Dr. R. J. White Response to “Draft Boardman River Assessment”

Comments on the Michigan DNR Fisheries Division’s 2014 Draft Boardman River Assessment

A Report to the Adams Chapter of the Michigan Council of Trout Unlimited

Ray J. White, Ph.D.[1]

August 10, 2014


In March 2014, the Michigan Department of Natural Resources issued a Boardman River Assessment draft (Kalish & Tonello 2014) for public comment.  Although the comment period ended May 16, the Adams Trout Unlimited Chapter, Traverse City–part of a local and extended stakeholder community–asked for my comments.  I agreed to make a cursory review.  Because the draft has so many problems, my review became more than that, even though I didn’t delve into all its topics and ignored many minor problems. In keeping with the Adams Chapter’s interest in the fishery for river-resident trout, namely brook and brown trout, I give special attention to how the draft relates to that fishery.

The word assessment has a complex of meanings.  The following definitions from Merriam-Webster[2] apply to the situation at hand: “the act of making a judgment about something” or “an idea or opinion about something,” and for the verb assess: “to make a judgment about (something); to officially say what the amount, value, or rate of something is” or “to determine the importance, size, or value of something.”

Does the draft meet these definitions?  Not well enough; not by a long shot.

River and watershed assessment, an eminently worthwhile effort, can provide essential information and analyses for rational conservation and management.  Probably, few states have undertaken so much work in that regard as Michigan.  MDNR began river assessment with admirable intent about 20 years ago, but reportedly, funding for the program has dwindled, it no longer has a central coordinator, and as the draft shows, its quality has suffered.

The draft follows standard MDNR format for river assessments, although subtopics may vary.  Its Executive Summary states the assessment program’s intent to compile information about the watershed and “demonstrate how the river is influenced by the physical landscape and the rivers [sic] relationship to biological communities” (bold is mine).  The document’s main section, labeled River Assessment, contains sub-sections on river and watershed characteristics (Geography, Geology, Hydrology, Biological Communities, etc.), and on Fishery Management, Recreational Use, and Management Options, as well as on some other largely human-related topics.

The draft’s sections on Biological Communities, Fishery Management, and Recreational Use should pertain to the river-resident fishes and fishery more than they do.  Also of concern for that fishery are many of the items in the section on Management Options, which the draft describes as “alternative actions that will protect, rehabilitate, and enhance the integrity of the river system. . . [and] are intended to provide a foundation for discussion, setting of priorities, and planning the future of the river system” (p 15, bold mine).  As I’ll discuss later, certain options are missing that could benefit the river-resident trout population and fishery.

After the assessment’s completion, the Fisheries Division will write a fisheries management plan that includes options chosen by the Division, based on its analysis and comments received (p 15).  I hope this means after completion of a much improved document, not just modest revision of the 2014 draft.

Prospects for removing or modifying dams on the main stem of the Boardman River provide impetus for the assessment.  About 20 or more dams have profoundly affected the river and its tributaries, some since the 1800s.  Two mainstem dams have already disappeared: the Keystone Dam, which washed out in 1961, and the Brown Bridge Dam, dismantled in 2012.  Three mainstem dams, Boardman Dam, Sabin Dam, and Union Street Dam remain.

General Comments

The draft contains a compilation of information but does not fulfill purposes stated or implied in the above bolded quotes, namely, of demonstrating how the physical landscape influences the river, demonstrating the river’s relationship to biological communities, and supporting the Management Options section’s purposes of serving as a proper foundation for discussion, for setting priorities, and for planning.  It is not properly ecosystem- and landscape-based.  Rivers and their organisms depend on the surrounding landscape, especially that of the riparian corridor, but the draft, as is common in stream fishery biology, takes an overly channel-constrained approach and does not even do justice to that.  Toward remedying this, I suggest that the MDNR Fisheries Division coordinate with wildlife biologists and foresters in developing the assessment.[3] Where does interaction with other government agencies, such as MDNR’s Wildlife Division and Forest Resources Division, enter into the river assessment?  Maybe I missed it.

For reasons such as those and many that follow, the draft falls far short as a basis for a comprehensive management plan.  One can expect drafts to show a certain amount of incompleteness, error, poor logic, and unclear wording, but the present document, apparently put together too hastily by too narrow a range of authors (undoubtedly burdened by many other duties), having too little checking, and lacking in editorial service and quality control by the Fisheries Division, is so deficient that MDNR should retract it and issue a revision improved by a team from several pertinent fields, not just fishery biology.[4]

I regret having to say that and hasten to add that the draft’s weakness must stem largely from decline in the funding for river assessment–and for MDNR in general.  Reportedly, MDNR attempted the assessment without sufficient resources. The root causes of poor funding may lie not only with parties who wish to impede the MDNR’s regulatory and management functions, but also with the public, with state legislators, and with other elected officials, whose appreciation of natural values has diminished, and whose commitment to conservation, science-based management, and support of the MDNR has shrunk.  MDNR has suffered deterioration of morale, departure of key personnel, and disintegration of programs, a tragic trend also taking place in Wisconsin.  In Minnesota, by contrast, the public rose up a few years ago and voted increased funds for their state natural resources agencies.

The draft lacks clarity in various respects and does not yet have enough meaningful analysis and interpretive discussion on many subjects, including key policy and management issues.  Adding further topics from recent advances in ecology and stream science would make the document more useful, drawing on substantial, readily-available sources of pertinent information and insight, some from Michigan studies, others from elsewhere.  In addition, further fieldwork should be done.

One elementary kind of fieldwork that the Fisheries Division apparently has not yet undertaken is getting acquainted with the whole channel and streamside features, for the draft does not show it.  Burroughs (2008) seems to have done this for part of the river, but the draft did not make much if any use of that information.  The draft doesn’t show that the Fisheries Division has examined its 9 electrofishing stations on the Boardman River in detail, but we would surely assume that it has and can see by instructions in the Stream Status and Trends Program (SSTP) sampling protocols (Wills et al. 2006) that it must have.  Those 9 stations, however, are each too short and cover in total only a small fraction of the river system’s 179 channel-miles.  I found no station lengths in the draft (it shows acres), but based on information in Burroughs (2008) and Wills et al. (2006), I estimate that the 8 sampling stations upstream from Boardman Dam cover only 1.2% of the about 155 miles of coldwater channel above that point.[5]  Or call it 2 or even 3% to allow for inaccuracy of my 155-mile estimation, and this still isn’t enough of a sample.  It’s miniscule.

So small a sample can’t be considered significant coverage of the channel, unless it could be shown that the sampling stations truly represent conditions for the whole river system or for just the coldwater part of it. I doubt that this could be shown–despite material in Burroughs (2008) which does not cover the river’s entire trout zone and may otherwise not be conclusive in all respects.  To properly sample the river-resident trout resource, MDNR would have had to select stations according to a randomized statistical design, and there’s no indication of this.  (I bring up further problems  of the sampling in comments on the draft’s Resident Brown and Brook Trout Section under the heading Fisheries Management.)

A complete visual survey of the stream is essential, and the draft gives no indication that the Fisheries Division has done this.  A biologist should, during autumn baseflow discharge, simply (yes, I mean simply because I’ve done lots of this) walk upstream along the whole mainstem and all tributaries, wading in the water wherever possible, preferably accompanied by a fluvial geomorphologist, and taking notes on stream and riparian features.  Covering 10 or more miles per day should be possible, hence perhaps a 2- or 3-week task for the 179 channel-miles unless interrupted by high water.

Doing this as a redd survey during the spawning season for brook and brown trout (perhaps about October 15 to December 7) and plotting redds on a map, or tallying them between major landmarks or breaks in stream characteristics will reveal where the most trout reproduction occurs.[6]  Knowing the spawning hotspots gives important insight for further investigation and management of the trout and their habitat.  (The word, redd, appears nowhere in the draft, by the way.)  It’s best for one biologist, preferably one of the assessment authors, to get a comprehensive impression of the system by doing the whole walking-wading survey him- or herself (accompanied by one or two others for at least parts of the walk), and then reporting it.  Having different people do parts of it results in a less consistent overview.

Assessment of a large, multi-faceted system like a river and watershed requires a team of experts from diverse fields.  The core of the 2014 draft has text on 13 “important ecosystem components.”  Further on, I will suggest splitting one of them and adding one or two others.  Those dealing with physical and biological attributes and processes are: Geography, History, Geology, Hydrology, Soils and Land Use, Channel Morphology, Dams and Barriers, Water Quality, Biological Communities, and Fishery Management.  Three others, Recreational Use, Special Jurisdictions, and Citizen Involvement involve primarily human aspects.  Having just two authors (both fishery biologists?) can’t be expected to result in proper compilation, analysis, and interpretation of information for all these topics.  The team should at least add a fluvial geomorphologist, a riparian plant ecologist, and a sociologist.  Others, such as a civil/environmental engineer, a water quality biologist, an upland plant (and soils) ecologist , and a stream benthos ecologist would be helpful.

The present document is superficial.  Many parts consist of mere compilations of information– uninterpreted fact-strings in the text, uninterpreted tabular lists, and uninterpreted maps, including a massive appendix of many maps, one for the distribution of each species of fish.  What are the meaningful characteristics of those distributions?  It isn’t informative enough to display river and watershed characteristics without discussing adequately, if at all, their significance–how those matters interact and affect ecosystem processes, fisheries, other biological resources, recreation, and so on.

Importantly in such respects, the draft fails to convert its information into clear diagnosis of various problems that certainly exist, that is, identification and analysis of them.  The Hydrology and Channel Morphology sections in particular need expansion and interpretation by professionals in those fields.  The topic of land use should be split from the Soils and Land Use section and treated in much greater detail regarding types of use and their effects on the ecosystem; also, some material that exists on land use elsewhere in the draft should be moved into this new section.  Numerous errors and inadequacies in the sections on Biological Communities and Fishery Management need remedy, as I discuss later.

