Executive Officer
David Leland
Mark Neely
On June 27, 2002, the North Coast Regional Water Quality Control Board (Regional Water
Board) directed staff to convene a team of independent scientific experts to provide input
regarding sediment impairment in the Freshwater, Bear, Jordan, Stitz, and Elk River watersheds.
The panel was convened in August of 2002, and, produced the above-noted document on
December 27, 2002.
"The energy challenge facing California is real. Every Californian needs to take immediate action to reduce energy
consumption. For a list of simple ways you can reduce demand and cut your energy costs, see our Web-site at:
www.swrcb.ca.gov."
EXECUTIVE SUMMARY................................................................................... 01
INTRODUCTION
THE ASSIGNMENT
Introduction .................................................................................................. 09
FINDINGS
Background ................................................................................................. 13
Longer-term (4 to 5 years).......................................................................... 32
CONCLUDING REMARKS................................................................................ 35
BIBLIOGRAPHY ................................................................................................ 37
APPENDICES
A. Terms of Reference for the Panel........................................................... 47
LIST OF ACRONYMS
CDF California Department of Forestry
CGS California Geologic Survey
CWE Cumulative Watersheds Effects
GIS Geographic Information Systems
HCP Habitat Conservation Plan
ISRP Independent Scientific Review Panel
MWAC Mass Wasting Areas of Concern
NCRWQCB North Coast Regional Water Quality Control Board
PALCO Pacific Lumber Company
PBDM Physically Based Distribution Models
SYP Sustained Yield Plan
THP Timber Harvest Plan
TOR Terms of Reference
TMDL Total Maximum Daily Loads
USDA United States Department of Agriculture
WEPP Water Erosion Prediction Project
EXECUTIVE SUMMARY
Physically based distributed models (PBDMs), such as vest rates that will not impede recovery of impaired
the Modeled Sediment Budget approach, usually rep watersheds:
resent environmental systems via differential equa
tions where all the terms in the equations are GIS data acquisition and analysis to stratify
measurable physical quantities (for example rainfall, watersheds into sediment production land
soil erodibility, etc.). Their inputs and outputs are spa classes.
tially distributed. This means, for example, that soil Development of land class rate factors based
erodibility must be calculated for all different parts of on empirical sediment production information.
the watershed and separate values applied for all sub- Review and refinement of land classes and
as a spatially varying amount in the sub-watersheds. Independent peer review of harvest rate
The Panel concluded that although the Modeled calculations based on land class rate factors.
Sediment Budget method has promise over the long- Performance monitoring to measure the
term, major problems are inherent in the use of a com reliability of harvest practices and mitigation
plicated sediment budget approach to establish measures in terms of limiting sediment
allowable timber harvest rates. For example, many production and meeting land class rate factor
qualitative judgements about sources of sediment are estimates.
poorly supported by quantitative studies. Also, the Adaptive revision of allowable harvest rates
quality and quantity of empirical data that are avail based on outcomes from performance
able or likely to become available over the short-term, monitoring measurements.
particularly regarding the efficacy of proposed mitiga
tion techniques, are key limiting factors. The Panel notes that due to the complexity and vari
ability of the physical systems involved and to the lack
EMPIRICAL SEDIMENT BUDGET of measurements, precise prediction of harvest rates
The Empirical Sediment Budget approach (Reid, 1998a that will not impede recovery is not possible. This is
& Reid, 2000), represents the physical system through further compounded by the untested nature of some
empirical relationships that relate inputs to outputs. of the innovative mitigation measures proposed. The
Empirical models use relatively simple regression- most reasonable approach to ensure recovery is to
derived equations to relate inputs to outputs. They are develop an adaptive policy that relies on monitoring
not distributed, meaning that they produce outputs and the measurement of reliability of measures to
that carry a value for the whole system (e.g. sediment limit sediment production during timber harvest or
yield for a whole watershed). The equations contain restoration practices.
constants and coefficients that are not directly related
to physically measurable properties of the system.
However, they generally have the advantage of being QUESTION 2
able to produce reliable results with sparse data. What options are available (e.g. dredging, and
Overall, the Panel found the Empirical Sediment Bud modification of activities resulting in, or reducing,
get approach to be fundamentally sound and at a level sediment delivery) that can be immediately
of detail commensurate with the kinds and amounts of implemented and will be effective in lessening
data that are available, or can be made available, in the the adverse flooding conditions and impacts to
near future. The field-based land class rate factors nec beneficial uses? Please discuss the potential ben
essary to use this approach can be estimated from efits, limitations, and tradeoffs of these options
existing aerial photographs, publicly available geologic for each watershed.
hazard maps, and published studies of representative
similar watersheds such as Caspar Creek.
Key Findings
RECOMMENDATIONS The Panel has evaluated the options available to
The Panel suggests that the following steps be taken lessen adverse flooding conditions and impacts to
to refine the Empirical Sediment Budget approach to beneficial uses of water. It has also developed a matrix
the point where it can be used to calculate timber har of benefits and impacts that may be used to prioritize
QUESTION 1 Please review the provided documents, and any other relevant information, regarding calcula-
tion of appropriate rates of timber harvest that would not impede recovery, meaning that the
water body can support all designated beneficial uses of water and meet the water quality
standards as outlined in the Basin Plan, from excess sediment loads and would not cause or
contribute to exceedence of water quality objectives. Please discuss the technical strengths
and weaknesses of the varying approaches described in some of these documents to address
harvest rate and flood severity, as well as any other reasonable approaches to calculate a rate
of harvest for each of the five watersheds that is protective of water quality, which considers
natural and other anthropogenic sediment sources.
QUESTION 2 What options are available (e.g. dredging, and modification of activities resulting in, or reduc
ing, sediment delivery) that can be immediately implemented and will be effective in lessen
ing the adverse flooding conditions and impacts to beneficial uses? Please discuss the
potential benefits, limitations and tradeoffs of these options for each of the five watersheds.
QUESTION 3 What additional data or piece(s) of information, if any, will be useful in the future for refining
approaches to address the above issues? This can include monitoring information, modeling
exercises, etc.
In addition, CONCUR recommended a Convening pendent Scientific Review Panel to provide guidance
Committee be established to determine the appropri on a suite of possible interim options that could be
ate structure and focus for the mediation process. used to protect beneficial uses and address flooding in
The NCRWQCB agreed to ask CONCUR to establish the four watersheds.
a Convening Committee, consisting of representatives During the four-week process, the Convening Com
from the Pacific Lumber Company, Humboldt Water- mittee members worked diligently but ultimately
shed Council, the Freshwater Working Group, the reached an impasse while attempting to frame an ini
Environmental Protection Information Center, and sev tial set of questions to guide the scientific review
eral non-industrial timber interests. NCRWQCB staff process. This result was reported at the June 27, 2002
participated as consultants to the Convening Commit NCRWQCB meeting. After hearing a report from NCR
tee (April through June, 2002) in order to assist the WQCB staff and testimony from Convening Commit
group in developing measures likely to be acceptable tee members and facilitators, the NCRWQCB directed
to the Regional Water Board, while maintaining the staff to: (1) initiate its own facilitated Scientific Review
Regional Water Boards independent authority over Process which would build upon the work on the Con
measures or other actions that it will be required to vening Committee to address issues in all five water
approve. The Convening Committee was originally sheds, and (2) invite the original members of the
focused on resolving issues in the Freshwater, Bear, Convening Committee, as well as Elk River representa
Jordan and Stitz watersheds only. tives, to assist the NCRWQCB staff in finalizing an ini
tial TOR that laid out the objectives, timeline, and
History of the Convening Committee deliverables for the proposed ISRP.2
The Convening Committee met five times from May Consistent with this motion, CONCUR and the
30th to June 26th. Its self defined and adopted mission Regional Water Board staff organized a July 12th
was to develop and agree on a set of interim meas meeting among the original members of the Conven
ures, by August 31, 2002 aimed at protecting and ing Committee, along with several Elk River represen
restoring beneficial uses and to mitigate nuisance in tatives to provide advice to the Regional Water Board
the Freshwater, Jordan, Bear, and Stitz watersheds staff on finalizing the TOR.
prior to TMDL development. One primary recom
mendation of the Committee was to establish an Inde 2. Refer to TOR (Appendix A).
Andrew Collison, Ph.D. Philip Williams & Associates Fluvial Geomorphology, Hydrology,
San Francisco, CA Slope Stability, Environmental Modeling
David Tarboton, Sc.D. Utah Water Research Laboratory, Hydrology, Civil Engineering
Utah State University
Logan, UT
Panel Nomination and Selection Panel that could collectively provide analysis and
Members of the Convening Committee were invited to understanding in Hydrology, Hydraulics and Fluvial
propose selection criteria as well as potential candi Geomorphology, Aquatic Ecology/Fisheries Biology,
dates for the Panel. They were then asked to offer Civil Engineering/Water Quality, Geotechnical/Slope
feedback on the suitability of candidates based on the Stability, Restoration Ecology, and Forestry/Silvicul-
following jointly developed selection criteria outlined ture. NCRWQCB staff took into consideration the need
in the TOR: 1) technical capability in their respective for representation on all of these disciplines in the final
disciplines plus an ability to work across disciplines; 2) selection process. The Panel selected by the NCR
objectivity, as reflected by their willingness and ability WQCB, including their affiliations and expertise, is
to integrate diverse viewpoints; 3) ability to work col shown in Table 2.
laboratively; 4) track record of science advising for The Independent Scientific Review Panel was con
environmental decision-making; 5) availability; 6) vened by the NCRWQCB to address three specific
experience evaluating cumulative watershed effects in questions defined in the final TOR document dated
a forested setting; 7) proven track record of meeting August 6, 2002. The Panel was officially appointed on
deadlines; 8) experience with practical application; August 10th, 2002 and began communication by email
and 9) broad acceptability by the stakeholders. Final and conference calls during the August, September
selection was to be made by the NCRWQCB staff. and October time frame. A compendium of 22 docu
Informed by the advice of the expanded Convening ments was transmitted to the Panel on August 14th
Committee and its facilitators, the NCRWQCB staff then along with a list of additional documents to be made
evaluated each of the proposed candidates to recruit a available upon request.
Site Visit and Deliberations site, and the bridge at Bear Creek. The site visit was
The Panel conducted face-to-face deliberations from followed the next morning with an aerial surveillance
October 9th to 11th. During that time, the Panel partic of all five watersheds. In this way, the Panel met with
ipated in a one-day site visit. The itinerary 3 for the site Dr. Matt OConnor, consultant to Pacific Lumber Com
visit was arranged by CONCUR with considerable pany (PALCO), and Dr. Jeff Barrett of PALCO. Addition
input and support from local stakeholders. A majority al briefings were provided by Mr. John Munn of CDF,
of the day was spent in the Freshwater Creek and Elk Dr. Leslie Reid of RSL, and Ms. Trinda Bedrossian, Mr.
River watersheds. There was also a brief tour of the Bill Short, and Mr. Gerald Marshall of CGS.
Bear Creek watershed. Representatives from various
stakeholder groups accompanied the Panel and Organization of this Report
afforded the opportunity to answer questions and This report is organized into seven sections. It begins
interact with the Panel throughout the day. While it with an executive summary followed by the introduc
was not possible to meet with all who expressed inter tion section. The introduction outlines the historical
est, Panel members considered additional material context, the impetus for the Panel, and the method by
provided during and following the meeting.4 which Panel members was convened. The third sec
The objectives of the site visit were to familiarize the tion presents the Panels perspective on its assign
Panelists with the watersheds and the differences ment. This section includes an overview of the
between them and to provide the Panelists with the physical setting, a description of the Panels methodol
opportunity to see and hear first hand from local ogy and a commentary on the relationship between
stakeholders where impacts have occurred and what science and policy. The fourth section describes the
types of monitoring and mitigation efforts have been Panels response to each of the three questions out
undertaken to date to address the issues. The site visit lined in the TOR.
included several stops in the Freshwater watershed, To improve the written flow of its final report, the
including the Howard Heights bridge, a fish trapping Panel elected to address the questions in a different
station, and the monitoring station at Dr. Terry Roelofs order than appears in the Terms of Reference. The
home. In addition, there were stops along the road to report first addresses the TOR Question 2 regarding
Elk River, Bridge Creek, Kristi Wrigleys Ranch, the calculation of appropriate rates of harvest. Second,
USGS gauging station, a properly abandoned road the report addresses TOR Question 1 concerning eval
uation of short-term options to mitigate flooding and
3. See Appendix B.
4. Two documents received from PALCO on December 16 were too late
promote recovery of beneficial uses. Finally, the report
for inclusion in this report but could be considered as part of a phase addresses TOR Question 3 with a series of recommen
two effort.
dations for additional analysis and data collection.
particular watershed, the orientation of structural and especially during unusually wet winters, is not pre
stratigraphic elements relative to topography can be dictable. The impacts of management activities are
expected to influence susceptibility to erosion and subject to normal climate variability and need to be
mass wasting. Episodic tectonic activity and a fluctuat evaluated against a background that recognizes this
ing but generally wet climate would produce large variability. For example, if a 100-year storm occurs
amounts of sediment, sometimes catastrophically, during early monitoring, it should not necessarily lead
even in the complete absence of human activity (e.g., to further restrictions on timber harvest. Likewise, if
Dumitru, 1991; Clarke, 1992; Li and Carver, 1992; Carv no significant storms occur in the near future, it should
er, 1994; Jacoby and others, 1995; Merritts, 1996; Som not lead to an increased rate of timber harvest.
merfeld and others, 2002). There is evidence in the When formulating policy in light of the issues of
Holocene geologic record, however, suggesting that natural variability and lack of adequate data, two con
major floods during the 20th century have delivered trasting options emerge. A conservative approach to
considerably more sediment than those of previous timber harvest would limit the rate of harvest until
centuries. This difference has been attributed to a short-term effects of timber harvest and mitigation
combination of changing precipitation patterns and measures can be demonstrated by performance effec
land use practices (Li and Carver, 1992; Sommerfield tiveness monitoring, especially as influenced by site
and others, 2002). specific geologic conditions affecting mass movement
Data are lacking to help define the background level and land surface erosion. A more aggressive timber
of activity prior to intense human disturbance of the harvest strategy would accept claims of the effective
landscape during earlier logging eras. This makes it ness of all mitigation measures as outlined in the Tim
difficult to accurately assess the impact due to current ber Harvest Plans (THP) and HCP and allow higher
land management and mitigation practices. In recent rates of harvest, while also acknowledging that timber
years a serious effort has been made to initiate more harvest may negatively impact this unstable geologic
extensive mitigation practices and make use of new setting with unpredictable climate.
and improved road construction, road maintenance,
and logging methods. The impact of these practices, Evaluating Science through Peer Review
however, has not been monitored long enough to For at least 200 years, scientists have been called upon
demonstrate their effectiveness. While the long-term to review the works of colleagues before the results of
benefits have been estimated the short-term conse scientific studies are published in the technical litera
quences have not been adequately quantified. For ture. The system, known as peer review, is still consid
example, improving roads for all-weather use by ered a linchpin of scientific investigation (Altman
installation of adequate culverts and gravel surfaces 2002). Peer review is designed to identify weaknesses
can have short-term negative effects such as the gen in study design, methodological errors in data collec
eration of sediment. Likewise, the long-term benefits tion and analysis, possible biases by scientific investi
of decommissioning roads by removing fill at stream gators, claims by authors that cannot be supported by
crossings and seeding grass or planting trees will pro the evidence they present, and other possible prob
duce additional sediment in the short-term. Therefore, lems. Although peer review is considered to be an
monitoring offers a rich opportunity to evaluate the imperfect process by some scientists (see e.g., Rus
effectiveness new management practices. tum 1993, Garfield 1993), it remains the foundation for
The spatial and temporal variability of precipitation evaluating the validity of scientific findings before the
intensity, duration, and frequency make it difficult to works are published or used in formulation of policy
assess risks and evaluate the effectiveness of mitiga by regulatory agencies.
tion efforts. A major storm event may occur the year Peer review follows a generally accepted process in
immediately following harvest or restoration, or there which authors submit manuscripts to journals whose
may not be one for 15 years. Therefore, it may take 15 editors send the most promising ones to other experts
years to realize the benefits of preventative actions (peers) in the appropriate disciplines to solicit their
taken today. Alternatively, restorative actions taken advice on the scientific merits of the works. The peers
today could be destroyed by a major flood event next then check the legitimacy of all aspects of the work and
year. Likewise, the occurrence of major earthquakes make suggestions that the editors subsequently use in
that further destabilize an already unstable landscape, deciding whether to publish or reject a manuscript.
