific Northwest. Three separate field studies have documented fire history in the area (Teensma 1987, Morrison and Swanson 1990, and Weisberg 1998). The most recent and complete analysis was conducted by Weisberg (1998). The fire regime parameter for which we have the best information available is fire frequency. Weisberg calculated fire frequency using a variety of statistical measures, and then built a linear regression equation to predict fire frequency based on a number of environmental variables. These equations were then used to create a map of three predicted fire frequency regimes (Weisberg 1998, Fig. 4.26) and associated fire frequency measures (Weisberg 1998, Tab. 4.11). Of these frequency measures, mean fire return interval (MFRI, variable "MFI.nlow") is the measure most useful for estimating the historical frequency of stand- and partial stand-replacing fires detected by Weisberg (1998).

Differences in fire severity by fire frequency regime have proven more difficult to quantify. Variability in fire severity has been reported in this watershed by Morrison and Swanson (1990) and Weisberg (1998), but quantitative evidence has not demonstrated a close association with fire frequency. Weisberg (1998) did show that high frequency fire regimes produced more cohorts of Douglas-fir than did low frequency fire regimes, but the relationship was not strong. The Blue River landscape management approach recognized variability in fire severity by prescribing three levels of overstory canopy retention, one for each fire frequency regime: low frequency/high severity (mean of 15% overstory canopy cover retention), moderate frequency/moderate severity (mean of 30% overstory canopy cover retention), high frequency/low severity (mean of 50% overstory canopy cover retention). While the quantitative basis for this link is weak, studies have shown an inverse association among frequency and severity in disturbance regimes of many types (Sousa 1984), including forests in this watershed (Morrison and Swanson 1990). For purposes of this assessment, I will continue to assume the same association of fire severity with fire frequency regime as was assumed in the Blue River landscape management plan. This assumption results in a reasonable portrayal of variability in fire severity at the watershed scale, but likely misrepresents historical fire severity at the scale of an individual site. This association is in need of further study.

III. Riparian areas and lower slopes

Information need:

amount and proportion of riparian areas historically disturbed by fire in each 20-year time period

Method:

Data necessary to estimate the amount and proportion of riparian and lower slope areas historically disturbed by fire are scarce. The best source is fire frequency data from paired upper and lower slope plots from Weisberg (1998, and personal communication - 7/22/99). Of 289 fires-at-plots reported, 264 (91%) burned upper slope plots and 242 (84%) burned lower slope plots, meaning that upper slope plots burned 9% more often than lower slopes. Weisberg performed a second type of analysis to further address this question by pooling his plots across slope positions and then determining slope position per plot via a GIS layer. Regression equations were generated to predict the effects of slope position, aspect and elevation on fire frequency. When averaged across slope aspects at a 900-meter elevation upper slope positions burned 13% more often than lower slopes. Weisberg also comments that his sampling method was not specifically designed to address this question, and that he believes the data under represent differences in fire frequency he observed in the field, particularly on lower, north-facing slopes. Although analysis of available data showed an approximate 10% higher occurrence of fires on upper slope plots as compared to lower slopes, I increased the differential to 20% in recognition of sampling design weaknesses inherent in the available data and the observations by Weisberg (1999). Accordingly, I applied a multiplier to adjust mean fire return intervals (MFRI) by fire regime calculated in Weisberg (1998) so that upper slopes burned 20% more often than lower slopes (Table 1).

The fire frequency regime map from Weisberg (1998) was then overlain with the stream class map from the Willamette National Forest database to determine the miles of stream in each stream class by fire frequency regime (Table 2). The proportion burned in an average 20-year time period was calculated by dividing 20 years by the MFRI for each fire regime. These proportions should be understood as a general average; in reality fire frequency has been highly variable over time (Teensma 1987, Morrison and Swanson 1990, Weisberg 1998). The proportions burned in a 20-year period were then multiplied by the miles of stream in each stream class for each fire frequency regime to estimate the miles of stream where adjacent riparian and lower slope areas burned in an average 20-year period (Table 3).

Differences in fire severity by slope position have not been demonstrated quantitatively to date. However, Weisberg (personal communication - 7/22/99) believes that fires sometimes burned less severely in riparian and lower slope positions than on upper slopes. This observation has been echoed by other observers, although the opposite relationship has also been observed (e.g., some drainages in the Warner Creek fire, personal observation). For purposes of this assessment, and because of the current emphasis on aquatic ecosystem conservation and past practices involving clearcutting to streams edge, I will assume lower fire severity in lower slope positions by decreasing expected mortality by 20% in lower slope positions (Table 3). The relationship among fire severity and distance from a stream remains a key area in need of further study.

Results:

Fire frequency regime	MFRI (Weisberg 1999)	Adjusted MFRI - lower slopes	Adjusted MFRI - upper slopes
High frequency	129	141	117
Moderate frequency	176	192	160
Low frequency	247	269	225

Table 1. Adjusted fire frequency (MFRI) for lower slope positions

Stream Class	Fire frequency regime (mean fire return interval (MFRI))			Totals
	High (141 years)	Moderate (192 years)	Low (269 years)
I	6.22	14.29	00.12	20.63
II	5.03	30.09	18.81	53.93
III	9.52	22.18	35.29	66.99
IV	78.24	90.87	85.63	254.74

Table 2. Miles of stream by fire regime and stream class

Stream Class	Fire regime Frequency (proportion burned each 20 years)/ Severity (overstory mortality)			Totals
	High (.142)/ Low (30%)	Moderate (.104)/ Moderate (50%)	Low(.074)/ High (70%)
I	.88	1.49	0.0	2.37
II	.71	3.13	1.39	5.23
III	1.35	2.31	2.61	6.27
IV	11.11	9.45	6.34	26.9

Table 3. Miles of stream where adjacent riparian and lower slope areas burned each 20 years by fire frequency and severity regime

IV. Earthflow soils

Information need: amount of earthflow soils burned in an average 20-year time period

Methods: A GIS map of earthflow soils taken from the Willamette National Forest Soil Resource Inventory (SRI) was combined with the Weisberg (98) fire frequency regime map to yield the amount of acres of earthflow soils in each fire regime. The proportion burned each 20 years (20/MFRI) by fire regime was then multiplied by the acres of earthflow soils in each fire regime to calculate the acres burned by fire regime (Table 4) per average 20-year time period. Average overstory mortality proportions are displayed for each fire regime.