Unclear wording pervades the draft.  Technical terms often appear as vague jargon rather than in precise senses.  Loosely used, for example, are the words habitat, productivity, diversity, stability, development, and indeed ecosystem.  This will cause perplexities and misconceptions, especially among lay readers of the document.  For public transparency, the assessment should define each technical term just before its first use (as well as in the Glossary), and then apply it throughout the text in specific senses that describe ecosystem features, processes, or conditions.  This will help readers understand the assessment’s material and will foster the informed, broad-based decision-making that should underlie policy and management.  In addition to serving MDNR personnel in their duties, the assessment should clarify issues for stakeholders especially, including the state’s conservation and outdoor recreation communities.

It would help in framing the subject of the river and its watershed if the assessment were to state at the outset–and reiterate throughout–that human actions (importantly land uses and introduction of exotic plants and animals) have significantly artificialized the Boardman River and its watershed, but that a wealth of natural areas and attributes remain, and that some of the damaged features can be helped to heal, although with respect to other features, processes, and conditions, undoing the legacy of human-caused harm won’t be practical.

The Introduction’s  first paragraph states: “We have approached this assessment from an ecosystem perspective, as we believe that fish communities and fisheries must be viewed as parts of a complex ecosystem.”  The next sentence, “However, this assessment is admittedly biased towards the aquatic components of this ecosystem,” and its manifestation in much of the draft conveys too limited a view, one not befitting a department of natural resources.  Ecosystem perspective is in keeping with exactly the right science-based approach to take, but it is only implicit in parts of the draft, whereas it should be made explicit throughout by discussing crucial relationships within the watershed’s ecosystem and sub-ecosystems–for example, specific effects of the various major land uses on certain biota and their communities and interspecific relationships within the communities.

For most of the public, “ecosystem” surely stands as a vague concept.  It would help to define at the outset what ecosystem means (not just in the Glossary, where it doesn’t yet exist, either).  The draft’s stage-setting statement that healthy aquatic ecosystems “have communities that are resilient to disturbance, are stable through time, and provide many important environmental functions” is important, although the term “stable” is outdated in ecology and could better be replaced by the concept that biological communities are always in flux and keep evolving in forms and functions.

A general characterization of the ecosystem at hand would usefully set the scene before the document goes into detail.  In a short paragraph, what are the Boardman River system and watershed like?  How do they compare with other river-and-watershed systems of the same region and of other regions in Michigan and North America?  Pointing out the well vegetated lowland setting and the strong, dependable groundwater-fed stream flow, as opposed to steeper streams having other kinds of water supply (e.g., mountain runoff), or streams that humans have severely damaged would provide useful context.  The Boardman River’s general characteristics strongly affect the biota and human uses, importantly the fishes and fishing.  The assessment should not assume that readers know the context.

The complexity alluded to in the draft’s introduction poses great difficulty in following through on an ecosystem perspective.  The complexity of an ecosystem, at least a major one like that of the Boardman River watershed, makes describing all its parts and workings impossible.  Ecologists say that no existing super-computer system could handle the task.  Therefore, river assessment must take a practical, reduced approach, but to so greatly exclude terrestrial biology as the draft does (lists of organisms won’t do) is inappropriate.  The draft treats most aquatic aspects too skimpily, as well. In taking an ecosystem perspective more fully, the assessment could repeatedly refer back to its initially-stated framework of fundamental ecosystem concepts when treating the most pertinent sub-topics (often sub-ecosystems) in useful detail.

The Introduction states (p 13) that MDNR river assessments “are based on ten guiding principles in the Fisheries Division Strategic Plan” (it’s not listed in the References), as follows, my comments in brackets:

1) recognize the limits on productivity in the ecosystem [what is “productivity”?];

2) preserve and rehabilitate fish habitat;

3) preserve native species;

4) recognize naturalized [exotic] species [intent of “recognize”?];

5) enhance natural reproduction of native and desirable naturalized fishes [“desirable” subject to debate];

6) prevent unintentional introduction of exotic species;

7) protect and enhance threatened and endangered species;

8) acknowledge the role of stocked fish [acknowledge?];

9) adopt the genetic stock concept, that is, protecting the genetic variation in fish stocks;

10) recognize that fisheries are an important cultural heritage.

These set the stage for the draft’s subsequent material, which ought to refer back to the principles more than at present.  The concept “productivity” in item 1 needs explanation.  It, together with “biological production,” has several, often confused meanings in ecology and fish or wildlife science,[7] depending on context.  When applied loosely, as in item 1, it appeals to the farmer in us but conveys too vague an idea.  Sounds great, but productivity of what and in what units?  The word occurs 5 more times in the draft. In most cases, its meaning could be made more specific, or more suitable words could be substituted, such as fish abundance, or what most anglers really seek, simply more larger fish of the species they desire.

The draft often applies the word habitat too loosely.  Habitat for what?  The word has little meaning and can lead to misconception by readers unless you specify the organism (brook trout, white cedar) or organism group (trout, raptors) to which it applies.  Showing the purpose that the habitat serves for the organism makes the term even more meaningful (trout spawning habitat).  People tend to say habitat when referring in general to a complex of features that favor an organism (water temperature, proper current, presence of pools and other hiding cover, taken together).  It’s best to spell those items out, where appropriate.

The same sometimes applies to use of the word diversity. The draft should specify diversity of what, and then what effect the diversity of that subject, stream property, or whatever has on the biota.  Also, some kinds of diversity within a biological community are not necessarily beneficial, for example, diversity for the vague sake of greater diversity that comes from introducing non-native species into an ecosystem–or for more diverse fishing opportunity, as a western state fishery manager once recommended.

Core Issue–Whether to Allow Upriver Migration of Lake Michigan Salmonids

Obviously, the Fisheries Division intends the assessment to serve as a case for letting Lake Michigan salmon and steelhead migrate into much more of the river system, removal of the Boardman and Sabin Dams being key to this.  Nationally, the general rationale for removing dams to benefit coldwater fisheries and stream ecosystems is well founded scientifically and well established in practice, so justification for removal of either one or both of the dams may require little assessment beyond feasibility and engineering studies, not anything approaching a full river assessment. Whether or not to allow upstream migration of the Lake Michigan salmon and steelhead, however, needs a lot of assessment and public discussion.

A case might be made for relatively fast but prudent benefit by proceeding as soon as possible–without river assessment–to remove the Boardman Dam, while continuing to block upstream passage of non-native Lake Michigan fishes by means of the Sabin Dam until later full river assessment and decision-making settle the upstream migration matter.  At that time, the decision might be to keep the Sabin Dam in place permanently or replace it with some other permanent barrier or eliminate it, opening the river to salmon and steelhead migration.  See also in the section on Management Options, below, a suggestion by the US Corps of Engineers to remove only the Sabin Dam.

The section on Management Options includes suggestions that, of the remaining three Boardman River main stem dams–Boardman, Sabin, and Union Street Dams–Boardman and Sabin Dams be removed to eliminate inundated stream reaches and help restore more natural characteristics and processes of the river, for example, hydrologic regime, rare high-gradient habitat, natural coldwater temperatures, and longitudinal aquatic connectivity for the biota.  This would also enable upstream passage of migrant salmon and steelhead from Lake Michigan.  A modification of the Union Street Dam, according to the options, would be left in place to block migration of non-salmonids.

Salmon and steelhead anglers, charter boat operators, and the MDNR Fisheries Division want Lake Michigan salmonids to migrate into as much as possible of the Boardman River system for reproduction and juvenile-rearing and to provide angling, but some anglers for river-resident brook and brown trout express concern that Lake Michigan migrants into the upper river and tributaries may harm their traditional fishery (Appendix F and Gary Marek, Traverse City, MI, personal communication).  In this regard, consider item 10 concerning “important cultural heritage” in the above list of guiding principles.

Obviously, removing the two dams will improve river conditions for coldwater fishes. Data in the draft and a large body of scientific literature support this.  But reasons exist to consider barring Lake Michigan migrants from the parts of the Boardman River system that they don’t now use, those reasons discussed at further points in my comments.  Removing the two dams and blocking upstream migration of all salmonids at the Union Street Dam would seem practical because one of the draft’s management options suggests using the Union Street site for blocking migration of other fishes , anyway; namely “Retrofit Union Street Dam to allow passage of select species, while preventing the migration of sea lamprey” (p 47).  To use this retrofit to block all non-native migrants from Lake Michigan–perhaps allowing passage of native fishes, such as lake trout (Salvelinus namaycush), certain whitefishes (coregonids), and lake sturgeon, species that used the river before dams blocked them–would be an option to consider.  We should keep in mind, however, that any selective-passage structure is subject to occasional failure in operation or maintenance, and by outright violation of management plans, as has happened in Michigan.[8]  Therefore, maintaining the Sabin Dam or the Boardman Dam without modification for fish passage should also be considered as an option, in order to preserve the river-resident trout fishery.

At this point, it occurs to me to simply ask: Has MDNR ever considered preserving some major trout river systems as refuges from Pacific salmonids that migrate from the Great Lakes?

Other Specific Comments

Among the draft’s many inadequacies, I see the following major, substantive shortcomings:

(1) It sorely needs a new section that deals thoroughly with the conditions, problems, and potentials of the river system’s riparian corridors, emphasizing their vegetational structure and function in particular; a plant ecologist should perform this task.

(2) Landform in the watershed requires discussion by a geomorphologist and could constitute a new section or a part of the Geography section;

(3) There is inadequate assessment of threats to river-resident trout and the broader ecosystem by contamination with toxic substances from Lake Michigan via migrating fish.