Peer reviews are generally conducted in one of Clearly and simply communicate key science
three formats: open, blind, and double blind. In the findings to all participants.
open format, the author(s) of the paper and the peer Evaluate whether of not the final policy
reviewers are each aware of the others identity. In the decisions are consistent with the science
the authors, but not vice versa. In a double blind Avoid advocacy of any particular policy
wood production. Thinning and delaying the final Caspar Creeks, Lewis (1998) and Lewis et al. (2001)
clearcut harvest in the highly productive redwood for showed that timber harvesting during the 1970s
est type should not markedly reduce the quantity or increased the annual suspended sediment load at the
quality of wood produced. From a water quality stand South Fork weir by 212 percent. Subsequent data from
point spreading the cut among different watersheds so individual drainages within the North Fork watershed,
that cutting is not concentrated within one watershed which were logged under conditions more representa
over a short time will help reduce peak flows and sedi tive of the California Forest Practices Rules of the
mentation. 1990s, showed an 89 percent increase in summed
storm loads relative to that predicted for undisturbed
Mechanisms for Sediment Delivery conditions, most of which was due to a single land
The rate at which sediment is produced and removed slide. Data from all but one of the watersheds within
from a watershed is controlled by three factors (e.g., the North Fork drainage suggested an increase in sus
Knighton, 1998): the rate at which erodible material is pended sediment load following timber harvest, but
produced by physical and chemical weathering of downstream effects appeared to be small or absent
bedrock, the rate at which sediment is transported because the sediment had not yet reached the down
downslope by mass wasting or slope wash processes, stream measurement stations. Lewis (1998) and Lewis
and the rate at which sediment is removed from valley et al. (2001) suggested that much of the suspended
bottoms by fluvial processes. Wet and tectonically sediment increase attributed to timber harvesting was
active areas underlain by weak rocks will produce related to an observed increase in peak flow volume in
large amounts of sediment even in the complete the years immediately following timber harvest, which
absence of human activity. Reid (1998b), for example, they expected to be short-lived. They further conclud
used climatic records to demonstrate that a five-fold ed that the effects of multiple disturbances within a
increase in debris flows in a northern California study watershed were approximately additive. Finally, they
area could be attributed to an increased frequency of concluded that sediment increases could have been
high-intensity storms during the late 1930s. Therefore, reduced by avoiding activities that degrade the banks
it is not a simple matter to distinguish the amount of of small channels and that sediment loads are affected
sediment that would have been produced by natural as much by channel conditions as sediment delivery
processes, and particularly extreme natural events, from adjacent hillsides.
from that produced as the result of land management Other studies have examined the relationship
activities such as timber harvesting in the watersheds between timber harvesting and landsliding (or other
of interest to the Panel. forms of mass wasting) and found a positive correlation
An abundance of published scientific literature between the two (e.g., Furbish and Rice, 1983; Sidle et
points to the role of timber harvest on increasing sedi al., 1985; Pyles et al., 1998; Montgomery et al., 2000;
ment production rates. Lewis (1998) suggested that Sidle and Wu, 2001). Other studies (e.g., Brardinoni and
the relationship between timber harvesting, sediment Slaymaker, 2001) suggest that modern logging prac
production, and sediment delivery is a complicated tices produce no detectable increase in landsliding. The
web that includes many feedback loops. Important increase in landsliding is generally attributed to a com
factors include timber felling (which can increase soil bination of increased soil moisture and reduced root
moisture, decrease tree root strength, increase the strength. Although soil moisture increases probably
likelihood of blowdown, and increase channel rough play an insignificant role in wet season landsliding,
ness through the addition of woody debris), yarding because the susceptible slopes are already saturated or
and skidding (which can compact soils, decrease infil nearly so, it may increase the temporal window during
tration rates, and increase runoff), road and landing which slopes are susceptible to sliding (Sidle and Wu,
construction (which can increase runoff as well as 2001). Most studies of timber harvesting and landslid
increase the likelihood of landsliding), burning (which ing have concentrated on clearcut harvesting, but data
can increase erodibility by creating bare ground and, collected in Humboldt County on behalf of PALCO sug
in some cases, create hydrophobic soils that increase gest that selective harvesting can also increase land
runoff), and site factors (such as geologic predisposi slide activity (Pacific Watershed Associates, 1998a).
tion to landsliding regardless of land management). OConnor (2002) developed a sediment budget for
In a paired watershed study of North and South Freshwater Creek based on a combination of computer
models and field observations. Sediment sources for ber harvesting as they relate to the NCRWQCBs mis
the years 1988 through 1997 included streambank sion. Watershed analysis is conducted on a scale
slides, soil creep, bank erosion, shallow landslides, appropriate for the assessment of cumulative impacts,
deep-seated landslides, scour of tractor fill, harvest sur which stands in contrast to the ownership-based SYP
face erosion, road-related landslides, and road surface or narrowly focused THPs.
erosion. He further attributed various fractions of each From the Panels perspective, watershed analyses
of these categories to background processes that would would need to have the following attributes to be an
have acted even in the absence of human activity, lega effective tool. Watershed analyses would at minimum:
cy sources that exist as the result of prior logging prac 1) be completed prior to the approval of THPs, 2) con
tices, and management-related sources that contribute sider the rate of disturbance, 3) consider the propor
sediment even under modern forest practice rules. tion of the watershed to be disturbed in a given time
period, and 4) consider the location of current and
Adequacy of the Existing future THPs relative to slope, geology, landslide risk,
THP-SYP-HCP Framework and related site conditions. Further, because the analy
The Panel considered each of the existing regulatory ses will require the application of scientific and techni
processes to determine if the existing framework could cal judgement under conditions of wide uncertainty,
provide a means for ensuring recovery of beneficial they should be prepared by an independent, third
uses. The first regulatory process considered by the party. The methodology, assumptions, and conclu
Panel was THPs. However, because each THP is evalu sions will need to be subjected to rigorous and inde
ated on an individual basis, it is not possible to assess pendent scientific review.
the combined impact on water quality and thus THPs This is particularly so if the analyses are to incorpo
for areas within impaired watersheds continue to be rate innovative but untested concepts such as the
approved. The cumulative effects considerations in the PALCO Mass Wasting Areas of Concern (MWAC) or
THP do not include a firm projection of additional near- other computer models to delineate areas of potential
term harvest plans, nor do they appear to offer assur instability. Watershed analyses conducted in a manner
ances as to the proportion of a watershed to be that respects the disparate time scales upon which
disturbed within a given time period such as the next geologic processes and humans work would also be
five or ten years. Further, assumptions in the THP that beneficial. The establishment of background rates, as
mitigations outweigh impacts leads to a conclusion discussed below, that take into account the magnitude
that there need be no limit on rate of disturbance or the and periodicity of geologic events that contribute sedi
total proportion of disturbance of any given watershed. ment regardless of any land management activities
The Panel then considered the SYP. One of the main would render the analyses more robust.
goals is to maximize sustained timber production by
ensuring that more timber is produced than harvested Assessment of Sediment Credit/Debit Analyses
(California Department of Forestry, 2002). The SYP The following discussion concerns the use of sedi
does rely on environmental documentation that con ment credits and debits. Sediment credits refer to
siders potential impacts to wildlife habitat and water future, planned reductions in sediment from mitiga
quality. However, because the SYP is based on owner tion measures, such as storm-proofing roads and
ship, it does not provide a means to assess the cumu removing Humboldt crossings. The notion is that
lative effects of timber harvest in any one watershed. reductions in sediment be credited against sediment
The HCP focuses on wildlife habitat conservation, in increases (debts) from other timber harvest activities.
which water quality per se is by definition a secondary PALCO and CDF argue that more harvesting will
concern, and does not include an integral science- lead to more recovery because their calculations show
based monitoring component that fully addresses that the credits associated with mitigation exceed the
water quality. From our limited review it appears that debits associated with timber harvest. (Jeff Barrett,
the THP-SYP-HCP structure lack some of the key ele personal communication, 10/9/02; John Munn, per
ments needed to move toward and assure attainment sonal communication, 10/10/02). However, mitigation
of water quality standards. activities such as the storm proofing, decommission
The Panel supports the general concept of water ing, and proper abandonment of roads will, despite
shed analysis to assess the cumulative impacts of tim their probable long-term benefits, almost certainly
have short-term negative impacts. A study of road be expected to reduce long-term sediment production
removal in Redwood National Park (Madej, 2001) esti by a factor of not much more than four or five and that
mated that each mitigated stream crossing produced the sediment savings realized from the proper aban
50 m3 of sediment, which is about one-fifth of that donment of roads is likely to be less.
which might have been produced had the mitigation Given the scarcity of scientifically robust studies that
not occurred. Unfortunately, short-term impacts such might reduce or constrain the uncertainty associated
as these have not been included in the debit side of with potentially effective but untested mitigation tech
PALCO and CDF estimations. Furthermore, although niques, one option would be for PALCO to test promis
mitigation activities may reduce the impact from pre ing methods and validate their effectiveness through
viously improperly abandoned harvest areas the Panel the use of carefully designed and executed monitoring
questions the concept of whether these mitigations programs. Paired watershed studies could shed con
should be credits relative to a natural background rate siderable light on the effectiveness of road treatments
of sediment production. and removal of Humboldt crossings. One approach to
The Panel also questions the veracity of the extreme be considered is to limit harvest rates until the effec
ly high estimates of credits associated with mitigation tiveness of the promised mitigation approaches has
apparently calculated with worst-case assumptions. been validated through such mitigation reliability per
The Panel made efforts to seek out measurement- formance monitoring. The alternative approach of per
based studies that might support those high estimates, mitting accelerated timber harvest rates based on the
but was unable to find sufficient evidence in support of untested promise of mitigation strategies will, in the
the values suggested by PALCO (cf., Madej, 2001; opinion of the Panel, lead to a significantly uncertain
McCashion and Rice, 1983). Madej (2001) wrote that outcome in terms of downstream sedimentation and
decommissioned stream crossings in Redwood water quality standards being met.
National Park reduced sediment production by at least
a factor of four, and possibly more, compared to that Background Rates of Sediment Input
which might have been expected from untreated cross Assessing the impacts of human activities by compar
ings. Completely decommissioning roads in the park ing water quality, sediment production, or any other
reduced sediment production by a factor of three to ten environmental variable to a naturally occurring back
compared to sediment production from untreated ground rate is an approach that seems intuitively
roads in an adjacent watershed. The decommissioned attractive and logical. Background rates, however, are
roads, however, still produced an average of 480 m3 of hard to define. Sediment production (and, conse
sediment per kilometer of road. Similarly, McCashion quently, turbidity) is a function of episodic tectonic
and Rice (1983) concluded that only about one-quarter uplift; bedrock type, including the influence of fractur
of the erosion from forest roads in northwestern Cali ing and folding; climate and weather, particularly pre
fornia can be mitigated using conventional engineer cipitation; fire frequency; and human activities. Thus,
ing methods. The remaining three-quarters were sediment production and turbidity can be expected to
attributed to site conditions and alignment choices that vary considerably in space (i.e., among watersheds
could not be changed. Finally, the Panel notes that and within watersheds as a function of geology and
some PALCO estimates of mitigation effectiveness are topography) and time regardless of human activities.
based on worst-case scenarios; for example, that every Background conditions are also likely to be episodic,
mitigated channel crossing would have completely with long periods of relative quiescencelasting
failed and contributed a large amount of sediment to decades, centuries, or even millenniapunctuated by
the watershed (PALCO, 2002). A more likely situation is events of short duration but large magnitude. The fur
that only some of the crossings would fail, that most of ther back one looks, moreover, the more uncertain the
the failures would not have been complete (Best et al., geologic record becomes and the more difficult it is to
1995; Madej, 2001), and that some crossing which infer anything more than gross averages that fail to
failed would not contribute the maximum amount of capture the episodic nature of watershed change. The
sediment. Thus, a 100% failure rate assumption will result is that any rate or formula will ultimately be a
probably lead to overestimates of the amount of sedi product of professional and scientific judgment, made
ment that is actually saved. It is therefore the Panels within a context of high variability and will not simply
opinion that completely decommissioned roads should emerge from the data.
It became apparent during the Panels work that the predict outputs, given inputs. In this case, inputs may
relevant working definition of background includes include the area of forested land that is harvested or
most of the 20th Century, a time during which Hum the rate of cut, for example, while outputs may include
boldt County watersheds were being heavily logged. water yield or sediment yield. There are two general
Scientists do not know how late 20th century sediment approaches to numerical modeling, and the models
production in Humboldt County watersheds compares described below fall into both categories.
to peaks in the geologic record, but the Panel is confi Empirical models represent environmental systems
dent that it is above the baseline levels that existed in terms of relatively simple regression-derived equa
between rare sediment-producing events before the tions that relate inputs to outputs. They are not distrib
advent of commercial logging. The NCRWQCB must uted, meaning that they produce output that carries a
decide whether this is an appropriate definition of back value for the whole system (e.g. sediment yield for a
ground. More work is needed to document the histori whole watershed). The equations contain constants
cal water quality for the five streams at the time the and coefficients that are not directly related to physical
standards were adopted and subsequent trends for ly measurable properties of the system. When devel
each stream. Since the entire watershed has been dis oping an empirical model the emphasis is usually
turbed at various times in the last 150 years, one option placed on obtaining the best possible fit between pre
would be to establish background levels with respect to dicted and observed output. Much of the complexity of
nearby minimally disturbed or reference watersheds. the environmental system is concealed within the
One promising approach to developing water quali black box of the empirical relationships, which means
ty standards in such complex landscapes would rely that empirical models are less used in research envi
on a relationship between suspended sediment (or ronments where the primary focus is on understanding
turbidity) and stream discharge for less-disturbed the internal processes and their interaction, rather than
watersheds that are otherwise similar to the five on getting exactly the right answer. Because of their
watersheds of concern. At the same time, suspended simplicity and the fact that internal processes are not
sediment-discharge rating curves could be developed directly simulated, empirical models have relatively
for Freshwater, Elk, Stitz, Bear, and Jordan water low data requirements and are easy to apply.
sheds. Paired watershed analyses of shifts in the rat PBDMs usually represent environmental systems via
ing curves would help agencies separate significant differential equations where all the terms in the equa
increases of sediment due to timber harvest from tions are measurable physical quantities (e.g. rainfall,
those that would have occurred without human activi soil erodibility, etc.). Their inputs and outputs are spa
ties. Residents in the Freshwater Creek watershed tially distributed, meaning, for example, that soil erodi
have been collecting suitable data for just this type of bility must be calculated for all different soils in the
analyses and PALCO has indicated a desire to begin watershed and separate values applied for all sub-
similar water quality monitoring efforts. The Panel watersheds. Also, sediment yield is calculated as a spa
underscores the need to develop separate sediment tially varying amount in the sub-watersheds. When
rating curves for each of the five watersheds. The Jor developing a PBDM, the emphasis is generally placed
dan, Bear, and Stitz watersheds are not occupied by on having the most scientifically rigorous representa
permanent residents but are more tectonically active tion of the main physical processes and their intercon
and steeper and have experienced a notably different nections. While accuracy of output is highly desired
history of timber harvest. and sought after, PBDMs are generally more valued for
accurately representing the internal workings of the
Approaches to Calculation environmental system than for producing the right
of Allowable Harvest Rates answer. Because of their complexity and the fact that
In order to clarify its discussion on specific approaches all equations require physically measurable parame
to calculation of allowable harvest rates, the Panel has ters, they have very high data requirements and are
prepared the following summary of general approach very unreliable where these data are not available.
es. The methods discussed below are models of water As a general rule, empirical models are used in
shed behavior and response to timber harvest. Such application environments where the emphasis is on
models seek to represent complex physical environ getting the most accurate answer with minimal data
mental systems as less complex numerical systems, to collection, while PBDMs are used in research envi-
ronments where large data collection efforts can be budgets such as those proposed by Dr. OConnor for
justified because of the insight into the internal behav the Freshwater watershed require large amounts of
ior of the environmental system that is gained by run data and calibration relative to other potentially useful
ning such a model. While PBDMs are conceptually approaches. The WEPP model, for example, relies on
more realistic and have the potential to be more pow very detailed site characteristics for small watersheds
erful predictive models than empirical models, obtain that require careful field measurements not available
ing a more accurate answer from a PBDM is rare in from secondary sources. The models used by Dr.
complex environmental systems because there is OConnor on behalf of PALCO are also very sensitive
almost never sufficient data to parameterize the to the models used to calculate sediment transport
model. In most settings, empirical models are more rate and methods used to estimate the residence time
likely to be accurate. of sediment stored in the channel. Predictions based
The peak stream discharge approach and the Empir purely on modeled results are subject to one to two
ical Sediment Budget approach described below are orders of magnitude variation in estimated sediment
examples of empirical models. The Modeled Sedi transport rates depending on the rates of sediment
ment Budget approach incorporates a large number of transport and storage that are assumed.