Results:

Earthflow soils	Fire regime Frequency (MFRI)/ Severity (overstory mortality)			Totals
	High (129 years)/ Low (50%)	Moderate (176 years)/ Moderate (70%)	Low (247 years)/ High (85%)
Acres	1982	1765	1794	5541
Proportion burned per 20 years	.155	.114	.081
Acres burned per average 20-year period	307	201	145	653

Table 4. Amount of earthflow soils burned per average 20-year time period

V. Shallow soils on steep hillsides

Information need: amount of soils identified as high potential for shallow, rapid landslides burned in an average 20-year time period

Methods: A map of soil types classified as shallow (<3 feet deep) from the Willamette National Forest SRI was combined with a terrain map stratified to identify hillsides with >70% slope steepness. This map was combined with the Weisberg (98) fire frequency regime map to yield the amount of acres of steep, shallow soils in each fire regime. The proportion burned each 20 years (20/MFRI) by fire regime was then multiplied by the acres of steep, shallow soils in each fire regime to calculate the acres burned by fire regime (Table 5) per average 20-year time period. Average overstory mortality proportions are displayed for each fire regime.

Results:

Steep, shallow soils	Fire regime Frequency (MFRI)/ Severity (overstory mortality)			Totals
	High (129 years)/ Low (50%)	Moderate (176 years)/ Moderate (70%)	Low (247 years)/ High (85%)
Acres	2520	2373	1419	6312
Proportion burned per 20 years	.155	.114	.081
Acres burned per average 20-year period	391	271	115	777

Table 5. Amount of earthflow soils burned per average 20-year time period

VI. Planning subdrainages (PSUB)

Need: area disturbed in each planning subdrainage by 20-year time period

Methods:

A GIS map of planning subdrainages (PSUBs) was combined with the Weisberg (98) fire frequency regime map to yield the amount of acres in each fire regime in each PSUB (Table 6). The proportion burned each 20 years (20/MFRI) by fire regime was then multiplied by the acres of each fire regime in each PSUB to calculate the acres burned in each PSUB and the proportion of the PSUB burned per average 20-year time period (Table 7).

Results:

Planning subdrainage (PSUB)		Fire frequency regime (MFRI)						Totals
	High (129 years)		Moderate (176 years)		Low (247 years)
Blue Quartz		2756		2072		864		5692
Cook		1391		2492		1462		5345
Lower Blue		1488		538		13		2039
Lower Lookout		2100		1430		819		4349
Mack		186		717		1698		2601
Mann		455		1335		1851		3641
McRae		1218		1383		1250		3851
Mona Scout		1910		566		52		2528
Quentin		2218		2468		1766		6452
Simmonds		847		534		57		1438
South Blue Lake		1204		350		0		1554
Tidbits		1760		2907		2060		6727
Upper Blue		781		1511		2645		4937
Upper Lookout		1529		1760		1925		5214

Table 6. Acres in each fire regime in each planning subdrainage (PSUB).

Planning subdrainage (PSUB)	Fire frequency regime (proportion burned in an average 20-year time period)			Total burned per 20 years	% burned per 20-years
	High (.155)	Moderate (.114)	Low (.081)
Blue Quartz	427	236	70	733	12.88
Cook	216	284	118	619	11.57
Lower Blue	231	61	1	293	14.39
Lower Lookout	326	163	66	555	12.77
Mack	29	82	138	248	9.55
Mann	71	152	150	373	10.25
McRae	189	158	101	448	11.63
Mona Scout	296	65	4	365	14.43
Quentin	344	281	143	768	11.91
Simmonds	131	61	5	196	13.66
South Blue Lake	187	40	0	227	14.6
Tidbits	273	331	167	771	11.47
Upper Blue	121	172	214	507	10.28
Upper Lookout	237	201	156	594	11.38

Table 7. Acres and proportion burned in each PSUB per average 20-year time period.

VII. Literature cited:

Morrison, P., and F. J. Swanson. 1990. Fire history and pattern in a Cascade Range landscape. USDA Forest Service General Technical Report PNW-GTR-254. Pacific Northwest Research Station, Portland, Oregon, USA.

Sousa, W. P. 1984. The role of disturbance in natural communities. Annual Review of Ecology and Systematics 15:353-91.

Teensma, P. D. A. 1987. Fire history and fire regimes of the Central Western Cascades of Oregon. Dissertation. University of Oregon, Eugene, Oregon, USA.

Weisberg, P.G. 1998. Fire history, fire regimes, and development of forest structure in the central western Oregon Cascades. Ph.D. dissertation, Oregon State University. 256 p.

Weisberg, P.G. 1999. Email message to John Cissel, dated 7/22/99.

Topic #3 - Spatial variability of retention trees

Date:   10/21/99

Reviewers: Geary, Mayo, Seitz, Overton, Ford

Question:

Are we gaining enough variability at the block scale?

Reason for Question:

Variability at the block scale is attained by using the rules for retention to respond to the variety of site conditions within any given block. During each project-level analysis that has been conducted thus far questions have arisen as to whether the 'rules' result in enough variability. Initial monitoring efforts on a small sample of units appear to show that block scale variability is being achieved

In a review of the rules for retention distribution it was noted that there was no language addressing the need for gaps within blocks.

Language addressing the placement of retention trees appears in a variety of places in the document, which has led to confusion when applying the prescriptions during project planning.

Relevant Study Component:

Spatial pattern of Retention trees, page 13 states that "the intent is to create a variable pattern of retention trees within landscape blocks".

Adaptation Options:

A)Create additional rules specifically one for gaps

B)Reorganize the document addressing rules for retention (complete after integration of the watershed process update)

C) Develop an example showing how to apply the rules.