(4) Analysis and discussion of the river-resident brook and brown trout populations and fishery are inadequate and erroneous.

(5) The draft fails to consider a full enough range of potential effects by in-river life history phases of migrant salmonids from Lake Michigan, for example, superimposition of their redds on those of river-resident trout (dislodgement of embryos and sacfry) and effects of competition or predation by their age-1 and -2 juveniles (ranging perhaps from 5 to 8 or more inches long) on age-0, -1, or -2 river-resident trout. Consider that 5- to 8-inch steelhead can easily outcompete 1-to 2.5-inch age-0 brown or brook trout, as well as eat them.

(6) The draft fails to adequately identify and discuss such resource-stakeholder groups or communities as anglers for river-resident trout, Lake Michigan anglers (who seek stream-migratory fishes there), and Lake Michigan charter boat operators, whereas other stakeholders receive mention, for example, on p 43, paragraph 3.

(7) The Management Options section lacks various possible options for dam removal and blocking all Lake Michigan non-native migrant fishes from passage into the river beyond the point they now reach.

I will discuss these and other matters in the course of comments below.  Other reviewers, such as stream and watershed ecologists, hydro-geomorphologists, sociologists, and economists, might suggest other missing or insufficiently treated topics.

The Dams and Barriers section contains much discussion, as it should.  Most other sections contain less.

The draft refers to a 2005 Boardman River creel census, first on page 40 and many times thereafter, its data presented in Appendix E.  Previous MDNR creel surveys are briefly mentioned, their data in Appendix G.  It would be helpful to include description of the 2005 creel census method, perhaps as text on Appendix E’s title page, with reference to that in the body of the assessment.  Performing a valid creel census is difficult, especially where anglers spread out along a long river (although contact at parking areas reduces this problem), and where angling begins early in the morning and lasts late into the night.  Many of the skilled anglers do their fishing mainly in early morning or in late evening, and specialized midnight trout fishing occurs.  It is important to census early and late anglers to the same degree as other anglers, and in my experience, it can be hard to assure that census-takers patrol the stream early and late.  Sub-sampling can reduce the labor-intensiveness and expense of a full-time census, and if that is the case, then the statistical design must be described for readers to evaluate the procedure and results.  Even when full census is intended, some anglers will be missed, so the results must be adjusted for that and the adjustments explained.

Further specific comments follow the draft’s scheme of section or sub-section headings.

Executive Summary

The Executive Summary deals with some matters a bit more completely than the body of the document does.  Strange!

page ix – paragraph 5 – river divided into 3 major sections, based on a gradient of physical attributes – is this described later?

page x – paragraph 4 – lines 6-8 – “The Keystone Rapids are indicative of the high quality aquatic habitat that is currently impounded by the three dams.”  In what way(s) are the rapids “high quality habitat”?  Habitat for what organism(s) performing what life process(es)?  Rapids specialists?  Rapids provide scenic beauty and recreational navigation challenge, but what for which fishes and other organisms? Has MDNR measured fish abundance in any rapids?

page x – paragraph 5 – “Dams degrade aquatic species and habitat through . . . introduction of invasive species.”  It is illogical to say or imply that dams introduce species.

Geography – no comment

History– no comment


p 20 – Doesn’t the quotation belong in the Land Use subsection? Its sentences on lines 9-14 express dangers from oil and gas operations merely as “concerns” and describes safeguards in vague terms.  This exemplifies bureaucratic timidity.  The river assessment should discuss the kinds and severities of damage to the Boardman River ecosystem that oil and gas development can cause.  Catastrophic examples exist elsewhere, at least a couple of them in Michigan; namely, the oil spills (or spill and seepage) into the Muskegon and Kalamazoo Rivers. Even out here in Washington state I hear about these.


The document should emphasize that the Boardman River and other streams of this region in Michigan owe the high quality of their water supply for coldwater fishes to a much more favorable geologic and climatic situation than exists in most other states or regions, perhaps more favorable than in any other U.S. state.  The 100-1,000 feet of glacial sediment (highly permeable for the most part) that overlie the bedrock (p 20) provide for a superb groundwater aquifer that feeds the stream systems.  Depth of this sediment is probably even a bit greater. The area’s abundance of lake-effect rain and snow (p 21) from Lake Michigan supplies that aquifer but isn’t, I suspect, what “moderates spring and summer [water] temperatures” so much as does passage of the water through the aquifer.  In comparison, the similar sandy-gravelly glacial sediments that supply Central Wisconsin trout streams have a thickness of only 80 to 120 feet, and flow in those streams fluctuates more in volume and temperature.

Daily, Seasonal, and Annual Streamflows Subsection

Each of the three paragraphs in this subsection emphasizes and explains the Boardman River’s stable streamflow.  Why is relatively stable flow important?  How does the Boardman River compare with other streams in this respect?  How does relatively stable flow affect trout and other organisms?

p 21, paragraph 4 – “The average annual discharge for the Boardman River . . . is 111 cubic feet per second . . .”  (supposedly at the Ranch Rudolf USGS gage, but it should say so).  It’s of little use to know the average annual flow without comparing it to that of other parts of this river (a longitudinal baseflow “pick-up” profile) or to other rivers.  The information in Tables 7 and 8 about the variability or relative stability of flow serves well, but should be explained in the text.  Maybe adding a total flow exceedence curve (if that is the right term) would illustrate more clearly that flow is lower than the mean most of the time.  Data on annual summer and winter extreme low flows give much better indication of conditions for fish than average discharge does.  Fish must survive extremes and do not live so much according to averages.

p 22, paragraph 1 – For the sake of public readers, say what the jargon, river gradient, means; better yet, use plain English instead.  “Streambed slope” or “slope of the stream course” describe the situation more clearly for most people.

Now we come (paragraph 2) to that word diversity, hydraulic diversity in this case.  So what?  If hydraulic diversity is worth mentioning, explain why it matters ecologically, and specifically to the organisms of most interest.

Next paragraph, next to last line – “These data will be available in 2008 and will be incorporated into this document during the next revision.” Well then, where are they in this 2014 version?

Soils and Land Use Patterns

Combining these two subjects is unnecessary, even illogical in various ways.  It can generate confusion in considering the highly important topic of land use because much of land use, for example, urbanization, siting of vacation homes, transportation facilities, and power line routes, depends more heavily on factors other than soil type.  Splitting this section could help in more fully treating the subject of land use.

And why “patterns” of land use?  It is not just in its patterns that land use affects the river.  The very word, patterns, conveys superficiality, just a map view.

p 25 – paragraph 5 – This paragraph on the problem of perched culverts blocking of fish movement belongs in the draft’s section, Dams and Barriers.

p 26 – The Land Use sub-section – This topic deserves far more attention.  The lead sentence, “Land use practices within the Boardman River watershed directly affect aquatic habitat and species,” sets the stage for meaningful assessment.  The 2-paragraph section, however, simply lists amounts of various kinds of land use, land cover, and land ownership.  Nowhere does it assess how and to what extent the various kinds of land use affect or could affect aquatic habitat and species.

p 26 – Paragraph 1 says that agriculture occupies 17% of the watershed.  What kinds of agriculture – dairy farming, row crops, haying, or what? Which kind predominates, and what are its known or likely effects on the river?  Do significant legacy effects exist from past, perhaps more extensive farming?  This section should include discussion of such matters or present evidence that they lack significance.

Same paragraph – About “54% of the Boardman River frontage is commercially or privately owned.” What land uses have resulted, and what are their effects on the river?  Pollution by wastewater and by runoff from roofs, pavement, and other poorly permeable surfaces? Extensive replacement of natural plant communities with lawns?  Removal of trees and brush from the shore to allow river view (or facilitate boating?) is alluded to elsewhere in the draft (cutting) but should be discussed in the land use section.

The Land Use sub-section should deal with the matter of oil and gas wells and the roads, pipelines, other infrastructure and the traffic, pollution, and other ecosystem disturbance involved.



Water Quality

This section lacks standard data on chemical parameters of the river’s water–and interpretation of that information, especially in terms of its meaning for the biological community.  Has no agency undertaken basic chemical analysis of the river’s water?  It could be, for example, meaningful to show values of alkalinity, conductivity, phosphorus, nitrogen, calcium, and other elements, as well as toxic substances, in various parts of the river system during different seasons, particularly during summer base flow, and to discuss the implications for fish and other organisms.  See Table 2 on page 4 of Avery and Hunt (1981) http://digital.library.wisc.edu/1711.dl/EcoNatRes.DNRBull121for a pertinent display of data. Measurements of chemical and biological oxygen demand downstream from wastewater outfalls at critical times of day and night could reveal problems.

The Water Quality section should give much attention to thermal quality, especially to interpreting summer extremes of warm water and winter extremes of cold water (and ice conditions) in terms of their effects on the biota.  At present, it mentions temperature only once and seeming in just passing, and it does not refer to Appendix D, which contains water temperature data for the river.  Strangely, the draft’s Introduction and some other sections deal with water temperature more than the Water Quality section does.  Only the Fisheries Management section has reference to Appendix D.  The draft needs to be reorganized a bit with respect to water temperature material.

p 29 – Paragraph 1 reports exceedance of water-quality standards for PCBs in a river section just upstream from Boardman Lake–without pointing out that this section lies downstream from Sabin Dam, which salmonids migrating from Lake Michigan have not been able to pass.  Drawing on the abundant scientific literature that exists, the authors should indicate the most likely process(es) that put the PCBs in the contaminated section, discuss how those chemicals probably affect the biota, give reasonable explanation(s) why similar PCB levels were not found upstream of that river section, and discuss the likelihood that migration of Lake Michigan salmonids into the river will bring contaminants that harm the ecosystem (see further comment on the PCB situation in Fisheries Management section, below).

p 29 – Paragraph 2 refers to http://www.liaa.info/crabmp/.   Apparently, that website no longer exists. It does not show up on my computer.