both empirical and physically based distributed mod The large number of categories (management
els. As an overall approach, it has many of the charac sources, legacy sources, and background sources) and
teristics of a PBDM. scarcity of empirical data required Dr. OConnor to
estimate the sediment input for eleven types of
PEAK STREAM DISCHARGE APPROACH sources (soil creep, bank erosion and small stream
The approach to allowable harvest rate calculation side landslides, deep seated landslides, shallow land
articulated by Mr. Munn of CDF is based primarily on a slides in harvest units, surface erosion of landslides,
consideration of the influence of proposed timber har surface erosion in harvest units, erosion of tractor-
vest activities on peak stream discharge (Munn, filled channels, erosion of low order valley fill, road-
1/14/2002). It does not take into account sediment pro related shallow landslides, surface erosion of roads,
duction or changes in the sediment transport capacity gullies and culvert-fill failures) using at least five dif
of channels that might result from harvest. During the ferent methods that range from field surveys to aerial
Panels discussion with him, Mr. Munn also indicated photograph inventories to computer models. It is
that the CDF approach to evaluating cumulative therefore very difficult to assess the degree of reliabili
impacts is designed to maintain the current level of ty or uncertainty associated with each sediment
impact rather than promote the recovery of impaired source estimate. The Panel questions, for example,
watersheds. As such, this approach, administered by whether soil creep truly accounts for an order of mag
CDF, yields a high risk that current harvest rates will nitude more sediment than harvest unit surface ero
not achieve recovery of beneficial uses of water in the sion and whether it is truly a background process.
impaired water bodies. Likewise, it is difficult for the Panel to understand the
logic behind classifying 40% of shallow landslides in
MODELED SEDIMENT BUDGET APPROACH harvest units as naturally occurring events. This is not
The approach taken by Dr. OConnor on behalf of to say that any given landslide would or would not
PALCO was to develop a comprehensive sediment have occurred in the absence of harvest, or that one
budget (OConnor, 2002). The Panel reasons that the could not have made a reasonably good estimate
idea of developing a comprehensive sediment budget given enough time and money, but rather that it is
is in principle a good one that should be pursued over impossible to evaluate the veracity of such claims
the long-term. Moreover, the sediment input cate from the information provided to us. The Panel also
gories used by Dr. OConnor on behalf of PALCO notes that soil creep and deep-seated landslides were
appear to be useful with regard to the effects of timber interpreted as background and indeterminate sedi
harvesting on water quality (OConnor, 2002). Howev ment sources, respectively. One can easily argue that
er, the reliance on complex models based on limited either might have been accelerated by timber harvest
data calls into question some of the findings, especial ing, because it is physically plausible to infer that ups
ly when these findings disagree with data based on lope harvesting increases the amount of water
empirical sediment budgets. Model-based sediment infiltrating into the slope. Therefore, one cannot infer,
in the absence of supporting data, that deep landslides impact relative to background less than a threshold
are never triggered by harvest-related practices. ratio (Reid used 1.2 which amounts to a 20% increase
Finally, the Panel notes that if Dr. OConnors sedi in sediment based on the NCRWQCB Basin Plan policy
ment budget is accepted at face value, it shows that that turbidity shall not be increased more than 20%
management-related sources produced 51%* more above background levels). Important assumptions in
sediment than background sources between 1988 and this Empirical Sediment Budget approach as applied
1997. This figure increases to 70% if legacy sources to Bear Creek are:
are added to the effects of recent timber harvesting, as
they probably should, and even higher if some of the A fixed recovery period exists, denoted n,
arbitrarily assigned background or indeterminate sed during which sediment production is at the
iment sources are actually management-related. enhanced rate quantified by rate factor L.
Although these percentages cannot be directly con Following this period of n years, sediment
verted into an estimate of excess turbidity, they clearly production reverts back to the background rate.
imply that modern timber harvest practices imple Reid used n = 15 years because that had been
mented by PALCO still resulted in significantly used in studies by Pacific Watershed Associates
increased sediment production in the Freshwater (Pacific Watershed Associates, 1998a, 1998b,
Creek watershed in the last few decades. 1999a, 1999b) in categorizing watersheds as
The Panel concluded that, although the Modeled harvested or not, even though she believed that
Sediment Budget method has promise over the long- 15 years was an underestimate of the true
term, major problems exist that are inherent in the use recovery time. If it is true that the recovery time
of a complicated sediment budget approach to estab is substantially greater than 15 years, this
lish allowable timber harvest rates. For example, assumption will tend to overestimate the
many qualitative judgments about sources of sedi background landslide rate and produce higher
ment are poorly supported by quantitative studies. allowable timber harvest rates.
Also, the quality and quantity of empirical data that
are available or likely to become available over the The sediment production is proportional to
short-term, particularly regarding the efficacy of pro landslide volume observed on aerial
posed mitigation techniques, are key limiting factors. photographs. This assumption will tend to
overestimate allowable harvest rates because
EMPIRICAL SEDIMENT BUDGET APPROACH landslides are not the only process for increased
The approach suggested by Dr. Reid of the Redwood sediment production due to timber harvest.
Sciences Laboratory (Reid, 1998a & Reid, 2000), is also Increased sediment production due to harvest-
a sediment budget approach, but it is empirically related sources other than landslides visible on
based on observed differences between harvested aerial photographs has not been accounted for.
and non-harvested areas. Each watershed is divided Any overestimate is at least partially offset by an
into a series of land classes. Reid initially used two underestimate induced by interpretation of aerial
classes: harvested and non-harvested. Each land class photography. Aerial photograph inventories are
has an assumed background sediment production rate known to underestimate the number of old
(tons/yr/mi2). An empirical rate factor then quantifies landslides in heavily forested terrain, thereby
the effect of harvest on sediment input. The rate factor overestimating the influence of recent timber
is denoted generally as L hereafter (as used by OCon- harvest activities and, as a consequence,
nor, Reid used the specific number 9.6). Based on the underestimating the allowable harvest rate
fraction of an area subject to harvest and fraction of (Pyles, 2000; Brardinoni and Slaymaker, 2001).
sediment (as inferred from aerial photograph land
slide volume estimates) from harvested areas, L can Any increase in turbidity is equal to the
and climate, the relative ratios for different land slide-derived sediment delivery (see Appendix C) and
classes are constants independent of weather found that the proportion of sediment from high haz
and climate. Stated another way, the rate factor, ard areas is clearly related to the proportional area of
L, that quantifies relative increased sediment the high hazard zones. The allowable n year harvest
production from harvested areas is constant. proportion can be expressed equivalently as
Reids strategy means that the total volume of
sediment from both classes may be expected to
(SRT - 1)[(R1 / R2)ah + 1 - ah]
NT = (1)
increase during wet years, but a recently (L - 1)(1 - ah)
harvested area will contribute L, times as much or
sediment per unit area as a completely healed
SRT - 1
area - regardless of weather and climate. This NT = (2)
assumption and the use of relative sediment (L - 1)(1 - fh)
input rates counters a principle criticism that Dr. in which SRT is the allowable threshold sediment yield
Reids estimates of increased landsliding on ratio (SRT = 1.2 in Reids calculations), L is the sediment
recently harvested land are biased because they production rate factor for recently harvested areas (L =
are based upon data from unusually wet years. 9.6 for Bear Creek in Reids calculations), and fh is the
proportion of sediment supplied by high hazard areas
A second, slightly more complicated, analysis for the (0.9 for Bear Creek in Reids calculations). R1 and R2
North Fork Elk River was expanded to include the time- are, respectively, the background sediment production
dependent effects of non-landslide sediment delivery rates per unit time for the high-hazard areas and the
in terms of annual proportion of canopy removal. This lower hazard harvestable areas. ah is the proportion of
extension used land class rate factors adapted from the the watershed classified as high hazard and declared
results of post-harvest sediment production from the unavailable for harvest. NT must be interpreted care
Caspar Creek experimental watershed. fully because it is the proportion that can be harvested
Dr. Reid kept the Empirical Sediment Budget approach in an n year period of the proportion of watershed
simple by using three land classes: available for harvest. The proportion of the total
watershed area that can be harvested in the same n
High hazard areas that will never be harvested year period without exceeding the threshold sediment
but, even without harvesting may produce production ratio is NT (1 - ah). Furthermore, NT must be
sediment at a rate above the background rate divided by n in order to calculate the single year allow
of lower hazard areas. able harvest rate.
Equations (1) and (2) are, as demonstrated in
Average to low hazard areas that were Appendix C, related to each other by the equation
harvested more than n years ago, are
ah(R1 / R2)
completely healed, and produce sediment at fh = (3)
background rates. 1 - ah + ah (R1 / R2)
Equation (3) shows that fh is a non-unique quantity
Average to low hazard areas that were
that is controlled by a combination of the high-hazard
harvested less than n years ago, are not
area withheld from harvest and the ratio of background
completely healed, and therefore produce
sediment production rates for high-hazard areas and
sediment at rates above background.
harvestable areas. Thus, there are an infinite number of
combinations of R1, R2, and ah that can give rise to any
One of the criticisms of Dr. Reids approach to estimat particular value of fh. Equation (3) reduces to fh = ah for
ing landslide-derived sediment has been that it does the special case of R1 = R2. The Panel does, however,
not explicitly account for the proportion of the water note that Dr. Reid probably overestimated the value of L
shed that is declared unavailable for harvest because (9.6 for Bear Creek and 13.0 for North Fork Elk River) by
of high landslide hazards (Opalach 1998). Instead, it including landslides resulting from timber harvest in low
specifies the proportion of the total watershed sedi to average hazard areas as well as high hazard areas in
ment production that comes from high hazard areas. which future logging presumably would not occur. In the
The Panel has re-derived Reids equations for land- studies (Pacific Watershed Associates, 1998a, 1998b,
1999a, 1999b) that Dr. Reid based her calculations on, tainty in the initial rate factors and consequences in
the sediment production from low and high hazard terms of downstream sedimentation and water quali
areas that may be specifically excluded from harvest is ty, and economic impacts on the land owners (both
not separated out when the volume of sediment from timber and downstream residents).
harvested areas is reported. Therefore, the rate factor L The Panel suggests that the following steps be taken
that Reid used is an aggregate from both low and high to refine the Empirical Sediment Budget approach to
hazard areas. To calculate an allowable harvest on low the point where it can be used by the NCRWQCB, other
hazard areas only using equations (1) or (2) requires that regulatory agencies, PALCO and other timber harvest
L be estimated for the low hazard areas. In Appendix C entities to calculate timber harvest rates that will not
we explored the sensitivity to L and found that unless impede the recovery of impaired watersheds.
the logging rate increase factor L is reduced consider
ably from Reids estimates of 9.6 or 13 down to values First, conduct exploratory data search to deter
approaching 4 that the proportion of area available for mine the availability, preferably in digital format
harvest per year is not significantly increased. amenable to GIS analysis, of geologic maps,
Overall, the Panel found the Empirical Sediment Bud landslide hazard maps, landslide inventories, aer
get approach to be fundamentally sound and at a level ial photographs, and topographic maps.
of detail commensurate with the kinds and amounts of
data that are available, or can be made available, in the Second, develop families of land class rate factors
near future. The field-based land class rate factors nec according to bedrock geology, geomorphology
essary to use this approach can be estimated from (e.g., inner gorge, swale, planar-convex slopes),
existing aerial photographs, publicly available geologic and harvest methods (e.g., in terms of canopy
and geologic hazard maps, and published studies of removal percentages) for each of the five water
watersheds such as Caspar Creek. The use of sediment sheds. Sediment production associated with roads
production ratios, rather than the absolute rates, allevi should, if there are sufficient data, be considered
ates much of the difficulty associated with background as a separate category. The Panel strongly recom
rate estimation because it is generally easier to esti mends that publicly available information be used
mate one ratio of rates than two independent rates. to develop land class rate factors. In cases where
The Empirical Sediment Budget approach is suit the necessary information is the subjective and
able for use with an adaptive performance-based interpretive product of professional judgement, for
allowable timber harvest calculation. The calculations example landslide hazard maps, the Panel further
that Reid performed have been criticized for reliance recommends that the NCRWQCB give strong pref
on data from locations claimed not representative of erence to documents that have undergone rigor
the watersheds at hand. However, although these criti ous and independent peer review. It would be
cisms have merit, in many cases this was the best data instructive to compare the land class rate factors
available. PALCO consultants have justified THP generated using alternative hazard zonation
requests based on model estimates of sediment pro schemes such as the PALCO mass wasting avoid
duction. Models of sediment production, although ance strategy or other computer models. The first
state of the art, have considerable uncertainty. Fur generation of estimates, however, should be
thermore, the claims of effectiveness of mitigation based on publicly available and peer-reviewed
measures are also uncertain and untested. A compre maps rather than potentially useful but as yet
hensive monitoring program in the watersheds being untested sources.
harvested could be used together with the Empirical
Sediment Budget approach developed by Reid and Third, review the first generation of land classes
expanded in Appendix C to adapt allowable harvest and their rate factors to determine what addition
rates in each watershed as the monitoring of sediment al information can be used to refine the classes,
yields leads to refined estimates of the rate factors consolidating or expanding them as appropriate.
involved. This would address criticisms of both
approaches by using data from the actual watershed Fourth, calculations of initial allowable harvest rates
being harvested. Policy makers setting initial allow be rigorously and independently peer reviewed.
able harvest rates need to weigh together the uncer
Fifth, permitted timber harvests be subject to either a reduction in sediment supply or an increase in
careful performance monitoring to measure and stream transport capacity or storage capacity. Increas
refine the reliability of mitigation measures and ing storage capacity by dredging is a short-term solu
rate factors that have been used in the allowable tion that will likely be countered by subsequent
harvest rate calculations. delivery of sediment. A longer-term solution would be
to reduce the sediment supply from the upper water
Sixth, adapt and refine allowable harvest rates shed. In addition to modifying the sediment system,
(upwards or downwards) based on the outcomes roads, bridges and structures in the flood prone area
from performance monitoring measurements. may be raised to increase the river stage at which they
can be used without inundation.
These suggestions are targeted at setting initial allow
able harvest rates as well as longer-term allowable Moderate Increase in Surface
harvest rates based on performance monitoring. Water Runoff/ Patterns of Peak Flows
Regarding peak flows and the pattern of peak flow, the
scientific literature converges on two points (e.g.,
QUESTION 2 Anderson et al., 1976; Satterlund, 1972; Brooks et al.,
What options are available (e.g. dredging, and 1991; Reid, 1993; Luce, 1995; Ziemer, 1998; Wohl, 2000):
modification of activities resulting in, or reducing,
sediment delivery) that can be immediately The effect of timber harvest on the frequency
implemented and will be effective in lessening of smaller peak flows depends on the
the adverse flooding conditions and impacts to sequence of storm events. If clear-cut
beneficial uses? Please discuss the potential ben watersheds experience a sequence of rain
efits, limitations, and tradeoffs of these options events one after another, with little time in
for each watershed. between for soil moisture to decline, then the
effect of timber harvest (including tree
removal and roads) will be to increase the
The Panel has identified two fundamental processes frequency of lower magnitude peak flows.
contributing to flooding problems and impacts to ben The effect of timber harvest on larger peak
eficial use of water: (1) a large increase in suspended flows is more tied to the effects of roads and
sediment yield, and (2) a moderate increase in surface compacted skid trails than to vegetation
in proportion to the area that was clearcut recently. In ing mitigation options by deciding on a weighting for
addition, impervious surfaces such as roads or com the attributes, and multiplying by the benefit or cost to
pacted skid trails and landings shed water much faster determine a ranking of options. Depending on which
than fully vegetated areas. attribute or group of attributes is given the greatest
Some research analysts have argued that peak runoff weight, different options will appear more or less
from clearcuts during high magnitude rainfall events is attractive. For example, in the option and attribute
not affected by forest removal, because the size of the matrix below (Table 3) selection of speed of benefitas
lost interception store is small relative to the total the most heavily weighted attribute will probably result
amount of water delivered in a large event. Thus, some in dredging and raising infrastructure having the high
hydrologists have dismissed the significance of inter est score on the matrix, whereas if long-term benefit to
ception, or its loss when the vegetation is removed by water quality is the most important attribute assessed
clearcutting (e.g., Patric, 1999). PALCO (1999) points out then options that reduce the sediment yield will have
that the Elk River and Freshwater Creek watersheds are the highest score.
not being widely clearcut over short periods of time, as Identification of which of those attributes should
was the case for the often cited Caspar Creek experi carry the most weight is a policy decision that requires
mental study. Also, PALCO notes that steep terrain is value judgments from the participants in the planning
not being tractor-logged and clearcuts are being process.
replanted, two steps that would quickly counter any In considering possible short-term options to lessen
increase in water and/or sediment production. the severity of flooding and impacts to beneficial uses,
In a watershed where a high proportion of the area the Panel focused on five attributes: effectiveness (as
had been clearcut recently and where the road surface defined by the amount by which flooding and turbidi
area including skid trails is substantial, the Panel ty is reduced), implementation speed (as defined by
believes there is a strong possibility that peak flows the time needed to achieve improvement), impact (as
from early fall storms or from smaller storms would defined by the degree to which options cause short-
be increased to the point where flood frequency and term damage to water quality or habitat during imple
magnitude was increased. In addition to canopy inter mentation), initial cost, and recurrent cost.
ception, forest cover significantly lowers soil moisture Based on a review of literature, meetings with
through transpiration. Forest removal or thinning will stakeholders, site visits, and discussions among its
significantly increase soil moisture levels, potentially members, the Panel identified three categories of
contributing to faster and larger amounts of runoff, options (in no particular order) with a variety of specif
and increasing pore-water pressures that are a trigger ic options related to each: Increase Channel Transport
in the activation of some landslides. Capacity, Reduce Sediment and Water Supply From
the Watershed, and Other Measures.