Recommended Adaptations:

Include language to address gaps in the rules for retention. Rewrite the rules for retention bringing in language that appears in the ACS analysis and in Appendix A. (attached below)

Create a guide showing how to apply the rules for retention at the block scale. (attached below)

Spatial pattern of retention trees:

These guidelines are intended to help translate spatial objectives for retention of live overstory trees at the time of timber harvest from the landscape level to the stand level, and to provide a basis for evaluation of the landscape plan. The intent is to create a variable pattern of retention trees within landscape blocks. Final placement of retention trees should integrate these criteria and fit on-the-ground conditions assessed at the time of timber sale planning. To the degree allowed by the need to protect ecological values, spatial patterns of retention trees should use site-specific disturbance patterns as a general template.

Retention trees are intended to maintain a more natural forest pattern, to provide wildlife habitat, and to integrate upslope and riparian management. Placement of retention trees along edges of cutting blocks should be designed to 1) minimize edge contrast, 2) avoid sharp boundaries with high wind throw potential or abrupt microclimate shifts, 3) emulate common post-fire patterns, and 4) maintain nutrient uptake capacity across the hillslope down to the riparian zone, Hardwood trees should generally be left standing where feasible, but are not considered part of the retention tree component of these prescriptions.

Overall guidelines:

• Retention trees should be both clumped and scattered individuals. Retention trees should be left out of some areas to create gaps (see #9).
• Clumps should range in size from _ acre to 5 acres.
• Rock outcrops, wet areas or other special or unique habitat could be used to anchor retention clumps.
• Larger blocks should have larger clumps.
• Scattered individual trees can range from 40 to 70% of the total retention trees (see Appendix a).
• Retention trees should include the largest, oldest live trees, decadent or leaning trees, wolf trees, and hard snags.
• Retention tree species mix shold meet the goals outlined in Appendix a.
• Spatial patterns of retention trees should consider the structure and timing of future cutting in adjacent blocks, and minimize edge contrast where feasible.
• There should be gaps in each landscape area. Landscape Area 1 should have twice as many 100% retention areas as 0% retention areas. Landscape Area 2 should have the same ratio of 100% rretention areas as 0% retention areas. Landscape Area 3 should have twice as many 0% retention areas as 100% retention areas.
• Minimum size of gaps is one tree height by one tree height. This will allow for removal of trees and for understory regeneration.
• Small landscape blocks should have relatively smaller gaps and large landscape blocks should have larger gaps.
• When unsuited or unavailable lands occupy a significant portion of the landscape within a block it may be appropriate to reduce the general target green-tree retention guidelines (see section titled "Land Unsuited or Unavailable for Timber Management").
• When a significant portion of the landscape block is currently in a clear-cut, or young conifer plantation that wont’ be harvested in the first entry, it may be appropriate to increase the overall green-tree retention level for the remainder of the block.

Riparian guidelines:

• No trees should be cut on any floodplain or streambank, nor should trees directly contributing to streambank stability be cut.
• Higher levels of retention should generally be located near streams and lower slope positions. Lower levels of retention should generally be located on upper slope areas.
• Fishbearing streams are mapped as reserved areas. The distance of the reserve depends on whether the channel is constrained or unconstrained. The trees left within this distance do not count towards the prescription retention levels.
• Perennial streams are not mapped as reserve areas but have an additional retention requirement above the prescription retention levels. The prescriptions for these streams vary by Landscape Area. In Landscape area 1, 70% canopy closure shall be maintained within _ tree height (86 feet) of the stream. In Landscape area 2 50% canopy closure shall be maintained within that same distance. These retention levels are prescribed to meet the Aquatic Conservation Strategy objectives. These trees do not count toward overall prescription retention levels.
• In areas identified as water quality source areas 70% canopy closure will be retained within one site potential tree (172 feet). On streams flowing east-west, the entire buffer will be situated on the south side of the stream. On streams flowing north-south, the buffer will extend for one half site potential tree height on each side of the stream. These trees do not count toward overall prescription retention levels.
• Areas of blocks where soils are shallow and occupy slopes greater than 70% are considered large wood and coarse sediment source areas. Retention trees will be left scattered evenly on these areas to achieve 50% canopy closure.
• On streams within earthflow terrain, or within areas of glacial deposits on potentially unstable earthflow terrain retain a one-site tree no harvest buffer. This will maintain a supply of large wood while maintaining stability and retention of fine sediments.
• In Landscape Area 3 for every 1000 feet of stream reach approximately _ acre of 100% canopy closure would be retained to provide for riparian and terrestrial processes. The extra trees provided will count toward overall retention levels.
• Intermittent streams are not mapped as reserve areas and in general have no additional retention requirements. However, a portion of the overall retention trees should be placed near these streams. 50% canopy closure should be retained along these streams in Landscape area 1, 30% canopy closure in landscape Area 2 and 15% canopy closure in Landscape area 3.
• Where intermittent streams flow through earthflow areas, a _ to full site tree no harvest buffer is prescribed. These trees do not count toward the overall retention level. This will maintain a supply of large wood while maintaining stability and retention of fine sediments.
• In Landscape Area 3 for every 1000 feet of stream reach approximately _ acre of 100% canopy closure would be retained to provide for riparian and terrestrial processes. The extra trees provided will count toward overall retention levels.
• Retention trees and clumps should be placed on sites of potentially unstable ground, and on localized areas adjacent to streams prone to streamside slides, to the degree needed to minimize mass movement risks.
• If on-the-ground conditions indicate that higher levels of retention trees are needed to meet ecological objectives, prescriptions should be modified accordingly. Similarly, reductions in retention levels may be appropriate in some instances to improve operational feasibility, as long as ecological objectives are met.

Example: Landscape Area 1

Perennial (Class III) and Intermittent (Class IV) Streams

Given:

100 acre Landscape Block

4000 ft Class III perennial stream, 20 acres in reserves within either side of stream

30 acres in north facing slopes

20 acres in south facing slopes

Step 1 -  Determine if block is in a Sediment/Large Wood Source Area or a Water Quality Source Area.

