Starrigavan Creek Watershed Restoration Plan
USFS TONGASS NATIONAL FOREST
SITKA RANGER DISTRICT
OCTOBER 2007
1
Overview
The Starrigavan Watershed Restoration Plan (WRP) area (Figure 1) includes the
Starrigavan Creek watershed which has a high priority for protection and restoration
among Sitka Ranger District watersheds. A Hydrologic Condition Assessment,
incorporating this area was completed as part of the Sitka Sound Landscape Assessment
(USDA-FS, 2004). Elements from that assessment will be included in this report.
The Starrigavan watershed Analysis Area, hereafter known solely as the Analysis Area in
this document, consists of one 6th field a Hydrologic Unit Code (HUC) watershed that
drains into saltwater at the head of Starrigavan Bay on Baranof Island in Southeast
Alaska. The watershed encompasses approximately 4,097 acres (6.4 miles2). It is
located about 7 miles north of downtown Sitka on northern end of the Sitka road system.
The majority of Analysis Area (3,644 acres) is administered by the Sitka Ranger District
of the Tongass National Forest, with the remaining lands in the lower watershed falling
under State of Alaska jurisdiction (453 acres). Historically it was used primarily for
subsistence purposes prior to European settlement. Typically, only current conditions on
Federal lands are described in documents such as this, however due to importance to
view the overall picture of watershed condition; most discussions will describe total
management impact summaries for the entire watersheds for all ownerships. Tables in
each section will differentiate summations by ownership. Additionally, the
recommendations and project proposal sections at the end of this document will outline
and discuss restoration and enhancement opportunities on both Federal and State lands.
The Starrigavan Creek watershed supports three species of anadromous salmon: pink
(Oncorhynchus gorbuscha); chum (Oncorhynchus keta); and coho (Oncorhynchus
kisutch); as well as; rainbow/steelhead trout (Oncorhynchus mykiss); and Dolly Varden
char (Salvelinus malma) (ADF&G Anadromous Stream Catalog, 2007). Many of the
streams within the watershed provide rearing and spawning habitat as well as low water
refugia for these species.
Like so many areas in the Sitka vicinity, the Starrigavan watershed has a long history of
human use dating back around 5,000 years to habitation by the first native people. Early
Russians established their first settlement, “Old Sitka”, near the mouth of Starrigavan
Creek around 1799 or 1800. Beginning in the late 1960s to early 1970s, timber
production occurred within the Analysis Area under management by the USDA Forest
Service. Today, it continues to provide both important subsistence and natural resources
to local residents.
The USDA Forest Service has determined that the Analysis Area is vital to the
subsistence, recreation, and ecosystem integrity of the area. The Analysis Area and its
components have changed significantly since timber harvest activities ended and as a
result, the USDA Forest Service has identified several predominant issues affecting the
current and future landscape and its uses. The issues described in this analysis serve as
the basis for recommending actions to rehabilitate many of those ecosystem components
in accordance with the Forest Plan.
2
Today, approximately 18 percent of the overall Analysis Area and 35 percent of the
overall riparian old-growth habitat is in a second-growth, even-aged forest structure,
which previously served as valuable deer winter habitat. It is recognized that much of
that forest structure will continue to be even-aged until thinning occurs. Wildlife
emphasis thinning treatments to enhance upland deer and bear habitat are recommended
in this analysis. 233 acres of the total 589 acres of upland harvest are considered high
priority and are recommended for type of thinning.
Timber production from the Analysis Area has not occurred in the last 30 plus years. The
Analysis Area is now in the Semi-remote Recreation Land Use Designation (LUD) status
which only allows for very limited forms of future commercial timber harvest. State
lands management for the Starrigavan valley falls under two designations: Public
Facilities-Retain (Pr) and Public Recreation and Tourism-Undeveloped (Ru). These
designations state that this parcel is to be retained for public recreation purposes.
Development authorizations other than those related to this purpose are not considered
appropriate.
Hydrologic connectivity and wetlands are integral parts of watershed function in the
Analysis Area. Landslides and soil erosion from harvest have not been identified as a
major source of resource damage to downstream ecosystems. Erosion from roads is
contributing some sedimentation to streams due to poor drainage and tire abrasion of the
road surface. Currently, 297 acres of harvest occur within the overall 2,713 acres of
Mass Movement Index (MMI) 3 and 4 soils and 218 acres of MMI 3 and 4 soils and 1.8
of the total 5.6 miles of roads occur with Riparian Management Areas (RMA). Through
field reconnaissance, it has been determined that sufficient regrowth of vegetation has
occurred within these MMI soil sites and no stabilization efforts are recommended at this
point. Furthermore, the majority of the 5.6 miles of roads within this watershed,
including the 1.8 miles within RMAs are for the most part stable. All of these remaining
structures are currently functioning for their current use. None of the remaining stream
crossing structures are recommended for removal
The use of the Analysis Area has always been valued by people for its important
subsistence, and more recently recreation opportunities. Restoration of stream channels
and riparian and uplands stands will almost certainly bring greater recreational,
subsistence, and economic importance to the area. Currently, one Forest Service
campground facility exists near the mouth of Starrigavan creek, as well as approximately
4 miles of trails and two shooting ranges. Projects to both directly enhance recreational
or commercial opportunities are recommended through this analysis.
Finally, fisheries habitat and aquatic ecosystem function has been impaired along many
reaches within the watershed due to riparian harvest and the conversion from coniferdominated riparian areas to red alder-dominated riparian areas. Approximately 96 acres
of riparian area that is along class 1 and 2 streams are recommended for thinning.
Additionally, 3.5 miles of stream channel have been identified for in-stream rehabilitation
of fisheries habitat and other aquatic ecosystem components.
The Forest Service assembled an interdisciplinary team (IDT) composed of a watershed
program coordinator, fisheries biologist, wildlife biologist, recreation planner, landscape
3
architect and staff officers. This IDT identified several key issues critical to the future
management of the watershed. The issues identified were:







Protection and restoration of fisheries habitat with an emphasis on coho and
steelhead salmon
Maintain and ensure protection of water quality throughout the watershed
Protection of the Starrigavan Creek Watershed and associated values
Protection and restoration of wildlife habitats with an emphasis on Sitka
Blacktail deer
Access and travel management within the watershed with an emphasis on
OHV use
Public education with an emphasis on resource protection and land
management
Maintenance, restoration and enhancement of recreation facilities, with an
emphasis on trails and campgrounds.
To address these issues, the IDT conducted the following:




