Case Study 1: Satsop Forest
The first case study uses a Habitat Evaluation Procedure
(HEP, USDI 1980) within LMS to meet the requirements of a
wildlife mitigation agreement on Satsop Forest in southwest
Washington (Ceder, 2001, URL: http://silvae.cfr.washington.edu/satsop-plan
). The agreement focused on the habitat needs of 5 species,
using previously defined Habitat Suitability Index (HSI) models.
The species were chosen to track changes in a variety of habitat
types: with the spotted towhee tracking changes in brush habitats;
the Cooper's hawk tracking changes in mixed hardwood conifer
forests; southern red-backed vole tracking changes in closed
canopy forests; the pileated woodpecker tracking changes in
mature forests; and black-tailed deer, a habitat generalist,
tracking overall changes. Twenty potential management alternatives
for Satsop Forest were developed ranging from 'no harvest'
to 40-year clearcut rotations with varying amounts, timings
and levels of thinning between these extremes. Assessments
of each alternative determined the amount of habitat and wood
volume that could be produced over an 80-year planning horizon.
Results indicate amounts of available habitats, similar to
no management or passive management, could be created through
active management (Figure 1). Through multiple thinnings,
which tree size and develop multi-storied stands, harvest
increased from 1.4 MMBF under the current mitigation agreement
to 5.4 - 30.1 MMBF with active management. Some species may
not be sensitive to forest management, even under the highest
level of harvesting. Cooper's hawk, southern red-backed vole,
and spotted towhee habitat values changed little as harvesting
increased. In contrast, habitat available for the pileated
woodpecker, which is associated with older forest structures,
decreased with higher harvest levels.
 |
| Figure 1: |
Habitat and volume production
for 20 potential 80-year management alternatives for Satsop
Forest. |
Case Study 2: Fuel Removal Strategies for Fremont and
Okanogan National Forests
Habitat modeling using LMS was included in the Investigation
of Alternative Strategies for Design, Layout and Administration
of Fuel Removal Projects (Mason, and others 2003). This project
examined effects of five fuel removal treatments and a wildfire
simulation using data from the Fremont and Okanogan National
Forests. Changes in wildlife habitat were assessed for northern
goshawk, Lewis woodpecker, white-headed woodpecker, and Williamson's
sapsucker using HSI models, while the wildlife habitat matrices
in the ICBEMP report were used for the Canada lynx, grizzly
bear, pileated woodpecker, northern flying squirrel, and Townsend's
big-eared bat were assessed using the . Assessments of lynx
and grizzly were done only for the Okanogan National Forest,
as they do not occur on the Fremont National Forest.
Initial habitat and fire risk relationships showed that stands
with high and moderate risk provided more habitat for the
majority of the species than the low risk stands on both forests.
This was particularly evident in species that are associated
with older forest structures such as the northern goshawk,
pileated woodpecker, and northern flying squirrel on both
forests and the lynx and grizzly on the Okanogan (Figures
2 and 3).
 |
| Figure2: |
Habitat levels for High,
Moderate, and Low risk stands on Fremont NF. Wildlife
species, left to right in each graph are: Northern goshawk,
Lewis woodpecker, white-headed woodpecker, Williamson's
sapsucker, pileated woodpecker, northern flying squirrel,
and Townsend's big-eared bat. |
 |
| Figure3: |
Habitat levels for High,
Moderate, and Low risk stands on Okanogan NF. Wildlife
species, left to right in each graph are: Northern goshawk,
Lewis woodpecker, white-headed woodpecker, Williamson's
sapsucker, Pileated woodpecker, northern flying squirrel,
Townsend's big-eared bat, Canadian lynx, and grizzly bear. |
Response to treatment from species varied among
species and treatments. Species are associated with older
forest structures had habitat levels more severely impacted
by the treatments than species associated with open forest
structures. As stands were opened more through thinning, habitat
decreased, when compared with no the no action alternative,
for most species. One exception was the Lewis woodpecker,
which thrives in open forests. When regeneration was included
available habitat increased, but still remained lower than
no action. Grizzly habitat on the Okanogan, though, which
was originally reduced by the thinning, returned to levels
higher than no action after 30 years. All treatments that
reduced fire risk also reduced habitat levels. Wildfire simulations
greatly reduced or eliminated habitat for all species associated
with older forest structures. Both forests are in fire regime
condition class 2 or 3 (FRCC, Hann, and others 2003), meaning
that the fire regime has diverged significantly from historical
conditions. With this in mind, questions can be asked about
historical habitats for some of the species now present in
the dry interior forests: Are current habitat levels, because
of fire exclusion and suppression, reflective of historical
levels? If forest managers perform fuel treatments to reduce
the current fire risk, how will habitat availability for old
forest species be effected? And, if habitats for some species
are at high risk and need to be preserved, what are the most
effective methods of creating low risk fuel and fire breaks
to protect the high risk areas from wildfire?