Block does not contain a Sediment/Large Wood Source Area.

Block is in a Water Quality Source Area.

Leave 70% CC on the south side of the east-west running stream for a distance of 1 site tree. Leave no buffer on the north side of the same stream. Leave 70% CC on both sides of the north-south running stream for a distance of _ half site tree on each side. The trees being placed for 100% CC will not count against the retention trees for the block.

Step 2 - Determine if there are any Class III perennial streams.

Since the south side of the east-west facing stream is being buffered due to the stream being in a Water Quality Source Area, then only the north side will need the standard Class III stream treatment of 70% CC retention a distance of _ site tree.

The north-south stream is also buffered because of the Water Quality Source Area. The buffer distance and canopy matches that of what is required with the standard Class III treatment.

These trees are considered extra to the retention trees.

Step 3 - Determine if there are any Class IV intermittent streams.

There are no Class IV streams.

Step 4 — Determine 100% Canopy Closure (CC) areas

IDT chooses 20% of the area to be in 100% CC using Appendix A.

20 % of 100 acres is 20 acres.

Both Class III streams will need the required 100% CC areas.

There are 19 acres remaining of 100% CC that need to be placed in block.

20 acres 100%CC - 1 acres along stream = 18 acres

IDT determines that the 18 acres of 100%CC will be placed around the lichen of concern, the steep headwall area, and scattered in the block in four clumps.

Step 5 - Determine 0% Canopy Closure (CC) areas.

Create 0% CC areas that total about half the area as the 100% CC areas.

Half of the 20 acres in 100%CC

= 10 acres in 0%CC.

IDT determines size of 0%CC to be large enough to facilitate the regeneration of Douglas-fir (approximately 1 tree height in size) which results in a _ acre size gap. The IDT decides to scatter the 0% CC gaps across the block in 9 to 10 clumps.

Step 6 - Determine number of trees to scatter

100          acres

- 20          acres along Class III

streams

- 20          acres of 100%CC area

- 10          acres of 0% area

50            acres to scatter trees across

20% of area in 100%CC results in 61% of the retention trees scattered. This results in a 37% CC of the scattered trees.

Since the average tree size is 20 inches, and a 20 inch tree equals 1% canopy per acre, there will need to be on the average, 37 trees per acre.

37 trees per acre x 50 acres = 1850 trees total to scatter.

Determine amount and area of trees to scatter.

The IDT decided that the north facing slopes should have a Canopy Closure of 40%. To accomplish this, 40 trees per acre will need to be left.

40 trees per acre = 1200 trees to scattered over 30 acres

The balance of the scattered trees should be placed on the south facing slopes.

1850 trees — 1200 trees = 650 trees/ 20 acres = 33 trees per acre (33% CC)

Step 7 — Determine number and location of wildlife and down wood trees.

Scatter 5% of the original stand of the trees for wildlife and down wood trees.

Example: Landscape Area 2

Given:

100 acre Landscape Block

20 acres unsuitable soil

11 acres in Sediment/Large Wood Source Area

Step 1 -  Determine if block is in a Sediment/Large Wood Source Area or a Water Quality Source Area.

Block contains a Sediment/Large Wood Source Area of approximately 11acres. Leave a canopy closure of 50% evenly scattered across the Sediment/Large Wood Source Area. The trees being scattered will not count against the retention trees for the block.

Average Tree is 20 inches, which has a 1% canopy per acre. 50% x 11 acres x 1% CC avg per tree = 550 trees needed.

Block is not in a Water Quality Source Area.

Step 2 - Determine if there are any Class III perennial streams.

There are no Class III streams.

Step 3 - Determine if there are any Class IV intermittent streams.

There are no Class IV streams.

Step 4 — Determine 100% Canopy Closure (CC) areas.

IDT chooses 9% of the area to be in 100% CC from Appendix A.

100 acres - 20 acres unsuitable not counted in block acres = 80 acres

9 % of 80 acres is 8 acres.

There are no streams, which will need any 100% CC areas.

c) IDT determines that all of the 8 acres of 100% CC should be placed in area of scattered unmappable unsuitable areas and that it should tie in with the larger unsuitable area.

Step 5 - Determine 0% Canopy Closure (CC) areas.

Create one 0% CC area that is approximately the same size as the 100% CC area.

2 times 8 acres = 16 acres

Place 0% CC area on upper slopes.

Step 6 - Determine number of trees to scatter.

80 acres — (8 acres of 100% and 16 acres of 0%) = 56 acres.

9% of area in 100% results in 40% of the retention trees scattered. This results in a 7% CC of the scattered trees.

Since the average tree size is 22 inches, and a 22 inch tree equals 1% canopy per acre, there will need to be on the average, 7 trees per acre.

7 trees per acre x 56 acres = 392 trees total to scatter.

Determine location of trees to scatter.

Place 60% of the scattered trees in the lower half of the slope.

392 trees x 60% = 235 trees to scatter over 28 acres (8.4 trees per acre)

Place 40% of the scattered trees in the lower half of the slope.

392 trees x 40% = 157 trees to scatter over 28 acres (5.6 trees per acre)

Step 7 — Determine number and location of wildlife and down wood trees.

Scatter 10% of the original stand of the trees for wildlife and down wood trees.

Example: Landscape Area 3

Given:

100 acre Landscape Block

3000 ft Class III perennial stream, 10 acres within a half site tree either side of stream

2000 ft Class IV intermittent stream, 7 acres within a half site tree either side of stream

30 acres in north facing slopes

42 acres in south facing slopes

Step 1 -  Determine if block is in a Sediment/Large Wood Source Area or a Water Quality Source Area.

Block does not contain a Sediment/Large Wood Source Area or Water Quality Source Area.

Step 2 - Determine if there are any Class III perennial streams.

Leave 30% retention within _ site tree (75 feet) both sides of creek in Landscape Area 3. These trees are considered extra to the retention trees.

Step 3 - Determine if there are any Class IV intermittent streams.