Reviewed existing watershed information
Review of existing and proposed watershed restoration projects
Review of existing OHV trail information
Scoping of public, private and State agencies and groups
4
Figure 1. Starrigavan Analysis Area Vicinity Map
Sitka
5
Project Area Description
Location
The Starrigavan watershed is located in the northern portion of Sitka Sound in, in the
western half of central Southeast Alaska. It is located at the northern end of the City of
Sitka road system (Halibut Point Road, known locally as HPR), about 7 miles from the
town’s center. Starrigavan Creek is a relatively small coastal watershed approximately
4,097 acres in size (Figure 1). HPR crosses the mouth of the creek within the estuary
(Figure 2), which then dead ends less than a quarter mile past the bridge at the entrances
to the USFS managed campgrounds.
Figure 2. Starrigavan Estuary as Seen from Halibut Point Road.
Land Use
Land Use Designation
Land Use Designations (LUDs) are categorized into two broad categories: development
and non-development. Development LUDs are those that “permit commercial timber
harvest (Timber Production, Modified Landscape, and Scenic Viewshed) and convert
some of the old-growth forest to early-to-mid-successional, regulated forests” (USDA FS
1997, p. 7-9). Non-development LUDs are “land use designations that do not permit
commercial timber harvest and generally maintain the integrity of the existing old-growth
ecosystem” (USDA FS 1997, p. 7-25).
6
The Analysis Area contains land allocated to primarily one of the non-development
LUDs and Non-National Forest lands under State of Alaska management (Table 1 and
Figure 2). Though small acreages of other LUDs are mapped within the Analysis Area,
due to the fact that only one (Timber Production) is a development LUD and it is of very
small size and located along the ridgeline, it will not be considered or described in this
report.
Table 1. Land Use Designations within the Analysis Area.
LUD
Development Status
Acres1
Percent of
Assessment Area
Semi-Remote Recreation
Mostly Natural
3579
87.4
Municipal Watershed
Mostly Natural
45
1.0
Timber Production
Intensive Development
21
0.5
Non-National Forest
N/A
453
11.1
Source: Sitka Ranger District 2007 GIS Coverage.
The goals of the Semi Remote Recreation LUD described below. Chapter 3 of the Forest
Plan contains a detailed description of each land use designation (USDA FS 1997).
Semi-remote Recreation
The goals of this designation are: 1) to provide predominantly natural or naturalappearing settings for semi-primitive types of recreation and tourism; and 2) to
provide opportunities for a moderate degree of independence, closeness to nature, and
self-reliance in environments requiring challenging motorized or non-motorized
forms of transportation.
State lands management for the Starrigavan valley falls under two designations: Public
Facilities-Retain (Pr) and Public Recreation and Tourism-Undeveloped (Ru). These
designations state that this parcel is to be retained for public recreation purposes.
Development authorizations other than those related to this purpose are not considered
appropriate.
Value Comparison Units
Table 2 lists the one Value Comparison Units (VCUs) located within the Analysis Area.
VCUs are parcels of land that generally encompass a drainage basin or watershed
containing one or more large stream systems. VCU boundaries usually follow easily
recognizable watershed divides. These units delineate areas for resource inventory and
interpretation. For the purpose of this analysis, VCUs were not uses and are referenced
only for future landscape level assessments purposes; all analyses were based instead on
watershed area.
Table 2. VCUs within the Analysis Area.
VCU Number
3110
VCU Name
Sitka
Source: Sitka Ranger District 2007 GIS Coverage.
7
Figure 3. Land Use Designation within the Analysis Area.
Source: Sitka Ranger District 2007 GIS Coverage.
8
Ownership
Land ownership within the Assessment Area is not complex, with the majority of the
Area (89%; 3,644 acres) in federal ownership and managed by the USDA Forest Service.
The State of Alaska owns and manages approximately 453 acres (11%) in the lower
portion of the watershed. Historically, it was used primarily for subsistence purposes
prior to European settlement, while today; it continues to provide both important
subsistence and natural resources to local residents.
Climate
The Analysis Area has a maritime climate that has affected the physical and biological
characteristics and the human uses of the area. Temperatures are moderated by the
Alaska Current, which circulates counterclockwise up the coast (Johnson and Hartman
1969). Data from the nearest climatic station in Sitka indicate there is only 22.6ºF
difference between the mean average temperatures of the warmest (August, 56.7ºF) and
coldest (January, 34.1ºF) months (Figure 4). The climate is predominantly cloudy, cool,
and wet throughout the year. This station also indicates that the average yearly
precipitation at Sitka is 96 inches. Precipitation occurs throughout the year, with June
being the driest month (3.7 inches) and October the wettest (14.5 inches). This station in
Sitka is located near saltwater, at less than 50 feet in elevation, on Japonski Island. The
actual climate data within the Analysis Area varies and is likely to be much colder and
wetter at higher elevations and further from saltwater.
Figure 4. Climate Data for Sitka, AK
Climate Data For Sitka Alaska
T
e
m
p
e
r
a
t
u
r
e
60
16
14
50
12
40
10
30
8
6
20
4
10
Mean (F)
Avg. Precip. Inches *
2
Month
Source: www.weather.com
* Source: Sitka Airport Records 1842-1996
9
r
r
em
be
D
ec
em
be
r
ob
e
N
ov
O
ct
m
be
r
us
t
Se
pt
e
A
ug
ly
Ju
ne
Ju
ay
M
A
pr
il
M
nu
a
Ja
ar
ch
0
ry
Fe
bu
ra
ry
0
I
n
c
h
e
s
Ecological Classification
The Analysis Area is comprised of one ecological subsection according to Nowacki and
others (2001). Ecological subsections, which help to define the ecosystems of Southeast
Alaska, are based upon physiography, lithology, and surficial geology due to their
interactions in processing water.
The only ecological subsection in the Assessment Area is the Sitka Sound Complex.
This subsection is defined by the past activities of Mt. Edgecumbe. Nowacki and others
(2001) describe it as:
… highlands encircling Sitka Sound include the mountains on northwest Baranof
Island, Halleck, Krestof, and Partofshikof Island, and the northern third of Kruzof
Island. This area was blanketed by 2 to 6 feet of ash about 9-12,000 years ago
(Riehle et al. 1992). Over time, portions of these volcanic deposits have washed
downslope exposing the underlying Sitka graywake, granite, and low-grade
metamorphic rocks such as phyllites. Much of this area is considerably lower in
elevation than surrounding subsections, particulary Halleck, Krestof, and
Partofshikof Island. This subsection has no glaciers, although it does contain a
few permanent snowfields. Volcanic ash and cinders are the principal parent
materials, covering over 50 percent of the subsection. Landslides are common on
the ash-coated surfaces, particularly in areas roaded during the 1960-70s.
Hemlock-spruce and hemlock forest dominate shorelines and low elevations.
Forested wetlands of lodgepole pine and mixed conifers are relatively abundant
compared to bordering subsections to the east. The alpine and coastal forest
habitats support brown bear, Sitka black-tailed deer, mountain goat (introduced),
marten (introduced), common shrew, Keen’s mouse, and tundra vole (p. 106).
Landscape Processes
A complete characterization of a landscape or landform must contain three components: a
description of the feature, the processes involved in its formation, and its development
through time (Chorley and others 1984). There are five primary processes that influence
landscapes: tectonism (geological plate movement), glaciation, hill slope processes
(landslides and surface erosion), fluvial processes (stream flow and sediment transfer),
and wind. Tectonic and glacial processes operate on a geologic time scale. Landforms
within the Analysis Area are generally less than 12,000 years old. Hill slope and fluvial
processes have the greatest potential to affect resource conditions on a time scale of years
to decades. This report has attempted to treat the Analysis Area holistically, discussing
the development of the landscape, soils, and vegetative types in relation to the major
disturbance factors in the watershed.
Tectonic Processes
Tectonic activity affects the Analysis Area on different temporal and spatial scales. On
the geologic time scale, the movement of large terranes has resulted in the many different
assemblages of bedrock in Southeast Alaska (Brew 1990). A geologic fault runs along
Silver Bay, through Sitka Sound, and up along Partofshikof Island and the outer coast of
Chichagof Island. On a time scale of thousands of years, some movement has probably
occurred along this fault.
10
Glacial Processes
Glaciation has exerted the most profound effect on the soils and plants of the Analysis
Area. The Wisconsin glaciation, which ended 12,000 to 13,000 years ago (Miller 1973),
along with earlier glaciations, has resulted in U-shaped valleys and higher elevation
cirque basins. The glaciers scoured some areas down to the bedrock and deposited basal
till and ablation till elsewhere. The Wisconsin deglaciation resulted in a sea level that
was much higher than it is today. This accounts for the presence of marine silts and
sands in many of the low-lying valleys of northern Southeast Alaska. It is likely that
these marine silts and sands now underlie many wetlands in the low-lying areas of the
watersheds in the Analysis Area.
The Little Ice Age was a period of worldwide cooling and glacial advance from the
middle of the 13th through the late 19th century (Porter 1986). Deep winter snow pack
and severe avalanching likely influenced Baranof Island’s upper tree line and forest
composition.
Hill Slope Processes
Erosion has had a large effect on local topography since the Wisconsin glaciation. Many
colluvial and alluvial fans (i.e., landforms partially formed by debris torrents) were
deposited on the valley floors during this time. Recent landslides suggest that this
process is continuing within the Analysis Area. Initiation of landslides in an undisturbed
environment is linked to temporary water table development during high-intensity storms
(Swanston 1969). Landslides in timber harvest areas are generally on gentler slopes and
are significantly smaller than those in undisturbed environments (Swanston and Marion
1991). Though one does not yet exist for Analysis Area, the Forest Service is in the
process of completing a forest-wide landslide inventory for the Tongass National Forest.
Fluvial Processes
Fluvial processes, or moving water processes, created the flood plains and alluvial fans in
the Analysis Area. Fluvial processes have varying effects depending on water and
sediment volumes; however, materials carried by the water are always sorted and
deposited according to size and weight. Today, the streams in the Analysis Area are not
overloaded with material and streams generally have one channel with fluvial deposits
such as point bars (on the inside of meanders) and levees (fine sands on the upper stream
banks).
Wind Processes
Southeast Alaska’s temperate rainforests are susceptible to wind damage because of the
combination of shallow root systems, poorly drained soils, and high winds, which often
occur during peak rain events. Most commonly, single trees or small groups of trees are
blown down; however, entire tree stands sometimes blow down. Currently no inventories
of stands that have regenerated after a large blowdown event exist for the Analysis Area.
11
Soils
Soils on mountain and hill slopes are formed of decomposed bedrock and colluvial
material (deposited by gravity). Bedrock soils are generally shallow, while colluvial soils
are deeper and better drained. In addition, soils formed of glacial till occur in patches
plastered along mountain and hill slopes to elevations of about 1,000 feet. In the valley
bottoms, soils have formed of river deposits, colluvial material, and marine sediments.
The cool, wet climate in the Analysis Area causes organic matter to decompose slowly,
creating soils characterized by organic surface layers. Where drainage is restricted by
topography or an impermeable layer, such as bedrock or glacial till, peatlands composed
of organic matter are common. In coarse alluvium (gravels and cobbles) the soils are
well drained and support forests. Where the alluvium is finer and restricts drainage,
nonforested vegetation communities such as fens and bogs form. Tree root depth is
shallow, primarily in the nutrient-rich organic layers and the first few inches of the
mineral layers. Typically the root zone is moist, acidic, and contains most of the
nutrients available for plant growth (Heilman and Gass 1972).
Soil Stability
Swanston (1969) counted more than 3,800 landslides, which occurred in the last 150
years in Southeast Alaska. Most slides occur on steep slopes and when heavy rainfall has
saturated the soil. In addition, wind associated with these storms can blow down trees,
which may help trigger slope failure.
Landslides typically begin on open slopes and are a mixture of rock, soil, and vegetation.
Swanston and Marion (1991), in their study of landslides within Southeast Alaska,
observed that only about 3 percent of all landslides reached fish streams. No in-depth
landslide inventory was completed for this analysis; however existing GIS inventories
indicate there are no mapped landslides within the watershed. However, aerial and
ground reconnaissance in 2007 observed that there was one landslide in the upper portion
of the south fork subwatershed. This slide appeared to have originated near the upper
boundary of a harvest unit and reached the stream channel. The slide appeared to be
approximately 7-10 years old by the age of vegetation regrowth.
Soil type also influences landslide occurrence. The soils in the Analysis Area are
mapped and described in the Chatham Area Integrated Resource Inventory (USDA
1986). In order to describe their relative instability, soils are grouped into mass
movement hazard categories: MMHAZ 1 (low hazard), MMHAZ 2 (moderate hazard),
MMHAZ 3 (high hazard), and MMHAZ 4 (extreme hazard). These categories are based
on a number of factors that influence landslides, including slope, landform, parent
material, and drainage. Sixty-six percent of the total Analysis Area is rated as either
MMHAZ 3 or 4.
Figure 5 shows the distribution of MMHAZ 3 and 4 soils and management activities
within them throughout the Analysis Area. Table 3 lists the MMHAZ 3 and 4 soils and
the extent of management activities that has occurred in them. Within the Starrigavan
Creek watershed there are 2,713 acres of Mass Movement Hazard class 3 and 4 soils
(MM-Haz3-4), of which approximately 218 occur within riparian management areas
(RMAs). These MM-Haz 3-4 soils have a high likelihood to become unstable when
12
disturbed by either natural events or management induced activities (timber harvest or
road construction). When disturbance of these soils classes exists in close proximity to
stream channels or directly within RMAs, the risk of water quality and aquatic habitat
degradation is highest. To date, 218 acres (8.0% of total) of MM-Haz3-4 soils have been
harvested of which 20 acres lies within RMAs. Additionally, when compared to other
watersheds within the Sitka Ranger District (37% average), this watershed has relatively
greater percentage of overall area (66%) with these hazard soil types.
Table 3. High Hazard Soils within the Analysis Area.
