| [ HOME | Forest Faunal Systems Home | Table of Contents | Gamma Home | The Finder | Glossary ] |
Decisions and Decision Making This chapter is land and it has been split into three parts. You are now in Part 1. You may go to Part 2 or to Part 3.
I worked for many years on a systems approach to wildlife management and assumed that most of the concepts of population and habitat management were concepts that could be called system processes. I had jumped too quickly to the animals and their homes. This is a book about decisions, not about animals. It is about resource management, not animal biology or ecological complexities. There is a need to understand the decision process itself and to understand it well, for then it can be applied to the old as well as not-yet-conceived aspects of this dynamic field. I knew that wildlife management was primarily decision making and then I realized that I had not addressed it specifically. I treated the parts of the system, chapter by chapter, but omitted "processes."
I remember the day that I read that statistics was for decision making. Of course!, and since wildlife management is decision making, a primary tool is statistics. Somehow that viewpoint had been lost to me (and many of my students) among the topics of data, esoteric symbols, and the need, it seemed, for squaring everything. I can only continue to encourage involvement in learning more of that broad field. There are other aspects of decision making that can be learned. It is not a mystical art form.
Decisions are little systems. You take a systems approach to decisions. You analyze them as systems and use the parts, with their techniques, to improve outcomes. The people recognized as good decision makers are those who have a pattern in their decision making activity. Odds are that the pattern will be easily analyzed as being like a "systems approach."
Most faunal system decisions are difficult. They have evident conflicts. Something that improves conditions for species x degrades the conditions for species y. What shall I do? Decisions in this field are not only difficult because they are complex (ecologically, energetically, economically, esthetically) but because they place the decision maker in a special social environment. They allow people to name the decision maker, to allow him or her to be classified. This is one of the consequences of non-trivial decisions (like deciding to get a glass of water or not). That decision making has costs and risks cannot be ignored. It is the chief reason why people are paid. The higher the risks, the higher the rewards.
In public agencies, the concept of the responsible decision maker slips away as more decisions seem to be made by committees and boards. Decisions are made by a person in an instant (no matter how long the decision making system has taken). Fixing responsibility, naming the person who will receive the praise or the blame (the profits or the losses) for the decision, is part of defining the system context, of naming the subsystem.
Some people start analyzing situations in which a decision seems needed and assert that the problem has to be identified. My analyses suggest that a problem is seen in the difference between objectives and the current situation. (I described that difference, the gap, in Chapter 4.) To improve decisions:
There is a tendency to list only three alternatives. How three became the standard, I know not! In many situations, I have found that forcing myself to seek out a fourth alternative (one for another vertex of the tetrahedron) has provided important insights and advantages. Realize too, that the computer can evaluate thousands of large and complex alternatives rapidly. To surrender to "three" almost assures that the best one available (somewhere) is not selected.
Information on possible secondary and cumulative effects is part of this input part of the decision system. It also includes listing the alternatives. The public, general advisors, and local experienced people can aid in producing this list.
"Sure" is an expression of risk, (i.e., 1.0 - the probability of failure).
All decisions are made with an element of risk; there is no certainty (I shall not discuss religious certainty). There are risks of physical system failure such as floods, storms, insect attack, and simple failure to perform as expected (as from fertilization or when a new variety of plant is tried on an area.) There are harvest failures (e.g., hunting success), social disruption (e.g., unseen secondary consequences of a road crossing, providing road access, or changing a season), and personal risks (e.g., reputation, promotion, opportunities for later tasks). In the private sector, risks can be tallied as direct loss of investment dollars. Accounting risks within public agencies is more subjective but needs to be done. Estimates need to be attempted.
Use expected value as described in Chapter 4. Multiple the best estimate of the likely consequence(s) by (1.0-risk) (resulting in the expected value and select the alternative with the highest value. The value is not "real"; it is a bunch of estimates, aids to making decisions. The value is a guide, not the decision it self. There are likely to be other things that say the alternative with the highest expected value is the wrong one to select. If so, explain why, insert the reason as an objective (something to be avoided) and start the procedure over again at a reasonable point.
 |
| Watersheds are unique. High variance prevents adequate sample sizes for such areas. Alternative strategies are needed. |
|
|
| Fig. 7.1. The manager exists in the force-field created by the interplay of three factors that are decided. Efforts to move as far away as possible from a zero value are opposed by time, available funds, tendencies to increase variance, needs for equipment, trained personnel and the difficulties of analyzing and communicating the results. |
As you read further, you may want to return to a place in the text well past this place in the unit. Ignore this statement for now; click on it later to return easily and quickly to a place later in the text.