Leave 15% retention within _ site tree (75 feet) both sides of creek in Landscape Area 3. Calculate the trees needed and subtract from retention trees.

Average Tree is 20 inches, which has a 1% canopy per acre.

15% x 7 acres x 1% per avg tree

= 105 trees needed

Step 4 — Determine 100% Canopy Closure (CC) areas

IDT chooses 4% of the area to be in 100% CC from Appendix A.

4 % of 100 acres is 4 acres.

Both the Class III and Class IV streams will need 100% CC areas along the streams.

Class III and IV streams need one _ acre 100% CC area for every 1000 reach.

5000 feet of stream/ 1000 feet = 5 _ acre 100% CC areas needed.

5 x _ acre = 1.25 acres in 100% CC along streams.

There are 2.75 acres remaining of 100% CC that need to be placed in the block.

4 acres 100%CC — 1.25 acres along stream = 2.75 acres

The IDT has determined that the 2.75 acres will be placed around seeps.

Step 5 - Determine 0% Canopy Closure (CC) areas.

Create one 0% CC area that is twice as large as the 100% CC area.

2 times 4 acres = 8 acres

IDT places the 0% CC area on ridge top in two separate areas.

Step 6 - Determine number of trees to scatter

100          acres

- 10          acres along Class III

stream

- 4            acres of 100% CC area

- 7            acres along Class IV

stream

- 8            acres of 0% area

71            acres to scatter trees across

4% of area in 100%CC results in 70% of the retention trees scattered. This results in an 11% CC of the scattered trees.

Since the average tree size is 20 inches, and a 20 inch tree equals 1% canopy per acre, there will need to be on the average, 11 trees per acre.

11 trees per acre x 71 acres = 781 trees total to scatter.

Determine location of trees to scatter.

Place 60% of the scattered trees on the north facing slopes.

781 trees x 60% = 469 trees to scatter over 30 acres (15.6 trees per acre)

Place 40% of the scattered trees in the south facing slopes.

781 trees x 40% = 312 trees to scatter over 42.2 acres (7.4 trees per acre)

Step 7 — Determine number and location

Scatter 15% of the original stand of the trees for wildlife and down wood trees.

Topic #8 - Commercial Thinning Regimes

Date: 3/10/00

Reviewers: Geary, Mayo, Seitz, Overton, Ford

Compiler: Jim Mayo

Question: How many entries fit with the low frequency fire regimes of LA-2 and LA-3? Should the number of entries be limited to minimize "disturbance"? Does it matter if the stand is of natural origin or managed, when prescribing treatments in a refugia?

Reason for Question: The concept of refugia and the number of entries was not fully discussed in the original paper, and thinning regimes were based on limited analysis. Natural stands of commercial thinning size and age class were not analyzed separately from managed stands.

New Information: A report by Steve Garman, OSU Department of Forest Science: "Accelerating Development of Late-Successional Conditions in Young Managed Douglas-fir Stands: A Simulation Study". The gap model ZELIG.PNW (3.0) was used in the simulations.

Relevant Study Component: Management prescriptions, Table 2 (BRLS).

Adaptation Options:

Minimize number of disturbances (thinning entries) over the harvest cycle.

Use a mix of prescriptions based on site and stand characteristics, LSI objectives and risk factors.

Evaluation:

Zelig Runs: Three timing choices for four levels of thinning intensity, resulted in 64 experimental prescriptions analyzed by the Zelig model. Of these, 35 had volumes less than 700 cubic feet per acre being removed (usually a light thin) which would not be an economically viable prescription. Twelve others resulted in periods of time greater than 200 years for the stand to reach late-successional habitat conditions. The late-successional index (LSI) is based on three live tree factors: 1) density of large boles > 40 inches dbh is four or more per acre; 2) the canopy height diversity index is eight or greater; and 3) the density of shade-tolerant species >16 inches dbh is four or more per acre. Snags and coarse wood are also important components, but they can be created if not present in the stand. This left 17 viable commercial thinning prescriptions of the 64 possible. Of these, eleven include two thinning entries, four have one entry, and two have three thinning entries. These prescriptions can be divided into three general groups based on the intensity of the initial thinning entry at age 40: no thin, moderate thin and heavy thin (see Tables LA-2 and LA-3).

No thin at age 40: There are ten prescriptions in this group (experiments 49-57, and 61). Generally this group has good volume at each thinning entry, and the best or near-best volume (95-100%) for the rotation. The time it takes to reach LSI ranges from 130 to 181 years. Postponing the first thin till age 60 can result in undesirable stand conditions. Relative densities will be greater than 55, a point where competition causes mortality. Trees will have a height/diameter ratio approaching 100 (height in feet divided by dbh in feet) which is an unstable condition. This makes trees prone to ice and snow damage, and when the stand is opened up there is a much higher risk of blowdown. A light thin can minimize the risk of blowdown, but prolongs the time to LSI. A heavy thin should only be prescribed in areas with a low risk of blowdown (lower slopes and broad valley bottoms).

Moderate thin: There are only three prescriptions in this group (experiments 17, 18, 20). This group has good volume at each of the thinning entries, and good overall volume (89-92%) for the rotation. The time to LSI is relatively short, ranging from 138-167 years. A heavy thin is needed at the second entry (age 60) to develop LSI in less than 200 years. A moderate first entry reduces the risk of blowdown, and allows the crown and root systems to develop before implementing the heavy thin and starting a new cohort at age 60.

Heavy thin: There are four prescriptions in this group (experiments 4, 13, 14, 16). Generally these have good to fair volume at each thinning entry, but the lowest overall volume (82-86%) for the rotation. The time to LSI includes the fastest prescription and ranges from 116-174 years. This group includes only two entries, skipping either the second or third entry. A heavy first thin has a higher risk of blowdown, and should be used sparingly. It also leaves the crown closure at or below 40%, for a time, which could affect its use for some wildlife species.