Ownership
Total WS
Area
(acres)
Total
MM-HAZ
3&4
(acres)
MM-HAZ
3&4 within
Previous
Harvest
(acres)
MM-Haz
3&4
within
RMA1
(acres)
MM-Haz
3&4
RMA
Harvest
(acres)
Percent
of Total
WS in
MM-Haz
3&4
Percent
of Total
MM-Haz
3&4
Harvest
Percent
of Total
RMA1 in
MM-Haz
3&4
USFS
3,644
2,604
231
206
14
71.5
8.9
59.2
State
453
109
67
12
6
24.1
61.5
15.0
Total
4,097
2,713
297
218
20
66.2
8.0
51.0
Source: 2007 Sitka Ranger District GIS Coverage
1
RMA refers to a Riparian Management Area.
13
Figure 5. High Hazard Soils within the Starrigavan Watershed.
Source: Sitka Ranger District 2007 GIS Coverage.
14
Surface Water and Stream Habitats
Drainage Basin Morphology
The Analysis Area ranges in elevation from sea level to a maximum of 3,010 feet.
Drainage patterns of the Starrigavan Creek watershed runs generally east to west and
empties directly into saltwater (Figure 6).
Steam density, also referred to as drainage density, is a measure of stream length per
square mile of watershed. This measurement is useful in determining a stream’s potential
for runoff and sediment transport. The same factors that influence channel type, geology,
landform, climate, and vegetation also influence drainage density. Drainage density
within the Analysis Area averages 3.02 miles per square mile (mi/mi2) (Table 4).
Table 4. Stream Miles and Drainage Densities for the Analysis Area.
Ownership
Area (mi2)
Total
Stream
Miles
Drainage Density
(mi/mi2)
USFS
5.7
15.3
2.68
State
0.7
4.0
5.71
Total
6.4
19.3
3.02
Source: 2007 Sitka Ranger District GIS Coverage.
Note: Total stream miles denote only mapped class I-III stream channels.
Unmapped stream channels including Class IV channels would likely increase
total stream miles and drainage densities.
Hydrology
Watershed delineations enable land managers to evaluate the effects of various
management activities on fish habitat and an aquatic system’s capability to produce fish.
The lower half of the Starrigavan Creek watershed has a moderately well-developed
flood plain that prior to valley bottom timber harvest activities, supported stands of large
Sitka Spruce (Figure 14). The uppermost portion of the watershed is generally steeper
and dominated with avalanche chutes with a relatively small floodplain. Throughout the
watershed, transport and transitional channels drain the moderate to higher gradient
reaches and transport sediment and organic debris downstream to the valley bottom
depositional streams. In addition to providing much of the available fish habitat, these
flood plain depositional stream channels provide short- and long-term storage for
sediment and are sensitive depositional reaches. The Starrigavan Creek watershed, like
many watersheds on the Sitka Ranger District, has a quick response to storm runoff and is
efficient in routing runoff to the mainstem channel and out of the watershed to saltwater.
15
Figure 6. Starrigavan Creek Watershed Streams.
Source: Sitka Ranger District 2007 GIS Coverage.
16
Stream Flow
Stream flows for the Analysis Area are typical of island watersheds in Southeast Alaska.
Steep slopes along with well-drained, shallow soils and high drainage densities
characterize the Analysis Area. Most watersheds in the region respond rapidly to
rainstorms, which can cause large daily fluctuations in stream flow. Stream flow is
highly variable during the year. River discharge generally peaks in September thru
November and gradually declines throughout winter and early spring. Though there is
only a short period of record for the stream gage within the Starrigavan watershed, data
over this time frame shows that maximum mean stream discharge is 148.5 cfs/mi2.
Snowmelt at high elevations results in moderate flow increases in May and June and
results in a second discharge peak. Infrequent winter storm freshets may result from
warm rain-on-snow events. Mean low flows for the Starrigavan gage is 1.1 cfs/mi2 and
generally occur between June and August, although low flows can also occur during
prolonged winter cold periods.
Overland flow is seldom observed in Southeast Alaskan coastal forests, except from
compacted sites such as roads and landings, rock outcrops and ice fields. Nearly all
runoff occurs by soil infiltration and subsurface routing to streams. Stream networks
expand during storms, especially storms continuing for several days to weeks. As the soil
becomes saturated, live flow reappears in low-order intermittent channels.
The majority of precipitation entering the Analysis Area exits through runoff. The
remaining percentage is lost to soil recharge, transpiration, and evaporation. Steep slopes
and stream gradients, combined with low groundwater storage capacity cause quick
hydrographic response and flashy flow after the onset of rain. Stream hydrographs for an
individual storm event underscore this short lag-time with a steep rising curve and rapid
recession. Though only limited stream flow data exists for this individual watershed,
flow response from runoff can be seen in data from this gaging station (Figure 7).
Other factors which influence water flow and conditions in the Analysis Area include
groundwater recharge from fens, bogs or ‘muskegs’, and shallow aquifers and seeps.
Shallow aquifers and seeps associated with valley floor wetlands and alluvium help
sustain summer and winter base flow in main stream channels.
Stream segments in the Analysis Area were mapped and classified using the Alaska
Region Channel Type Classification System (USDA FS 1992). The area contains
roughly 19.3 miles of significant streams with an average stream density of 3.02 miles
per square mile (Table 4). For this report, stream class (a measure of fish habitat) and
channel type (a measure of sediment transport) were analyzed. There is one small alpine
lake and four manmade burrow ponds on NFS lands within the Analysis Area.
17
Figure 7. Typical Storm Event Hydrograph (Starrigavan, September, 2007).
Source: http://water.usgs.gov/data.html
Management Effects on Stream flow. In large basins where timber harvest activities are
dispersed in space and over time, relatively small changes in stream flow can be expected
(Duncan 1986). Studies in Oregon showed increased magnitude of small and moderate
peak flows associated with logging (Harr 1986). Salmon have adapted to average flow
regimes for all stages of their freshwater life history. Seasonal low flows and peak flows
can affect migration, channel conditions, water quality and egg survival (Hicks et al.
1991).
Reduced low flow in watersheds that have been converted from old-growth forest to
second-growth forest is a relatively new issue. This reduction in summer and winter
flows is from increased canopy interception of precipitation and increased
evapotranspiration rates. Myren and Ellis (1984) speculated that converting old-growth
watersheds to second-growth forests may significantly reduce summer low flows in
Southeast Alaska streams and impair summer rearing and spawning for salmonids. This
decrease would be evident in the intermediate stages of forest succession. However,
streamflow analysis for Staney Creek, a large watershed on Prince of Wales Island near
Ketchikan, indicated an increase in summer low flows after at least 25% of the watershed
was harvested. Low flow changes are most likely to occur where a significant portion of
the stream riparian area has been harvested (Hicks et al. 1991).
18
Peak flow increases from timber harvesting in rain-dominated runoff regimes will be
minor, assuming minimal soil compaction and low road density in a watershed.
However, clearcut harvest practices have the potential to increase the magnitude of peak
flows under a rain-on-snow runoff regime (Harr 1986).
The sustained baseflow and thermal cover found in palustrine channel types are important
to winter survival of juvenile fish. Low streamflow during extreme cold weather may
freeze gravel riffles and incubating eggs. Low flows in the summer and winter can
adversely affect adult spawners, rearing juveniles, and egg incubation. Low summer
flows may shrink and occasionally dry up rearing pools used by juveniles; this most often
affects young-of-the-year coho, steelhead, cutthroat and Dolly Varden and occurs in the
smaller tributaries and side channels of the main stem stream.
Changes in the magnitude and duration of winter peak flow can adversely affect rearing
salmonids and the integrity of spawning beds. Flooding reshapes and redistributes gravel
bars and large woody debris, causing eggs to be washed away, buried, or crushed.
Annual peak streamflows and rain-on-snow storm flows consistently occur during egg
incubation. Debris flows, landslides, alluvial fan and flood plain channel migration and
stream crossing failures usually occur during peak streamflows. All of these processes
have the potential to dramatically affect egg survival and alter habitat features.
There are two current gage sites within this Analysis Area. Lack of long-term stream
gaging information for the Analysis Area streams precluded us from doing a quantitative
analysis of streamflow condition and trends in the watershed. Kadashan River is the only
drainage near the Analysis Area with adequate stream gauging information to track
annual flow levels. However, little timber harvest has occurred and it is essentially an
unmanaged watershed. Eight years of stream gauging data were collected for the upper
Indian River (Tenakee) for the Indian River Watershed Analysis (IRWA) (USDA-FS
1996), these two sets of data were compared to evaluate trends over time and possible
changes to the Indian River hydrology associated with timber harvest practices. The
following results can be extrapolated to some degree for watersheds with similar harvest
levels (<18%).
Peak Flows. As mentioned, rain-on-snow peak flow events are the most susceptible to
change as the result of timber harvest in Southeast Alaska watersheds. Areas with
shallow winter snowpack and large canopy openings such as clearcut units are the most
important source zones for rain-on-snow floods (Harr 1986). For the IRWA, maximum
daily flows from November through February for the period prior to and following timber
harvest (at Indian River) were compared. An analysis of the two regression lines
indicated no significant difference (P=.95) between pre- and post-timber harvest winter
peak flows. The IRWA concluded that it was unlikely that 10% harvesting of the
transient snow zone resulted in measurable changes in Indian River peak flows. This
same conclusion should also hold true for this Analysis Area as harvest level is less than
10%, harvest occurred below this transient snow level, and subsequent regeneration of
harvest is beyond that of the Indian River harvests.
Low Flows. The month of August is considered to be a critical period for summer low
flows in the Analysis Area. August typically has warm temperatures and periods of one
19
to two weeks with no or little precipitation. Alpine snowpack runoff contributions to
base stream flow are minimal. Adult salmon are also migrating and spawning during this
time. Similar to peak flows, the IRWA team analyzed summer low flow conditions and
trends for Indian River. Mean monthly flows and minimum daily flows in August for
Upper Kadashan were compared to flows for the Indian River. They concluded there was
a consistent relationship between Kadashan and Indian River over most of the period that
both stream gauges were operated and they discounted the possibility of measurable
changes to low flow levels in Indian River resulting from timber harvest. Again, this
same conclusion should also hold true for this Analysis Area as the harvest level is less
than 10%, harvest occurred below this transient snow level, and subsequent regeneration
of harvest is beyond that of the Indian River harvests.
The factors affecting both low and peak flows can be moderated somewhat by thinning
young growth and recovering riparian stands to a more pre-management stand and
vegetation state.
Figure 8. Monthly Mean Discharges for Starrigavan Creek: (USGS 15087618)
Discharge (cfs)
700
600
Mean
500
Mean Max
Mean Min
400
300
200
100
Au
gu
st
Se
pt
em
be
r
O
ct
ob
er
N
ov
em
be
r
D
ec
em
be
r
Ju
ly
Ju
ne
M
ay
Ap
ri l
h
M
ar
c
Ja
nu
ar
y
Fe
br
ua
ry
0
Month
Source: http://water.usgs.gov/data.html
Note: period of record for calculations: 3 years (2003-10-23 to 2006-09-30)
Stream Channel Types
Stream channel types are determined by their size, location in the watershed, adjacent
landforms, gradient, hydraulic control, and riparian vegetation. Channel type and stream
class are influenced by geology, landform, climate, and vegetation. Table 5, 6 and 7
display the stream miles by channel type and stream class within the Analysis Area.
Table 8 summarizes these stream miles into individual channel process groups.
20
Table 5. USFS Lands Streams by Stream Class and Channel Type
Stream Class
Channel Type
Grand Total
I
II
III
ES4
0.1
FP0
0.2
FP3
0.8
FP4
0.7
HC0
HC2
0.5
HC3
1.0
HC5
HC6
MC1
0.9
MM0
0.6
MM1
0.1
MM2
1.7
Grand Total
3.4
3.0
Source: 2007 Sitka Ranger District GIS Coverage.
0.3
2.3
6.2
8.8
0.1
0.2
0.8
0.7
0.3
0.5
1.0
2.3
6.2
0.9
0.6
0.1
1.7
15.3
Table 6. State Lands Streams by Stream Class and Channel Type
Stream Class
Channel Type
Grand Total
I
II
III
AF1
0.2
FP0
0.6
FP3
1.1
FP4
1.1
HC0
HC6
MM0
0.3
Grand Total
3.2
0.0
Source: 2007 Sitka Ranger District GIS Coverage.
0.3
0.5
0.7
21
0.2
0.6
1.1
1.1
0.3
0.5
0.3
4.0
Table 7. Total Analysis Area Streams by Stream Class and Channel Type.
Stream Class
Channel Type
Grand Total
I
II
III
AF1
0.2
ES4
0.1
FP0
0.8
FP3
1.8
FP4
1.8
HC0
HC2
0.5
HC3
1.0
HC5
HC6
MC1
0.9
MM0
0.3
0.6
MM1
0.1
MM2
1.7
Grand Total
6.7
3.0
Source: 2007 Sitka Ranger District GIS Coverage.
0.2
0.1
0.8
1.8
1.8
0.6
0.5
1.0
2.3
6.7
0.9
0.9
0.1
1.7
0.6
2.3
6.7
9.6
19.3
Table 8. Stream Miles by Process Group for the Analysis Area.
Watershed
AF
USFS
State
0.2
Total
0.2
ES
FP
HC
MC
MC
Grand Total
0.1
1.6
10.3
0.9
2.3
15.3
2.8
0.7
0.3
4.0
4.4
11.0
2.6
19.3
0.1
0.9
Source: Sitka Ranger District 2007 GIS Coverage.
Stream channels can also be classified into three main types: transport, transitional, and
depositional channels (Table 9). Transport channels have low sediment retention and
include high gradient contained (HC), moderate-gradient contained (MC), and low
gradient contained (LC) channels. HC channels are located on steep headwater slopes and
are the primary sediment conduit to the low-gradient valley bottom and footslope
streams. Transitional channels, in contrast, have moderate sediment retention and include
moderate-gradient mixed control (MM), estuarine (ES3), glacial (GO5), and some
alluvial fan (AF2) channels. Finally, depositional channels have high sediment retention
and include the valley bottom flood plain (FP), palustrine (PA), estuarine (ES2 and ES4),
and some alluvial fan (AF1) channels. As mentioned above, the Analysis Area contains
19.3 miles of mapped streams: 11.9 miles (62 percent) are transport channels, 2.6 miles
(14 percent) are transitional channels, and 4.7 miles (24 percent) are depositional
channels. Depositional channels which contain the flood plain and palustrine streams
22
process group generally have the most anadromous (Class I) fish spawning and rearing
habitat.
Table 9. Analysis Area Transport, Transitional and Depositional Stream Miles.
Transport
Transitional
Depositional
Total
Miles
Watershed
Miles
% of WS
Miles
% of WS
Miles
% of WS
USFS
11.2
73.2
2.3
15.0
1.8
11.8
15.3
State
0.7
17.5
0.3
7.5
2.9
72.5
4.0
Total
11.9
61.7
2.6
13.5
4.7
24.4
19.3
Source: Sitka Ranger District 2007 GIS Coverage.
Stream Habitat
Four stream designations are used on the Tongass National Forest to classify stream
channels (USDA-FS1997).