Evaluation of Options to Address
Flooding and Impacts to Beneficial Uses Increase Channel Transport Capacity
As outlined in the terms of reference and noted on DREDGING
page 11 (the relationship between science and policy), The most immediate (short-term) option is to dredge
the panel does not believe its role is to advocate spe the channel in Freshwater and Elk creeks from the
cific options. We have outlined a number of potential point where flooding begins to the sea. The principal
options and sought to objectively evaluate them as advantage of this approach is that it would immediate
positive or detrimental in terms of different attributes. ly lower flood levels for a given flow. In addition it
We have not sought to add more detailed scores or would increase water and sediment transport capacity,
values to the options as this would require additional and could jump-start ecological recovery of the chan
investigation beyond the scope of the Terms of Refer nel by removing fines and exposing spawning gravel.
ence, and individual values would vary from water However, dredging has many disadvantages and is
shed to watershed. However, we have identified the an option that would fly in the face of current recom
actions that would need to be taken in order to calcu mendations for watershed management and restora
late these values. tion. Principally, it would be treating the symptom
Once these values have been determined, the result rather than the cause of the problem, and would entail
ing framework can provide a methodology for evaluat a large amount of medium term damage to the channel
and riparian corridor, which would have a major nega IMMEDIATE REPLANTING OF CLEARCUTS AND
tive impact on fish and other wildlife. Unless sediment SEEDING OF HIGHLY DISTURBED AREAS INCLUDING RIPPED
supply is reduced at the same time, dredging will only LANDINGS, SKID TRAILS, AND DECOMMISSIONED ROADS
be a short-term benefit, requiring repeated intervention The fifth option intended to improve erosion control is
and consequently high damage and costs. Benefits efficient revegetation of areas disturbed by logging or
could be nullified in short order by one or a few major road decommissioning. The Panel did not receive suf
storms or debris flow events. Thus, the Panel sees this ficient information on the replanting of clearcuts and
a high-risk strategy that would require further detailed seeding of disturbed areas to determine the potential
study and cost benefit analysis before acceptance. of this option. Replanting of clearcuts with tree
seedlings during the first post-cutting opportunity
Reduce Sediment and (within one year) helps reduce the area needing herbi
Water Supply From the Watershed cide treatment and allows the natural revegetation
DECREASE THE RATE OF TIMBER HARVEST process to go forward. Also, seeding grasses on
The second option is to decrease the rate of timber ripped landings and skid trails and on decommis
harvest until a threshold of recovery is reached. The sioned roads so that the grasses provide protection
anticipated benefits of the approach are that it would before the first big winter storm events has potential
reduce risk of increased peak flows and therefore to mitigate the large sediment loads when heavy rains
reduce risk of accelerated erosion from roads, harvest hit bare soil. PALCO has practiced both techniques
units, and landslides. The anticipated disadvantages with some success; however further fine-tuning of
are that its effects would not be immediate and that it these efforts may have potential to significantly
would be costly in terms of reduced timber harvest reduce the erosion potential. The anticipated disad
revenues and their associated economic impacts. vantages are that the fast-growing grass species may
spread to regeneration areas, require subsequent her
REDUCE TRACTOR AND SKIDDER YARDING bicide treatments, and depending on the type of vege
The third option is to reduce tractor and skidder yard tation used, interfere with timber regeneration.
ing by substitution of helicopter and cable harvest
methods, thus reducing ground compaction and ROAD DECOMMISSIONING/WEATHERPROOFING
reduced infiltration capacity. Both methods are now The sixth option is road decommissioning/weather-
being substituted frequently for ground-based logging proofing. In the Panels view, PALCO is currently doing
in current operations. The anticipated benefit of the an admirable job of maintaining, repairing and
approach is that it would reduce surface runoff, which decommissioning roads. Accelerating the rate of
is in turn anticipated to reduce the risk of accelerated decommissioning or proper abandonment, however,
erosion. The anticipated disadvantages are that its would more rapidly decrease the risk of erosion and
effects would not be immediate and that it would be sediment production in the affected watershed. The
costly in terms of reduced timber harvest revenues advantage of this method is that it addresses a very
and their associated economic impacts. significant sediment source that is accessible and
readily mitigated. The disadvantage is that there is
RIP PREVIOUSLY COMPACTED AREAS likely to be a short-term increase in sediment produc
The fourth option is to rip previously compacted areas, tion due to road reconstruction and the combined,
including skid roads with or without a concomitant immediate increase in sediment may be significant. In
decrease in harvest rate. The anticipated benefit of the addition, benefits may be slow to accrue even under
approach is that it would increase infiltration and an accelerated decommissioning program.
reduce surface runoff, which is in turn anticipated to
reduce risk of accelerated erosion. The positive impact STABILIZING LANDSLIDES
of this approach depends on the amount of compacted The seventh option is to stabilize landslides, particu
area available for ripping. The anticipated disadvantage larly shallow slope failures that enter watercourses.
is that it is likely to increase short-term erosion and sed Landslides are known to be a major sediment source
iment yield depending on on-site mitigation efforts. in the area and are currently not mitigated. Possibili
ties include:
Development of a program for prioritizing and basins would require frequent maintenance (sediment
stabilizing landslides as soon as they become excavation and disposal) and that land would have to
apparent. be acquired for their construction. There would also
have to be consideration of the sediment yield and
Toe armoring and biotechnical stabilization of provision for fish passage at low flows. Also, large
landslides (especially landslide toes adjacent capital investments in structures are vulnerable to the
to stream channels). next large storm/debris flow event.
A B C D E F G H I
Flooding Water Flooding Water Water Channel Riparian Initial Recurrent
METHODS Quality Quality Quality Habitat Habitat
3: Reduce Ground-Based
+ + + + + + + $ $
Logging
6: Road Weatherproofing + + + + $ $
9: Construct Sediment
+ + + + NONE + + $ $
Trapping Basins
channel habitat. A literature review could reveal typ channel habitat. A literature review might reveal typ
ical expected values and durations. ical expected values and durations.
G) Short-term impact/riparian habitatNo detrimental G) Short-term impact/riparian habitatNo detrimental
impact would be expected. impact would be expected.
H) Initial cost An estimate of the cost of road decom H) Initial cost An estimate of the cost of road weather
missioning should be obtained from a forest engineer. proofing could be obtained from a forest engineer.
I) Recurrent cost No recurrent costs are expected. I) Recurrent costAn estimate of the cost of road weath
erproofing could be obtained from a forest engineer.
6: ROAD WEATHERPROOFING
A) Speed of benefit/floodingThe benefit would be 7: LANDSLIDE TREATMENT
expected to start as soon as roads are weather A) Speed of benefit/floodingThe benefit would be
proofed, and to increase over time, as more roads expected to start as soon as landslide treatment
are disconnected from the stream channel network, commenced, and to increase over time, as more
reducing runoff and landslides. A literature review landslides are prevented and runoff and sediment
might reveal typical recovery rates. Runoff monitor yield from existing landslides is reduced. Calcula
ing from weatherproof and non-weatherproof road tion of speed of benefit would be difficult, and an
plots could be used to measure the effects over estimation would have to be made from a literature
time, as a basis for future planning. review. Field monitoring should be carried out to
B) Speed of benefit/water qualityThe benefit would provide data from future plans.
be expected to start as soon as roads are weather B) Speed of benefit/water qualityThe benefit would
proofed, and to increase over time, as more roads be expected to start as soon as landslide treatment
are weatherproofed, reducing runoff, erosion and commenced, and to increase over time, as more
sediment delivery. A literature review might reveal landslides are prevented and runoff and sediment
typical recovery rates. Sediment yield monitoring yield from existing landslides is reduced. Calcula
from field plots should be used to measure the tion of speed of benefit would be difficult, and an
effects over time, as a basis for future planning. estimate would have to be made from a literature
C) Long-term benefit/floodingThe literature and field review. Given the likely magnitude of costs, field
monitoring approach described in sections 6A and monitoring should be carried out to provide data
6B can be used to provide water and sediment for future plans.
inputs to a combination of models that can be used C) Long-term benefit/floodingAn inventory of land
to assess downstream flood impacts. One-dimen- slides could be carried out, and measurements of
sional hydraulic modeling coupled with sediment sediment and water delivery to the stream network
erosion and transport models (e.g. HEC-RAS com made. This information can be used to provide
bined with HEC-6) can be used to predict the rate of water and sediment inputs to a combination of mod
channel scour or deposition that would occur if sed els that can be used to assess downstream flood
iment and water yield from the watershed was impacts. One-dimensional hydraulic modeling cou
reduced. Given these, it would also be possible to pled with sediment erosion and transport models
estimate the resulting channel bed elevations and (e.g. HEC-RAS combined with HEC-6) can be used to
flood frequencies. predict the rate of channel scour or deposition that
D) Long-term benefit/water qualityData from 6A and would occur if sediment and water yield from the
6B may be used to predict the long-term benefit on watershed was reduced. Given these, it would also
water quality. be possible to estimate the resulting channel bed
E) Short-term impact/water quality Road weather elevations and flood frequencies.
proofing would be expected to produce a small D) Long-term benefit/water quality Data from 7C
short-term increase in sediment yield, increasing may be used to predict the long-term benefit on
turbidity. A literature review might reveal typical water quality.
expected values and durations. E) Short-term impact/water quality No detrimental
F) Short-term impact/channel habitat Road decom impact would be expected.
missioning would be expected to produce a small F) Short-term impact/channel habitat No detrimental
short-term increase in sediment yield, impacting impact would be expected.
10: RAISING FLOODPLAIN INFRASTRUCTURE the Panel suggests that PALCO develop and use a silvi
A) Speed of benefit/floodingThe benefit would be culture regime that uses thinning or partial cutting to
expected to start on completion of construction. maintain relatively long rotations.
B) Speed of benefit/water qualityNo benefit would
be expected as the work would not affect the sedi
ment yield from upstream. QUESTION 3
C) Long-term benefit/flooding One-dimensional hyd What additional data or piece(s) of information, if
raulic modeling (e.g. HEC-RAS or MIKE-11) can be any, will be useful in the future for refining
used to predict the rate reduction in flood frequen approaches to address the above issues? This
cy achieved by raising infrastructure. can include monitoring information, modeling
D) Long-term benefit/water quality No benefit would exercises, etc.
be expected as the work would not affect the sedi
ment yield from upstream.
E) Short-term impact/water quality No detrimental The following steps may help the Board take actions
impact would be expected. appropriate to the current level of knowledge and move
F) Short-term impact/channel habitat No detrimental sequentially over time to buttress its science base.
impact would be expected.
G) Short-term impact/riparian habitatNo detrimental Immediate period
impact would be expected. The Immediate Period is defined as the period of
H) Initial cost An estimate of the cost of raising infra time including the next few months. The following
structure should be obtained from a civil engineer. steps should be possible with a small amount of staff
I) Recurrent costNo recurrent costs are expected. time, and should be seen as interim measures.
Potential to Recover Beneficial Uses of Water CLARIFYING THE DEFINITION OF BACKGROUND LEVEL
Because the fundamental problem can be attributed to The current sediment standard in the North Coast
an abundance of water and sediment, restoring the Water Quality Control Plan states that turbidity must not
flux of water and sediment to a level that more closely be increased by more than 20 percent above natural
resembles their pre-logging levels would increase the background levels. The Panel, stakeholders, and some
probability of restoring the beneficial uses of water in regulatory agencies are bothered by a standard that
the long-term. If water and sediment yields are in bal depends on defining natural background conditions.
ance with transport capacity, a self-sustaining system Natural is generally assumed to mean pre-settlement,
should develop again, with all the associated benefits and therefore pre-logging conditions. The NCRWQCB
in terms of water quality, flood frequency, and stream could decide now to set an interim definition of Back
biological and physical function. ground levels that recognizes the variability in
Although natural background levels of runoff, ero response of geologic units to timber harvest activities.
sion and landslide activity are not known, the Panel The interim definition would become more fully devel
believes the inevitable consequences of logging activi oped and formalized in the TMDL process and other
ties, even with the best mitigation efforts, are increased Board actions; but some immediate clarification is
runoff, erosion and landslide activity. Therefore, while needed. The Panel is concerned that other agencies
mitigation should form an important part of any effort definitions will not meet the Boards mandate, and in
it is not likely to achieve the desired levels of sediment fact, will frustrate the attainment of water quality stan
reduction alone. A concurrent reduction in the harvest dards. The Panel is especially concerned that any defi
rate would likely increase the probability of long-term nition that progressively subsumes changing legacy
recovery. Reid (1999) has estimated the hydrologic sediment sources over time cannot serve as a baseline
recovery as a function of stand age. After 20 years, 56% for achieving water quality standards.
recovery is expected; after 50 years, 89% recovery is Performance of this task is beyond the Panels
expected; after 80 years, 100% recovery is expected. If purview, but a good definition is essential to neces
timber is harvested over shorter time cycles, water sary regulations. The Panel recommends that the
shed impacts will be cumulative because recovery time Board consider a definition that is based upon esti
will be interrupted by a new set of impacts. Therefore, mates of watershed performance under essentially
undisturbed conditions, recognizing variation in geol work may have priority); assistance from CGS and
ogy, climate, fire, and mass wasting. While no single CDF as well as private stakeholders would be helpful.
control watershed will serve to define undisturbed Considerable information is potentially available now
conditions, it would not be unreasonable to use the from a number of organizations, but data-sharing,
Headwaters drainage as a starting point. One although begun, has not built a readily available set of
approach that deserves investigation is to describe a reference material. The essence of the GIS work would
new standard based on measurement of suspended be to help the NCRWQCB and its staff visualize and
sediment differences from a suspended sediment-dis- track sediment-causing impacts in the five water
charge rating curve. Since turbidity or suspended sed sheds. Appendix D of this report depicts some prelimi
iment concentration vary considerably with water nary views, and a list of data layers that are available
discharge, any background level should be defined now, and gives some suggestions for further acquisi
with relation to discharge rate. In other words, it tions from cooperators.
should take the form of a discharge v. sediment dis
charge or turbidity rating curve, with exceedence FIELD DATA
defined as deviation more than a prescribed percent It would be desirable to initiate a limited set of field
age away from the line. studies now, even though data might not be forthcom
ing for use in immediate decisions. Items noted during
SETTING A RATE OF TIMBER HARVEST the Panels field trip and discussions were:
FOR THE FIVE WATERSHEDS
Any approach to setting rates of timber harvest should Longitudinal profiles and notation of
allow for adjustments over time based upon new streambed conditions. Some profiles exist, but
information. But it is essential that corrective actions they were prepared for fisheries purposes.
be started soon and not postponed awaiting research There is considerable controversy about
and monitoring that would take place over a period of pulses of sediment. Do they exist? How large
years. In the Panels opinion, it should be possible are they? How frequently do they move? Are
within this immediate time period to refine the Empiri they composed of fine or coarse sediment?
cal Sediment Budget Approach along the lines sug How persistent are channel bed changes
gested in Appendix C. Using this approach, it should caused by these sediment pulses? Some of
be sufficient to set interim rates for each of the five this dissention could be reduced by simple
watersheds. These rates then could be adjusted up or field surveys; other items would require more
down as information and a more robust management substantial inventories and monitoring.
structure evolve during the longer-term period
described below. The Panel prefers this approach to Core samples of streambed sediment could be
one that would rely on the Modeled Sediment Budget taken to not only authenticate sampling of
Approach because of the uncertainties with each step sediment depth, but possibly to help identify
in the detailed analyses. sediment sources by reference to the location
of geologic features.
Short-term (6 months to 2 years)
DATA AND ANALYSIS DIGITAL LIBRARY SERVICES
In the short-term (e.g., 6 months to 2 years), it should It is obvious to the Panel that a very large and growing
be possible to collect and re-assemble information body of literature, reports, maps, and memoranda
that already exists, but is not currently being used for must advise the decision process. Creation of a digital
decision support. repository for key documents, as well as a web site to
list and provide access to files, would be of great help.