Recommended Adaptations With Rationale:

Non-refugia: A moderate first thinning entry (similar to experiment 18) provides a good balance between volume, time to LSI, and risk of blowdown. The stand is opened up enough to promote development of crowns, diameters and root systems. This will reduce the risk of blowdown when the stand is opened up to promote development of the second cohort at age 60.

In areas with a low risk of blowdown, a heavy first thin (similar to experiment 4) would be an option that accelerates development of LSI, but at a significant cost to overall volume production.

Refugia: The main concern in these areas is limiting the number of entries over the 40 year refugia period. The option of skipping one or two thinning entries in dynamic young stands involves some tradeoffs. Not thinning at age 40 and/or 60 can produce good overall volumes, including the maximum. However, the stand becomes very dense, with small crowns and small diameters. After delaying entry, the first thin can have a high risk of blowdown, and care must be taken during the planning process so as not to exacerbate the risk. These prescriptions should be used in combination with other choices to produce relatively long periods (ten years or more) of quiet within the refugia. This means that some stands will have a thinning at age forty, and some will be thinned at ages older or younger, so that periods of disturbance are minimized (perhaps three entries over the 40 year interval). These changes will affect volume production, risk of blowdown, and the time it takes to reach LSI, but the impact should be relatively minor.

Another consideration for stands in refugia, is whether they originated as a result of natural disturbance or management activities.

Managed stands: These stands have been impacted by past harvest, burning of slash, reforestation, precommercial thinning, etc. Unnatural conditions resulted from planting heavily to insure regeneration within five years as required by law. A commercial thinning entry is a relatively small disturbance compared to what the site has been through.

Natural stands: These stands most commonly originated after a stand replacement fire, and regeneration took place over a longer time period and with a significant brush component. Stocking levels were generally much lower, and stands may not have gone through a stem exclusion seral stage. These stands should be evaluated on a case by case basis to see if a commercial thin is desired. Stand characteristics like stocking level, structure, species diversity of the overstory and understory, and canopy height diversity should be considered. Other factors include the existing road network, stream conditions, and how long until the scheduled regeneration harvest for the block? The main reason for commercial thinning is to accelerate the development of late-successional habitat, and maintaining it for a significant period of time.

Topic #11 - Crown Closure

Date: 3/10/00

Reviewers: Geary, Mayo, Seitz, Overton, Ford.

Compiler: Jim Mayo

Question: Is crown closure the appropriate criteria for implementing the Blue River Landscape Strategy? If so, what is the appropriate measurement technique and implementation procedure that should be used to ensure the expected outcome?

Reason for question: There are a variety of forest measurements that could be used as goals for desired conditions including trees per acre, basal area per acre, density per acre as well as crown ratio. Using crown closure as the desired outcome requires a translation into one of these other measurements to implement on the ground. There have been questions around the use of crown closure as the goal and questions around the measurement and conversions used to place the prescriptions on the ground.

Relevant Study Component: Crown closure is the basic desired outcome associated with each of the landscape area. Page 12 discusses the objectives of each landscape area and the section labeled "General landscape area prescriptions" explains that each of these areas are meant to approximate key elements of various frequencies and intensities of fire. Specific crown closure objectives are listed in the tables. The original landscape plan does not discuss crown closure or the reasons it was selected as the measurement of choice.

Adaptation Options:

Retain crown closure as the desired outcome, define it more specifically, add the rationale for its selection, outline measurement techniques and implementation procedures.

Use a different measurement, such as relative density, trees per acre, or basal area per acre..

Evaluation:

Outcome

Crown Closure

Pros:    Crown closure is used by stream temperature models

Many wildlife habitat conditions rely on crown closure

Crown closure is easy to visualize and communicate

Easy measure to help estimate light to the lower canopy/forest floor

It is a specific end result, independent of tree size and the number of trees

Cons:   Crown closure must be converted to another meaurement, such as trees per acre (spacing) to implement.

Pros:    Easy to implement

Cons:   Not easy to visualize

It could be all big trees or all little trees resulting in drastically different crown canopies

Basal Area

Pros:    Easy to measure

Cons:   Not easy to visualize

Hard to understand

Relative Density

Pros:    Combines basal area and diameter resulting in numbers which represent the density of the stand regardless of the stand’s age.

Cons:   Hard to understand and visualize.

Not used in wildlife habitat descriptions.

Implementation Techniques

The use of crown closure as an objective requires that it be converted to a different measurement that can easily be used by a marking crew. There are several studies that have tied crown width to diameter at breast height. In a normally stocked stand, crown width in feet is about equal to dbh in inches (CW/D = 1). A tree with a 20 inch dbh will have a crown width of 20 feet. This relationship varies with the stand density such that open grown trees will have a CW/D ratio of 2, and densely stocked stands will have a ratio of about 0.7. Most of the studies referenced below, had similar results from widely distributed areas ranging from southern Oregon to British Columbia. The empirical data from stand exams on Blue River are similar to the other studies, and were used for planning the Blue River Landscape Study timber sales.

Implementation can be accomplished with one or a combination of the following techniques:

Diameter Limit:

Crown closure has been converted to a diameter limit used in marking guidelines on three projects on the Blue River District thus far. Stand exams were used to develop tables of diameter distribution for each stand, with the amount of crown closure listed by one-inch diameter classes (Attachment 1). The desired crown closure was assigned from the largest trees down, until the target was reached. The largest trees are more likely to have survived a fire. This resulted in a general diameter limit to use for the stand. As a backup and check, the residual basal area and trees per acre were also given for marking guides. In some cases, where more consistent spacing was desired (wildlife habitat, or potentially unstable area), the marking guides included an average spacing. During marking, the crew could easily check the basal area from time to time, and each crewmember kept a tally of trees marked to leave, so the total number could be checked against the prescription.

Using diameter limits results in some randomness in spacing which is similar to what occurred naturally with a fire. Location of gaps and clumps are also important, and must be mapped out in advance and averaged into the total crown closure for the block.