Class I streams and lakes have anadromous or adfluvial (resident migration) fish
habitat.
Class II streams and lakes have only resident fish populations.
Class III streams do not have fish populations but have the potential to influence
the water quality of downstream aquatic habitat.
Class IV streams are small, intermittent and/or perennial channels with
insufficient flow or transport capabilities to have an immediate influence on the
water quality of downstream fish habitat.
Class IV streams have not been analyzed for this report because of a lack of data.
However, Class IV streams are analyzed during project-level planning and
implementation. In the Starrigavan Creek watershed, there are a total of approximately
6.7 miles of Class I streams (35 percent of all stream miles), 3.0 miles of Class II streams
(15 percent of all stream miles), and 9.6 miles of Class III streams (50 percent of all
stream miles) (Table 7 and Figures 6).
Management Effects by Stream Class. The Tongass National Forest channel type system
(USDA 1992) can be used for determining management considerations, management
opportunities and aquatic habitat capability (spawning/rearing) for fish species on the
Tongass National Forest. As mentioned before, timber harvest was not evenly distributed
throughout the watershed, with harvest occurring primarily in the valley bottom and
lowlands. Consequently, the vast majority of streams affected are Class I and II fish
channels. Tables 10, 11 and 12 list the mapped stream lengths by class and channel type
23
within harvested areas within the watershed. Within the Starrigavan watershed, 60% of
all the fish bearing streams have had harvest along their banks (Class I = 81%; Class II =
13%). The majority (61%) of Class 1 and 2 streams in the harvested area are channel
types that are sensitive to management activities carried out within RMAs or along
stream channels.
Table 10. Harvested Streams on USFS Lands by Class and Channel Type
Stream Class
Channel Type
Grand Total
I
FP0
FP3
FP4
HC3
HC5
HC6
MM1
MM2
Grand Total
II
III
0.2
0.7
0.6
0.4
0.3
0.5
0.1
1.2
2.6
0.2
0.7
0.6
0.4
0.3
0.5
0.1
1.2
0.4
0.8
3.9
% of Total Streams
76.5
13.3
Source: Sitka Ranger District 2007 GIS Coverage.
9.1
25.5
Table 11. Harvested Streams on Sate Lands by Class and Channel Type
Stream Class
Channel Type
Grand Total
I
AF1
FP0
FP3
FP4
HC6
Grand Total
II
III
0.2
0.6
1.0
1.0
2.8
0.1
0.1
0
% of Total Streams
87.5
0
Source: Sitka Ranger District 2007 GIS Coverage.
14.3
24
0.2
0.6
1.0
1.0
0.1
2.8
70.0
Table 12. Total Harvested Streams by Class and Channel Type
Stream Class
Channel Type
Grand Total
I
AF1
FP0
FP3
FP4
HC3
HC5
HC6
MM1
MM2
Grand Total
II
III
0.2
0.8
1.6
1.6
0.4
0.3
0.6
0.1
1.2
5.4
0.4
0.9
% of Total Streams
80.6
13.3
Source: Sitka Ranger District 2007 GIS Coverage.
9.4
0.2
0.8
1.6
1.6
0.4
0.3
0.6
0.1
1.2
6.7
Large wood (LW) is naturally introduced into stream channels during storms by flooding
(bank erosion), windthrow events or natural mortality. Trees enter the stream singly or in
small groupings from these sorts of disturbance events. These pieces, if small enough to
be transported downstream by current velocities, most often then accumulate into debris
jams downstream. Debris jams, as well as the largest LW pieces, called ‘key’ pieces,
dissipate stream energy. This dissipation is primarily through channel scour, which
creates pools that are an important component of fish habitat. Additionally, energy
dissipation through pools helps routing and distribution of substrates, stabilizing them
and maintaining channel dimensions, patterns and profiles. Beside pool habitat for fish,
wood in streams provides cover from predation and serves as a primary production
source of food for fish. A study in Southeast Alaska found lower fish population
densities in streams where smaller LW pieces were selectively removed (Dolloff 1986).
Influxes of large wood similar to pre-harvest conditions cannot be expected until mature
conifers are restored within the riparian areas that were harvested. Introduction of large
wood through restoration projects, within appropriate stream channels, may facilitate
habitat stability and improvements.
Fish Populations
Southeast Alaska is famous for its salmon runs. Most of the streams in the Analysis
Area are used for spawning and rearing by a variety of anadromous and resident fish.
Salmon are a main food source and the livelihood for many of the people who live in
the Sitka Sound area. The abundant salmon runs near Sitka made it attractive for
human settlement (USDA FS, Katlian WA, 2002). In prehistoric times, the Tlingit
clans in the area built their villages in close proximity to salmon streams, and salmon
comprised the bulk of their food supply. Salmon have exerted a strong influence on
local native culture. Native Alaskans protected salmon streams from invasion by
other clans, tribes, and Europeans. In the 1800s, Sitka and the surrounding area was a
Tlingit stronghold, but in 1804 Russian fur traders drove the Tlingit from the area.
25
The Tlingit clans returned to the area and continue to have strong cultural ties to the
salmon streams (USDA FS, Katlian WA, 2003).
In 2002, commercial fishers harvested over 56 million fish valued at 38 million
dollars from Southeast Alaska waters. Sport fishing in the Assessment Area is also
big business. According to the Alaska Department of Fish & Game (ADF&G) web
site, there are 138 saltwater fishing guides registered in the Sitka area (ADF&G,
2002). In 2000, there were over 22,000 salmon harvested by recreational fishers in
the saltwater surrounding the Assessment Area (ADF&G, 2002).
The Alaska Department of Fish and Game Anadromous Fish Catalog lists steelhead,
coho, pink, and chum salmon, sculpin and Dolly Varden char for streams in the Analysis
Area.
Fish Escapement Conditions and Trends
Escapement is defined as the number of fish that return to the spawning grounds of a
stream or lake during any given year. Escapement numbers are collected by aerial
estimates, stream/lakeside counts, and at weirs where fish are counted as they pass. Weir
counts are the most accurate means of estimating escapement, but it is a costly process
and is used only occasionally on key fish streams. For this reason, aerial, foot and
snorkel escapement estimates are primarily used. Though not as accurate as weir data,
these methods do provide escapement data that can be used to compare the year-to-year
variability in salmon escapement numbers.
Figure 9 below displays peak coho escapement counts for the Starrigavan watershed.
These counts were collected primarily by foot surveys and were completed during the
presumed peak timing of the run. As one can see, variability from year to year can be
quite dramatic, however as the red trend line indicates, there is a general downward trend
in coho returns.
Figure 9. Peak Coho Escapement Counts for Starrigavan Creek.
450
No. of Coho
400
350
300
250
200
150
100
50
19
81
19
83
19
85
19
87
19
89
19
91
19
93
19
95
19
97
19
99
20
01
20
03
20
05
0
Year
Source: ADF&G 2007
26
In addition to escapement data, a potential fish production model was used to determine
potential fish production within the watershed. Pink and coho salmon productivity
numbers were estimated based upon stream channel types. These numbers are based on
past stream and channel type studies which measured available habitat in each type of
channel to determine the number of pink and coho smolts that streams can produce.
First, GIS data is used to calculate the length of channel types that provide fish habitat.
This number is then multiplied by a smolt production value assigned to given channel
types. Finally, the total number of smolts is then multiplied by a survival rate to
determine the watershed’s potential fish production capabilities. Table 13 details the
results of this model.
Though fish production capability estimates are not 100 percent accurate, they do allow
us to compare the potential productivity in stream and between watersheds. For this
analysis, the assumption was made that fish could access all of the channel types where
fish production is possible. Estimates of the number of fish produced in ponds were not
included in this calculation. The model does not take into account the natural and
management-related factors that influence fish populations.
Table 13. Fish Production Capability Model Results
ADF&G Stream
Watershed Name
Number
Starrigavan River
113-41-10150
ADF&G
Cataloged
Species
Present1
Adult Pink
Capability2
Adult Coho
Capability2
P CH CO SH DV
35,459
565
1
P=Pink Salmon; CH=Chum Salmon; CO=Coho Salmon; SH=Steelhead Trout; DV=Dolly Varden
Char
2
Survival rate from smolt to adult estimated at 0.024 for pinks and 0.10 for coho
Sport Fisheries
The Analysis Area streams receive at least light sport fishing use. Much of this use is
concentrated in the estuary and bay area near the mouth of the stream. Though true
numbers of actual use are limited, Table 14 below displays reported sport harvest
numbers for the Analysis Area. This table displays information for people who received
and returned Alaska Department of Fish and Game survey forms, actual sport fishing
anglers and harvest are known to be much higher. Coho, Pink and Steelhead species are
also known to be harvested by sport anglers at Starrigavan. Sport harvest of coho salmon
however is restricted within the freshwater reaches of Starrigavan Creek and is only
allowed in saltwater beyond the HPR bridge.
Table 14. Reported Annual Sport Harvest of Freshwater Fish Species.
Year
Species/ #
Location
Caught
1996
1997
1998
1999
2000
Starrigavan
Dolly
0
25
0
Varden
2
2
1
Responses
2001
0
2
Source: Howe and others 2001a, b, c, d and e.
Note: Estimates based on fewer than 12 responses are only useful for documenting that fishing occurred.
27
In addition to these fish species, steelhead are present in within the Analysis Area.
Although there are few good estimates of escapement, foot surveys indicate that the
number of returning steelhead in area streams is inherently low.
Water Quality Concerns and Status
There are no state-listed water quality-impaired water bodies in the Analysis Area.
Propagation of fish and other aquatic species is the primary beneficial use of water in this
Analysis Area. Temperature, dissolved oxygen, pH, turbidity, and total dissolved solids
are the main parameters adopted by the State of Alaska as standards for assessing surface
water quality. As with streamflow, the only quantitative water quality data are available
primarily for the Kadashan River watershed.
Stream Chemistry. There are no indications of historic or future sources of chemical
contamination in the Analysis Area. Atmospheric sources of chemical pollutants are not
a major factor influencing water quality in the region. Due to the lapse of time since and
low-to-moderate intensity of past management activities, it is unlikely that stream
chemistry will be out of the natural range of variability.
Stream Temperature. The proportion of clearcut harvest within stream RMAs in the
Analysis Area can be used as a relative index of cumulative sunlight and temperature
changes associated with second-growth riparian stand development. Miles of clearcut
harvest by stream class for the Analysis Area were summarized previously in Table 12
and later on in the Vegetation section in Table 16 acres of riparian stands harvest will be
discussed. This index of past riparian harvest identifies watersheds most likely to have
experienced stream temperature changes and to experience future temperature changes.
Consequently, because of the lapse of time since past harvest and subsequent regrowth of
trees along impacted reaches, it is presumable that stream temperatures will be within
state standards and will not be out of the natural range of variability.
Vegetation
The Analysis Area is a diverse and dynamic landscape with considerable topographic
relief. It contains a mosaic of young and old forests, muskegs, avalanche brushfields, and
alpine areas. Forest vegetation structure, composition, and distribution are largely
determined by site productivity and soil drainage, as well as natural and human-caused
disturbance. The dominant tree species in the Analysis Area is western hemlock.
Varying amounts of Sitka spruce and Alaska yellow cedar are also found within the area.
The most productive forests are associated with deep, well-drained soils, many of which
are found in the alluvial fan and flood plain landforms. Sitka spruce favors these more
nutrient-rich and well-drained sites. Western hemlock dominates the less productive sites
with Mountain hemlock at higher elevations. Yellow cedar is often absent on the more
productive sites, but does occur in scattered pockets. Mixed conifer stands dominated by
small to medium-sized mountain and western hemlock and yellow cedar are typical of
wet, sparsely forested muskeg areas and low productivity sites. Shore pine, a variety of
lodgepole pine, is also common in these mixed conifer stands and open muskeg areas.
28
Red alder tends to grow on exposed and disturbed soil sites such as old roads, stream
banks and gravel bars. Sitka alder dominates exposed and disturbed high gradient stream
channels and avalanche slopes and chutes in the upper watershed.
The distribution and abundance of understory plants is highly variable and dependent on
soil drainage, the distribution of large organic debris as a rooting substrate, the amount of
light reaching the forest floor, and the type and amount of natural or human-caused
disturbance. Vaccinium (blueberry, huckleberry) tends to be the most prevalent
understory shrub. It is typically found with Menziesia, copperbush, and devil’s club.
Salmonberry is common on disturbed sites, and skunk cabbage occurs throughout the
area on wet micro-sites. The dominant forbs are typically five-leaf bramble and
bunchberry. Various species of ferns, lichens, and moss are also numerous. The
dominant plant associations1 are western hemlock/blueberry and western
hemlock/blueberry/devil’s club.
The plants in estuaries and along the beach fringe include red alder, Sitka alder,
crabapple, and various sedges and grasses.
Muskeg vegetation is a mixture of sedges, deer cabbage, sphagnum mosses, and low
growing herbs such as Labrador tea and bog laurel. Muskegs typically contain numerous
small ponds. Stunted, slow-growing shore pines grow on the less saturated areas.
Forest Vegetation Structure
Forest stand structures in the Analysis Area vary from single-storied, even-aged forests to
complex, multi-layered, uneven-aged forests.
Even-Aged Forest
Stand replacing disturbances such as clearcut timber harvest and/or windthrow are
responsible for most of the even-aged stands within the Analysis Area. These stands are
generally classified as young-growth. The Analysis Area contains 739 acres of
management induced young-growth forest. These stands follow a clearly defined pattern
of development beginning with rapid establishment of conifer seedlings, shrubs, and
herbaceous plants (i.e., stand initiation) and followed by canopy closure after about 25 to
35 years. These developing young forests are extremely dense, containing thousands of
trees per acre. They are also characterized by relatively uniform tree height and diameter
distributions that result in intense competition preventing new tree regeneration (i.e., stem
exclusion). During the stem exclusion stage, light is unable to reach the forest floor. The
absence of light prevents the growth of understory shrubs and herbs. The stem exclusion
stage can persist for 50 to 100 years before understory vegetation is reestablished and
new tree cohorts emerge (i.e., understory reinitiation). Understory reinitiation occurs as
wind disturbance, insects, and diseases create gaps in the forest canopy (Deal 2001, p. 2).
Intermediate silvicultural treatments such as thinning can reduce the duration of the stem
exclusion stage, encourage more rapid growth among a smaller number of trees, and
maintain or enhance understory vegetation. The majority of harvest generated younggrowth in the Analysis Area is currently in the early to middle stage of stem exclusion.
1
Plant association refers to the climax forest plant community type representing the end point of
succession.
29
The majority of young-growth forest in the Analysis Area is located in the valley bottoms
drainages.
Uneven-aged Forest
Uneven-aged stands are characterized by a patchy, multi-layer canopy; trees that
represent many age classes; larger trees that dominate the overstory; large standing dead
trees (snags) or decadent trees; and higher accumulations of large down woody material
(USDA 1997 [Forest Plan], p. 7-31). These multi-aged stands, which produce at least 20
cubic feet of wood fiber per acre per year or have greater than 8,000 board feet per acre,
are classified as productive old-growth forest.
The remaining forested acres of NFS Lands in the Analysis Area are characterized by
non-productive forest. Non-productive forest is associated with muskeg land types
including lowlands, fens, riparian areas, broken mountain slopes, plateaus, glacial
outwash zones, and other unproductive land types (e.g., steep, narrow canyons associated
with areas other than muskegs). Non-productive forest is characterized by very low
timber volume, mixed species, and old, defective, and stunted trees. Table 15 provides a
summary of harvest within the Analysis Area.
Table 15. Acres of Harvest by Watershed within the Analysis Area.
Watershed
Watershed Area
(acres)
Total
Harvest
(acres)
Total Watershed
Harvested
(%)
USFS
3,644
484
13.3
State of Alaska
453
255
56.3
Total
4,097
739
18.0
Source: Sitka Ranger District 2007 GIS Coverage.
30
Figure 10. Harvest within the Analysis Area.
Source: Sitka Ranger District 2007 GIS Coverage.
31
Harvest History and Regeneration
Approximately 739 acres have been harvested within the Analysis Area. This represents
an estimated 18 percent of the total land area. Clearcut regeneration harvest method was
the primary means of harvesting timber within the area and most harvest occurred
between 1968-1974. Though records indicate that harvest occurred in the valley over this