The proper location in this tetrahedral space is partially the manager's decision, partially dependent upon what administrators have been taught, what reviewers expect in journal reports, and, as always, available time, technology, and budgets. Precision is the category for measurement devices and metrics. The use of a millimeter rule is possible for measuring an area boundary; so is pacing. The ruler is very precise; pacing very gross or imprecise. The manager needs to decide.
Max and Snellgrove (1952) commented that the consequences of choosing a precision level for measurement can be critical. If too high, measurements may be too costly and time consuming to make. (See also Hamilton 1979.) If too low, very imprecise measurements may result in variability that contributes to the overall expression of error in statistical analyses.They observed that levels based on experience can stand questioning and new problems arise for which there is no experience. They presented a method - an analysis of the contribution of the maximum measurement error to the maximum total error - for selecting a level of precision. However, even it requires subjective comparisons of the resulting ratios.
A poorly trained or disinterested field assistant may not make measurements very accurately, no matter what unit is selected. Accuracy expresses the difference between the true measure and that observed and reported. A well-intentioned observer may be biased for many reasons. Bias may be purposeful (an observer wanting to "show something" like there not being enough forage for a species or there being too many resource users). Bias does not need to be purposeful. An instrument may malfunction at some levels, an observer's sight or hearing may not be appropriate for the assigned tasks, instructions may not have been understood, or there may be misperception (as in optical illusions). Estimating the percent covered by vegetation in a square plot is a typical situation where bias is frequent. It can be "trained away" but frequently it is as useful to determine what the bias is (e.g., by having the observer look at 30 plots of known percent coverage). He or she may report what was observed. A comparison is made and, for example, they consistently report 0.93 of the true coverage. Then all of their field work is divided by 0.93 to bring it to the true value. Whether being 0.93 from the true value is too inaccurate is a decision for the manager. It depends ... on many factors. "Close enough" is a frequently heard expression.
Habitat or Faunal Space
"Habitat" is no longer a useful word in faunal system management (Moen 1973, Coulombe 1978, Hoover and Wills 1987, Harris and Kangas 1988). An animal is inseparable from its surrounding. A toxic molecule in a lung - is it inside the animal or in its environment? The moisture in a cell of an animal's nose - inside or out? The union of a frog or earthworm with its over-wintering conditions suggests that all is one. Habitat, niche, community - landscape, management area, forest - all are discussed and debated because of confusion in meaning, ignorance of or unwillingness to accept first-use definitions, unwillingness to move to common terminology and to develop alternative taxonomy as new knowledge suggests meaningful change.
There are multiple meanings and assumptions about"habitat" and whether it expresses (1) individual, population, or species assemblage needs; (2) differences in seasonal and migratory needs; (3) fundamental requirements or their associates; (4) needs in disjunct areas, or (5) whether the animals themselves are considered habitat. In the landscape ecology sense (Rodiek and Bolen 1990), all factors may be present for"good" habitat for a life group, but the size of the area or its nearness to other areas may be inadequate. Nearness to a large suitable area may make a"bad" area capable of supporting a population. "Habitat" is too elastic. Coulombe (1978) found four themes for the concept: (1) life requirements for single species or life groups, (2) resource allocation between or within species (as in"partitioning"), (3) spatial distributions and types (as in"nesting"), and (4) quality for a species or group. He noted that"habitat" rarely related to scale since that for an organism could be described as (a) being the bark of trees in general, or that (b) of a tree species, (c) of a mountain ravine, (d) of a topographic zone, (e) of many spaces (as for migrants), or (f) of a biome.
A replacement of the word is needed. To improve the health or well-being of an individual animal, you hire a sensitive veterinarian. To improve on the well-being of a population of animals you hire a sensitive modern forester or faunal system manager. Rarely does the forest faunal manager deal with individual animals, only the population, and usually by manipulating where it exists. The forest is a very complex and complicated place and the manager must master the complexity of this place.
"Habitat" has its roots in"home" and thus is misleading as the basis for a viable full-scale management system that deals with animals that migrate, and that are influenced by global changes. The habitat of many animals is faunal (microorganism of a wide range, parasites, and other animals, predators, nest mates, and pack and herd associates) as well as flora. I find feathers or the outer 2 centimeters of hair very"habitat like" but yield reluctantly to convention; these are a part of animal studies, rarely of habitat. Rather than habitat, the"faunal surround" or"faunal space" are useful phrases, implying the outside-of-the-animal environment - everything.