Biological Communities

Huge neglect: the draft has insufficient attention to the characteristics and functions of vegetation in the ecosystem.  Plants compose a major part of the ecosystem, and the plant communities of the watershed, especially of the riparian corridor deserve much discussion.  Vegetation of the riparian corridor tremendously influences stream channel form and various kinds of habitats for fish.  Riparian vegetation also supports many food organisms for stream fish.

The concept of live vegetation occurs 37 times in the draft,[9] including duplicate references, but is accompanied by no real description of its functions.  State forest land is mentioned.  How is that land managed, and how does this affect the river?  Disregard for effects of vegetation and its management constitutes terribly insufficient treatment of an ecosystem component that matters much to the river and its fish.

The draft also fails to deal with beaver, except to list the animal in Table 9.  Beaver modify streams in ways that can tremendously benefit or degrade habitat for trout and other organisms, depending largely on the context of climate, landscape, vegetation, and hydrology.  Even if beaver may scarcely affect the Boardman River nowadays (is that true, and aren’t they significant in tributaries, at least?), they must have exerted much shaping of the watershed’s streamscape before trapping for the 17th-18th century fur trade almost exterminated them from this part of the continent.  What legacy of their former abundance remains in the riparian corridor and small tributary creeks, and to what extent have their populations recovered in the Boardman River watershed.  What is their present influence?  Future prospects?  The assessment should discuss this aquatic animal, key to various physical and biological processes in many streams.

p 31 – Paragraph 8 indicates that one should see the Soils and Land Use section for information about oil and gas development.  It is, however, in the Geology section that oil and gas development are covered.

p 32 – paragraph 1 – line 5 – Many readers will not know what the jargon, flashy, means.  Use a clearer word or phrase.

p 32 – paragraph 1 – last sentence: “Unnatural amounts of sand entering a trout stream like the Boardman River can inundate high-quality gravel habitat, thereby reducing trout populations and insect productivity.” Too equivocal.  Say whether or not you or other biologists believe this a significant problem in the Boardman River system, and present evidence for the opinion (measurements or other observations of sediments, trout reproduction, insect abundance, or whatever–even if in a similar river of the region). The document should discuss ways in which the sediments (of what size?) affect fish and other organisms, not just a blanket statement that they reduce populations and productivity.

p 32 – Current Fish Communities section – The text mentions that 59 species of fish have been recorded in the watershed; says they can be grouped as residents of the river, migrants from Lake Michigan, or dwellers of lakes in the watershed; and refers to Appendix C (1-page maps showing geographic occurrence of each species).  The text names 14 fish species or hybrids that live in lakes of the watershed.  The text also shows names of 9 river-migrant species but strangely no mention of any river-resident fish, not even brook, brown, or rainbow trout–except that brook and brown trout appear in parentheses as parental for the tiger trout hybrid.  A reader might conclude at this point that trout have minor status in the river-resident fish community–or are simply unimportant to MDNR.

The fish communities section typifies the draft’s general flaw of presenting simple facts without discussing their interrelationships (for example, the roles species play in their communities) and other significance.  Which species do humans use or otherwise especially value?  Which important species require what special habitats at certain life stages? Which species prey on or compete with which others?  Many more aspects probably deserve discussion.  That the assessment characterizes the assemblage of fish species as belonging within communities makes discussion of relationships especially appropriate.

In contrast, note, for instance, that the 1-paragraph sections on Arthropods and Birds (p 33) do talk, albeit briefly, about importance of those animals in the watershed and to humans.  Oddly though, the Birds section does not mention bird predation on fish, probably a major source of trout mortality in the Boardman River system.  The draft’s detailed list of birds (Table 8) includes herons, Osprey, Bald Eagle, and Belted Kingfisher, all significant fish predators.  And I wonder to what extent birds are really “a critical component of the watershed because they provide forage for species of . . . fish” (sentence 2).

The sections on Reptiles and Mammals also do not mention the well known predation on fish by northern water snake, otter, raccoon, and possibly other species, including black bear.

p 32 – paragraph 3 – line 6 – “niches” is the wrong word.  Change the sentence to: Some species occur in more than one of those groups.

Fisheries Management

p 36 – paragraph 1 – line 6 – “MDNR fisheries index station” appears here for the first time.  The term hasn’t been defined (and doesn’t exist in the Glossary).  What is it?  The next two paragraphs also contain the term.  Actually, most of the stations (or all of them?) are part of MDNR’s Stream Status and Trends Program (SSTP), which the body of the draft does not mention.  The assessment should show the relationship of its stations to the SSTP, briefly explain what the SSTP is, and refer to sources of SSTP information.

p 38 – paragraph 1 – lines 5-6  – “Due to the warmer water temperatures, this reach [Sabin Dam to Boardman Lake] appears to be marginal for the survival of other salmonid species, and natural reproduction and survival only occur in cooler summers.”  Which of the salmonid species at issue reproduce in summer?

Resident Brown and Brook Trout Section (p 39-41)

The draft fails to adequately discuss potential harm to the resident trout population from intrusion by Lake Michigan migrant salmonids as bearers of toxic contaminants and disease, as competitors, as predators, as spawning redd superimposers, etc.  The draft completely neglects the well known problem of genetic harm to wild salmonid populations (reduced fitness) via interbreeding with hatchery-bred conspecifics, for instance, interbreeding of the wild, stream-resident brown trout population with hatchery-bred brown trout that migrate from Lake Michigan.  I don’t know whether MDNR stocks hatchery brown trout in Lake Michigan, but Wisconsin DNR does.

As said in my previous comments, there’s no sign that MDNR has investigated brook and brown trout reproduction in the Boardman River system by surveying the distribution of redds.  This should be done for the river assessment.  It reveals the major and minor spawning areas and is easy to do by walking (wading wherever possible) the length of the channel and recording redds on a stream map (or recording counts of them between significant points along the channel) at 2- or 3-week intervals beginning mid or late October until the first week or so of December.  Surveying at intervals during the spawning season is needed to determine which redds the trout complete, and which diggings are just trials.  Even some sand-bedded areas can have redds, especially those of brook trout.  Brook trout redds often differ from brown trout redds in placement and form.

The assessment should express its data on trout density, both numerical density and biomass density (the latter often called standing crop), not only as amounts per unit of channel area, as done, for example, in Table 12 as number and pounds per acre, but also as amounts per unit of channel length (number and pounds per mile).  Density per mile often better reflects habitat quality than does density per acre because the trout depend so greatly on features of the stream banks, and most stream reaches have just two banks no matter how wide they are, so it shouldn’t be surprising when narrow reaches equal or exceed wider ones in fish density per acre.  At stream margins, cover is provided by undercut banks, by closely overhanging grasses, sedges, and bushes, and by downed logs and tree branches that typically lodge against current-bearing banks. Also, the food supply from terrestrial invertebrates falls close to the bank, and the drift of all invertebrates borne in the stream’s flow veers close to the current-bearing banks (lateral-scour pools with ledges and lodged wood or overhanging plants) that large trout favor as habitat, especially during daytime.

See Fausch (1984) on the advantages that trout derive from occupying such positions in streams.  I remember finding in the 1960s (I won’t take time to dig the data out now) that trout density per mile in the narrow upper reaches of streams was about the same as in the wider downstream reaches.  Part of this could be due to better spawning habitat in the upper reaches, but extensive electrofishing, angling, and other observation always revealed the great effect of stream-edge habitat on the spatial distribution of trout.  This seemed to apply more to brook and brown trout than to rainbow trout, which tended to occupy the upstream parts of rather open pools.  Showing trout densities per mile would add further meaning to the assessment.

The draft mentions pools and wood debris once or twice but has little substantive information on trout habitat.  As intimated in the previous paragraph, such features as undercut banks and closely overhanging vegetation are tremendously important for brook and brown trout.  Abundance of those features depends on the condition of the riparian plant community.  This is why the assessment should have a section that analyzes the riparian corridor in detail–and should show the amounts of important habitat features along the river system’s length.

The draft discusses Boardman River findings from a comprehensive MDNR multi-stream report (Gowing & Alexander 1980) in paragraph 2 (also in paragraph 3?–but not referenced), and should make greater use of that report, especially by comparing its results with more recent data and pondering reasons for differences.

Paragraph 3 – first 2 sentences – “Since 1985, the trout populations of the Boardman River have been intensively studied. Mark-recapture electrofishing surveys have been used to obtain population estimates for trout populations at nine stations in the Boardman River watershed (Table 12).”  On what basis was each of the nine stations selected, how representative is each of the larger area in which it lies, and what are the channel characteristics of each (length, mean width, bed forms, bed material, etc., as well as the acreage, which is shown)?  Channel characteristics can be shown in a separate table.

Without knowing the length of stream channel electrofished at each station, readers cannot determine how thorough the fish population study has really been.  It cannot have been extensive enough because, as I explained on page 3, the total amount of sampling covered only about 1%  to 2 or 3% of the trout water above Boardman Dam.  The calculation (again): the Boardman River system contains 179 miles of perennial stream (draft page ix), but on the maps in Appendix C it looks like brook and brown trout occupy about 90% of channel length.  Subtracting the 6.4 miles downstream from Boardman Dam and subtracting 10% of the remainder as non-trout water gives a result of 155 miles of trout water.  The sampling station lengths add up to only about 1.9 miles, which is 1.2% of the estimated 155 miles of trout-harboring channel length above Boardman Dam.  Even if, due to inaccuracy of  my 155-mile estimate, the percentage is off  by 100 or 200%, which could make it as large as a 3% sample, trying to draw conclusions from the MDNR station data resembles the parable of the blind men trying to find out what an elephant is like, each having touched only a much different part (trunk, toe, tail. etc.) of the strange animal.