6 year period, no accurate records are available to depict when individual stands or areas
were harvested. Figure 10 above displays the locations of harvest with the Analysis
Area.
Regeneration
National Forest Management Act (NFMA) regulations state that “when trees are cut to
achieve timber production objectives, the cuttings shall be made in a way as to assure that
the technology and knowledge exists to adequately restock the lands within five years
after final harvest” [(36 CFR 219.27c (3)]. Regeneration of harvested acres on National
Forest System (NFS) lands within the Analysis Area has been successful; all previously
harvested areas have been certified as regenerated.
Young-Growth Management
The management of young-growth stands is a responsibility that comes with timber
harvest and is an important element of timber and land management. At present, only 25
acres of the harvest generated young-growth stands within the Analysis Area have been
precommercially thinned to improve the growth and yield of timber, to change the
species mix to favor more profitable species, or to improve wood quality. These 25
acres, however were thinned a part of study to determine the affects of thinning and
pruning prescriptions on wood production and understory development. Additionally,
61 acres of multiple emphases thinning, primarily to improve fish habitat, and to a lesser
extent, wildlife habitat has taken place. Multiple emphasis prescriptions typically occur
within stream RMAs and are designed to maintain, enhance, or restore understory
vegetation by delaying canopy closure; maintaining greater species diversity; and
restoring riparian structure and/or instream fish habitat by decreasing the time needed to
grow large trees that will eventually serve as large wood for instream habitat. Figure 11
below shows a picture of a typical unthinned young growth stand within the Analysis
Area. Figure 12 shows an example of one of the experimental thinning blocks.
32
Figure 11. Typical Young Growth Stands.
Figure 12. Experimental Thinning Block within the Analysis Area.
33
Future Logging
Since the end of timber activities in 1974, no commercial harvest has occurred or is
scheduled within the Analysis Area. Current LUD designation for Federal lands allows
for only limited types of commercial timber harvest within the Analysis Area. State
lands management for the Starrigavan valley states that this parcel is to be retained for
public recreation purposes. Development authorizations other than those related to this
purpose are not considered appropriate.
Precommercial Thinning
Past timber harvest has generated 739 acres of young-growth within the Analysis Area,
which constitutes approximately 18 percent of the entire Analysis Area. Only 25 acres
within the Starrigavan watershed are known to have been precommercially thinned.
Consequently, the unthinned young-growth in the area is approaching 40 years old and
bumping up against the window of opportunity for precommercial thinning. Also, the
harvest that is within riparian buffers would not be thinned under the precommercial
thinning program. Opportunities exist and should be further developed to remove and
use thinned trees from upland and riparian stands. Removal of thinned trees will reduce
slash on the forest floor and speed recovery of understory development and wildlife
habitats. Recently the Forest Service has completed such a project by utilizing thinned
second growth trees to construct a cabin within the Starrigavan Campground.
Commercial Thinning
To date, the commercial thinning or other harvest of young-growth timber has been
limited in Southeast Alaska due to the small size of the trees, the lack of a market for
small logs, and high logging costs. Commercial thinning in the Analysis Area is not
likely to occur in the near future for these reasons. However, this could change as new
markets develop and technology advances. The Forest Service Alaska Wood Utilization
Research and Development Center based in Sitka is conducting research in primary and
secondary wood processing in an effort to enhance economic opportunities for the Alaska
timber industry. Also, as above, the vast majority harvest by is currently within nondevelopment Land Use Designations (LUDs) or within riparian or beach buffers and
therefore would not be thinned under the commercial thinning program.
Riparian Vegetation
Disturbance patterns and soil moisture adjacent to streams and lakes create unique
riparian vegetation types. The streams and vegetation influence each other. During high
flows, streams disturb soils and vegetation, creating opportunities for early successional
species such as alder to grow and persist. In addition, soil moisture, which ranges from
wet to dry, influences species composition and growth rates (Malanson 1993). The
vegetation, in turn, contributes to fisheries habitat by stabilizing riverbanks; partially
controlling sediment entry into streams; providing shade, temperature control, and cover;
and contributing organic material (woody debris, leaf litter input, and insects) to the
channel.
As described above, we classify streams into different process groups, which reflect the
interaction of watershed runoff, landform, geology, climate, and glacial and tidal
influences (USDA-FS, 1992). These process groups each interact with the adjacent
34
vegetation in different ways. Information on stream channel process groups can be found
within the Stream Channel Types section and in Region 10: Channel Type Users Guide
(USDA-FS 1992).
Based on the average widths for different channel types, stream riparian management
area (RMA) acres encompass 428 acres or 10.4% of the Assessment Area (Table 16).
The distribution of the riparian areas and the harvest that has occurred within them in the
Analysis Area is shown in Figure 13.
Natural Disturbance in Riparian Areas. In addition to disturbance caused by flooding,
wind also affects riparian areas. Small-scale windthrow is the most important natural
disturbance factor in the Tongass (DeMeo et al. 1992). Ott (1995) found that canopy
gaps occupy about 9% of old-growth western hemlock/blueberry/shield fern
communities. Most of these were less than 540 ft2 (50 m2) and formed by three or fewer
trees.
Harvest in Stream Riparian Zones. Of the 428 acres of riparian zones, 150 acres have
been harvested (Table 16). Total harvest acres equal approximately 35% of the stream
riparian area in the Analysis Area.
The Starrigavan watershed has had over a third of its stream riparian areas harvested,
including harvest along main valley bottom channels (Table 16, Figures 13). The most
extensive streamside harvest and possibly most significant cumulative effects to fish
habitat historically within watersheds with significant amounts of fish habitat (> 5 miles),
occurred along Class I streams in the Starrigavan watershed, where harvest totaled
approximately 5.8 miles (61% of total) of Class I streams (Table 12).
In addition, there has been some harvest along the banks of Class III streams that directly
influence downstream Class I and II channels. The condition of stream habitats in these
channels currently appears to be stable due to regrowth of stands over the past 30 plus
years.
35
Figure 13. Riparian Harvest within the Starrigavan Creek Watershed.
Source: Sitka Ranger District 2007 GIS Coverage.
36
Any decline in instream large wood from decomposition and downstream migration will
likely be offset in the near term by recruitment of streamside stumps, unimpacted
upstream reaches and residual old growth trees left during harvest. Long term prognosis
indicates there will likely be a gap in instream large wood as existing pieces decay and
potential recruitment pieces are limited.
Table 16. Riparian Harvest within the Analysis Area
Ownership
Watershed
Area
(acres)
Total
Harvest
(acres)
Total
RMA
In
WS
(acres)
Total
RMA
In
WS
(%)
RMA
Harvest
(acres)
Total RMA
Harvest
(%)
USFS
3,644
484
348
9.6
84
24.1
State of Alaska
453
255
80
17.7
66
82.5
Total
4,097
739
428
10.4
150
35.0
Source: Sitka Ranger District 2007 GIS Coverage.
Riparian vegetation surveys were not conducted within the Analysis Area due to funding
and time constraints with this project. Several reconnaissance stream walks were
conducted throughout the valley. These surveys indicate that the current condition of
harvested riparian stands today is typical of other harvested riparian areas on the Sitka
Ranger District and Southeast Alaska.
Historic stand density for these areas was inferred by measuring average spacing of
harvested stumps. These data showed that the harvested riparian stands had an average
tree spacing of around 25-30 feet between old growth conifer trees. This spacing gives
an average of approximately 65 trees per acre (tpa). Though scattered areas of larger
residual and dominant second growth trees exist in both the main channel and tributary
riparian stands, the numbers of these large trees are currently below that of historic levels.
Subdominant young growth conifers are extremely dense in most riparian stands and are
typical of the photo in Figure 11 displayed above.
Figure 14 below shows a time sequence of aerial photos of the lower Starrigavan valley.
Though the scales and angles of each photograph are slightly off, changes in stand
compositions and characteristics are clearly evident. Figure 15 shows a photo of riparian
thinning that took place in 2005. This existing prior condition on the far right side of this
photo is typical of unthinned conifer dominated RMA stands.
37
Figure 14. Time Sequence Aerial Photos of the Lower Starrigavan Watershed.
Pre Harvest: -12 years
Overstory dominated by large,
widely spaced mainly Sitka
Spruce, lesser Western
Hemlock and a few scattered
Red Alder. Well developed
understory.
1956
Post Harvest: +2 years
Vegetation dominated by
understory brush species and
conifer seedling trees. Red
Alder seedlings dominate along
disturbed stream channel banks
and skid trails.
1976
Post Harvest: +31 years
Overstory consists of a patchy
mosaic of dense conifer
thickets and alder dominated
stands. Little to no understory
development in conifer stands.
Minor understory development
in alder stands.
2005
38
Figure 15. 2005 Starrigavan Riparian Thinning.
Wildlife
The availability and distribution of productive old growth (POG) in lower elevation
habitats is important to some species. Goshawks, bald eagles and other raptors prefer to
nest in POG habitat below 1000 feet in elevation. Sitka black-tailed deer prefer highvolume old-growth stands with southern aspects located in areas below 800 feet in
elevation for winter habitat use. Figure 16 shows locations of high quality deer winter
habitat, productive old growth habitat and harvested areas of watershed.
39
Figure 16. High Quality Winter Deer Habitat and POG within the Analysis Area.
Source: Sitka Ranger District 2007 GIS Coverage.
40
Roads
Public use of Forest roads that are connected to the Sitka road system, like Starrigavan, is
high. The Nelson Logging Road which accesses the valley is used primarily for
recreational purposes such as sightseeing, berry picking, subsistence hunting and to gain
access to trailheads and to the public and State Trooper Academy shooting ranges.
Besides the public, various Federal, State and local agencies use this road and trail system
for utilities maintenance, training, research, inventories, and field monitoring for projects
involving fish, wildlife and forest vegetation. Commercial users, including tour
operators, water distributors, and maintenance contractors also use the area. Use by all
user groups within the Analysis Area is expected to increase in the future.
There are a total of approximately 5.6 miles of road in the entire watershed. Of this total,
only 1.9 miles are open to standard vehicle traffic and 3.7 miles have been converted to
off-highway vehicle use (OHV). The portion of the Nelson Logging Road which is open
to standard vehicles is entirely on State lands and is controlled by the City and Borough
of Sitka. The Forest Service however, owns and maintains the bridges and stream
crossing structures on this and all other portions of the road and trail system. The
remaining miles of road in the valley are overgrown with vegetation and closed to all
forms of motorized vehicle traffic. Table 17 below displays the road summaries for the
watershed. Due to the fact that the majority of the OHV trails within the valley are
converted roads, all trails within the valley will be considered roads for summation and
analysis purposes. Previous restoration work within the watershed has replaced all fish
stream crossing structures with passable ‘fish pipes’. Additionally, there are several
drivable fords that have been installed on the OHV trail system on Class III and IV
streams. There are eight bridges remaining within the Analysis Area, one on HPR, two
on the drivable portion of the Nelson Logging Road and five on the OHV trail system.
Of these, only two along the OHV trail system are log stringers, one across the main
channel and one across a tributary channel. Of all the bridges, only the main channel
bridge is suffering any degradation in integrity, mainly loss of some surface fill. None
are currently becoming sediment sources or posing resource damage. Tongass GIS data
shows 1.8 miles of roads in RMAs in Starrigavan watershed (Table 17). The total road
density average for the Analysis Area is 0.9 mi/mi2 (Table 17).
Table 17. Road Summaries for the Analysis Area.
Ownership
Watershed
Area
(mi2)
RMA Area
(mi2)
Total
Miles of
Road
Miles of
Road
within
RMAs
Total
Watershed
Road
Density
(mi/mi2)
Stream
RMA Road
Density
(mi/mi2)
USFS
5.7
0.54
4.0
1.3
0.70
2.41
State
0.7
0.13
1.6
0.5
2.29
3.85
Total
6.4
0.67
5.6
1.8
0.88
2.69
Source: Sitka Ranger District 2007 GIS Coverage.
41
The effects of roads on water resources vary by the type of road as well as its location in
the landscape. Roads and associated ditchlines can intercept surface and groundwater
flows, thereby serving as first order streams during wet weather. Roads can also divert
water from stream channels where they cross roads. This expanded stream network can
serve to increase peak flow and sedimentation to stream channels if their densities and
proximity to channels is high and close enough. Though there are no set thresholds for
road densities within RMAs, we summarized that data to analyze the potential effects on
stream channels, water quality and fish habitats. This analysis shows that the overall
average road density within RMAs for the Analysis Area is 2.7mi/ mi2 (Table 17). Field
surveys during 2006 and 2007 found that OHV use outside of designated trails within the
RMAs is presently occurring in several areas along the trail system. Resource damage in
these areas ranges from moderate to severe.
Restoration Projects Completed
Initial work in the Starrigavan watershed occurred in 1986 with placement of LWD
structures in a tributary stream mostly devoid of structure. This site specific work did not
address the larger scale problems, but did improve tributary habitat and provided long
term monitoring sites for LWD structure effectiveness.
The Starrigavan restoration effort really developed starting in the late 1980s and took
several years of planning and coordination efforts between Sitka District staffs, and with
other land owners and interest groups. A lack of appropriated funding and periods of
wading through red tape and misunderstandings among different interests were overcome
during this early period by dedication on many people's part, responsive and timely
volunteer partners, on-going communication among cooperators, and general
perseverance. The big challenge then, as well as today, is to accomplish our apparently
conflicting goals of restoring watershed ecological functions while still providing a much
needed motorized recreation area for Sitka residents. Watershed protection and
restoration were the first priority then and continue to be today.
About the same time the original watershed inventory and assessment were taking place,
a growing number of local ATV riders were asking SRD to provide a place where they
could safely and legally ride within the short Sitka road system. There was increasing
public demand for motorized trail opportunities in Sitka. Illegal use was occuring
elsewhere, including near residential areas. A local ATV group (some members owned
construction companies and heavy equipment) volunteered labor and equipment to help
stabilize and improve the Starrigavan road system, while also improving the ATV trail
network. The City of Sitka upgraded and maintains the first several miles of road and
shooting range area. The Sitka Sportman's Association donated labor and money to help
upgrade the range. With this support, the Sitka District successfully competed for a
National Forest Foundation grant in 1995, and a "10% Roads and Trail Funding" grant in
1998. Since 1995, these grants, as well as USFS project dollares have allowed the Sitka
District to join forces with the various community cooperators and volunteers to
implement major restoration work in the Starrigavan watershed.
42
Project Work completed to date:
 Between 1995 and 1998, the local ATV group (Sitka Recreation Vehicle Association
[SRVA]) and associated volunteers donated $98,000 worth of skilled labor and heavy
equipment to complete the following:
--Helped restore several key miles of the valley bottom road, including replacing a
delapidated bridge, placing riprap at existing bridges to stop erosion, ditching,
replacing failed drainage structures, filling in washout areas, removing artificial fish
barriers, and stabilizing road segments along stream floodplain areas.
--Placed large rootwads as barriers at key locations to block ATV access to
Starrigavan streams.
--Removed junk cars, washing machines and other assorted garbage from the
valley.
--Constructed a parking/ATV training area near the shooting range.
--Constructed 500 ft of new trail away from stream areas.
--Helped Sitka District place a bridge across the main stream to connect 1 km of old
logging road into the new ATV trail network.