Pros:    Better chance to get best crown, since largest trees usually have the biggest crown

Larger trees should be more windfirm

More spatial variation

Meets wildlife needs

Leave trees that the fires left

Cons:   Added emphasis on stand exam. If data is incorrect, the wrong diameter limit could be selected

Trees per Acre

The target crown closure is translated into the number of trees necessary, given the crown diameter for trees in the stand. The resulting trees per acre are expressed as a spacing or as a total tree count.

Pros:    Easy to monitor implementation

Cons:   Spacing will result in less stand variation

Basal Area

The target crown closure is translated into the number of trees necessary, then into basal area, given the crown diameter for trees in the stand. The resulting basal area is expressed as a basal area per acre.

Pros:    Easy to monitor implementation using a prism or relaskop.

Cons:   Harder to implement

Selection by Canopy

The stand is marked by selecting trees that will achieve the target crown closure regardless of tree spacing or basal area.

Pros:    Will most closely attain desired crown closure compared to all the other alternatives

Cons:   May result in less stand variation

Difficult to implement on 15 and 30% crown closure prescriptions (harder for markers to visualize)

More difficult to monitor

Information Sources:

"How to Estimate Canopy Cover Using Maimum Crown Width/DBH Relationships". Warbington and Levitan, 1992.

"Studies of Crown Development are Improving Canadian Forest Management". J.H.G. Smith, 1966.

"Tables for Quantifying Competitive Stress on Individual Trees". James D. Arney, 1973.

"Maximum Crown Width Equations for SW Oregon Tree Species". D. P. Paine and D.W. Hann, 1982.

Measuring techniques

a) Aerial photos — large scale aerial photos that allow individual live tree crowns to be seen is a method that could be used to measure crown widths after harvest.

b) Moosehorn densiometer- allows a vertical look upward from the ground

c) Spherical densiometer — measures light that reaches the ground but when used to measure crown cover it measures it from the sides as well as vertically.

Recommended Adaptations With Rationale:

Outcome

The review team agreed that crown closure is the best criteria for measuring what is to be left on the site to represent the various fire intensities that are being used in the landscape strategy. Crown closure represents a desired outcome that is independent of pretreatment stand conditions.

However, a clear definition of crown closure is needed, as well as implementing procedures and measurement techniques.

Definition: Crown closure is the percent of area covered by live tree crowns as measured vertically from above or below.

Measurement: Aerial photos, at a large enough scale that individual tree crowns can easily be seen, is a method that could measure percent crown closure from above. From the ground, the Moosehorn densiometer allows a vertical look upward to measure crown closure.

Implementation: The diameter limit procedure used for the past three projects appears at this point to result in meeting the crown closure objectives outlined in the prescriptions. Desired crown closure also needs to be converted to a measure that can easily be used by a marking crew given their experience (TPA and BA) with the final guidelines usually including a blend of all of the techniques.

Topic # 15 - Prescriptions for LA-1

Date: 3/10/00

Reviewers: Geary, Mayo, Seitz, Overton, Ford

Compiler: Jim Mayo

Question: Do we need to evaluate RX for LA1? Is there a need to change prescription elements for retention and regeneration rules, or configuration of gaps and clumps to ensure Douglas fir reproduction and growth?

Reason for Question: Ground observations and discussions as to how DF will grow under various retention levels. Johnston/Barbour modeling shows a shift from shade intolerant (SI) to shade tolerant (ST) species at crown closures greater than 40%.

Relevant Study Component: s: Management prescriptions, Table 2.

Current LA-1 RX elements, Table 2

BRLS Weisberg

RET.    SI 65%                                    SI-90%

ST-35%                       ST-10%

REF.    SI- 40%                       SI-70%

ST-60%                       ST-30%

Spatial Pattern of Retention Trees — page 13. Specify clump/gap ratio.

Adaptation Options:

Retention Mix:

Use data from existing table 2. (SI=65%, ST=35%)

Use data from Weisberg. (SI=90%, ST=10%)

The species mix of retention trees should reflect the current condition of the overstory (dominants and co-dominants) for the stand being harvested.

Reforestation Mix:

Use data from existing table 2. (SI=40%, ST=60%)

Use data from Weisberg. (SI=70%, ST=30%)

Plant only shade intolerant species and rely on natural regeneration for shade tolerant species.

Configuration of Gaps and Clumps:

Use an equal area of gaps and clumps so that the matrix will be an average of 50% canopy closure.

Use a ratio of 2:1 for clumps to gaps, so that the matrix will be an average of 40% canopy closure or less.

Recommended Adaptations (with rationale):

Retention Mix: The review team agreed that the species mix of retention trees should reflect the current mix of dominant and codominant trees in the overstory of the stand that is being harvested. This provides flexibility to recognize the range of conditions found in the natural landscape. This change will apply to Landscape areas 2 and 3 as well.

Reforestation Mix: The review team agreed that the main concern is for maintaining Douglas fir in these stands. There is also a desire to introduce rust-resistant western white pine into these stands. Western white pine is also more shade tolerant that Douglas fir, and should do well in LA-1 and 2. Western hemlock and western redcedar are expected to seed in naturally over time and grow well under the relatively heavy overstory. Planting is not required and the genetic program is not as advanced as for Douglas fir and white pine. The recommendation is to plant Douglas fir and white pine to ensure their presence in Landscape areas 1 and 2. Planting will take place in the gaps and matrix only. Total trees to be planted will average 140 per acre in LA-1 and 200 per acre in LA-2 (75% DF and 25% WP), and will be somewhat variable with more being planted in gaps and less in the matrix. Spacing will range from 12 feet in the gaps to 20 feet in areas with 30-40% crown closure.

Configuration of Gaps and Clumps: The ratio of clumps to gaps will generally be 2:1. This will allow the overall block to average 50% canopy closure, while the matrix will average 40% or less. This is based on modelling with Organon that showed Douglas fir could survive and grow relatively well at 40% crown closure, but tended to drop out of the stand at 50% or more. Western white pine, being a little more shade tolerant, should also do well.

These changes will be reflected in Table 2.