The method(s) used in electrofishing, in recording characteristics of the catch, and in processing the resultant raw data can greatly affect the accuracy, precision, and interpretation of the population estimates derived.  It is difficult to perform the procedures for population estimates properly, and fisheries personnel in all states commonly do them improperly.  Reports of population characteristics cannot be considered credible without thorough description of the methods used to obtain data, process them, and make consequent calculations.  The draft contains no description of the methods used to generate the material shown in Tables12 through16.  SSTP protocol information (Wills et al. 2006) would indicate certain shortcomings of the methods that were probably used in the Boardman River electrofishing, for example, not waiting at least 2 days (for trout to regain physiological ability to respond to electric current) between the first and second sampling runs, not carrying captured fish back to the downstream end of the station before releasing them (to reduce upstream wandering), and not stratifying the data by length group when calculating the population estimates.  And to reiterate the point, station lengths were too short to obtain meaningful population data.[10]

The draft doesn’t refer to other reports that would show the required description of methods, and does not even show the dates of electrofishing.

Improper method can be seen in Table 12, which shows population densities (number/acre and pounds/acre) of brook, brown, and rainbow trout at the electrofishing stations, all ages combined for each species.  To show number of fish, all ages combined, represents the population so inadequately that unknowing readers will get false impressions, and those who understand the problem will say, “So what?”  The tremendous difference in body size between young (age 0 or 1) and older trout makes combining them in numerical expression of their population well neigh meaningless. Moreover, fish at different ages (hence different sizes, especially age 0 or 1 versus the rest) in the same population tend to exist at much different life stages, performing different ecological roles (habitat use, other behaviors, predator-prey relationships, etc.) that render them functionally different species.  Therefore, the population densities in Table 12 should, at a minimum, have been divided into those for age 0 and those for older fish (Table 15 indicates that the data for this exist).  The same criticism does not apply to population densities expressed as pounds per acre in Table 12, if they have been calculated properly, but we cannot know that because the draft lacks description of methods.

Because the draft does not describe the population estimate procedures, one cannot tell whether the estimate calculations were stratified by length groups of the fish, as must be done because the capture efficiency of electrofishing depends largely on body size of the fish.  Larger fish intercept more of the electric field (and are more visible and enticing to crew members), thus are more likely to be caught.  For example, given optimal electrofishing conditions (temperature, clarity, depth, current, conductivity of the water), good weather, proper gear, and a skilled and well motivated crew, the capture rate will typically vary from more or less 40% for age-0 trout (2.5-4.5-inch length) to more or less 80% for trout that are 18-20 inches long.  Therefore, we need a two-pass, mark-and-recapture method (or multiple-pass method) for the field work in order to account for variation in catch efficiency, and then calculation of the estimates by a method like the Chapman-Petersen formula mentioned in the Table 12 caption.  Because the draft fails to describe procedure, we cannot know whether or not data were stratified by size group for calculation of separate estimates (and then the sub-estimates added for the combined-age estimates in Table 12, which, as said, should not have been expressed that way, but could be usefully shown there in more stratified form).

Matters of the electrofishing operation’s timing and the crew’s technique can also affect validity of resultant population estimates, especially with regard to enabling physiological recovery of the fish (for proper response to electric field during the second pass) and minimizing their movement into and out of the sample station between electrofishing passes (which will be great, anyway, because the stations are so small).  We can’t tell what the field operation was like because the draft doesn’t describe it.

In addition, methods exist for calculating the statistical confidence intervals of mark-and-recapture population estimates.  Apparently, these were not calculated but should have been and then shown in the tables.

Regarding Table 14, many of the above comments also apply.  Note that the three streams that contained juvenile coho salmon had no brook trout (save for just 1/acre during just one of the nine population inventories), and that the three streams that lacked salmon had substantial brook trout populations.  During electrofishing of the 1970s in part of the Salmon Trout River (Upper Peninsula of Michigan) which contained juvenile coho salmon and juvenile brook trout, I found that the coho were much larger and behaved more energetically than the brook trout.  Therefore, Kurt Fausch and I undertook a study in eight Lake Michigan tributary streams and in laboratory-stream experiments to evaluate competition between juveniles of coho salmon, brook trout, and brown trout.  The results suggested that the larger size and competitive superiority of coho should give them advantage over brook and brown trout juveniles in tributary streams of the Great Lakes (Fausch & White 1986).  Note also the MDNR report (referenced in the draft) that juvenile coho salmon depressed abundance of brook and brown trout in three Lake Superior tributaries (Stauffer 1977).

The fact that the draft pays only scant attention to literature on the negative effects of coho salmon on brook trout and does not even acknowledge what its own Table 14 indicates in that respect constitutes major grounds for considering the draft deficient.  Indifference of the MDNR Fisheries Division to stream-resident brook trout fisheries?

p 40 – Paragraph 3 contains the judgment that “there are few older brown trout in the Boardman River,”  which some local anglers dispute according to their catches (Gary Marek, personal communication).  Maybe MDNR would conclude otherwise if it were to inventory the brown trout population by electrofishing a greater amount of the river.  In paragraph 4, the authors write about “lack of older, large brown trout” and claim: “Survival to older age classes is not occurring in most stretches of the river.”  But MDNR has not sampled the trout population in most stretches of the river, and much movement into and out of the unsuitably short sampling sections must occur–see the statement on page 40: “Data from the Ranch Rudolf station (Table 15) indicate that immigration into the station is taking place.” Therefore, calculations of survival from the available data cannot be valid.  Rather than survival, the year-to-year changes in Table 15’s values represent the changes in numbers caused by death, emigration and immigration, and the amount of each process cannot be known, so the table should be discarded as meaningless.

That survival of brown trout to older ages in the Boardman River is poor, and that older trout are unusually few for a brown trout population might be so, but I doubt it, and for the draft to state or imply that is unwarranted in view of the draft’s own information. “Few” older brown trout, based on what evidence and compared with what?  That judgment in the text relies on a finding that “[o]f 1,157 Boardman River brown trout that have been aged since 2002, only 37 have been older than age 3 (Table 13)” (actually, Table 13 shows that 1,148 brown trout were aged).  This would amount to a 3% proportion of older-than-age-3 fish in the population.  Actually, however, the proportion of older-than-age-3 fish is probably much less than 3%, which may not be unusually low for a brown trout stream and, in fact, may be rather high.

The deduction in the draft may be erroneous because proportions of fish in the true population’s younger age groups, age 0 in particular, are probably much higher than the samples in Table 13 represent.  This is, as said before, because electrofishing catches relatively few of the smaller fish, and perhaps also because crews selectively measure a greater proportion of larger fish, a common tendency. (The draft tells neither how MDNR selected the fish for age analysis nor the age analysis method.)  Anyhow, the probable under-sampling of smaller fish would make the proportion of older-than-age-3 brown trout smaller than stated.  Totaling the figures for age groups in the first part of Table 15 would probably give a closer but still not good approximation of the proportions (408 or 38% in age 0; 311 or 29% in age 1; 297 or 27% in age 2; 51 or 4.7% in age 3; and 14 or 1.3% of age 4 and older).  The proportion of age-0 fish in the Boardman River system’s whole brown trout population must be much higher than the 38% represented in my totaling from the data in Table 15, which represents only one river reach, the Ranch Rudolf sampling station.  The brown trout sub-population at Ranch Rudolf probably depends on recruiting age-0 and age-1 fish from areas where the main reproduction occurs. The text contains observation that “immigration into the station is taking place.”

Even a value of 1.3% or lower for proportion of age-4-and-older, however, probably should not constitute grounds for characterizing older brown trout in the Boardman River as “few.”  The apparent proportion of older brown trout in the Boardman River matches fairly well–or exceeds–that of a classic study in New Zealand, where a biologist estimated that hundreds of thousands of fry emerged from brown trout redds in a 7.4-mile stream, resulting 12,000 age-0 fish by the next summer; 3,000 age-1 in summer 2; 800 age-2 in summer 3; 220 age 3 in summer 4; and 60 age-4 in summer 5, the latter figure being just 0.5% of the age-0 abundance, and there may have been some still older trout that weren’t detected (Allen 1952).  Consider also that in 2 years of brown trout population estimates in greater-than-one-mile sections of four excellent Central Wisconsin trout streams, somewhat similar in landscape and hydrology to the Boardman River area, mean autumn age-4-and-older abundance of 11.5/mile was 0.4% of the mean age-0 abundance of 2,880 (Avery & Hunt 1981, Table 8).  Therefore, abundance of older-than-age-3 brown trout in the Boardman River may be perfectly normal or even better than average.

If the draft were to include comparison of its information on brown trout abundance with that in Gowing & Alexander (1980), some added insight might develop.

Table 15 contains a major glitch: it shows 100% survival of brown trout from age 0 to age 1 in 2002 and 2003.  That’s an impossible situation.  It can arise from inadequate sampling, chance error, and the movement of age-1 trout from other parts of the river into the section for which the age-0 population was estimated the previous year.  In view of the admitted trout immigration, trying to calculate survival from the available data  is futile.  Display of figures like these brings the credibility of the whole trout population study into question.

p 41- paragraph 2 – “The reason for the lack of older, larger brook trout in the Boardman River is unknown.”  The authors go on to discuss possible reasons and related matters.  They neglect to mention the well known possibilities that same-aged brown trout out-compete brook trout for space and food, that larger brown trout eat them, and that reproductive interference and other disruption occur.  Brown trout are displacing brook trout in many North American streams.  The fishery biology field has known this for many years.  See, for example, Waters (1983) and Sorenson et al. (1995).