In 1995, riparian thining of 20 acres of dense, second-growth riparian stands
dominated by alders to release spruce and hemlock trees to improve bank stabilization
and provide future sources of instream LWD.

In 1995, instream large woody debris structures were placed in two more miles of
stable stream reaches identified (basinwide habitat survey) as debris deficient and key
fish habitat.

In 1995, the City of Sitka helped install a new bridge, removed trash, and upgraded
and maintained the first several miles of road.

In 1996, four old borrow ponds were enhanced to provide 1 acre of valuable juvenile
salmonid rearing habitat. The ponds were enlarged and dug deeper to prevent ATV
use and provide over-winter habitat; LWD was added to provide cover; and
connections to fish streams were improved. The gravel removed from the ponds was
stockpiled for trail construction.

In 1997, the City of Sitka provided $40,000 and the Sitka Sportman's Assoication
donated labor and additional money to help upgrade the public shooting range, some
of which was underwater during higher stream flows.

In 1998, Sitka Ranger District competed for and received special "10%" Roads &
Trails funding used to:
--Redirect stream flow from a 100 meter section of old road back into the main
channel near the upstream end of the valley, ending an ongoing sediment source
caused by a landslide.
--Ditch several locations along the road and parking lot to improve high flow
drainage.
--Replaced a bridge connecting an old logging road spur. This eliminated the major
ATV stream crossing problem in the watershed, as ATVs were driving through
salmon spawning habitat to reach this isolated 1km road segment.
43

The ATV group donated labor and equipment to help construct a bridge and new trail
section to connect an old logging road as a loop trail in the new ATV trail network.
This directed ATV activity away from streams.

Each winter and spring from 1986-2004, students from Sitka High School and
Sheldon Jackson College have volunteered to help Sitka District biologists monitor
the long term effectiveness of large woody debris structures and ponds (since 1995)
in providing salmonid rearing habitat. Results over 10+ years show a 2 to 5 fold
increase in the number of rearing juvenile salmon in restored stream reaches.

In 1997, the Pacific Northwest Forest and Range Experiment Station established 5
pre-commercial thinning plots on the footslope area of Starrigavan valley as part of a
long term study.

In 2005, riparian thining of 10 acres of dense, second-growth riparian stands
dominated by alders to release spruce and hemlock trees to improve bank stabilization
and provide future sources of instream LWD.

In 2006, riparian thining of 15 acres of dense, second-growth riparian stands
dominated by alders to release spruce and hemlock trees to improve bank stabilization
and provide future sources of instream LWD.

In 2007, contract riparian thining of 36 acres of dense, second-growth riparian stands
dominated by alders to release spruce and hemlock trees to improve bank stabilization
and provide future sources of instream LWD is scheduled to be completed.
III. Problem Identification
Hydrology/Fish
Human influences do not appear to be greatly contributing to declining hydrologic
condition. However, the Analysis Area roads and associated ditchlines do capture and
redistribute water, which could be influential at the stream reach scales. The high extent
of forest canopy loss to clearcuts and may have altered timing and quantity of flows when
initially harvested, however subsequent regrowth of vegetation most likely has subsided
these affects. The coarse valley bottom alluvium deposits where the most roads and
timber harvest have occurred are most sensitive to these factors and their influence on
groundwater reserves. There are approximately 6.7 miles of mapped streams within
managed stands in the Starrigavan watershed (Tables 18-20). Of this total, 87% (5.8
miles) occurs along Class I and II fish streams. The majority of these Class 1 and 2
harvested stream channels in Starrigavan watershed are channel types that are sensitive to
management activities, especially when those actions are carried out within RMAs or
along stream channels. Table 7 earlier, showed stream channel types in the watershed,
while Tables 18, 19 and 20 show stream channel types within managed stands and their
sensitivities to various management impacts. Escapement counts for coho salmon
conducted by ADF&G show a downward trend in coho salmon. These numbers, when
44
compared with potential productivity indicated that returns are far below that which the
valley’s streams are capable of producing.
Table 18. USFS Lands: Stream Miles within Managed Stands in the Analysis Area.
Stream Channel
Sediment
Miles LWD
Class
type
Retention
I
II
III
FP0
FP3
FP4
MM2
MM1
HC3
HC5
HC6
Total
0.2
0.7
0.6
1.2
0.1
0.4
0.3
0.5
3.9
H
H
H
H
H
M
M
M
H
H
H
M
M
L
L
L
Flood
Stream
Culvert
Sideslope
Plain
Bank
Fish
Sensitivity Protection
Sensitivity
Passage
Need
M
N/A
M
M
M
N/A
M
M
H
N/A
H
H
H
L
M
H
H
L
M
H
M
H
N/A
L
M
H
N/A
L
M
H
N/A
L
Source: Sitka Ranger District 2007 GIS Coverage.
Table 19. State Lands: Stream Miles within Managed Stands in the Analysis Area.
Stream Channel
Sediment
Miles LWD
Class
type
Retention
I
III
Total
AF1
FP0
FP3
FP4
HC6
0.2
0.6
1.0
1.0
0.1
2.8
H
H
H
H
M
H
H
H
H
L
Flood
Stream
Culvert
Sideslope
Plain
Bank
Fish
Sensitivity Protection
Sensitivity
Passage
Need
H
N/A
H
H
M
N/A
M
M
M
N/A
M
M
H
N/A
H
H
M
H
N/A
L
Source: Sitka Ranger District 2007 GIS Coverage.
Table 20. Total Stream Miles within Managed Stands in the Analysis Area.
Stream Channel
Sediment
Miles LWD
Class
type
Retention
I
II
III
Total
AF1
FP0
FP3
FP4
MM2
MM1
HC3
HC5
HC6
0.2
0.8
1.6
1.6
1.2
0.1
0.4
0.3
0.6
6.7
H
H
H
H
H
H
M
M
M
H
H
H
H
M
M
L
L
L
Source: Sitka Ranger District 2007 GIS Coverage.
45
Flood
Stream
Culvert
Sideslope
Plain
Bank
Fish
Sensitivity Protection
Sensitivity
Passage
Need
H
N/A
H
H
M
N/A
M
M
M
N/A
M
M
H
N/A
H
H
H
L
M
H
H
L
M
H
M
H
N/A
L
M
H
N/A
L
M
H
N/A
L
Riparian Harvest
Harvest occurred along approximately 6.7 miles of streams (Table 20) and 150 acres
(approximately 35%) of RMAs (Table 16) in the Starrigavan watershed. Harvest
occurred predominantly in the main valley bottom of the watershed in depositional
stream channel zones.
Many of the stream channel types in the Analysis Area are sensitive to disturbances and
are dependant on large wood for proper functioning (Tables 18-20). Fish habitat and
aquatic ecosystem function has been impaired along the watershed due to riparian
harvest, stream cleaning and the conversion from old growth conditions to dense conifer
and red alder-dominated riparian areas. Timber harvest in RMAs and along streams in
the watershed directly impacted streams by erosion and sedimentation during logging
activities and has caused a reduction of LW currently available to streams by removing
trees in RMAs that could have fallen into streams. Red alder and conifer now growing in
many of these areas are too small to effectively function as LW. The loss of LW in
tributary streams has not a noticeable impact on instream habitats, primarily in tributary
channels. Key LW levels in the main channel appear to be adequate, however total
numbers are decreasing. While this would indicate a relatively good current condition,
questions remain about future trends as current LW habitat forming features decay and
future sources of large wood as indicated in the riparian vegetation section are lacking.
Impacts foreseen if large wood supplies are not replenished in stream channels are:
increased stream gradients, reduced pool quality and quantity and reduced gravels
available for spawning and rearing of anadromous and resident salmonids.
Soils
Within the Starrigavan Creek Watershed 297 acres of MM-Haz 3-4 soils have been
harvested, 20 acres of which lies with the riparian areas of stream channels.
Additionally, 0.9 mile of road construction has occurred on these soils (Table 21 and
Figure 5). As noted earlier, MM-Haz 3-4 soils have a high likelihood to become unstable
when disturbed. Aquatic resources are also threatened when disturbance of these soils
occurs in close proximity to stream channels or directly within RMAs. To date, one
landslide has been identified within the watershed. This landslide originated in harvested
MM-Haz 3-4 soils. Field reconnaissance indicated there was no water resource
degradation was currently evident from this slide.
Table 21. MM-Haz Soils within the Starrigavan Creek Watershed.
Ownership
Total Acres
MM Haz 3-4
Soils
Total Acres
MM Haz 3-4
Soils within
RMA
Total Acres
of MM Haz 34 Soils
Harvested
Total Acres
MM Haz 3-4
Soils within
Harvested
Stream RMA
Total Road
Miles within
MM Haz 3-4
Soils
USFS
2,604
206
231
14
0.7
State
109
12
67
6
0.2
Total
2,713
218
297
20
0.9
Source: Sitka Ranger District 2007 GIS Coverage.
46
Roads/Trails
There is a total of 5.6 miles of road in the watershed. According to road reconnaissance
surveys, the 2 log stringer bridges within USFS lands pose the nearest concern for future
possible resource damage. Also, poor drainage along much of the OHV trail, as well as
off-trail OHV use in several areas are contributing sediment to water resources and
stream habitats.
Wildlife/Silviculture
The availability and distribution of productive old growth (POG) in lower elevation
habitats is important to some species. Goshawks, bald eagles and other raptors prefer to
nest in POG habitat below 1000 feet in elevation. Sitka black-tailed deer appear to prefer
high-volume old-growth stands with southern aspects that receive little snowfall and are
located in areas below 800 feet in elevation for winter habitat use (Doerr et al. 2005,
Suring et al. 1992).
Clearcut harvest was conducted over 30 years ago with thinning activities completed on a
very small portion of the managed area which previously contained productive old
growth habitat and served as valuable deer winter habitat. Red alder dominates or is a
large component of the harvested area. Much of the upland areas of conifer re-growth are
in the stem exclusion stage. With a lack of gaps in the forest canopy conifer regeneration
is extremely slow. In addition, the lack of light reaching the forest floor prevents the
growth of herb/shrubs causing reduced food source for foragers. Other problems
associated with these harvested and unthinned units are poor winter habitat for deer and
the reduction in connectivity of productive old growth. Valley bottom stands are much
the same as those in upland habitats. In each of these cases, understory development of
so forage species is poor.
There are approximately 739 acres of clearcut harvest in the watershed considered as Tier
3 and 4 young growth stands as identified in the Sitka Ranger District Strategy for
Prioritizing Stands for Treatment (2005). Tier 3 and 4 stands are defined as the north,
east or west aspect, under 800 foot elevation harvested 20+ years ago. It is important to
note that entire stands were assigned the aspect and elevation that represented the
majority of the stand area using 60m resolution Digital Elevation Models. This resulted
in some stands getting a lower rating because they were not south aspect, but portions