Topic #: 19 - Complex prescriptions - fire history

Date: 4/10/00

Reviewers:

Blue River Landscape Team, J. Cissel compiler

Question:

Should the prescriptions be modified to include concepts of multiple frequency and intensity timber harvests for each landscape area to more accurately reflect the fire history of the watershed?

Information source(s):

Recommendations contained in Weisberg, 1999 ("An evaluation of the Blue River landscape project: how well does it use historical fire regimes as a model?", unpublished report on file at the Blue River Ranger District.)

Relevant study component:

Landscape area prescriptions

Background:

The Weisberg evaluation report (99) contained several recommendations to more closely align management practices with information reported in the Weisberg fire history thesis (98). Pete recommended that we consider adding more complexity to the prescriptions by employing multiple cycles of timber harvest with varying intensity in each landscape area. He felt that varying timber harvest intensity and frequency over time in a given landscape area would more closely approximate the highly variable fire patterns he observed in the watershed. For example, landscape area one currently has a more frequent (100 year rotation) - lower intensity (50% canopy cover retention) timber harvest prescription. Under a more complex prescription there might be two cycles: one for a 50% canopy cover retention cut every 100 years, and a second with a 15% canopy cover retention cut every 400 years.

Evaluation:

The landscape team considered this recommendation in informal discussions and in team meetings.

Recommendations:

The team recommendation is to not make these changes at this time, and to keep these ideas in mind for future modifications of the landscape management plan. The team feels there is merit to the recommendation, but that it is premature to significantly modify our existing prescriptions since we have not yet been able to fully implement and evaluate the prescriptions in the first iteration of the plan.

Topic # 14 - Adjustments to landscape area one boundaries

Date:

4/10/00

Reviewers:

Blue River Landscape Team, J. Cissel compiler

Question:

Should the landscape area one boundary be moved further up the watershed to conform with the recommendation in the Weisberg evaluation report (99)?

Information source(s):

Recommendation contained in Weisberg, 1999 ("An evaluation of the Blue River landscape project: how well does it use historical fire regimes as a model?", unpublished report on file at the Blue River Ranger District.)

Relevant study component:

Landscape area mapping and related analysis

Background:

The Weisberg evaluation report (99) contained several recommendations to more closely align management practices with information reported in the Weisberg fire history thesis (98). Pete recommended that we consider moving the boundary between landscape areas one and two further up the watershed to better match the fire regime mapping in his thesis, and he provided a map with recommended boundaries.

Evaluation:

The landscape team considered this recommendation in informal discussions and in team meetings.

Recommendations:

The team recommends that we adopt Pete's suggestion. No issues surfaced with this recommendation.

Topic # 16 - Wetlands

Date: April 13, 2000

Reviewers: Ford, Kretzing

Question: How will wetlands be managed and incorporated into the Landscape Management and Monitoring Strategy?

Information sources: Willamette National Forest Land Resource Management Plan, Management Area 15 (WNFLRMP, 1990), the Northwest Forest Plan, ROD, C-30-31 (1994), Blue River Landscape Management and Monitoring Strategy (4/18/97), Special Habitat Management Guide (1996), several research papers by Chen and Chan, and project NEPA.

Relevant study component: Blue River Landscape Management Strategy, Aquatic Reserves, pages 19-20, Inclusions, pages17-18, and stream monitoring pages 69-70. These sections do not contain specific guidelines regarding wetlands and the management of these areas.

Background: The Blue River Landscape Management Strategy developed guidelines for aquatic reserves, and inclusions to meet the Aquatic Conservation Strategy Objectives. However, no direct guidelines address the management of wetlands.

Evaluation: Experience gained during the project analysis for Wolfmann Environmental Impact Statement determined that the aquatic reserves did not adequately provide for wetlands, and large wetlands did not fit the intent of inclusion.

Recommendations: The WNFLRMP standard and guideline FW-211, special wildlife and plant habitats, and the Northwest Forest Plan Riparian Reserves can be applied to wetlands within the landscape. However, modifications to the Landscape Strategy should be made to provide for the management of wetlands under the context of the Landscape Strategy objectives and to meet the Aquatic Conservation Strategy Objectives.

Topic # 20 - Prescribed Fire

Date:               2/1/01

Reviewers: Swetland, Cissel

Question:

Does more information on use of prescribed fire need to be included in the landscape strategy?

Reason for Question:

Through planning efforts on the Wolfmann Timber Sale, questions arose on the specific objectives and desired results for the burns. Review comments asked questions ranging from why we would risk large-scale fires to why aren’t we burning more. What changes to the description in the plan would answer these questions.

Relevant Study Component:

Prescribed Fire P. 15,16,17 and table 2

Adaptation Options:

A) Include or reference demonstrations of model outputs to graphically demonstrate the spatial and temporal implications of implementing the prescribed fire as written in the plan.

B) Strengthen description of scheduling and area selection criteria. First cut of criteria was done for Wolfmann T.S. Some of those considerations will affect actual implementation of the prescribed burn projects.

C) Strengthen the connection to monitoring and research to clarify objectives. Emphasize integration of all resource objectives with fire disturbance.

D) Leave it alone until information from monitoring of some activities drives need for modification of plan.

Recommended Adaptations With Rationale:

Clarification of prescribed fire objectives is a complex issue. As is stated in the strategy already, "…knowledge of the influence of fire on habitats and other ecological processes is incomplete, especially for low- and moderate-severity fires." The plan as written outlines justification of use of fire process and suggested burn area selection criteria from Wolfmann planning effort were based on finding low complexity, less controversial areas to begin using prescribed fire. Future implementation plans should continue along the course planned in the Strategy. Selection of areas within other project proposal areas for burning could be driven by other factors than those used in finding "easy" areas for the Wolfmann plan.

Some scheduling that is constant could be further elucidated. 1) Minimum stand age at the time of application of prescribed fire is at least 100 years for low stand mortality. 2) There should be at least two time periods (40 yrs) between the burn and other activities. 3) Feasibility of accomplishment to meet objectives could be defined in terms of percentage of block being treated, boundary locations being functional fire perimeter or other subjective measures such as political or policy driven limitations.