The forest faunal system manager creates and maintains space for animal populations and for users of these populations. A hunting"area" is managed for animals (usually to increase them) but also for users so that animals produced may be hunted or otherwise used in ways to maximize benefits. The resource system manager does not manage"animal homes," except at personal risk of mismanagement. Management is of spaces for desired populations and their legal users. It is available forage and protection from energy loss; legal protection; adjacent animals hiding places; special landscape patterns; and species-specific seasonal provisions. Habitat is a word which readers may lay aside for use only in general discussions.
There are few topics more difficult to discuss than the spaces for forest wildlife. There are so many species and life groups, so many biomes, so many different requirements of each life group, so many techniques (Thomas 1979, Cooperrider et al. 1986, Bonham 1989, DeGraaf et al. 1989, Patton, 1992) and, if those were not enough reasons for the difficulty, then there can be added the potential conflicts between tree cutters and people with a wildlife interest. Resolving or easing these difficulties is possible, but it is very hard work. Knowing that there are long-standing, deep-seated disagreements, and that resolution is hard work may help reduce the frustrations and inspire concerted effort for people of the forested world ahead. McAtee (1936) of the Bureau of Biological Survey that preceded the U.S. Fish and Wildlife Service said they had "...signed a truce with the Forest Service that neither will criticize the other publicly." But he was speaking of his experience that "...indicates that difficulties are likely to arise in combining forest management and wildlife management on the same area." He went so far as to suggest the option that the two should "...'agree to disagree' and plan for substantial or complete separation of their spheres of activity." He made a remarkable statement for its date, 1936:
As has been shown in the preceding discussion, however, stand improvement practice steadily diminishes the value of forests for wildlife and maintenance of the closed canopy virtually destroys it. This is so obviously the case that authorities freely assert that the interior of a dense forest is practically barren of wildlife. The reputation of our forest reserves as wildlife producers is really due chiefly to their un-forested portions, to open glades and parks, to brushlands and slashings, to cut-over tracts, and to burns. When, therefore, we speak of forested areas as the home and creative center of wildlife, we are using a figure of speech and we could more literally and truthfully say that they are the conservators and the source of wildlife about in proportion to their departure from the standards of an ideal forest. Departing from those standards is a necessity for attaining any considerable degree of coordination with worthwhile policies of wildlife management. In this country, forestry, and its practices are established while real wildlife management is an innovation. Necessarily, therefore, concessions essential to good coordination of forestry and wildlife management must come from the forestry side. The fact certainly is clear that if we sincerely wish to favor wildlife in forest reserves, we must curb the universal growth of trees, we must preserve openings of varied types, and liberally foster brushland vegetation.Immediately looms the specter of 'highest use' and if that use irrevocably is decided to be timber production, then we must seek other reasons for wildlife production. Forests and foresters must be supplemented by coverts and keepers. Where the production of wildlife is accepted as the primary objective, it is futile, nay it is fatal, to adopt standard forestry practices, as in the long run they do not favor, but on the contrary definitely discourage, achieving specific faunal populations.
Many, probably most, foresters desire to favor wildlife so far as may be practicable, but both forest growth and forestry practice of the first class limit their opportunities. In fact wildlife must ever be a by-product on areas primarily devoted to growing timber. It is well for wildlife managers to take these facts into consideration and make necessary allowances. On the other hand when in charge of projects, the prime purpose of which is wildlife production, they should avoid trying to combine forest and wildlife management, but forestry precepts to the contrary notwithstanding, they should firmly put into practice any and all policies necessary for the highest welfare of wildlife. It is to be hoped that such ranges will increase rapidly in number and area so that wildlife shall at last receive the recognition so desperately needed and to which for a long time it has been fully entitled (McAtee 1936:422-423).
 |
| Moose on Wyoming winter range. Moose, elk, and deer move out of the high-elevation, deep-snow areas and forage on shrubs. |
The big game has an abundance of summer range within the forests, but the year-long problem brings into the picture the ever-troublesome question of winter range. Only 6 percent of the winter game range is within the national forest boundaries, the other 94 percent is mostly privately owned lands along with a small amount of public domain and is generally overused by domestic stock. This winter range situation controls the numbers and distribution of big game, especially in the case of deer ...
In southern Idaho the public demand has resulted in an expensive although inadequate feeding program, especially in the case of deer herds and to a lesser degree for a few isolated elk herds.