Boardman River Dams Project Subsection

Paragraph 1 – next-to-last sentence – “. . . the optimal situation is a barrier that permits controlled migration so current and future managers will have the ability to limit or pass species as deemed appropriate.”  Many kinds of such artificial barriers occasionally fail due to lapse of maintenance or other human error.  And can MDNR  install a barrier that will block upstream migration during the worst foreseeable floods?  Note also that the phrase “optimal situation” in the statement demonstrates bias for enabling rather than blocking upstream migration of fish.

Paragraph 2 – The first sentence refers to Appendix F concerning pros and cons of allowing migratory salmonids further access to the watershed than they already have.  Appendix F contains a hypothesis presented in support of letting migrant salmonids access more of the Boardman River than at present.  I find material in Appendix F incomplete and the hypothesis unconvincing.  See my comments below under the heading for Appendix F.

Draft p 42 – last paragraph – last two sentences – “There are Michigan streams that support excellent populations of resident brook trout, even though they have large wild runs of migratory salmonids. Bear Creek, a larger tributary to the Manistee River, is one example (Tonello, unpublished data). Some reaches of the Jordan River also support abundant populations of both rainbow trout and brook trout (MDNR file data).”  These statements are based on unpublished, non-peer-reviewed material, therefore to be regarded with a grain of salt.  Again, the draft does not refer for comparison to its own Table 14.  Also, the text does not draw close contrast of those statements with the one near the end of the next paragraph that “. . . Stauffer (1977) suggested that juvenile coho salmon depressed abundance of brook and brown trout in three Lake Superior tributaries.”

p 43 – paragraph 2- The last two sentences briefly mention possible detriments from migration of Great Lakes salmonids into rivers.  Regarding “organic toxins such as PCBs in resident trout populations,” the draft merely refers to Janetski et al. (2012) and Zorn & Sendek (2001), but it should inform the public with a thorough discussion of the consequences to resident trout and the broader ecosystem, drawing on those two publications and other pertinent sources, such as Giesy et al. (1994, 1995–the lead author at Michigan State University), Freeman et al. (2002), Krümmel et al. (2003), O’Connor (2003), Missildine (2005), and Alexander & Pagano (2008).

The MDNR’s Au Sable River Assessment (Zorn & Sendek 2001) has, within its Fisheries Management section, a subsection, Considerations Regarding Upstream Passage of Great Lakes Fishes, that takes broad ecosystem perspective in dealing with pros and cons of the matter and includes an extensive discussion of the PCB issue.  Now, over a dozen years later, the Boardman River Assessment should have an even more extensive treatment of Great Lakes migrant fish considerations. It could well include the Au Sable River Assessment’s material on this subject (and some others) almost verbatim, and then build on it, including information from literature published in the meantime and other pertinent material that Zorn & Sendek (2001) did not happen to reference for whatever other reason.

I am told that PCB concentrations in Lake Michigan fish have fallen (Paul Seelbach, U.S. Geological Survey, Ann Arbor, MI, personal communication), and I see reference to this on the Internet, so the potential for PCB contamination by migrants from Lake Michigan may not be as great now as it was about 10 to 20 years ago, when some of the above references were published.  The assessment draft, however, does not contain that information, and the authors should contact Dr. Seelbach for published sources of evidence.  The reduced PCB levels in Great Lakes salmonids that migrate into Michigan streams may have eliminated immediate overt harm to eagles and others of the ecosystem’s top consumer species, but there may be risk of long-term problems, such as via genetic damage and persistent bioaccumulation.  The relatively recent publication, which the draft references but neglects to discuss (Janetski et al. 2012), may indicate that the PCB contamination remains a serious problem.  To sort that out is beyond the scope of my review (and I’m not a toxicologist); the MDNR should do that in the assessment via extensive discussion.

p 43 – Paragraph 3 covers other pros and cons of allowing access for salmon and steelhead throughout the river system.  Here, with regard to the cons, the text takes predominantly commodity and social orientations, to wit: “Allowing an upriver salmon or steelhead fishery to develop would likely create more fishing pressure at peak run times. While passing migratory salmon and steelhead to upstream reaches of the Boardman River could create a dynamic new fishery, it could also increase the potential for social conflict. The increased fishing pressure that might come with the new fishery could also be accompanied by an increase in conflicts with riparian landowners, fish law violations, and potential resource degradation from angler traffic. Chinook salmon runs in particular on some Michigan streams have a history of creating a number of law enforcement and social problems.” These are important issues, but the text of this section discusses neither effects on the Boardman River’s traditional fishery for stream-resident trout (or any other fishery) in terms of aesthetics and other  personally-experienced values of sport fishing, nor effects on broader ecosystem services that contribute to human quality of life.  Also regrettably, the draft does not cover in this section, other than previous superficial mention of the PCB problem, the matters I mention elsewhere (including re Appendix F) concerning possible deleterious effects of salmon and steelhead on river-resident trout populations via competition, predation. interference with reproduction, etc.

Management Options

The options within each subsection should be numbered in order to facilitate referencing.

Geology and Hydrology Subsection

Option 2 mentions “best management practices.” Management of what?

Channel Morphology Subsection

The introductory paragraph says adequate data on channel morphology are lacking.  Do they not exist in the papers by Burroughs?–to which I don’t have access.  If the data exist there, the assessment should use them; if they don’t, MDNR should make measurements before revising the draft.

The paragraph also says lack of adequate woody debris and pools may be a result of “historic peaking operations of the dams which would homogenize channel morphology.”  What about the former log drives and the present riverside tree cutting, both mentioned elsewhere in the draft?

Dams and Barriers Subsection

Add another option: Maintain the Union Street Dam to prevent upstream passage of any fish.

And add yet another option: Maintain the Sabin Dam to prevent upstream passage of any fish.

I also see the following option in US Army Corps of Engineers Detroit District (2014, page iii): “6. Remove Sabin Dam – This alternative provides habitat improvements for coldwater species including brook trout and dace. It removes Sabin Dam and allows a free-flowing river to be restored from the Boardman Dam to Boardman Lake including the draining of Sabin Pond.”


Water Quality Subsection

None of the options deals with preventing contamination of the river system with toxic chemicals.

Biological Communities Subsection and Fisheries Management Subsection

The Fisheries Management subsection lacks an explicitly stated option to maintain and enhance the Boardman River’s fishery for wild, river-resident trout–although option 2 in the Special Jurisdictions subsection has something vague along that line by stating: “Protect coldwater tributaries by designating appropriate reaches as Designated Trout Streams to ensure proper management and environmental protection.”

The introductory paragraph for the Biological Communities subsection mentions detriments from certain activities but fails to mention the adverse effects of introduced and invasive exotic species (which were touched upon elsewhere in the draft but should be reiterated here).

A new management option should be added: Manage state-owned parts of the riparian corridor to maintain optimal plant communities for trout habitat in the river channel, and encourage such management among other riparian landowners.

Contradictions among the following options (covered on page 49 of the draft) should be acknowledged and discussed, namely that the last three options (items 2, 3, and 4) listed below contradict the first one.

Under Management Options for Biological Communities:

(1) Option: Protect native and naturalized aquatic species from predation, competition, and habitat destruction from invasive species, by suppressing the spread and population expansion of invasives.

(2) Option: Restore the potential for fishes to migrate through the watershed by removing dams whenever feasible. (The migrating fishes–primarily the Pacific salmon and steelhead–would constitute invasive species, resulting in the detriments mentioned in Option 5.)

Under Management Options for Fisheries Management:

(3) Option: Restore aquatic connectivity of the Boardman River by removing Boardman and Sabin dams. (Same comment as above, except the problem could be solved by keeping the Union street Dam or other facility as a barrier.)

(4) Option: Enhance the fishery of the Boardman River by developing and operating a selective species passage barrier prior to the removal of Sabin and Boardman Dams. (Same comment.)

The following option should be added: Protect the river-and-stream-resident populations of brook and brown trout — and the sport fishing they provide — by maintaining a barrier to upstream migration of Lake Michigan salmonids after the planned removals of dams.

A further, more specific option should be added, as well: Maintain the Sabin Dam without modification for fish passage in order to preserve the river-resident trout fishery.

If anyone is interested in having a superb brook trout fishery in the Boardman River, add an option to eradicate brown trout from the river above Sabin Dam or some other barrier.  It can be done.  Wisconsin DNR has a program to rid streams of brown trout and restore pure brook trout fisheries.  Does Michigan?


Appendix A: Boardman River Dams Settlement Agreement

Does this document state the purposes of the proposed action, removing some dams?  In a quick skim of it, I don’t find such a statement.  What I see deals with performance of an activity, not why to do it or what the effects should be.  This seems an example of management by activity (MBA) rather than management by objective (MBO), as described in Drucker (1954) and Odiorne (1965), perhaps symptomatic of much that underlies the draft river assessment.

Appendix C

Its index (p 180) is labeled APPENDIX B INDEX and shows wrong page numbers for (all?) species.

How did MDNR determine the distributions of river-resident fishes?  Brief description of method needed.

Appendix F

Appendix F is a 2007 report by MDNR Research Biologist Andy Nuhfer (now retired), entitled “An analysis of Potential Effects of Upstream Passage of Steelhead on the Boardman River.”  It presents findings from a several-year evaluation of effects of steelhead on brown trout in Hunt Creek, Michigan, and from data from other Michigan streams.  Nuhfer concludes that “significant adverse effects of steelhead on resident trout populations in the Boardman River are unlikely to occur, and if there are unwanted effects they could be mitigated by limiting the numbers of adult steelhead passed upstream.”  I raise the following issues (others may exist) in connection with appendix F:

The report in Appendix F gives no indication of having been peer-reviewed, so its material should not be considered to carry much weight until it has gone through that process and been revised accordingly.  Standard peer-review consists of scrutiny and comment by at least 3 independent reviewers (which in this case would mean persons outside MDNR), recognized as competent within the profession (in this case trout and salmon scientists), followed by revision of the paper, overseen by an independent editor of research publications.