actually were south aspect. The majority of harvest was conducted in the valley bottom
of Starrigavan watershed. This is a very low gradient area so aspect is not important.
There are 1,360 acres of POG and HPOG adjacent to managed stands in the Starrigavan
watershed, of which 416 acres is considered high quality deer habitat.
A large portion of the harvested stands have the potential to become POG or HPOG and
provide connectivity for old growth dependent species and high quality deer winter
habitat.
47
IV. Summary and Recommendations
Historic management activities may be contributing to declining hydrologic condition
along individual reaches. Roads and ditches capture and redistribute water, which could
be influential at sub-basin or stream reach scales. The high extent of forest canopy loss
to clearcuts in riparian area over the past forty years may have altered timing and quantity
of flows during earlier stages of succession. The valley bottom and lowland alluvium
deposits where the most roads and timber harvest have occurred are most sensitive to
these factors.
The following summaries were created to move the existing conditions where problems
were identified toward desired future conditions and restoration objectives outlined in the
Tongass Forest Plan (1997).
Hydrology/Fish
Thinning activities in riparian areas can in the long term restore riparian structure and/or
instream fish habitat by decreasing the time needed to grow large trees that may serve as
LW in the future. In the short-term, the addition of LW as large, single pieces or
aggregates of smaller pieces into streams will bring more immediate benefits to quality
and quantity of fish habitat thereby creating conditions to support larger and healthier
populations of anadromous and resident fish.
In Starrigavan watershed, there are four small ponds in the mid watershed that may be
enhanced to improve valuable off-channel rearing habitat. These ponds were connected
and enhanced in 1996, however due to aggradation and vegetation growth in the outlet
channel; all of these ponds have become shallower than designed and are at risk of posing
a winter kill problem. Recent field visits have also discovered several locations that
additional ponds could be created and connected to the main channel to provide
additional off-channel rearing habitat.
Wildlife/Silviculture
The availability and distribution of productive old growth (POG) in lower elevation
habitats is important to some species. Goshawks, bald eagles and other raptors prefer to
nest in POG habitat below 1000 feet in elevation. Sitka black-tailed deer appear to prefer
high-volume old-growth stands with southern aspects that receive little snowfall and are
located in areas below 800 feet in elevation for winter habitat use.
In the Starrigavan watershed there are approximately 739 acres of clearcut harvest in the
watershed considered as Tier 3 and 4 young growth stands as identified in the Sitka
Ranger District Strategy for Prioritizing Stands for Treatment (2005). There are 416
acres of high quality deer winter habitat adjacent to young growth stands. High quality
deer winter habitat is defined as areas containing productive old growth (POG) or highly
productive old growth (HPOG) at elevations under 800 feet and is based on the Tongass
deer winter model (Doerr et al. 2005, Suring et al. 1992). There are 1,360 acres of POG
and HPOG adjacent to managed stands in the Starrigavan watershed.
48
All harvest was conducted over 30 years ago with very little upland thinning activities
afterward. The management of young-growth stands is a responsibility that comes with
timber harvest and is an important element of timber and land management.
Management of harvested areas can involve replanting and/or thinning. Thinning should
be completed on all of the uplands stands outside of the 1997 thinning study area to
improve wildlife habitats Monitoring of the 1997 study stand conditions should continue
periodically to assess stand development and the need for future silvicultural treatments
or prescriptions.
Road/Trail Maintenance
Roads within the Analysis Area are, for the most part, are in pretty good condition, with
the exception of some areas of poor drainage. Restoration work should involve
resurfacing trails and improving drainage to reduce erosion and sedimentation.
The public has expressed a desire for more roads and better quality roads to be used for
recreation purposes, and as this desire and use (of all kinds) continues to increase, the
existing open road systems on the District will become even less adequate and users will
likely branch out for new opportunities.
Access and Travel Management (ATM) planning and Off Highway Vehicle (OHV)for
the Analysis Area is currently taking place for the entire Sitka Ranger District, including
road, foot travel and OHV use. This effort will determine what road systems are
necessary to meet access objectives and follow with maintenance and rehabilitation plans
consistent with protection of soil and water resources. The Forest Service recently
announced a proposed rule to require each forest to designate a system of roads, trails and
areas slated for motor vehicle use. Once the designation process is complete, ATV use
would be confined to designated routes and areas, and ATV use off these routes (crosscountry travel) would be prohibited. The development of an OHV plan for the District
must include the education and cooperation of ATV users.
Inventory Needs
Additional riparian area and stream site visits and surveys are recommended periodically
in the future to determine if habitats are continuing to be stable and moving toward old
growth habitat characteristics. Existing, as well as any additional stream survey data
information should used to update Tongass GIS stream data, to determine fish habitat
quality (compared to reference reaches) and to determine where additional LW placement
would benefit aquatic ecosystems based on stream channel types and condition. Post
thinning riparian vegetation survey information should determine if additional thinning
activities would further enhance riparian and upland stand function.
49
A summary of recommendations for the Analysis Area at large, and specific to the
Starrigavan watershed, follows. Recommendations are not listed in order of priority or
with regard to cost.
Starrigavan Creek Watershed
 Complete road/trail rehabilitation plans focused on maintaining natural distribution of
surface and groundwater, as well as improve/restore fish passage.
 Improve signage within and along OHV trail to educate users on safe riding and the
protection of resources.
 Create a new OHV trail user parking area adjacent to the existing one to expand and
create a larger ‘play area’ for OHV users.
 Install relief culverts/bridge along the lower Nelson Logging Road to alleviate
flooding issues.
 Consider second growth management objectives in harvested riparian areas. Primary
objective should be recovery of old growth structure and canopy for wildlife and
fisheries habitat.
 Consider second growth management objectives in harvested upland areas. Primary
objective should be recovery of old growth structure and canopy to restore wildlife
habitat.
 Update the existing stream and riparian GIS layers using field verification, digital
orthophoto overlays, and aerial photo interpretation. Use this to update the
information presented in this analysis for the future updates to the Sitka Sound
Landscape Assessment.
 Focus silvicultural prescriptions to minimize windthrow forest canopy alteration. All
silvicultural activities should also include objectives to minimize windthrow.
 Maintain habitat connections by utilizing innovative thinning prescription techniques
to replicate natural disturbances.
 Thin 96 acres in previously harvested RMAs for Fisheries and Watershed
improvements.
 Thinning 233 acres of previously harvested upland areas for wildlife habitat
improvements.
 Improve instream habitat through the placement of LW along approximately 3.5
miles of tributary channels.
 Improve stream and riparian habitat and channel function by channel restoration
below the first bridge on the Nelson Logging Road.
 Construct 4 new burrow ponds for off channel rearing habitat for coho salmon.
 Enhance 4 existing burrow ponds for off channel rearing habitat for coho salmon.
 Work with the Alaska Department of Fish and Game (ADF&G) and the US Fish and
Wildlife Service (USFWS) to identify key wildlife habitat connectivity
 Work with Alaska Department of Natural Resources (AK-DNR) to complete
cooperative restoration activities to improve habitats within the entire watershed.
50
Monitoring and Information Needs
A variety of hydrologic information needs are briefly identified here.
1. How does seasonal and annual streamflow vary in response to continued climate
change? Maintain stream gages at Starrigavan.
2. How do low flows vary during rainless weather in valley bottom and lowland
areas? Maintain/add district stream gages.
3. How does groundwater influence low flows in watershed with and without
management activities? Install and maintain monitoring wells on the District.
4. What is the stream temperature regime in these watersheds and their tributaries
with respect to state water quality criteria (focus on low flows and harvested
reaches)? Install continuous temperature instruments (and/or maintain those near
stream gages) and add air temperature.
5. What is the condition of all drainage structures and/or removed structures on
roads with respect to flow conveyance, diversion (seasonal or perennial), fish
passage, and sediment sources? Continue and expand field inventories and
monitoring of removed structures.
6. Are OHV trails diverting streams or causing resource degradation? Continue and
expand field inventories.
7. What are the long term trends in channel morphology and habitat features along
harvested reaches within the Analysis Area? Complete and repeat Tier II surveys
and establish monumented Tier III surveys and cross sections.
8. How is LW recruitment in the Analysis Area watersheds affecting LW
distribution and function? Tag and monitor key pieces.
9. Future: Has upland and riparian thinning activities in the watershed achieved the
desired conifer species mix and spacing, and has the connectivity of existing
productive old growth been enhanced by these management actions or is
additional thinning or planting necessary? Continue to monitor previously
thinned stands to verify that species mix and spacing has been obtained. Thin
unthinned stands to obtain resource objectives
10. Have previous watershed/in-stream restoration measures in Starrigavan Creek
obtained their enhancement objectives. Monitor existing in-stream LW and pond
enhancement structures and projects.
51
V. Restoration Plan
This section outlines the restoration strategy designed to meet the objectives for the
Analysis Area previously completed in this document.
Restoration Objectives
The Tongass Forest Plan (USDA 1997) designates standards and guidelines for the
management of different forest resources. The following objectives are pertinent to the
Starrigavan Creek watershed:

Hydrology/Fish
o Restore stream banks and stream channel processes (i.e.: flow regime,
sediment dynamics)
o Maintain or restore natural quantities of LWD
o Reconnect streams and restore opportunities for fish migration
o Restore water quality to provide for fish production and sustain soil
productivity
o Move physical characteristic (i.e.: width-depth ratio, pool spacing,
incision) and aquatic habitat (i.e.: spawning, rearing habitat) toward premanagement conditions expected for channel types.

Wildlife/Silviculture
o Provide productive old growth habitat and connectivity for dependant
species (i.e.: goshawk, brown bear, marten), create deer winter habitat, and
enhance forage component.
o Improve timber growth and productivity.

Roads
o Minimize and/or mitigate adverse effects to fish and wildlife habitat and
populations.
o Maintain the present and continued productivity of anadromous fish and
other fish habitat.
Table 22 displays the criteria used to prioritize watershed improvement activities. The
following sections provide detailed project descriptions, objectives, benefits, timelines
and estimated project costs.
52
Table 22. Criteria for Prioritizing Analysis Area Watershed Improvement
Activities.
Driving Factor
Restoration Issues/Concerns/
Objectives
Relative
Degree of
Influence
Relative
Probability
of Success
for
Restoration
Rehab
Priority
Moderate
to high at
the stream
reach
scale.
High
#1
Moderate
to high at
the local
stream
reach
and/or
stand
scale.
Low
understory
development
in the very
short-term
due to slash
accumulation,
High in longterm
#2
Low
Low in the
short-term
#3
Reduced riparian tree size and
stand age due to harvest resulting
in future source of LW deficit
Timber harvest
and Young
Growth Mgt
(Stream
Habitat).
Timber harvest
and Young
Growth Mgt
(Wildlife
Habitat).
Timber harvest
and Young
Growth Mgt
(Stream Flow).
Objective: Implement thinning
treatments for dense, young
growth stands to accelerate
development of mature forest
canopy structure. Increase tree
diameter upon snagging will
increase Key LW counts,
improving Stream Habitat.
Reduced tree size and stand age
due to harvest resulting in stem
exclusion structure and reduced
understory vegetation in riparian
and upland.
Objective: Implement wildlife
emphasis thinning treatments
for dense, young growth stands
to accelerate development of
mature forest canopy structure
to improve deer winter range
and bear habitat.
Reduced canopy may accelerate
snowmelt, resulting in earlier
depletion of groundwater
reserves. Rapid release of shrubs
may increase evapo-transpiration
loss.
Objective: Implement thinning
treatments for dense, young
growth stands to accelerate
development of mature forest
canopy structure.
53
Pond Creation /
Enhancment
Roads and
Runoff
Diversions
A few burrow ponds created for
road construction are located
immediately adjacent to stream
channels.
Objective: Enhance and Create
Ponds and connect to stream
channels creating off-channel
rearing habitat for fish.
Some roads intercept
groundwater and may have
altered hydraulic gradients,
reducing groundwater available to
streams. Some roads capture and
divert surface water. Surface
erosion from poorly drained or
deteriorating road/trail surfaces is
contributing sediment to stream
channels and fish habitat. Off
trail OHV use is also causing
resource damage.
Objectives: Restore adequate
stream flow conveyance, cross
and surface drainage and fish
passage along all roads. Block
non-designated OHV trails.
54
Moderate
to high at
stream
reach scale
Moderate to
high
#4
Moderate
to high at
sub-basin
or stream
reach scale
High
#5
Project Descriptions and Implementation Schedule:
1. Starrigavan Watershed Young Growth Riparian Treatments.
Site Type/Description: Current riparian stand compositions consist of dense second
growth conifer stands. Conifer size distribution shows the majority of trees are small in
diameter and suppressed by other conifers and high density alders.
Treatment Objective/Description: Implement thinning strategies that will improve
second-growth canopy conditions to improve low flows, riparian wildlife habitat and
accelerate dominant tree growth for future sources of instream LWD. Objective will
involve treatment of 96 acres of previously harvested riparian stands (USFS: 30 acres;
State: 66 acres) to reduce tree density and improve understory development. Thinning
treatments should consist of a combination of girdling and thinning alders to release
conifers to a minimum 16 foot by 16 foot, with an ultimate long-term goal of 20 foot by
20 foot spacing.
Benefits: : Restored riparian habitat and increased conifer growth for future sources of
LWD along Class 1 and 2 fish streams, improved fish rearing habitat in natural stream
channels, improved bank stability and watershed function.
Outputs: 96 acres of riparian habitat restored
Project Phase/FY: Design and Restoration, FY 2009
Estimated Cost: $32,700
Funding Type(s): NFVW
Activity Type: Watershed Stewardship
Potential Partnership Contributors: AK-DNR, TU, SCS, City of Sitka
55
2. Starrigavan Watershed Young Growth Upland Treatments.
Site Type/Description : Current upland stand compositions consist of dense second
growth conifer stands with little understory vegeation. Conifer size distribution show the
majority of trees are small in diameter and suppressed by other conifers and alders.
Treatment Objective/Description: Implement thinning strategies that will improve
second-growth canopy conditions to improve wildlife habitat. Objectives will involve
treatment of 233 acres of high priority previously harvested upland stands (USFS: 149
acres; State: 94 acres) to reduce tree density and improve understory development.
Thinning treatments should consist of a combination of girdling, thinning and gap
creations to meat wildlife habitat objectives. Additional slash treatments should also be
implemented to provide and maintain wildlife and recreational corridors to upland and
alpine habitats.
Benefits: Restored wildlife habitat and increased understory development to enhance and
restore deer winter range habitat and survivability.
Outputs: 233 acres of wildlife habitat restored
Project Phase/FY: Design and Restoration, FY 2009
Estimated Cost: $99,500
Funding Type(s): NFWF
Activity Type: Wildlife Stewardship
Potential Partnership Contributors: AK-DNR, Stewardship contracts; Sitka Conservation
Society volunteer help to open and clear game trails.
56
3. Analysis Area Fish Habitat Improvements
Site Type/Description: 3 tributary channels totaling approximately 3.5 miles of stream
channel lack instream fish habitat. Scope of problems identified through field
reconnaissance.
Treatment Objective/Description: Place and install LW habitat and cover structures in
appropriate reaches and locations to create pools and rearing habitat. Source of LW
would come from thinned trees within adjacent RMA. Emphasis is on coho salmon.
Benefits: Increased number of pools, pool depth and rearing habitat for coho salmon.
Increased recreational opportunities for sport and commercial fishers. Increased wildlife
forage opportunities (bears).
Outputs: 3.5miles of stream channel
(USFS: 2 streams – 1 mile; State: 3 streams – 2.5 miles).
Project Phase/FY: Survey and Design FY 2008
Restoration FY09-11
Estimated Cost: FY08 $20,000
FY09 $27,000
FY10 $28,000
FY11 $15,000
Funding Type(s): NFWF
Activity Type: Watershed Stewardship
Potential Partnership Contributors: AK-DNR, TU, SCS, City of Sitka
57
4. Analysis Area Fish Habitat Improvements
Site Type/Description: Enhancement of four existing ponds and creation of four new
ponds for off-channel fish rearing habitat.
Treatment Objective/Description: the four existing ponds would be deepened and
widened to improve function and original design. Four additional new ponds would be
created near tributaries and the main channel. Ponds would be connected to neighboring
channels to provide off-channel rearing habitat for coho salmon and other fish species.
Excavated material when and where suitable would be used for other integrated projects
like trail resurfacing, OHV play area enhancement, floodplain reconstruction, etc.
Benefits: enhanced anadromous and resident fish habitat, reduced sedimentation and
improved watershed function and water quality.
Outputs: 8 ponds created enhanced; 4 acres.
Project Phase/FY: Design and Restoration, FY 2009
Estimated Cost: $20,000
Funding Type(s): NFWF
Activity Type: Watershed Stewardship
Potential Partnership Contributors: AK-DNR, TU, SCS, City of Sitka, Recreation, OHV
user groups
58
5. Analysis Area Water Quality and Fish Habitat Improvements
Site Type/Description: Restore channel function and fix flooding caused by location and
design of the Nelson Logging Road (NLR).
Treatment Objective/Description: Placement of floodplain drainage structures (culverts),
as well as raising the road surface elevation to eliminate flow constriction and flooding
issues at the first bridge on the NLR. Also, channel reconstruction of 200 meters
downstream of the bridge to reestablish a stable dimension, pattern and profile of the
channel through the only remaining old growth section in the lower valley
Benefits: Restored anadromous fish habitats, reduced sedimentation and improved
watershed function and water quality. Improved access to the established recreation
facilities in the valley.
Outputs:
Project Phase/FY: Design: FY 2010; Restoration, FY 2011
Estimated Cost: $90,000
Funding Type(s): NFVW, NFWF, Engineering, In-kind private
Activity Type: Watershed Stewardship
Potential Partnership Contributors: AK-DNR, TU, City of Sitka
59
6. Analysis Area Water Quality, Fish Habitat and Passage Improvements
Site Type/Description: Construct on a new parking area, access trail and expansion of
the OHV training/play area for the Starrigavan OHV trail. Project would be in
cooperation with the USFS Recreation group and in conjunction with other restoration
projects, namely the pond enhancement and creation project. A new parking area would
be installed to the west of the existing parking area. The now combination parking/play
area would be expanded and solely become a play area. A trail connecting the parking
area and play area would be constructed parallel to the road to keep OHVs and vehicle
traffic separated. Enhancement of the training/play area would include installation of
barriers to keep OHV users from riding in nearby fish streams, which is currently taking
place.
Treatment Objective/Description: Construction of one new parking area, one OHV
access trail and expansion and enhancement of the OHV play area.
Benefits: Reduced aquatic and riparian resource damage, improved fish habitat,
increased OHV recreation opportunities, and increased safety by eliminating mixed
traffic on the Nelson Logging Road.
Outputs:
Project Phase/FY: Design: FY 2011; Construction: FY 2012
Estimated Cost: $
Funding Type(s): NFVW, Recreation, In-kind private
Activity Type:
Potential Partnership Contribution: AK-DNR, TU, SCS, City of Sitka, Recreation, OHV
user groups
60
References
Brew, D.A. 1990. Plate-tectonic Setting of Glacier Bay National Park and Preserve and
of Admiralty Island National Monument, Southeastern Alaska. In A.M. Milner and J.D.
Wood, Jr., eds. Proceedings of the Second Glacier Bay Science Symposium, pp 1-5.
U.S. Department of the Interior, National Park Service, Alaska Regional Office,
Anchorage, Alaska.
Curtis, J.C. 1993. Southeast Alaska Forecasting Problems. In T. Brock, ed.,
Proceedings: Watershed ’91, R10-MB-217. US Department of Agriculture, Forest
Service, Alaska Region. Juneau, AK, 20-26 pp.
Deal, R.L. 2001. The effects of partial cutting on forest plant communities of western
hemlock-Sitka spruce stands in Southeast Alaska. Canadian Journal of Forestry
Resources. 31: 2067-2079.
DeMeo, T.E.; Martin, J.R.; and West, R.A. 1992. Forest Plant Association Management
Guide: Ketchikan Area, Tongass National Forest. USDA Forest Service, Alaska Region,
Juneau, AK, 405 p.
Doerr, J.G., E.J. DeGayner, and G. Ith. 2005. Winter habitat selection by Sitka blacktailed deer. Journal of Wildlife Management 69(1):322-331.
Dolloff, C. Andrew. 1986. Effects of stream cleaning on juvenile coho salmon and
Dolly Varden in Southeast Alaska. Transactions of the American Fisheries Society 115:
743-755, 1986.
Duncan, S.H. 1986. Peak Stream Discharge During Thirty Years of Sustained Yield
Timber Management in Two Fifth Order Watersheds in Washington State. Northwest
Science 60:258-264.
Farr, W.A. and Hard, J.S. 1987. Multivariate Analysis of Climate Along the Southern
Coast of Alaska-Some Forestry Implications, Res. Pap. PNW-RP-372. USDA Forest
Service, Pacific Northwest Research Station, Portland, OR. 38 p.
Harr, R. Dennis. 1986. Effects of clearcutting on rain-on-snow runoff in western
Oregon: a new look at old studies. Water Resources Research, Vol. 22, No. 7, pages
1095-1100.
Heilman, P.E. and Gass, C.R. 1974. Parent Materials and Chemical Properties of
Mineral Soils in Southeast Alaska. In Soil Science: Vol. 117, No. 1, p21-27.
Hicks, Brendan J., Robert L. Beschta, and R. Dennis Harr. 1991. Long-term changes in
streamflow following logging in western Oregon and associated fisheries implications.
Water Resources Bulletin Vol. 27, No. 2, pages 217-226.
Johnson, P.R. and Hartman, C.W. 1969. Environmental Atlas of Alaska. Inst. Arct.
Environ. Eng. and Inst. Water Resour., Univ. Alaska, College.
61
Jones, J. A. 2000. Hydrologic processes and peak discharge response to forest removal,
regrowth, and roads in 10 small experimental basins, western Cascades, Oregon. Water
Resources Research, Vol. 36, No. 9, pages 2621-2642, September 2000.
Jones, J. A., and G. E. Grant. 1996. Peak flow responses to clear-cutting and roads in
small and large basins, western Cascades, Oregon. Water Resources Research, Vol. 32,
No. 4, pages 959-974.
Keppeler, Elizabeth T. and Robert R. Ziemer. 1990. Logging effects on streamflow:
water yield and summer low flows at Caspar Creek in northwestern California. Water
Resources Research, Vol. 26, No. 7, pages 1669-1679, July 1990.
Malanson, G. P. 1993. Riparian Landscapes. University Press, Cambridge, U.K., 296 p.
Maser, Chris and James Sedell. 1994. From the forest to the sea: the ecology of wood in
streams, rivers, estuaries, and oceans. St. Lucie Press, Delray Beach FL. 200 pages.
Myren, R.T., and Ellis, R.J. 1984. Evapotranspiration in Forest Succession and
Longterm Effects Upon Fishery Resources: A Consideration for Management of Oldgrowth Forests. In Meehan et al (1984) p183-186.
Neal E.G., M. Todd Walter, C. Coffeen. 2002. Linking the Pacific decadal oscillation to
seasonal stream discharge patterns in Southeast Alaska. Journal of Hydrology 263
(2002) pages 188 – 197.
NOAA. 2003. National Climatic Data Center, National Oceanic and Atmospheric
Administration website. http://www.ncdc.noaa.gov/oa/climate/climatedata.html
Nowacki, Gregory, Michael Shepard, Patricia Krosse, William Pawuk, Gary Fisher,
James Baichtal, David Brew, Everett Kissinger, Terry Brock. 2001. Ecological
Subsections of Southeast Alaska and Neighboring Areas of Canada. USDA Forest
Service, Alaska Region. R10-TP-75.
Ott, R.A. 1995. Small-scale Natural Disturbance and its Influence on Vegetation Structure and Diversity in Temperate Rainforests of Southeast Alaska. Progress Report to
USDA Forest Sevice, PNW Research Station, Forestry Sciences Laboratory, Juneau, AK,
50 p.
Stednick, J.D. 1981. Precipitation and Stream Water Chemistry in an Undisturbed
Watershed in Southeast Alaska, Research Paper PNW-291. Pacific Northwest Forest and
Range Experiment Station, Portland Oregon, 8p.
Swanston, D.N. 1969. Mass Wasting in Coastal Alaska. USDA Forest Service,
Research Paper PNW-84, Pacific Northwest Forest and Range Experiment Station,
Portland, Oregon.
62
Swanston, D.N. and Marion, D.A. 1991. Landslide Responses to Timber Harvest in
Southeast Alaska. In Proceedings of the Fifth Federal Interagency Sedimentation
Conference, March 18-21, 1991, Lan, S.S. and Huang Kuo, Y. (editors), Las Vegas, NV,
p10-49 to 10-56.
Thomas Robert B. and Walter F. Megahan. 1998. Peak flow responses to clear-cutting
and roads in small and large basins, western Cascades, Oregon: a second opinion. Water
Resources Research, Vol. 34, No. 12, pages 3393-3403.
USDA-FS (see USDA Forest Service).
USDA Forest Service. 1986. Chatham Area integrated resource inventory handbook.
Unpublished Handbook. United States, Forest Service. On file with: U.S. Department of
Agriculture, Forest Service, Tongass National Forest, Sitka Ranger District, Sitka, AK.
USDA Forest Service. 1992. Channel Type Users Guide, Tongass National Forest. R10
tech. paper 26, S. J. Paustian (editor), Juneau, AK, 179 p.
USDA Forest Service. 1995. Ecosystem Analysis at the Watershed Scale: Federal Guide
for Watershed Analysis. United States, Forest Service.
USDA Forest Service. 1996. Indian River Watershed Analysis; Unpublished Report.
United States, Forest Service. On file with: U.S. Department of Agriculture, Forest
Service, Tongass National Forest, Sitka Ranger District, Sitka, AK
USDA Forest Service. 1997. Tongass National Forest Land and Resource Management
Plan (R10-MB-338dd, 1997), Land Management Plan Revision, Final Environmental
Impact Statement (R10-MB-338b, 338c, 338 e through 338h, and 338n, January 1997,
and Errata, May 1997, and Record of Decision, May 1997). United States, Forest
Service.
USDA Forest Service. 2005. Sitka ranger district wildlife habitat in young growth: A
strategy for prioritizing stands for treatment. July. 11 pp.
63