GIS
A number of useful data layers are now available and Longer-term (next four or five years)
simply need to be re-projected to a common base and BACKGROUND LEVEL
assembled for analysis. In the Panels view, the NCR The interim definition must be refined based upon con
WQCB has the staff capabilities and computing sideration of long-term climatic and geomorphic episod
resources to do this quickly (recognizing that other ic events, (e.g. floods, seismicity, tectonic uplift, fire).
This topic warrants formal scientific review, but the Monitoring of water and sediment runoff from
Panel would note that the final definition will not emerge roads, harvest units subject to different
from science alone, but come as a policy decision. harvesting methods (e.g. tractor logging, cable
An early step would be to develop further monitoring logging), log landing areas, and landslides
data for the Headwaters, and/or other selected water should be carried out to enable more accurate
sheds. Channel surveys and the development of dis- sediment budgets to be constructed. This work
charge-suspended sediment rating curves are foremost will also reveal the effectiveness of mitigation
among data needs. The importance of monitoring is methods and allow for more rigorous
already understood, and residents and companies are identification of problem areas.
involved. Some excellent monitoring is underway, and
the Boards reliance on monitoring will underscore the Longitudinal profiles of the five streams,
value of such efforts. combined with geomorphic analysis of
channel condition and sediment storage,
RATE OF DISTURBANCE would be helpful in identifying the current
In the mid- to longer-term, watershed analyses and condition of the watershed and providing
cumulative effects analyses, coupled with improved some insight into the future impact of the
data should permit re-evaluation of the interim rates of upper watershed and stream geomorphology
cut. Evaluation could be timed to follow from new on the lower channel and floodplain.
information, rather than a pre-set date; the interim rate
might in any event be re-evaluated in four to five years. Continued monitoring of on-site and near-site
disturbances should be intensified.
WATERSHED AND STREAM CONDITIONS
Detailed mapping of Mass Wasting Potential has already IMPROVING THE SCIENCE BASIS FOR FUTURE ACTIONS
been prepared by CGS in the form of maps and will be The Boards record of decision should be increasingly
exposed to scientific review in the short run. Watershed based upon peer-reviewed studies and documenta
Analyses will be improved and finalized during the tion. Reliance on gray literature commonly leads to
short-term for at least some of the five watersheds. conflicting interpretation. The Panel recommends that
if the Board adopts policies regarding background and
Sediment rating curves should be developed rate of disturbance, it should upon adoption also initi
for watersheds with different geological ate a process whereby the science basis for back
characteristics and different land disturbance ground and rate of disturbance is published and
histories, to provide background conditions subject to peer review. This can be time-consuming,
and to enable deviation from background level so the Panel recommends a three-step process as fol
to be objectively demonstrated in the future. lows: 1) policy adoption, 2) formal publication and
review, and 3) revision of the policy if peer review sup
ports alternative interpretation.
Documents Beschta, R.L., M.R. Pyles, A.E. Skaugset, and C.G. Surfleet. 2000.
Alm, L. R. 1990. The United States-Canada Acid Rain Debate: Peakflow Responses to Forest Practices in the Western Cas
The Science-Politics of Linkage. American Review of cades of Oregon, USA. Journal of Hydrology 233:102-120.
Canadian Studies, 20 (Spring 1990): 59-79. Best, D.W., H.M. Kelsey, D.K. Hagans, and M. Alpert. 1995.
Altman, L.K. 2002. When Peer Review Yields Unsound Sci Role of Fluvial Hillslope Erosion and Road Construction in
ence. The New York Times, June 11, 2002. the Sediment Budget of Garret Creek, Humboldt County,
Anderson, H.W., M.D.Hoover, and K.G. Reinhart. 1976. California. Chapter M of Geomorphic Processes and
Forests and Water. General Technical Report PSW-18, For Aquatic Habitat in the Redwood Creek Basin. K.M. Nolan,
est Service, U.S. Department of Agriculture. H.M. Kelsey, and D.C. Marron, editors. U.S. Geological
Andersen, S., W. Ostreng. 1989. International Resource Man Survey Professional Paper 1454-M. pp.M1-M9.
agement: The Role of Science and Politics. Belhaven Brardinoni, F., and O. Slaymaker. 2001. Identification of Nat
Press, New York. ural and Logging-Related Landslides in the Capilano River
Anderson, J.K. 2001. Review of Elk River Flood Analysis Basin, Coastal British Columbia. Eos, Transactions, Ameri
Summary Prepared by the Pacific Lumber Company. can Geophysical Union, Volume 82, Fall Meeting Supple
Unpublished report prepared for the North Coast Region ment, Abstract H32A- 2091. 2001
al Water Quality Control Board, Santa Rosa, California. Bray, B. 2000. Quantitative Assessment of Suspended Sedi
Anderson, J.K. 2001. Review of Elk River Flood Analysis ment Concentration on Coho Salmon in Freshwater Creek.
Summary. October, 2001. A senior project presented to Dept. of Environmental
Baca, B. R., and R. E. Tepel. 2001. Freshwater Creek Water- Resources Engineering, Humboldt State University.
shed Analysis Mass Wasting Assessment: Comments on Brooks, H. 1964. The Scientific Advisor. P. 73-96 in Scientists
Review Draft Dated January 2001. Memorandum to John and National Policy Making. R. Gilpin and C. Wright (eds.).
G. Parrish, State Mining and Geology Board. April 24, 2001. Columbia University Press, New York.
Barrett J., M. OConnor, and E. Salminen. 2001. Elk River Brooks, K.N., P.F. Folliott, H.M. Gregersen, and J.L. Thames.
Flooding Analysis Summary, Hydrologic Change, and 1991. Hydrology and the Management of watersheds.
Channel Assessment. Unpublished report prepared for Iowa State University Press: Ames, IA, 392 pp.
the Pacific Lumber Company, Scotia, California. Cafferata, P.H. 1997a. Hydrologic Review of the Jordan Creek
Barrett, J. 2000. Freshwater Flooding Analysis Summary. Watershed. Memorandum to Thomas Osipowich,
Memorandum to John Sneed. Pacific Lumber Company. Resource Manager, California Department of Forestry and
Scotia, California Fire Protection. November 11, 1997.
Bedrossian, T. L. 2001. Review of A Scientific Basis for the Cafferata, P.H. 1997b. Hydrologic Review of the Elk River
Prediction of Cumulative Watershed Effects. Memoran Watershed. Memorandum to Thomas Osipowich, Resource
dum from California Division of Mines and Geology to Manager, California Department of Forestry and Fire Pro
Ross Johnson, California Department of Forestry and Fire tection. December 10, 1997.
Protection. October 3, 2001. Cafferata, P.H. 1997c. Hydrologic Review of the Freshwater
Benedick, R. 1991. Ozone diplomacy: New Directions In Safe Creek Watershed. Memorandum to Thomas Osipowich,
guarding the Planet. Harvard University Press, Cam Resource Manager, California Department of Forestry and
bridge, MA. Fire Protection. October 31, 1997.
Cafferata, P.H. and H. Scanlan. 1998. Freshwater Creek cross- Falls, J.N. 1997a. Orphan Roads: A Sediment Problem for
section remeasurement. Memorandum to Mr. Dave Ebert, Fisheries. The Example of Cummings Creek Road, Hum
Ranger Unit Chief, Humboldt-Del Norte Ranger Unit, boldt County, California. In Abstracts and Program. Asso
dated 11 September 1998. ciation of Engineering Geologists 40th Annual Meeting.
California Department of Forestry. 2002. Forest Practice Rules. Portland, Oregon.
California Regional Water Quality Control Board, North Coast Falls, J.N. 1997b. Preliminary Engineering Geologic Review
Region 1. 1997. Water Quality Control Plan (Basin Plan), of a Debris Flow at Stafford, Humboldt County, California.
Including Amendment Adopted February 26, 1997. In Abstracts and Program, Association of Engineering
Carver, G.A., A.S. Jayko, D.W. Valentine, and W.H. Li. 1994. Geologists 40th Annual Meeting, Portland, Oregon.
Coastal Uplift Associated With the 1992 Cape Mendocino Falls, J.N. 1997c. Reconnaissance Engineering Geologic Eval
Earthquake, Northern California. Geology 22:95-198. uation of Watershed Conditions in the Bear Creek Drainage,
Clarke, S.H., Jr. 1992. Geology of the Eel River Basin and Pepperwood, California. Memorandum from California
Adjacent Region: Implications for Late Cenozoic Tectonics Division of Mines and Geology to Craig E. Anthony, Deputy
of the Southern Cascadia Subduction Zone and Mendoci Director, Resource Management, California Department of
no Triple Junction. American Association of Petroleum Forestry and Fire Protection. August 21, 1997.
Geologists Bulletin. Volume 76, (2): 199-224. Falls, J.N. 1997d. Reconnaissance Engineering Geologic
Cleland, C.E. 2001. Historical Science, Experimental Science, Evaluation of Watershed Conditions in the Elk River
and the Scientific Method. Geology. 29 (11): 987-990. Drainage, Humboldt County, California. Memorandum
Collingridge, D. 1980. The Role of Experts in Decision-Mak- from California Division of Mines and Geology to Craig E.
ing. P. 183-196 in The Social Control of Technology, D. Anthony, Deputy Director, Resource Management, Califor
Collingridge (ed.). Frances Pinter, London. nia Department of Forestry. December 14, 1997.
Conroy, B. 1999. A Comparison of Rainfall Runoff Relations in Falls, J.N. 1997e. Reconnaissance Engineering Geologic Evalu
Elk River, A Small Coastal Northern California Watershed. ation of Watershed Conditions in the Jordan Creek Drainage,
Masters Thesis. Humboldt State University. December 1999. Pepperwood, California. Memorandum from California Divi
Crowser, Hart. 2000. Geology and Stream Morphology, Bear sion of Mines and Geology to Craig E. Anthony, Deputy
Creek Sub-Basin, Lower Eel Watershed, Humboldt County, Director, Resource Resource Management, California Depart
CA. Prepared for Pacific Lumber Company. November 27, ment of Forestry and Fire Protection, August 21, 1997.
2000. J6923. Falls, J.N. 1998. An Introduction to Geology and Landsliding.
Curtis, R.O. and Marshall, D.D. 1993. Douglas-Fir Rotations- In Abstracts and Program. Geology and Mass Wasting in
Time for Reappraisal. West. J. Applied For. 8:81-85. Forested Landscapes Workshop, California Licensed
Custis, K.H., and T.E. Spittler. 2002. Evaluation of W atershed Foresters Association. August 27 and 28, 1998. Fortuna,
Recovery, Rockpile Creek, Sonoma County, California. California.
Memorandum from Californian Geological Survey to Forster, B.A. 1992. The Acid Rain Debate: Science and Spe
William Snyder, California Department of Forestry and cial Interests in Policy Formation. Iowa State University
Fire Protection. July 31, 2002. Press, Ames, IA.
Dumitru, T.A. 1991. Major Quaternary Uplift along the North Freshwater Working Group Memorandum and Appendices.
ernmost San Andreas Fault, King Range, Northwestern 2002 From FWWG to Independent Scientific Review Panel,
California. Geology 19: 526-529. October 9, 2002. Further contextual explanation of items
Dunne, T., J. Agee, S. Beissinger, W. Dietrich, D. Gray, M. referred to in FW WA.
Power, V. Resh, and K. Rodrigues. 2001. A Scientific Basis Furbish, D.J., and R.M. Rice. 1983. Predicting Landslides Relat
for the Prediction of Cumulative Watershed Effects. The ed to Clearcut Logging, Northwestern California, U.S.A.
University of California Committee on Cumulative Water- Mountain Research and Development. v. 3, p. 253-259.
shed Effects. University of California Wildland Resource Garfield, E. 1993. Despite Problems with Peer Review, Science
Center Report No. 46. Publishing is Healthier than Ever. The Scientist 7(18): 12.
Elliott, Richard. 1997. Fish Habitat Conditions in Bear Creek, Gilpin, R. 1962. American Scientists and Nuclear Weapons
Tributary to Lower Eel River, Humboldt County. Memoran Policy. Princeton University Press, Princeton, NJ.
dum from California Department of Fish and Game to Glen Goldstein, W., V. A. Mohnen. 1992. Global Warming Debate
J. Newman, Chief, Coast-Cascade Region, California Depart- in the USA: The Clash Between Scientists on Policy Pro
ment of Forestry and Fire Protection. October 8, 1997. jections. Futures (January/February): 37-53.
Everest, F., D Swanston, C. Shaw, III, W. Smith, K. Julin, S. Grant, G. USDA Forest Service Pacific Northwest Research
Allen. 1997. Evaluation of the Use of Scientific Information Station, Hayes, S., Department of Geosciences, Oregon
in Developing the 1997 Forest Plan for the Tongass State University. Poster: Geomorphic response to peak
National Forest. Gen. Tech. Rep. PNW-GTR-415. Portland, flow increases due to forest harvest activities, Western
OR: U.S. Department of Agriculture, Forest Service, Pacif Cascades, Oregon [http://www.fsl.orst.edu/wpg/semi-
ic Northwest Research Station. nar%20schedules/HJPoster(sm).pdf]
Jacoby, G., G. Carver, and W. Wagner. 1995. Trees and Herbs Lee, Jonathan. A Multimetric Analysis of Benthic Macroin
Killed by an Earthquake 300 Years Ago at Humboldt Bay, vertebrate Data Collected from Freshwater Creek Water-
California. Geology. 23(1): 77-80. shed. Humboldt County California. From 1994-1998.
Jasanoff, S. 1990. The Fifth Branch: Science Advisors As Pol Levine, A.G. 1982. Love Canal: Science, Politics, and People.
icy Makers. Harvard University Press, Cambridge, MA. Lexington Books, Lexington, MA.
Johnson, K. N., J. F. Franklin, J. W. Thomas, and J. Gordon. Lewis, J. 1998. Evaluating the Impacts of Logging Activities
1991. Alternatives for Management of Late-Successional on Erosion and Suspended Sediment Transport in the
Forests in the Pacific Northwest. A Report to the Agricul Caspar Creek watersheds. Pp. 55-70 in R.R. Ziemer (tech
ture Committee and the Merchant Marine Committee of nical coordinator), Proceedings of the Conference on
the U. S. House of Representatives. Coastal Watersheds; the Caspar Creek Story; 6 May 1998;
Jones, J. 2000. Hydrologic Processes and Peak Discharge Ukiah, CA. Gen. Tech. Rep. PSW-GTR-168; Pacific South
Response to Forest Removal, Regrowth, and Roads in 10 west Research Station, Forest Service, U.S. Department of
Small Experimental Basins, Western Cascades, Oregon. Water Agriculture, Albany, CA.
Resources Research. 36(9): 2621-2642. September 2000. Lewis, J. and R. Eads. 2001. Turbidity Threshold Sampling for
Jones, J., F. Swanson, B. Wemple, and K. Snyder. 1999. Suspended Sediment Load Estimation. From Proceedings
Effects of Roads on Hydrology, Geomorphology, and Dis of the Seventh Federal Interagency Sedimentation Confer
turbance Patches in Stream Networks. Conservation Biol ence, March 25-29, 2001. Reno, Nevada.
ogy 14(1): 76-85. Feb. 1999. Lewis, J., S. Mori, E.T. Keppeler, and R.T. Ziember. 2001.
Jones, J.A. and F.J. Swanson. 2001. Hydrologic inferences Impacts of logging on storm peak flows, flow volumes
from comparisons among small basin experiments. Invit and suspended sediment loads in Caspar Creek, Califor
ed commentary. Hydrological Processes 15:2363-2366. nia, in M.S. Wigmosta and S.J. Burges, editors, Land Use
Katz, J. E. 1984. The Uses of Scientific Evidence in Congres and Watersheds: American Geophysical Union, Water Sci
sional Policymaking: the Clinch River Breeder Reactor. Sci ence and Application 2, p. 85-125.
ence, Technology and Human Values. 9(1): 5162. Li, W.H. and G.A. Carver. 1992a. The Late Holocene Stratigra
Keppeler, E.T. 1986. The Effects of Selection Logging on Low phy of the Eel River Delta. Final Report, Department of
Flows and Water Yield in a Coastal Stream in Northern Geology, Humboldt State University. February 17, 1992.
California. Unpublished M S. Thesis. Humboldt State Uni Li, W.H. and G.A. Carver. 1992b. The Late Holocene Stratigra
versity. Arcata, CA. phy of the Eel River Delta. Final Report prepared for the
Keppeler, E.T., R.R. Ziemer, and P.H. Cafferata. 1994. Changes U.S.D.A. Soil Conservation Service, Humboldt State Uni-
in Soil Moisture and Pore Pressure after Harvesting a versity. Arcata, California.