The report does not refer to any other studies.  Pertinent research has been done elsewhere and published in peer-reviewed journals, for example, Hayes (1988, 1989), Kocik & Taylor (1995), Scott & Irvine (2000), and Taniguchi et al. (2000); some of these may support the report’s findings, others may conflict with them, and still others may augment them.  Some of the above references deal with redd superimposition, an important related issue.

The analysis deals only with some possible effects of steelhead that would migrate into and spawn in the Boardman River if allowed to do so, but effects of other “wild naturalized” and hatchery-reared Pacific salmonids, Chinook and coho salmon are also at issue in some respects.

The report addresses relationships between steelhead and river-resident trout only in terms of possible interactions between age-0 fish (called young-of-the-year or YOY in Appendix F) in a trout stream into which steelhead do not migrate but where adult steelhead (hatchery or wild origin?) were stocked in an experimentally blocked off section–the treatment zone (TZ)–to spawn, thereby producing age-0 steelhead to interact with resident brown and brook trout.  The report either ignores or fails to point explicitly to possible interactions other than between age-0 fish, other interactions which could include (but not be limited to) competition by age-1 and -2 steelhead with age-0, -1, and -2 resident trout, superimposition of steelhead redds on those of resident trout, and contamination of resident trout with toxic chemicals that steelhead bring into the Boardman River from Lake Michigan.

Juvenile steelhead commonly remain in streams for 2 or 3 years (Hart 1973, KRIS 2011, among many other sources), that is, during ages 0, 1, and 2, and may even spend up to 7 years in freshwater before migrating to estuarine areas as smolts (NOAA 1014).   Therefore, competition at least by age-1 and -2 steelhead with river-resident trout of ages 0, 1, and 2 must be considered besides that between age-0 fish, the only type that the Appendix F study addressed.  In addition, as already mentioned, age-1 and -2 steelhead probably prey on age-0 brown and brook trout. Steelhead juveniles 5 to 8 inches long can eat brown or brook trout that are only 1 to 3 inches long.

The Appendix F report (p 261, paragraph 4) states: “Density of cohorts of yearling-and-older brown trout in Hunt Creek that interacted with steelhead as young of the year (YOY) was about half that of allopatric cohorts [those not having been subjected to steelhead],” and the report attributes that situation to the finding that “mean annual survival of (YOY) brown trout declines from 38% to 23% when YOY steelhead were present.”  The report does not tell whether age-1 or -2 steelhead remained in the TZ, of if they did not, whether (as I suspect) they were artificially removed.  In any event, if presence of age-0 steelhead reduced the population density of brown trout, as stated in the first sentence of the above quotation, must this not be considered a negative effect on the brown trout?

When we consider the redd superimposition issue, which neither the report nor any other part of the 2014 assessment draft does, note that the report says that adult steelhead were stocked into the TZ each spring (to produce age-0 offspring to interact with age-0 brown and brook trout spawned the previous fall), but it does not tell how many were stocked, the sizes of these fish, nor the dates of stocking.  If steelhead spawners were stocked before all emergence of brown and brook trout juveniles from redds occurred, then redd superimposition could be an issue in the experiment.  And in any event, the report fails to take into account what the redd superimposition effects of steelhead would be under free-migration, rather than in the TZ’s experimental stocking situation.

The questions arises about whether and to what extent the redd-digging period of Lake Michigan steelhead overlaps the intragravel period of brown and brook trout–their incubation and sacfry period, which may last into February or March. If steelhead dig redds on top of stream-resident trout redds that contain developing eggs and sacfry, those reproductive products will be dislodged.  If steelhead dig beside or not too far upstream from those redds, smothering or entombment with fine sediments can become a problem.

Although wild steelhead that haven’t had hatchery influence in recent generations might not begin spawning until March or April, those in the Great Lakes apparently have been changed so much by generations of breeding in hatcheries for earlier spawning (to enable earlier release) that some now spawn in fall.  A Michigan DNR website reports that: “Great Lakes steelhead enter their spawning streams from late October to early May. At the present most spawning occurs in the spring, although more steelhead are beginning to spawn in fall” (Michigan Department of Natural Resources 2014).[11]

Also, the report does not say what share of the suitable gravel spawning area that steelhead redds covered in the TZ, or whether the investigators saw any superimposition.  Perhaps the density of steelhead spawning was so low (atypically low for Great Lakes steelhead streams?) that few of the redds would have been dug into brown or brook trout redds.

Appendix F does not adequately consider effects on brook trout.  The analysis focuses mainly on relationships of age-0 steelhead and age-0 brown trout, not on age-0 or other brook trout, but reveals that: “With a few exceptions such as Bear Creek (Manistee County) rainbow trout were rarely abundant in the same streams where brook trout were found in significant numbers” (p 262, paragraph 4, my italics), which is consistent with negative effects of steelhead on brook trout.  Appendix F’s observation that “Some of the highest standing crops of resident trout in the state are found in streams receiving runs of potamodromous salmonids (Figure 8)” ignores the situation for brook trout in Figure 8’s streams.  If the standing crops of brook trout were plotted against those of steelhead, we might see a negative relationship.

Nuhfer hypothesizes (p 262, paragraph 6) that “adverse effects of YOY steelhead on survival of YOY brown trout are less likely to occur when densities of brown trout YOY are lower than in Hunt Creek, as they are in most streams, including the Boardman River.”  As neither a complete electrofishing survey of age-0 trout in the Boardman River system exists, nor apparently does any survey of spawning redds, one cannot know whether the Boardman River’s (as said, unsuitably short) sample section or sections that Nuhfer included in his analysis (Figure F.4) are where major brown trout reproduction occurs.  Some other parts of the Boardman River system may have age-0 (YOY) brown trout densities similar to or greater than in Hunt Creek, but we cannot know that without complete survey of the system’s trout reproduction. Similarly, we cannot know from the information presented whether or not other streams in the Figure F.4 analysis have significantly higher age-0 densities than other parts of the same stream (primary sources for recruitment to the rest of the stream), hence may be subject to the density-dependent detriment of adding steelhead.  To the extent this is so, the analysis and the hypothesis deriving from it are questionable.

In view of these considerations, the Appendix F hypothesis seems unconvincing as support for letting migrant salmonids access more of the Boardman River than at present.



Appendix G

The term Catch Per Effort appears in the introductory paragraph and in table captions.  The unit of effort, an hour of angling or whatever, should be stated.


According to much of the 2014 draft, the MDNR Fisheries Division seems to intend the Boardman River Assessment largely as support for enabling Pacific salmon and steelhead to migrate from Lake Michigan throughout most of the river system, but the material presented is inadequate.  Although in various respects the document may try to take the ecosystem perspective stated at the outset, it is ineffectual in doing that.  The draft is so superficial and so deficient in other respects that MDNR should retract it and have a diverse team from pertinent professions develop a greatly expanded and improved revision.  This will require added survey of available information, added field work on the river and in its basin, and much greater analysis and synthesis of material. A panel of independent representatives of those professions should review and comment on the revised draft.

The draft is superficial not only because its material is incomplete, but because it has compiled information without interpreting it.  To quote Einstein: “Information is not knowledge.”  Someone expanded on this by pointing out that information as such can only be passed on; it must be converted into knowledge before it can be used, and it is converted into knowledge by relating it to its context–oneself, one’s experiences, environment, and other surrounding information.  Others have gone on to observe that knowledge is not wisdom.

In regard to fisheries, the draft takes a predominantly commodity outlook, rather than one based on ecosystem services and on the sporting and aesthetic values–indeed enjoyment–involved in human use of the river system. That disregard seems ironic in a state which has traditionally taken pride in–and energetically advertises–the recreational fishing opportunities that its waters offer.  The inadequate treatment of river-resident trout fishery issues, as seen, for example, in the draft’s failure to show management options applying directly to the recreational fishery, may signal a degree of inappropriate disinterest ingrained in the MDNR Fisheries Division’s culture.


Alexander, C. M., & Pagano, J. J. (2008). Evaluation of persistent contaminant levels in Great Lakes salmon eggs. In Proceedings of QUEST 2008 (pp. 15-23). Oswego, NY: SUNY.

Allen, K. R. (1952). A New Zealand trout stream – some facts and figures (Fisheries Bulletin No. 10A). Wellington, NZ: New Zealand Marine Department.

Avery, E. L., & Hunt, R. L. (1981). Population dynamics of wild brown trout and associated sport fisheries in four central Wisconsin streams. (Technical Bulletin No. 121). Madison: Wisconsin Department of Natural Resources. http://digital.library.wisc.edu/1711.dl/EcoNatRes.DNRBull121

Burroughs, B. A. (2008). Boardman River feasibility study: a report on the Boardman River fisheries habitat survey and data collection. Report of Environmental Consulting and Technology Inc. to the Boardman River Dams Committee, Traverse City, Michigan.

Burroughs, B. A., & Hayes, D. B. (2007). The rehabilitation potential of dam removal: temporal perspective from Michigan’s past dam removals. Chapter Three of a 2007 Dissertation entitled: Effects of Dam Removal on Fluvial Geomorphology and Fish. East Lansing. MI: Department of Fisheries & Wildlife, Michigan State University.

Dewberry, T. C. (1992). Protecting the biodiversity of riverine and riparian ecosystems: the national river public land policy development project. Transactions of the 57th North American Wildlife and Natural Resources Conference, 424–432.