It is evident that feeding of these game animals cannot be continued indefinitely. The more feeding is practiced the more certainly do these game animals return to these areas for winter feeding. This revolving process, if continued, will produce more and more game, and although game may find ample summer forage, the winter ranges rapidly deteriorate and eventually become so reduced in carrying capacity that destruction of the game herds result. I will admit that this is an argument against winter feeding. I believe feeding should not be done except in unusual emergency conditions. Certainly we should so manage our game that a herd can be stabilized in numbers to meet the ability of the winter range to carry them safely. This can be done only by a well-planned program of management based upon all of the facts bearing on each local problem and by unified action of all cooperating agencies backed by the support of an informed and interested public.
Another of our major problems in southern Idaho is the existence of extremely large game preserves established by the State legislature. In some instances these preserves include a large part of individual national forests. The enactment creating these preserves made no provision for managing the game within them and has resulted in creating a serious problem which is detrimental to the game and is resulting in injury instead of benefit.
One specific case is the South Fork of the Payette River game range for 1,000 deer, 1,000 elk, and 500 mountain goat; it now has 4,000 deer, 600 elk, and 250 goat. This has created a huge over-population of game which is in excess of, and has materially reduced, the carrying capacity of the inadequate winter range within the preserve. Where possible the game has overflowed into the only other available winter game range, namely, privately owned lands.
Due to the climatic conditions, migration from the game preserve to the winter range does not take place until mid-winter. The regular open season provided by the state game laws closes prior to the usual migration of the game from the preserve boundaries and consequently systematic reduction of the herd is not possible. One year out of every five, when the weather conditions are severe during the hunting season, the deer are slaughtered. As much as 50 percent of the herd has been killed during one season. On the other hand, for several average years the annual kill has not exceeded 10 percent.
Game simply cannot be properly managed under such conditions, and it is imperative that every interested organization and sportsman direct constant effort toward early legislative action which will make it possible to manage game properly.
The recent establishment of extensive primitive areas is another complication. Because of the few roads into such areas, hunting is of little importance as a means of controlling the harvest of game. The game herds actually increase toward a maximum in numbers. We can anticipate overpopulation, resulting in overuse of the available winter range within the next decade, unless some method of control is developed in the near future ...
We realize that there is a real problem in connection with our furbearers and that we have very inadequate information upon which to base any corrective action.
More study, planning and positive action must be taken in handling our mountain sheep, mountain goats, and remaining antelope herds ...
All of these problems become the more acute when it is realized that the largest human population centers of the State are in many cases immediately adjacent to the winter ranges as well as the summer ranges. During the winter period thousands of people visit the game herds and quite naturally raise a cry for more and more feeding, because of the inevitable presence of a few starving deer or elk.
There is always a demand for more game production but no thought of how to produce and maintain it. The first thought of 'John Citizen' seems to be more protection. In fact, it might be said that in many cases we are preserving our game that it may starve. One of the most needed things at the present time is a state game law under which it would be possible to take prompt action in managing all game wherever the need arises. Most certainly we wish to produce more game, but before we do so we must provide for its maintenance, and in the meantime we must properly provide for and manage what we now have (Varner 1938:9-11).
 |
| Ancient big game migration routes are blocked or range innundated by hydro-energy or irrigation projects. Winter range (often no more than 10% of the area once used by animals) is otherwise destroyed by human developments or otherwise made unsuitable by human presence in the lowlands |
There have been changes, though, since the 1930's in knowledge of faunal spaces and how they can be managed. The evidence is that knowledge per se is not what allows or causes the changes in the forest. The essentials are: knowing what is desired, what to do, why, doing something, then how to explain actions and their reasons, and then how to stabilize or enforce the actions.
The manager working with specific animal groups will often develop personal guides for analyzing areas and animals. For example, someone may use the guide of 2.5 kilograms of forage needed per day per 50 kilograms of herbivore (e.g., deer). This varies with season, dryness of forage, whether the animals are lactating, etc. It is a gross guide, a way of thinking quickly about what is generally needed by a large population. For example, for 2000 animals each weighing about 50 kg, the manager must supply 1,825,000 kilograms of forage. By knowing home range sizes (grossly), estimates can be made of how the forage must be distributed. If we use a clipper to sample forage and discover an average of 300 kg of forage per hectare, we had better have 6100 hectares like it to feed the 2000 animals. If the population has been stable at 2000 and we have more than 6100 hectares, then we can estimate the amount of unproductive land or estimate the losses due to poaching, predators, disease, etc. The usual assumption accompanying such gross analyses is that animals and plants will fill an environment, use every available resource to its limit. These are rough calculations and assumptions that are occasionally useful in the field. They need to be brought into the lab for careful thought, computer work, and detailed local observations and analyses. Before that, however, the decision maker, this manipulator of the faunal system, must understand the system context. Not some plea for an academic exercise, this understanding reduces frustrations, wasted effort, and suppressed projects; it may reduce sample sizes and costs because it allows meaningful assessment of desired confidence and accuracy. The understanding invariably improves communication about events which are often perceived as if they were people passing in dark agency halls after a power failure. It prevents direct violation of policy and law. It allows team efforts, i.e., many managers and owners, over many areas, to engage in loose coalitions to best an opponent (Chapter 17).