Forested Hillslope in Northern California. In Proceedings, Lisle, T.E., L.M. Reid, and R.R Ziemer. Review of Masters The
Annual Summer Symposium of the American Water sis authored by Mr. William John Conroy: A Comparison
Resources Association: Effects of Human-Induced of Rainfall-Runoff Relations in Elk River, a Small Coastal
Changes on Hydrologic Systems, June 26-29, 1994, Jack Northern California Watershed. USDA Forest Service
son Hole, Wyoming. Marston, R. A., and Hasfurther, V.R. Pacific Southwest Research Station, Redwood Sciences
(eds). Journal of the American Water Resources Associa Laboratory.
tion, pp. 205-214. Bethesda, Maryland. June 26-29, 1994. Lisle, T.E., L.M. Reid, and R.R Ziemer. 2000a. Review of:
Kittredge, J. 1948. Forest Influences. McGraw-Hill: New York, NY. Freshwater Flooding Analysis Summary. Unpublished
Knighton, D. 1998. Fluvial Forms and Processes: New York, report prepared by the USDA, Forest Service, Pacific
Wiley, 383 p. Southwest Research Station in Arcata for the California
Knudsen, Keith. 1993. Geology and Stratigraphy of the Freshwa Department of Forestry and Fire Protection, Sacramento,
ter Creek Watershed, Humboldt County, California. Unpub California.
lished M.S Thesis. Humboldt State University. April 20, 1993. Lisle, T.E., L.M. Reid, and R.R Ziemer. 2000b. Addendum:
Koehler, R.D., K.I. Kelson, and G. Matthews. 2001. Sediment Review of: Freshwater Flooding Analysis Summary.
Storage and Transport in the South fork Noyo River Unpublished report prepared by the USDA, Forest Ser
Watershed, Jackson Demonstration State Forest. Report vice, Pacific Southwest Research Station for the California
submitted to the California Department of Forestry and Department of Forestry and Fire Protection, Sacramento,
Fire Protection. June 26, 2001. California.
Lang, M, and E. Cashman. 2001. Quality Assurance/ Quality Luce, C.H. 1995. Forests and wetlands. Ch. 8 (pp. 253-284) in
Control Report Turbidity and Suspended Sediment Moni Ward, A.D. and Elliot, W.J. (eds.), Environmental Hydrolo
toring. conducted by Salmon Forever during Hyrdrologic gy. Lewis: Boca Raton, FL, 462 pp.
Year 2001. Humboldt State University Dept. of Environ Madej, M., B. Barr, T. Curren, A. Bloom, G. Gibbs. 2000. Effec
mental Resources Engineering Report. tiveness of Road Restoration in Reducing Sediment Loads.
Lave, L.B., E.P. Seskin. 1979. Epidemiology, Causality, and Agreement No. FG7354IF. U.S. Geological Survey Redwood
Public Policy. Am. Sci. 67(2): 176-186. Field Station. [http://were.usgs.gov/redwood/project-doc.pdf]
Madej, M.A. 2001 Erosion and Sediment Delivery Following Munn, J. 2002c. Response to Comments by Reid. Memoran
Removal of Forest Roads. Earth Surface Processes and dum to Dean Lucke dated April 12, 2002. California Depart
Landforms. 26: 175-190. ment of Forestry and Fire Protection, Sacramento, California.
Markle, G. E., J. C. Peterson. 1980. Politics, Science, and Can Natural Resources Management Corporation, 1998. Stitz
cer: The Laetrile Phonema. Westlake, Boulder, CO. Creek Sediment Source Assessment and Sediment Reduc
Marshall. G. 2002. Rapid Review of Engineering Geologic tion Recommendations.
Conditions for Specific Timber Harvest Plans in the Elk Neely, M.K., and R.M. Rice. 1990. Estimating Risk of Debris
River Watershed. Memorandum from the Department of Slides after Timber Harvest in Northwestern California.
Conservation, Division of Mines and Geology, to Ross Bulletin of the Association of Engineering Geologists
Johnson, California Department of Forestry and Fire Pro XXVII (3): 281-289.
tection, Sacramento, California. January 11, 2002. Nelkin, D. 1972. The University and Military Research: Moral
McCashion, J.D. and R.M. Rice. 1983. Erosion on Logging Politics at MIT. Cornell University Press. Ithaca, NY.
Roads in Northwestern California: How Much is Avoid- Nelkin, D. 1977. Scientists and Professional Responsibility:
able?: Journal of Forestry 81(1): 23-26. The Experience of American Ecologists. Soc. Stud. Sci.
Merritts, D.J. 1996. The Mendocino Triple Junction: Active 7:75-95.
Faults, Episodic Coastal Emergence and Rapid Uplift: North Coast Regional Water Quality Control Board. 2000.
Journal of Geophysical Research 101(B3): 6051-6070. Staff Report for Proposed Water Board Actions in the
Merritts, D.L., O.A. Chadwick, D.M. Hendricks. 1991. Rates North Fork Elk River, Bear Creek, Freshwater Creek, Jor
and Processes of Soil Evolution on Uplifted Marine Ter- dan Creek and Stitz Creek Watersheds. September 2000.
races, Northern California: Geoderma 51: 241-275 North Coast Regional Water Quality Control Board. 2001.
Merrits, D.L., O.A. Chadwick, D.M. Hendricks, G.H. Brimhall, Review - August 2001 Elk River Flooding Document, Pacif
and C.J. Lewis. 1992. The Mass Balance of Soil Evolution ic Lumber Company, Humboldt County. November 9, 2001
on Late Quaternary Marine Terraces, Northern California. North Coast Regional Water Quality Control Board. Rebuttal
Geological Society of America Bulletin 104:1456-1470. to proposed testimony submitted on behalf of Pacific
Mills, T.J., T.M. Quigley, F.H. Everest. 2001. Science-Based Lumber Company in the matter of Staff Report for Pro
Natural Resource Decisions: What are they? Renewable posed Water Board Actions in the North Fork Elk River,
Resources Journal (Summer 2001) 10-15. Bear Creek, Freshwater Creek, Jordan Creek and Stitz
Mills, T.J., F.H. Everest, P. Janik, B. Pendleton, C.G. Shaw, III, Creek Watersheds.
D.N. Swanston. 1998. Science/Management Collaboration: North Coast Regional Water Quality Control Board. 2002.
Lessons from the Revision of the Tongass National Forest Executive Officers Summary Report: Public Hearing for
Plan. Western Journal of Applied Forestry 13(3): 90-96. Consideration of Potential Requests for Report(s) of Waste
Monrone, J.G., E.J. Woodhouse. 1989. The Demise of Discharge for Timber Harvest Activities on and about Elk
Nuclear Energy: Lessons for Democratic Control of Tech River. April 8, 2002.
nology. Yale University Press, New Haven, CT. North Coast Regional Water Quality Control Board. 2002.
Montgomery, D. Peer Review of North Fork Cutting Rate by Executive Officers Summary Report: Public Hearing for
Dr. Leslie Reid. Consideration of Potential Requests for Report(s) of Waste
Montgomery, D.R., K.M. Schmidt, H.M. Greenberg, and W.E. Discharge for Timber Harvest Activities on and about
Dietrich. 2000. Forest Clearing and Regional Landsliding: Freshwater Creek, Bear Creek, Stitz Creek, and Jordan
Geology, v. 28, p. 311-314. Creek. April 8, 2002.
Munn, J. 2000. Review of Issues Related to Sediment Produc O Connor, M. 2000. Analysis of erosion and sedimentation
tion in the Addendum to the Redwood Sciences Laboratorys and its effects on flooding in Freshwater Creek: Freshwa
Review of Freshwater Flooding Analysis Summary. Unpub ter Creek between Graham Gulch and Little Freshwater
lished report dated December 20, 2000. California Depart Creek. Unpublished report prepared for the Pacific Lum
ment of Forestry and Fire Protection, Sacramento, California. ber Company, Scotia, California. June 28, 2000.
Munn, J. 2001. Elk River Channel Assessment Report. Memo OConnor, M. 2000. Proposed Testimony in Support of the
randum to Dean Lucke dated October 2, 2001. Department Pacific Lumber Company and Scotia Pacific Company LLC
of Forestry and Fire Protection, Sacramento, California. in regard to the Pacific Lumber Companys and Scotia
Munn, J. 2002a. Elk River Peak Flow Analysis. Memorandum to Pacific Company LLCs Timber Harvest and Related Activi
Dean Lucke dated January 14, 2002. California Department ties in the North Fork Elk River, Stitz Creek, Bear Creek,
of Forestry and Fire Protection, Sacramento, California. Jordan Creek and Freshwater Creek Watersheds. Decem
Munn, J. 2002b. Responses to Issues Raised by North Coast ber 1, 2000.
Regional Water Quality Control Board, Jeffery Anderson, OConnor, M. 2001a. Elk River stream channel assessment.
and Randy Klein about PALCOs Channel and Hydrology Unpublished report prepared by OConnor Environmen
Assessment for the Elk River Watershed. Memorandum to tal, Inc. for Pacific Lumber Company, Scotia, California.
Dean Lucke dated January 14, 2002.
OConnor, M. 2001b. Memo to John Munn: Response to Pacific Watershed Associates. 1998b. Sediment source inves
Comments, Elk River Channel Assessment Report. tigation and sediment reduction plan for the North Fork
November 5, 2001. Elk River watershed, Humboldt County, California. Unpub
OConnor, M. 2002. Quantitative assessment of erosion and lished report prepared for the Pacific Lumber Company,
sedimentation effects of forest management in northern Scotia, California.
California. Poster presented at Seventh Federal Intera Pacific Watershed Associates. 1999a. Sediment Source
gency Sedimentation Conference, March 25 through 29, Investigation and Sediment Reduction Plan for the Jordan
2001, Reno, NV. Creek Watershed, Humboldt County, California.
Ogle, B.A., 1953. Geology of the Eel River Valley Area, Hum Pacific Watershed Associates. 1999b. Sediment Source
boldt County, California: California Division of Mines Bul Investigation and Sediment Reduction Plan for the Fresh
letin 164. water Creek Watershed, Humboldt County, California.
Pacific Lumber Company. 1999. An Analysis of Flooding in Pacific Watershed Associates. 1999c. Memo to Ray Miller:
Elk River and Freshwater Creek Watersheds, Humboldt Differentiation of ASAP and High Treatment Immediacy
County. (priority) sites. October 6, 1999.
Pacific Lumber Company. 2000a. Memo to Craig Anthony: Pacific Watershed Associates. 2000a. Bear Creek Channel
Information Relevant to CDFs Moratorium in Elk River. Monitoring and Cross Section Surveys. June 2000.
January 31, 2000. Pacific Watershed Associates. 2000b. Jordan Creek Channel
Pacific Lumber Company. 2000b. Draft Freshwater Creek Monitoring and Cross Section Surveys. August 2000 .
Watershed Analysis. Pacific Watershed Associates. 2001. Memo: Road-related and
Pacific Lumber Company. 2001a. Freshwater Creek Water- non road-related erosion and sediment delivery to Clapp
shed Analysis. January, 2001. Gulch, Railroad Gulch, South Fork Elk River and lower
Pacific Lumber Company. 2001b. Pilot Turbidity Monitoring mainstem Elk River (interfluves). December 10, 2001.
Project: Winter 2000-2001 Completion Report. May 16. 2001. Pacific Lumber Company. 2002. Independent Science Review
Pacific Lumber Company. 2001c. Response to Preliminary Panel Visit: unpublished guidebook distributed during
Comments on PALCO Elk River Materials by P. Cafferata, field visit.
dated August 17, 2001. September 12, 2001. Patric, H.H. 1999. Letter to J.C. Barrett. In PALCO 1999. An
Pacific Lumber Company. 2001d. Memo: Response to CDF analysis of flooding in Elk River and Freshwater Creek
review of Elk River Preliminary Hydrologic Change watersheds, Humboldt County, California. Pacific Lumber
Assessment. October 23, 2001. Company: Scotia, CA.
Pacific Lumber Company. 2001e. Memo to Craig Anthony: Prellwitz, R.W., J. Oswald, and W. Adams. 2001. Manage-
Additional Materials for Freshwater Creek THPs. Decem ment-Related Landslides on Pacific Lumber Lands, Hum
ber 12, 2001. boldt Co., CA: A Geotechnical Perspective: unpublished
Pacific Lumber Company. 2001f. Memo to Craig Anthony: report prepared for Pacific Lumber Company, Scotia, Cali
Peer Review of RSLs Analysis [of Freshwater Flooding fornia.
Summary]. December 13. Prentice, C.S., D.J. Merrits, E.C. Beutner, P. Bodin, A. Shill,
Pacific Lumber Company. 2002. Compendium of Documents and J.R. Muller. 1999. Northern San Andreas Fault near
Compiled by Pacific Lumber Company for Independent Shelter Cove, California: Geological Society of America
Science Review Panel Site Visit. Includes various maps Bulletin (111) 4: 512-523.
and charts. October 9, 2002. Preston, L, D. McLeode, J. Schwabe, and PALCO Scientific
Pacific Lumber Company. Elk River and Freshwater Creek Collectors. 1999. Juvenile Salmonid Index Sampling,
Thalwegs. Freshwater Creek, Humboldt County.
Pacific Lumber Company. Trend Monitoring Information for Pyles, M., P. Adams, R. Beschta, and A. Skaugset. Forest
Elk River, Freshwater Creek, Bear Creek, Jordan Creek, Engineering Dept, Oregon State University. 1998. Forest
and Stitz Creek. Practices and Landslides: A report prepared for Governor
Pacific Lumber Company. Trend Monitoring results for Bear John A. Kitzhaber. January 1998, 49 pp.
Creek, Stitz Creek, Jordan Creek, Elk River, and Freshwater Pyles. 2000. Occurrence of Landslides on Forest Land: The
Creek. Devil is in the Details. Proceedings, Summit 2000, Wash
Pacific Lumber Company. Landscape Assessment of Geo ington Private Forests Forum, Electronic Proceedings,
morphic Sensitivity. Draft Sustained Yield Plan / Habitat http://www.cfr.washington.edu/outreach/summit/proceed-
Conservation Plan for the properties of The Pacific Lum ings.html.
ber Company, Scotia Pacific Holding Company, and Regens, J. L. 1984. Acid Rain: Does Science Dictate Policy or
Salmon Creek Corporation, Volume II, Part D. Policy Dictate Science? Economic Perspectives on Acid
Pacific Watershed Associates. 1998a. Sediment Source Deposition Control 8: 5-20, Crocker, T. D. (ed.). Butterworth
Investigation and Sediment Reduction Plan for the Bear Publishers, Boston, MA.
Creek Watershed, Humboldt County, California.
Reid and Lisle. Review of Freshwater Watershed Analysis. Assessment. October 23, 2001
USDA Forest Service Pacific Southwest Research Station, Salminen, E. 2002. Memo to Steve Horner: North Coast
Redwood Sciences Laboratory. Regional Water Quality Control Board comments on the
Reid, L.M. 1993. Research and cumulative watershed effects. Elk River Flooding Analysis. April 3, 2002.
General technical Report PSW-GTR-141, Pacific South Salmon Forever, Watershed Watch. 2002a. Elk River
west Research Station, Forest Service, U.S. Department of 2001/2002, Humboldt County California, Turbidity Sus
Agriculture, 118 pp. pended Sediment Plots.
Reid, L. M. 1998a. Calculation of Appropriate Cutting Rate in Salmon Forever, Watershed Watch, Freshwater Residents.
Bear Creek Watershed. Unpublished report prepared for 2002b. Freshwater Creek 2001/2002 Sites HH and MC,
the California Regional Water Quality Control Board, Humboldt County California, Turbidity and Suspended
Santa Rosa, California. Sediment Plots.
Reid, L. M. 1998b. Review of: Sediment Source Investigation Salmon Forever, Watershed Watch, Freshwater Residents.
and Sediment Reduction Plan for the Bear Creek Water- 2002c. Freshwater Creek Site FTR, Turbidity Threshold
shed for the EPA and the NCRWQCB. Sampling, Annual Load Plots Turbidity and Suspended
Reid, L.M. 1998c. Calculation of average landslide frequency Sediment Discharge Plots, Hydrologic Years 1999-2002.
using climatic records: Water Resources Research, v. 34, Sapolsky, H. M., I. Spiegel-Rosing, and D. de Solla Price
p. 869-877. (eds.). 1977. Science, Technology, and Military Policy. Sci
Reid, L.M. 1999. Review of the Final EIS/EIR and HCP/SYP for ence, Technology, and Society: A Cross Disciplinary Per
the Headwaters Forest Project. Report prepared for Con spective, p. 443-472 Sage, London.
gressman George Miller. Satterlund, D.R. 1972. Wildland Watershed Management.
Reid, L. M. 2000. Calculation of Appropriate Cutting Rate in Wiley and Sons: New York, NY, 370 pp.
North Fork Elk River Watershed. Unpublished report pre Sidle, R.C., A.J. Pearce, and C.L. OLoughlin. 1985. Hillslope
pared for the North Coast Regional Water Quality Control Stability and Land Use: American Geophysical Union,
Board, Santa Rosa, California. Water Resources Monograph 11, 140 pp.