Drucker, P. F. (1954). The practice of management. New York: Harper & Brothers.

Fausch, K. D. (1984). Profitable stream positions for salmonids: relating specific growth rate to net energy gain. Canadian Journal of Zoology, 62, 441-451.

Fausch, K. D., & White, R. J. (1981). Competition between brook trout (Salvelinus fontinalis) and brown trout (Salmo trutta) for positions in a Michigan stream. Canadian Journal of Fisheries and Aquatic Science, 38, 1220-1227.

Freeman, R., Bowerman, W., Grubb, T., Bath, A., Dawson, G., Ennis, K., et al. (2002). Opening rivers to Trojan fish: The ecological dilemma of dam removal in the Great Lakes. Conservation in Practice, 3(4), 35−40.

Giesy, J. P., Bowerman, W. W., Mora, M. A., Verbrugge, D. A., Othout, R. A., Newsted, J. L., et al. (1995). Contaminants in fishes from Great Lakes-influenced sections and above dams of three Michigan rivers: III. Implications for health of bald eagles. Archives of Environmental Contamination and Toxicology, 29, 309−321.

Giesy, J. P., Verbrugge, D. A., Othout, R. A., Bowerman, W. W., Mora, M. A., Jones, P. D., et al. (1994). Contaminants in fishes from Great Lakes-influenced sections and above dams of three Michigan rivers. I: concentrations of organochlorine insecticides, polychlorinated biphenyls, dioxin equivalents, and mercury. Archives of Environmental Contamination and Toxicology, 27, 202−212.

Gowing, H., & Alexander, G. R. (1980). Population dynamics of trout in some streams of the northern lower peninsula of Michigan. (Fisheries Research Report No. 1877). Ann Arbor: Michigan Department of Natural Resources.

Hart, J. L. (1978).  Pacific fishes of Canada. (Bulletin No. 180). Ottawa: Fisheries Research Board of Canada.

Hayes, J. W. (1988). Mortality and growth of juvenile brown and rainbow trout in a lake inlet nursery stream, New Zealand. New Zealand Journal of Marine and Freshwater Research, 22(2), 169 – 179.

Hayes, J. W. (1989). Social interactions between 0+ brown and rainbow trout in experimental stream troughs. New Zealand Journal of Marine and Freshwater Research, 23(2), 163 – 170.

Janetski, D. J., Chaloner, D. T., Moerke, A. H., Rediske, R. R., O’Keefe, J. P., & Lamberti, G. A. (2012). Resident fishes display elevated organic pollutants in salmon spawning streams of the Great Lakes. Environmental Science Technology, 46(15), 8035–8043.

Kalish, T. G., & Tonello, M. A. (2014). Boardman river assessment (Special Report Draft). Lansing, Michigan: Michigan Department of Natural Resources, Fisheries Division.

Kocik, J. F., & Taylor, W. W. (1995). Effect of juvenile steelhead (Oncorhynchus mykiss) on age-O and age-1 brown t rout (Salmo trutta) survival and growth in a sympatric nursery stream. Canadian Journal of Fisheries and Aquatic Sciences, 52(1), 105-114.

KRIS. (2011).  Hypothesis #2: The lower and middle mainstem Garcia River are in recovery with regard to its salmonid juvenile carrying capacity.  Krisweb http://www.krisweb.com/krisgarcia/krisdb/html/krisweb/analysis/hypothesis_2_garcia.htm.

Krümmel, E. M., Macdonald, R. W., Kimpe, L. E., Gregory-Eaves, I., Demers, M. J., Smol, J. P., et al. (2003). Delivery of pollutants by spawning salmon. Nature, 425, 255−256.

Landergren, P. (1999). Spawning of anadromous rainbow trout, Oncorhynchus mykiss (Walbaum): a threat to sea trout, Salmo trutta L., populations? Fisheries Research, 40(1), 55 – 63.

Lincoln, R. J., Boxhall, G. A., & Clark, P. F. (1993). A dictionary of ecology, evolution and systematics. Cambridge, UK: Cambridge University Press.

Michigan Department of Natural Resources. (2014). Steelhead, Oncorhynchus mykiss. Website: http://www.michigan.gov/dnr/0,1607,7-153-10364_18958-45692–,00.html

Missildine, B. (2005). Salmon carcass deployment: A potential pathway for PCB contamination. Fisheries, 30(1), 18-19  http://www.readabstracts.com/Mass-communications/Salmon-carcass-deployment-A-potential-pathway-for-PCB-contamination.html#ixzz36Td86037.

NOAA. (2014).  Steelhead Trout (Oncorhynchus mykiss).  Office of Protected Resources, NOAA Fisheries. http://www.nmfs.noaa.gov/pr/species/fish/steelheadtrout.htm.

O’Connor, A. (2003, September 23, 2003 ). Spawning salmon haul toxins to Alaska lakes, experts find New York Times http://www.nytimes.com/2003/09/23/science/23SALM.html.

Odiorne, G. S. (1965). Management by objectives; a system of managerial leadership. New York: Pitman Pub.

Scott, D., & Irvine, J. R. (2000). Competitive exclusion of brown trout Salmo trutta L., by rainbow trout Oncorhynchus mykiss Walbaum, in lake tributaries, New Zealand. Fisheries Management and Ecology, 7(3), 225 – 237.

Sorensen, P. W., Cardwell, J. R., Essington, T., & Weigel, D. E. (1995). Reproductive interactions between sympatric brook and brown trout in a small Minnesota stream. Canadian Journal of Fisheries and Aquatic Sciences, 52, 1958–1965.

Stauffer, T. M. (1977). Numbers of juvenile salmonids produced in five Lake Superior tributaries and the effect of juvenile coho salmon on their number and growth, 1967–1974 (Fisheries Research Report No. 1846). Ann Arbor: Michigan Department of Natural Resources.

Taniguchi, Y., Miyake, Y., Saito, T., Urabe, H., & Nakano, S. (2000). Redd superimposition by introduced rainbow trout, Oncorhynchus mykiss, on native charrs in a Japanese stream. Ichthyological Research, 47(2 ), 149 – 156.

Taube, C. M. (1975). Abundance, growth, biomass, and interrelationship of trout and coho salmon in the Platte River (Fisheries Research Report No. 1830). Ann Arbor: Michigan Department of Natural Resources.

US Army Corps of Engineers Detroit District. (2014). Feasibility study for the Boardman River, Grand Traverse County, Michigan (Environmental Assessment, Great Lakes Fisheries and Ecosystem Restoration Program). Detroit, MI: US Army Corps of Engineers.

Waters, T. E. (1983.). Replacement of brook trout by brown trout over 15 years in a Minnesota stream: production and abundance. Transactions of American Fisheries Society, 112, 137-146.

Wills, T. C., Zorn, T. G., & Nuhfer, A. J. (2006). Stream Status and Trends Program sampling protocols. Chapter 26 in Schneider, James C. (ed.) 2000. Manual of fisheries survey methods II: with periodic updates (Fisheries Special Report No. 25). Ann Arbor, MI: Michigan Department of Natural Resources.

Zorn, T. G., & Sendek, S. P. (2001). Au Sable River Assessment (Fisheries Special Report No. 26). Ann Arbor: Michigan Department of Natural Resources.


[1] 320 12th Avenue North, Edmonds, WA 98020; rw@seanet.com; Phone: 425-672-8268

[2] http://www.merriam-webster.com/dictionary/assessment

[3] I’ve observed that wildlife biologists and managers take a strong landscape ecology approach in their work, with great attention to plant communities, as one would expect in work on terrestrial animals, many of them herbivores, and most of them depend on vegetation for food, shelter, and concealment.  Often foresters and range managers cooperate in this work.  Stream fishery biologists and managers could achieve much by involving these other professions in their work and adopting new outlooks from them.

[4] According to an MDNR contact, the authors and acknowledged associates all have their main background in biology.

[5] I roughly estimate the Boardman River system’s total trout zone above Boardman Dam as 155 miles–the system’s entire 179 miles, minus the 6.4 miles below the dam, minus 10% (17.3 miles) for non-coldwater in the remainder.  According to Burroughs (2008) the lengths of 6 of the stations (one being a bit more than thrice the 1000-ft protocol length, and the others being near it) total 8,193 ft, and if the other two stations each have the standard 1000-ft length, then the total sampled length above Boardman Dam is 10,193 ft or 1.93 miles–which is 1.2% of the 155 miles.

[6] I give further detail on redd survey in comments in the Fisheries Management section.

[7] Rate of carbon fixation; potential rate of incorporation or generation of energy or organic matter by an individual, population or trophic unit time per unit area or volume; and other, more specific meanings (Lincoln et al. 1982).

[8] In exuberance at the outset of Michigan’s introduction of coho salmon into Lake Michigan tributary streams in 1966, fish distribution personnel stocked some of the fish in undesignated streams on their way to the three, and only three, streams where they had been authorized to plant the coho salmon (Howard Tanner, personal communication, ca. 1972).

[9] The words plant, vegetation, and vegetate appear only 8 times in the text, the word tree twice, reference to specific types of tree (maple, oak, pine, etc.) 17 times, and the word hardwood or conifer 10 times.  Grass, bush, brush, and sedge receive no mention. The word forest occurs about 7 times other than in the many references to state forest land.

[10] The SSTP protocol specifies station length of 1,000 ft (305 m), about what most of the Boardman River stations were.  Trout movement into and out of such a short station between the first and second electrofishing run will often render a population estimate invalid.  Standard contiguous sampling sections for trout surveys in some other states, for example, Wisconsin and Montana, are one to several miles long.

[11] In Montana, I’ve seen rainbow trout (certainly of hatchery ancestry) spawn in November.