Discussion and Conclusions
HEP, HSI, and the ICBEMP WHR matrix models implemented in
LMS are only the beginning of the possibilities for habitat
analysis. Other WHR approaches are the Johnson & O'Neil
(2001) WHR matrices and the California Wildlife Habitat Relationships
(CWRH, URL: http://www.dfg.ca.gov/whdab/html/cwhr.html) with
forest structures quantified by forest inventory measures.
Empirical models can derived from tree measures, as with the
bird population models of Hansen (1995), who generated regression
models relating trees per acre in specific diameter classes
to bird population. The Washington State Department of Natural
Resources quantified Nesting, Roosting and Foraging (NRF)
habitats for the northern spotted owl based on tree and snag
measures (WAC 222-16-085). Implementation of all these examples,
and other models based on tree and snag measures, is possible
within LMS and give managers and planners the ability to analyze
many alternatives quickly and easily while holding all assumptions
constant. This consistency in assumptions provides uniform
comparability between simulations so relative tradeoffs between
alternatives can be assessed.
Limitations to this approach are the lack of understory models
that are compatible with forest growth models and the need
to field verify the habitat models. Understory vegetation
is a key component for many wildlife species and associated
models. Local understory/overstory relationships can be developed,
as in the Satsop Forest project, which derived mean values
for understory measures for each forest cover type, but it
will increase the cost and complexity of an analysis. Until
regional models of understory/overstory relationships are
developed, the number models that can be implemented in LMS
is limited. Many of the available habitat models are theoretical
and have not been field verified. Without field verification,
outputs from habitat models may be suspect. With these limitations
in mind, habitat analysis using habitat models implemented
in LMS, or other forest simulation tools, can be a very useful
tool to assess habitat availability, risks to habitat, and
communicate the potential tradeoffs between management regimes.
References
- Hann, Wendel.J., Strohm, Diane J. 2003. Fire regime condition
class and associated data for fire and fuels planning: methods
and applications. p 337-443. In: Omi, Philip N.; Joyce,
Linda A., technical editors. Fire, fuel treatments, and
ecological restoration: Conference proceedings; 2002 16-18
April; Fort Collins, CO. Proceedings RMRS-P-29. Fort Collins,
CO: U.S. Department of Agriculture, Forest Service, Rocky
Mountain Research Station. 475 p.
- Hansen, A. J., W. C. McComb, et al. (1995). Bird habitat
relationships in natural and managed forests in the west
Cascades of Oregon. Ecol appl 5(3): 555-569.
- Johnson, David H. and Thomas A. O'Neil, managing directors.
2001. Wildlife habitat relationships for Washington and
Oregon. Oregon State University Press. Corvallis, OR.
- Mason, C. Larry, Kevin Ceder, Heather Rogers, Thomas Bloxton,
Jerrrey Comnick, Bruce Lippke, James McCarter and Kevin
Zobrist. 2003. Investigation of Alternative Strategies for
Design, Layout, and Administration of Fuel Removal Projects.
Rural Technology Initiative, College of Forest Resources,
University of Washington, Seattle. 91p. Available online
at: URL: http://www.ruraltech.org/pubs/reports/fuel_removal/
- USDI. 1980. Habitat evaluation procedures 102 ESM. Washington
DC, USDI Fish and Wildlife service.
|