Reid, L. M. 2002. Comments Concerning Differences Sidle, R.C., and W. Wu. 2001. Evalution of the Temporal and
Between Analyses by Lisle et al. and by Mr. John Munn. Spatial Impacts of Timber Harvesting on Landslide Occur
Memorandum to John Woolley, Third District Supervisor, rence, in M.S. Wigmosta and S.J. Burges, editors, Land
Humboldt County Board of Supervisors, Eureka, Califor Use and Watersheds: American Geophysical Union, Water
nia. March 25, 2002. Science and Application 2, p. 179-193.
Reid, L. M. Review of: An Analysis of Flooding in Elk River Simpson, G. 2002. Trench Investigation of Recent Sediment
and Freshwater Creek Watersheds, Humboldt County, Cal History Along the Elk River, Revision 1. Humboldt County,
ifornia (prepared by The Pacific Lumber Company, Scotia, California. Prepared for Stoel Rives LLP. February 2002
California). Skaugset, A., G. Reeves, and R. Keim. Landslides, Surface
Reid, L.M. Review of the Final EIS/EIR and HCP/SYP for the Erosion, and Forest Operations in the Oregon Coast
Headwaters Forest Project, Appendix 4, Discussion of Range. p. 213-242.
memo from PALCO concerning analysis of flooding in Slatick, E. 1994. Survey of Log Jams on the Lower Elk River.
Freshwater. USDA Forest Service Pacific Southwest Interoffice Memorandum to R. McCormick, Eureka Office,
Research Station, Redwood Sciences Laboratory. Dept. of Fish and Game. Includes Field Notes and graphs
Ricker, S. 2001. Study 1a1, Annual Report, Escapement and dating 1972-1994.
Life History Patterns of Adult Steelhead in Freshwater Smelser, M.G. 2001. Rapid Review of Aerial Photographs for
Creek, California, 2000-2001. Specific Timber Harvest Plans in the Bear Creek Water-
Ricker, S. 2001. Study 2a6, Annual Report, Results of Juve shed: Memorandum from California Division of Mines and
nile Downstream Migrant Trapping Conducted on Fresh Geology to Ross Johnson, California Department of
water Creek, California, 2001. Forestry and Fire Protection, June 27, 2001
Roelofs, T. 2002. Memorandum to Independent Scientific Smelser, M.G. 2002. Water Quality Complaint Investigation
Review Panel. Subject: Freshwater Creek Monitoring Station. of Cloney Gulch and South Fork Freshwater Creek: Memo
Ronayne, J. 1978. Scientific Research, Science Policy, and randum from California Resources Agency to Gerald Mar
Social Studies of Science and Technology in Australia. shall, August 2, 2002.
Soc. Stud. Sci. 8(3): 361-384. Smith, D.M., B.C Larson, M.J. Kelty, and P.M.S. Ashton. 1997.
Rushefsky, M.E. 1986. Making Cancer Policy. State Univ. of The Practice of Silviculture: Applied Forest Ecology. 537 p.
New York Press, Albany, NY. Sommerfield, C.K., D.E. Drake, and R.A. Wheatcroft. 2002.
Rustum, R. 1993. Science Publishing is Urgently in Need of Shelf record of climatic changes in flood magnitude and
Reform. The Scientist 7(17): 10. frequency, north-coastal California: Geology 30 (5): 395
Salminen, E. 2001. Memo to Steve Horner: Response to CDF 398.
review of Elk River Preliminary Hydrologic Change
Spittler, T.E. 1984. Landsliding in Forested Terrain, Southern Underwood, M.B. 1984. Franciscan and Related Rocks of
Humboldt County, California (abstract): Geological Soci Southern Humboldt County, Northern California Coast
ety of America Abstracts with Programs, Volume 16, Num Ranges; Analysis of Structure, Tectonics, Sedimentary
ber 6, p. 665. Petrology, Paleogeography, Depositional History, and Ther
Spittler, T.E. 1997. Preliminary Engineering Geologic Review mal Maturity: Unpublished Ph.D. Thesis, Cornell University.
of the Freshwater Creek Watershed: Memorandum from USDA Forest Service. 1992. Final Environmental Impact
California Division of Mines and Geology to Craig E. Statement on Management for the Northern Spotted Owl
Anthony, Deputy Director for Resource Management, Cal in the National Norests. Portland, OR.
ifornia Department of Forestry. November 7, 1997. USDA Forest Service. 1996. Status of the Interior Columbia
Spittler, T.E. 1998. Review of Report Sediment Source Inves Basin: Summary of Scientific Findings. USDA For. Serv.
tigation and Sediment Reduction Plan for the Bear Creek Gen Tech. Rep. PNW-GTR-385.
Watershed, Humboldt County, California: Memorandum USDA Forest Service. 1997a. Tongass Land Management
to James F. Davis, State Geologist. September 9, 1998. Plan Revision. Final Environmental Impact Statement. Part
State Mining and Geology Board, Technical Advisory Com 1: summary, chapters 1 and 2, and chapter 3 (physical and
mittee on Forest Geology. 2002. Technical Review of biological environment). U. S. Department of Agriculture,
Report: A Scientific Basis for the Prediction of Cumulative Forest Service R10-MB-338b. Washington, DC.
Watershed Effects, University of California Wildland USDA Forest Service. 1997b. Tongass Land Management
Resources Center Report No. 46 University of California Plan Revision. Final Environmental Impact Statement. Part
Committee on Cumulative Watershed Effects, June 2001, 2: Chapter 3 (Economic and Social Environment); Chap-
Memorandum to John G. Parrish, State Mining and Geol ters 4-7, and Appendix H. U. S. Department of Agriculture,
ogy Board, February 4, 2002. Forest Service R10-MB-338b. Washington, DC.
Surfleet, C.G., and R.R. Ziemer. 1996. Effects of Forest Har USDA Forest Service. 1997c. Integrating Science and Deci-
vesting on Large Organic Debris in Coastal Streams: In sion-Making: Quidelines for Collaboration Among Man
LeBlanc, John, ed., Conference on Coast Redwood Forest agers and Researchers in the Forest Service. Bulletin
Ecology and Management, Humboldt State Univ., Arcata, FS-608. U. S. Department of Agriculture, Washington, D. C.
CA, pp. 134-1. June 18-20, 1996. USDA Forest Service and USDI Bureau of Land Manage
The University of California Committee on the Scientific ment. 1994. Draft Environmental Assessment for the
Basis for the Analysis and Prediction of Cumulative Water Implementation of Interim Strategies for Managing
shed Effects. 1999. Review of March 1999 document An Anadromous Fish-Producing Watersheds in Eastern Ore
Analysis of Flooding in Elk River and Freshwater Creek gon, Washington, Idaho, and Portions of California.
Watersheds, Humboldt County, California by The Pacific Watershed Professionals Network. 2001. Freshwater Creek
Lumber Company. July 1, 1999. Watershed Analysis. Prepared for Pacific Lumber Company.
The University of California Committee on the Scientific White, A. 2002. Memorandum to Diana Henrioulle-Henry: Elk
Basis for the Analysis and Prediction of Cumulative Water River Peak Flow Analysis. North Coast Regional Water
shed Effects. Letter to Andrea Tuttle, Director, CDF, Re: Quality Control Board. February 1, 2002.
PALCO Report on Flooding in Elk and Freshwater Basins. Wohl, E. 2000. Mountain Rivers. Water Resources Mono
Thomas, R.B., and W.F. Megahan. 1998. Peak Flow Respons graph 14, American Geophysical Union: Washington,
es to Clear-Cutting and Roads in Small and Large Basins, D.C., 320 pp.
Western Cascades, Oregon: A Second Opinion: Water Wooster, John. Compilation of Stream Clearing Data in the
Resources Research 34(12): 3393-3403. North Coast, CA: California Department of Fish and Game.
Toppozada, T., G. Borchardt, W. Haydon, M. Pertersen, R. unpublished report.
Olson, Lagorio, and T. Anvik. 1995. Planning Scenario in Ziemer, R.R. 1981. Stormflow Response to Roadbuilding and
Humboldt and Del Norte Counties, California, for a Great Partial Cutting in Small Streams of North California: Water
Earthquake on the Cascadia Subduction Zone: California Resources Research 17(4): 907-917.
Division of Mines and Geology Special Publication 115. Ziemer, R.R. 1998. Flooding and Stormflows. Pages 15-24 in
U.S. Fish & Wildlife Service and California Department of Ziemer, R.R. (ed.), Proceedings of the conference on
Forestry and Fire Protection. 1999. Final Environmental coastal watersheds: the Caspar Creek story. General Tech
Impact Statement/Environmental Impact Report and Habi nical Report PSW-GTR-168, Pacific Southwest Research
tat Conservation Plan/Sustained Yield Plan for the Head Station, Forest Service, U.S. Department of Agriculture.
waters Forest Project.
June 27th meeting, the Regional Water Board directed and alterations throughout the watersheds and sedi
staff to (1) initiate its own facilitated Scientific Review ment input rates from sources, including but not limit
Process which would build upon the work on the Con ed to natural sources, timber harvest and related
vening Committee and which would include Elk River activities, and road rehabilitation activities.
and (2) invite the original members of the Convening
Committee, as well as Elk River representatives, to A. What options are available (e.g. dredging, and
assist the Regional Water Board staff in finalizing an modification of activities resulting in, or
initial set of Terms of Reference (TOR) for the pro reducing, sediment delivery) that can be
posed Scientific Review Panel. immediately implemented and will be effective
This document is the final version of the TOR. in lessening the adverse flooding conditions
and impacts to beneficial uses? Please discuss
Issue Statement and Objectives the potential benefits, limitations and tradeoffs
The Convening Committee originally recommended of these options for each of the five watersheds.
focusing on the following issues from which to derive B. Please review the provided documents, and
questions for the Independent Scientific Review any other relevant information, regarding
Panels review and analysis: calculation of appropriate rates of timber
harvest that would not impede recovery from
Sediment impairment/reduced channel capacity; excess sediment loads2 and would not cause
Flooding/peak flows; or contribute to exceedence of water quality
Riparian/flood plain management; objectives. Please discuss the technical
strengths and weaknesses of the varying
and the impacts of each of the above on beneficial uses. approaches described in some of these
The Independent Scientific Review Panels delibera documents to address harvest rate and flood
tions should be solution-oriented and will be used as a severity, as well as any other reasonable
foundation for developing the package of interim approaches to calculate a rate of harvest for
measures. The Panel will undertake the following each of the five watersheds that is protective
activities: (1) review and comment on current docu of water quality, which considers natural and
ments and reports that address the questions posed to other anthropogenic sediment sources.
the Independent Scientific Review Panel (2) identify
what further information and analysis, if any, is neces C. What additional data or piece(s) of information,
sary to assess these or other relevant questions, and if any, will be useful in the future for refining
(3) provide guidance on possible interim options that approaches to address the above issues? This
would address these issues. can include monitoring information, modeling
In carrying out its work, the Independent Scientific exercises, etc.
Review Panel shall address the following issues and
should consider time frames, natural sediment trans 2. Recovery is interpreted to mean that the water body can support all
designated beneficial uses of water and meet the water quality stan
port potential, existing infrastructure, development dards as outlined in the Basin Plan.
The following documents provide contextual informa The following documents represent approaches pro
tion regarding beneficial uses of water, water quality posed to date to evaluate timber harvest effects on
objectives and prohibitions, and watershed-specific flooding, and approaches to determining appropriate
sediment sources, sediment impairment, and timber rates of harvest. These documents are presented in
harvesting activities. chronological order.
California Regional Water Quality Control Board, North Coast Reid, L. M. 1998 Calculation of Appropriate Cutting Rate in
Region 1, Water Quality Control Plan (Basin Plan), includ Bear Creek Watershed. USDA Forest Service Pacific South
ing amendment adopted February 26, 1997. west Research Station, Redwood Sciences Laboratory.
North Coast Regional Water Quality Control Board. Septem Barrett, J. 2000. Memorandum to John Sneed: Freshwater
ber 2000. Staff Report for Proposed Water Board Actions Flooding Analysis Summary. Pacific Lumber Company.
in the North Fork Elk River, Bear Creek, Freshwater Creek,
Jordan Creek and Stitz Creek Watersheds. Reid, L. M. August 28, 2000. Calculation of Appropriate Cut
ting Rate in North Fork Elk River Watershed. USDA Forest
North Coast Regional Water Quality Control Board. April 8, Service Pacific Southwest Research Station, Redwood Sci
2002. Executive Officers Summary Report: Public Hearing ences Laboratory.
for Consideration of Potential Requests for Report(s) of
Waste Discharge for Timber Harvest Activities on and about Lisle, T.E., L.M. Reid, and R.R Ziemer. September 15, 2000.
Freshwater Creek, Bear Creek, Stitz Creek, and Jordan Creek. Review of: Freshwater flooding analysis summary.
Unpublished review prepared for California Department
North Coast Regional Water Quality Control Board. April 8, of Forestry and Fire Protection and the North Coast
2002. Executive Officers Summary Report: Public Hearing Regional Water Quality Control Board. USDA Forest Ser
for Consideration of Potential Requests for Report(s) of vice Pacific Southwest Research Station, Redwood Sci
Waste Discharge for Timber Harvest Activities on and ences Laboratory.
about Elk River.
Lisle, T.E., L.M. Reid, and R.R Ziemer. October 25, 2000. Adden
Natural Resources Management Corporation, 1998. Stitz dum: Review of: Freshwater flooding analysis summary.
Creek Sediment Source Assessment and Sediment Reduc Unpublished review prepared for California Department of
tion Recommendations. Forestry and Fire Protection. USDA Forest Service Pacific
Southwest Research Station, Redwood Sciences Laboratory.
Pacific Watershed Associates, 1998. Sediment Source Inves
tigation and Sediment Reduction Plan for the North Fork Munn, J. R. December 20, 2000. Review of Issues Related to
Elk River Watershed, Humboldt County, California. Sediment Production in the Addendum to the Redwood
Sciences Laboratorys Review of Freshwater Flooding
Pacific Watershed Associates, 1998. Sediment Source Inves Analysis Summary. California Department of Forestry and
tigation and Sediment Reduction Plan for the Bear Creek Fire Protection.
Watershed, Humboldt County, California.
Barrett, OConnor, Salminen. August 6, 2001. Elk River Flood
Pacific Watershed Associates, 1999. Sediment Source Inves ing Analysis Summary. Pacific Lumber Company.
tigation and Sediment Reduction Plan for the Jordan
Creek Watershed, Humboldt County, California. Munn, J. R. January 14, 2002. Memorandum to Dean Lucke:
Elk River Peak Flow Analysis. California Department of
Pacific Watershed Associates, 1999. Sediment Source Inves Forestry and Fire Protection.
tigation and Sediment Reduction Plan for the Freshwater
Creek Watershed, Humboldt County, California. White, A. February 1, 2002. Memorandum to Diana Henri-
oulle-Henry: Elk River Peak Flow Analysis. North Coast
Pacific Watershed Associates. December 10, 2001. Memo: Regional Water Quality Control Board.
Road-related and non road-related erosion and sediment
delivery to Clapp Gulch, Railroad Gulch, South Fork Elk Reid, L. M. March 25, 2002. Comments concerning differ
River and lower mainstem Elk River (interfluves). ences between analyses by Lisle et al. And by Mr. John
Munn. USDA Forest Service Pacific Southwest Research
Station, Redwood Sciences Laboratory.
Munn, J. R. April 12, 2002. Memorandum to Dean Lucke: by Dr. Leslie Reid.
Response to Comments by Reid. California Department of
Forestry and Fire Protection. Munn, J. October 2, 2001. Memo to Dean Lucke: Elk River
Channel Assessment Report. Department of Forestry and
The panelists will be expected to review the above- Fire Protection
mentioned documents. If panelists choose to review
additional background information, any of the follow Munn, J. January 14, 2002. Memo to Dean Lucke: Responses to
issues raised by North Coast Regional Water Quality Control
ing documents will be made available upon request.
Board, Jeffery Anderson, and Randy Klein about PALCOs
Please note the below list is not exhaustive.
Channel and Hydrology Assessment for the Elk River Water-
shed. Department of Forestry and Fire Protection.
Anderson, J.K. Review of Elk River Flood Analysis Summary,
October, 2001.
North Coast Regional Water Quality Control Board. Rebuttal
to proposed testimony submitted on behalf of Pacific
Cafferata, p. August 17, 2001. Preliminary Comments on
Lumber Company in the matter of Staff Report for Pro
PALCO Elk River Materials. California Department of
posed Water Board Actions in the North Fork Elk River,
Forestry and Fire Protection.
Bear Creek, Freshwater Creek, Jordan Creek and Stitz
Creek Watersheds.
Conroy, B. December 1999. Masters Thesis: A Comparison of
Rainfall Runoff Relations in Elk River, A Small Coastal
North Coast Regional Water Quality Control Board. Novem
Northern California Watershed. Humboldt State University.
ber 9, 2001. Review - August 2001 Elk River Flooding Doc
ument, Pacific Lumber Company, Humboldt County
Dunne, T., J. Agee, .S. Beissinger, W. Dietrich, D. Gray, M.
Power, V. Resh, and K. Rodrigues. 2001. A Scientific Basis
O Connor, M. June 28, 2000. Analysis of Erosion and Sedi
for the Prediction of Cumulative Watershed Effects. The
mentation and its Effects on Flooding in Freshwater Creek:
University of California Committee on Cumulative Water-
Freshwater Creek Between Graham Gulch and Little
shed Effects. University of California Wildland Resource
Freshwater Creek.
Center Report No. 46.
Pacific Lumber Company. October 23, 2001. Memo: Response Ricker, S. Study 1a1, Annual Report, Escapement and Life
to CDF review of Elk River Preliminary Hydrologic Change History Patters of Adult Steelhead in Freshwater Creek,
Assessment. California, 2000-2001.
Pacific Lumber Company. May 16, 2001. Pilot Turbidity Moni Ricker, S. Study 2a6, Annual Report, Results of Juvenile
toring Project: Winter 2000-2001 Completion Report. Downstream Migrant Trapping Conducted on Freshwater
Creek, California, 2001.
Pacific Lumber Company. Trend Monitoring results for Bear
Creek, Stitz Creek, Jordan Creek, Elk River, and Freshwater Salminen, E. October 23, 2001. Memo to Steve Horner:
Creek. Response to CDF review of Elk River Preliminary Hydro
logic Change Assessment.
Pacific Lumber Company. Elk River and Freshwater Creek
Thalwegs. Salminen, E. April 3, 2002. Memo to Steve Horner: North
Coast Regional Water Quality Control Board comments on
Pacific Watershed Associates. October 6, 1999. Memo to Ray the Elk River Flooding Analysis.
Miller: Differentiation of ASAP and High Treatment
Immediacy (priority) sites. Simpson, G. February 2002. Trench Investigation of Recent
Sediment History Along the Elk River, Revision 1. Hum
Pacific Watershed Associates. August 2000. Jordan Creek boldt County, California. Prepared for Stoel Rives LLP.
Channel Monitoring and Cross Section Surveys.
U.S. Fish & Wildlife Service and California Department of
Pacific Watershed Associates. June 2000. Bear Creek Chan Forestry and Fire Protection. 1999. Final Environmental
nel Monitoring and Cross Section Surveys. Impact Statement/Environmental Impact Report and Habi
tat Conservation Plan/Sustained Yield Plan for the Head
Preston, L, D. McLeode, J. Schwabe, and PALCO Scientific waters Forest Project.
Collectors. 1999 Juvenile Salmonid Index Sampling,
Freshwater Creek, Humboldt County. The University of California Committee on the Scientific
Basis for the Analysis and Prediction of Cumulative Water
Reid, L.M. Review of the Final EIS/EIR and HCP/SYP for the shed Effects. July 1, 1999. Review of An Analysis of
Headwaters Forest Project, Appendix 4, Discussion of Flooding in Elk River and Freshwater Creek Watersheds,
memo from PALCO concerning analysis of flooding in Humboldt County, California by The Pacific Lumber Com
Freshwater. USDA Forest Service Pacific Southwest pany, March 1999.
Research Station, Redwood Sciences Laboratory.
The University of California Committee on the Scientific
Reid, L. M. Review of the Sustained Yield Plan / Habitat Con Basis for the Analysis and Prediction of Cumulative Water
servation Plan for the properties of The Pacific Lumber shed Effects. Letter to Andrea Tuttle, Director, CDF, Re:
Company, Scotia Pacific Holding Company, and Salmon PALCO Report on Flooding in Elk and Freshwater Basins
Creek Corporation, Appendix 4. The influence of cross-
sectional changes on flood frequency, Freshwater Creek. Guiding Principles
USDA Forest Service Pacific Southwest Research Station, This initiative, to be launched in August, will be guid
Redwood Sciences Laboratory. ed by several key principles:
things, issues such as who pays for the solution, facilitated workshop format, open to the
source activity. Any such issues would in the end need review additional documents and conduct
to be addressed by the Regional Water Board and/or additional analysis, as needed; and
other agencies/stakeholder groups, as appropriate, as convene a final ISRP meeting to present the
Approach DELIVERABLE
PARTICIPANTS The Independent Scientific Review Panels primary
Disciplines of Technical Experts. The Regional Water deliverable will be a concise, final written report in
Board staff will recruit recognized technical experts response to the questions regarding water quality
who collectively can provide analysis and understand issues in these five watersheds, with recommenda
ing in the following subjects: tions of potential interim measures.
10:15 Point out Freshwater pool on the way to visit monitoring station at Roloefs
11:00 Begin driving along back road up divide between Elk and FW.
Stop at Vista #1 (Horse shoe) on Freshwater side to examine pre-HCP and post-HCP logging
(includes an example of recent operation and an operation that is at least 1 year old)
12:40 Brief stop at Vista #2 on Elk River side to get an aerial overview of the Elk River drainage
1:00 Lunch at Bridge Creek to see the channel, coarse substrate, road work, new bridge, geology
1:45 Stop at Wrigley Ranch to see gradient change. It is the first reach below and adjacent to PL property.
2:15 Brief stop at North Fork Bridge to see accreted channel and loss of fine sediment due to flooding.
area land class, obtained from this solution, gives the 37% of the area cut within the previous 15 years. This
maximum area that can at any time be impacted by har implies that 15% of sediment was derived from the
vest activities if the relative sediment production is to 63% of the area not cut. Letting represent the frac
be kept below the target threshold level. Solution of this tion of sediment derived from the cut proportion N,
requires knowledge of the normalized sediment pro
duction coefficients, wi, from each distinct land class S = NLR (7)
representing the geologic and geomorphologic vari
ability present in the area. In the empirical sediment for the cut proportion and
budget approach these coefficients are estimated based
on field measurements. With this general background, (1 )S = (1 N)R (8)
the calculations of sediment production and rate of har
vest limitations that have been proposed by Dr. Reid are for the uncut proportion of the watershed. Both R and
examined. S cancel when equations (7) and (8) are combined into
a ratio, which can then be solved for L in terms of and
REIDS FIRST CALCULATION (REID, 1998A) N. The result is
Dr. Reid calculates post-harvest sediment production
/N
using two classes: the fraction of area cut, N, and the L= (9)
remaining fraction that is left uncut, 1N. The fraction (1 ) / (1 N)
of area cut is defined as the area that has been harvest Values of = 0.85 and N = 0.37 yield a result of L = 9.6.
ed within the last n years (she used n = 15). Therefore, Using this result and a value of SRT = 1.2 in equation (6)
there is an n year recovery period built in to the calcu gives an allowable n year cut of NT = 0.023, or 2.3%,
lation and, under a uniform sustainable harvest rate, which corresponds to an annual rate of 0.15%.
the annual harvest rate is N/n. The sediment produc
tion coefficient for uncut or fully healed areas is the INCLUDING THE EFFECT OF NO-CUT ZONES
background rate R and the sediment production coeffi The land class approach of equation (1), may be
cient over the cut area is this rate multiplied by a factor applied to the sediment production from two classes: a
L, i.e. LR. Equation (1) can therefore be written as high-hazard area with the proportional area ah and the
remaining low-hazard proportional area (1ah). The
S = (1 N)R + NLR (4) definition of high-hazard is open to debate. A conserva
tive estimate might be all of the areas listed as high or
The pre harvest conditions are simply background (i.e., very high hazard on a CGS landslide hazard map,
S = R), so normalization of equation (4) by R yields whereas a liberal estimate might be those areas delin
eated by the PALCO mass wasting areas of concern
SR = 1 N + NL (5) (MWACs). Regardless of the definition used, however,
the background sediment production for a two-class
If the allowable relative impact is specified as a thresh reference state is
old (e.g., using a value of SRT = 1.2 to be consistent
with Reid), equation (5) can be solved for the threshold R = R1ah + R2(1 ah) (10)
value of N to yield
Reids calculation did not explicitly distinguish the pos
SRT 1
NT = (6) sibility of different background sediment production
L1 rates from these different areas, nor did it explicitly
The subscript T indicates an allowable threshold on SR incorporate the proportion of the watershed that falls
and the resulting threshold fraction of area N impacted within the high-hazard class. Instead, Reid wrote (equa
over n years. This result was used for Reids first calcu tion 5, Reid, 1998a) R = 0.9R + 0.1R with the 0.9R intend-
lation of allowable cut in the Bear Creek watershed ed to represent the sediment production from high
(Reid, 1998a). hazard areas and 0.1R intended to represent the sedi
To estimate L, Reid relied on the numbers reported ment production from the remaining area. For general-
by Pacific Watershed Associates, (1998a), namely that ity, the fraction of sediment production from the high
85% of landslide-derived sediment originates on the hazard area is denoted as h. Thus,
Combining equations (11) and (12), Equation (19) is equivalent to Reids (1998a) equation
(7) and equation (17). Using (11) and (12) to substitute
R1 h / ah
= (13) for R1ah and R2(1 ah) in equation (18), results in
R2 (1 h) / (1 ah)
(SRT 1)(hR + R(1 h))
This permits calculation of the different background NT = (20)
(L 1)R(1 h)
areas given the area fraction ah and ratio of sediment This is equivalent to Reids (1998a) equation (8).
production rates R1/R2, namely There are three dimensionless empirical parame
ters that enter in to this calculation:
ah (R1 / R2)
h = (14)
1 ah + ah (R1 / R2) R1 / R2, the dimensionless ratio of background
If no timber harvesting is allowed in the high-hazard sediment production rate from high-hazard
areas, then the land class approach of equation (1) con areas to that from low-hazard areas.
tains three classes: the high hazard zone with proportion
ah, the harvested fraction of the remainder N (1 ah), and ah, the proportion of the watershed classified
the unharvested fraction of the remainder (1 N)(1 ah). as high-hazard.
As before, N refers to the area harvested in the last n
years and it is assumed that harvested areas are com L, the increase in sediment production due to
pletely healed after n years. Therefore, the annual har logging on non high-hazard areas.
vest fraction is N / n of the harvestable area, or (N / n)(1
ah) of the total area. Again using L to denote the increase With these parameters equations (17) or (19) can be used
in sediment production from harvested area the sedi to calculate the relative increase in sediment production,
ment production is SR, given the harvest impact area proportion N. The com
plementary relationship expressed in equations (18) or
S = R1ah + R2(1 ah)(1 N) + R2(1 ah)LN (15) (20) can be used to calculate an allowable impact area
proportion NT given a threshold permissible relative sed
The ratio of equation (15) to equation (10) is the rela iment production rate SRT. Furthermore, the parameters
tive increase in sediment production R1 / R2 and ah can be combined into the single quantity h
(equation 14), which is the fraction of background sedi
R1ah + R2(1 ah)(1 N) + R2(1 ah)LN
SR = (16) ment produced in high hazard areas. Therefore, relative
R1ah + R2(1 ah)
sediment production or allowable impact area propor
which can be divided by R2 to yield tion can be calculated using only h and L.
Reids second calculation (Reid, 1998a equation 8)
(R1 / R2)ah + (1 ah)(1 N + LN)
SR = (17) used this approach with h = 0.9, and L = 9.6. The result
(R1 / R2)ah + 1 ah
yields N = 0.23 and with n = 15 an annual allowable
Solving equation (17) for the allowable cut rate given harvest N/n = 0.015 or 1.5%.
a threshold SRT, results in Some points that emerge from this derivation are:
(L 1)R2(1 ah)
to logging on non high-hazard areas. The
(L 1)(1 ah)
from the first calculation that did not separate As an example, the Panel presents some of these calcu
out non-high hazard areas, so may not be a lations, loosely based on the Bear Creek Pacific Water-
good estimate for non-high hazard areas. shed Associates (1998a) report that Reid (1998a) used
as her data source. The Pacific Watershed Associates
h represents the fraction of sediment report (1998a, page 18) indicates that 37% of the water
production from high hazard areas under shed was in a state of recently harvested (<15 year
reference (non-harvest) conditions. In this old harvested slopes) at the time of the 1996/97 storm.
context, reference conditions are areas not This report (table 5, page 19) gives sediment production
logged in the n = 15 previous years. It is not volumes for recently harvested and older harvested
clear whether the value of h = 0.9 that Reid slopes, and notes that 75% of the slides occurred on
used was for a combination of logged and inner gorge hillslopes (high hazard areas). The report
reference areas or reference areas alone. It is, does not separate out the sediment production from
however, a very conservative estimate that high hazard areas, either recently or older harvested
will help to maximize the annual cut rate for areas. With a geographic information system map of
low-hazard areas. high hazard areas and sediment volume estimates from
each landslide it is in principle simple to separate out
The data upon which the estimates of L = 9.6 and h the area and sediment production from the four possi
= 0.9 are based (Pacific Watershed Associates, 1998a, ble area classes, namely reference low hazard, refer
1998b, 1999a, 1999b) consists of landslide inventories ence high hazard, logged low hazard and logged high
derived from aerial photographs, with the estimated hazard. Because the Panel does not have this informa
volume of sediment production associated with each tion, estimates of these numbers have been used for
landslide included in the total. The area harvested in illustrative purposes and to examine sensitivity.
the previous n = 15 years was delineated to obtain the The base data used for Bear Creek was:
sediment production from harvested areas. There is
also in principle a delineation of high hazard areas Sediment
(although this may not have been explicitly done yet). Area Production
Therefore the information does in principle exist to [mi2] [yd3/interval]
delineate the watershed into four area classes and Area Class
estimate the sediment production coefficients needed Reference, low hazard 3.02 4020
for proper application of a source area based sediment Reference, high hazard 2.02 36180
production calculation and allowable harvest. The pri Logged, low hazard 1.78 22850
mary input quantities required are listed in Table C1. Logged, high hazard 1.18 205650
Total 8.00 268700
TABLE C1: QUANTITIES REQUIRED FOR SOURCE
AREA SEDIMENT PRODUCTION CALCULATION The total 8 mi2 area was divided in to high hazard
Sediment and low hazard assuming a ratio of 0.4 (a guess for
Area production illustrative purposes; the exact value is not critical
[L2] [L3/T] because the result is very insensitive to changes in this
Area class ratio). Then the area fraction of each that was logged
Reference, low hazard A1 S1 was taken as 37%. Pacific Watershed Associates
Reference, high hazard A2 S2 reported that this number applies to the whole area,
Logged, low hazard A3 S3 but here, we assumed that the split is equal between
Logged, high hazard A4 S4 low and high hazard areas.
The total sediment production of 286,700 yd3 from the
The primary calculation inputs are then obtained as 1996/97 photo interval (reported by Pacific Watershed
Associates) was split between logged and reference using
R1 = S2 / A2
Pacific Watershed Associates (1998a) table 5. These
R2 = S1 / A1
amounts were then split between high hazard and low
ah = (A2+A4) / (A1+A2+A3+A4)
hazard areas using a factor h = 0.9 (the same as Reid).
L = (S3 / A3) / R2
The resulting derived empirical input parameters are:
GIS can also serve as a basis for analysis to assist DWR CALwater Planning Watersheds and
GIS EXHIBIT 1
In this image, PALCO mass wasting zonation is shown as crosshatched lines over the mass wasting
map prepared by the California Geologic Survey (CDG CD 2002-06, August 2, 2002). This example
covers only a portion of the Freshwater Creek watershed, but was selected as being a balanced rep
resentation of the relative coverage of the two classification systems for of the entire watershed.
GIS EXHIBIT 2
Timber Harvest Plans (THPs) from the years: 19902000 are shown as semi-transparent overlays to
USGS 7.5 minute topographic maps; covering a sample area Northeast of the Elk River Scout Camp.
The colored polygons representing the areas covered by THPs can also be draped over other maps
and imagery, such as Mass Wasting Zonation, or aerial photography.
GIS EXHIBIT 3
This is a static image of an on-line Web-GIS display prepared by a member of the Panel (at:
http://calsip.regis.berkeley.edu/northcoast/). It is presented as an example of how maps and imagery can
be mounted for work over the internet. When working in the on-line browser, map layers can be added to
the list in the upper left, and toggled on and off. The user can zoom out to get an overview or zoom down
to see details, and the user can pan so as to fly around the region. This screen capture simply shows a
portion of the Jordan Creek watershed boundary (in purple) over panchromatic ortho-photos on which
recent harvest areas, landings, roads, and many other features can be seen. This tool could be useful for
project work involving multiple organizations, and for communication with stakeholders and the public.