The subject of faunal spaces is very large, the entire"forest" of forest faunal systems. Topics are interrelated. I start, suboptimally, with "history."
Vegetation History
I think a faunal system person must know "inside and out" (implying"past and future") the environment within which work is to be done. The past should include plate tectonics and the location of the management area relative to the equator in pre-history (Vermeij 1987). This knowledge, delightful unto itself, is potentially useful in explaining the paleoecology of an area, interpreting and protecting fossil evidence, and developing theories of modern geobotany and how they may help improve predictions about vegetation on the area.
Persistent Pleistocene and pre-Pleistocene environments are exciting places. Sometimes called relict communities, these areas are valuable for their special flora and fauna and their role in adding faunal richness to the larger, more modern communities within which they are located. Special places, they deserve attention, but like threatened and endangered species, they are jewels. Like jewels, they are important in a household, but the tasks of paying the rent and putting out the garbage are the on-going and overpowering topics of household concern. They, like jewels, have high security costs. The management of millions of hectares of other wildlife areas, the non-relics, is the overpowering topic for the faunal resource manager.
Recent vegetation history provides insight to the past system dynamics, particularly limits to production and production rates. Post- Pleistocene changes, say the past 10,000 years in the U.S., are suggestive of the sources, dynamics, and potentials of the present environment.
In the eastern U.S., the American chestnut (Castanea dentata) has been nearly exterminated by the fungus Cryphonectria parasitica (Murr.) Barr (formerly Endothia parasitica). Vast forests were composed of 30% or more of chestnut and now that tree is gone. The forests have restructured. The enormous food supply of nuts supplied by these trees is no longer present. The net change in hard mast, in replacing chestnuts with acorns and hickory (Carya) nuts, is not known. (The lack of knowledge suggests studies needed now to answer similar future questions.) Chestnuts flower after any danger of frost, thus these trees provided a relatively stable hard-mast supply, unlike their replacements, the oaks and hickories, that have mast crops that are largely dependent on spring climatic conditions ( Diamond et. al 2000).
The faunal manager will master the historical changes as much as possible. Art work, old maps, conversations with older people, old photographs, diaries, and publications are essential materials for understanding the past. Oral recordings are very useful, especially if keyed to animals that indicate successional stages, to production (as honey from bees, mast from trees), and to indicator plants (showing pictures, not asking names). Much other landscape history can be read using the techniques listed in Table 7.1.
| Table 7.1. Techniques of landscape interpretation to discover the age and dynamics of ecosystems |
|
1. Geology: Geological maps and site inspections for fossils, strata, etc. 2. Readings: Paleobotany and paleoecology sources 3. Topography: Inference about climate and vegetation from slopes, aspect, colluvial depth, land forms, and waterways 4. Radio-carbon Dating: Also related isotope and chemical dating 5. Pollen Analyses: Samples taken from bog cores to indicate species present and changes 6. Pre-settlement Human Site Studies: An animal's remains that are present in Indian middens imply requisite habitats 7. Soil Pit Layers: soil pits may reflect root penetration, root remains, layering, floods, etc. 8. Soil: Texture, color, and organic matter content may be studied relative to present vegetation 9. Rock and Surface Materials: Rock and soil particle size and thus water forces, depths, and salinity needed to move them 10. Deeds and Land Survey Descriptions: Corner trees and witness trees are frequently identified. Entire areas have been vegetatively mapped from such deeds 11. Dendrochronology: Growth rates of tree rings (standing trees and logs in historic buildings and bridges). Years of rain, drought, or insect attack. Time of fire scars in borings of the boles or tree stump observation. The technique is generally useful to about 11 A.D. Match rings in the field to those in wooden beams in dated (cornerstone) houses or churches 12. Plant Age: Mean age and age distribution of trees. Singular ages reflect cataclysmic events such as fires, storms. Succession and forestry curves enable retrospective and predictive vegetative analyses 13. Relict Study: Distribution and description of relict communities; interpretation of past climate 14. Trails: Paths of grazing animals on hillsides and depth of compaction or erosion 15. Plant Density: Density and distribution of trees. Reflecting forestry, stand manipulation, grazing, and fire 16. Lichen: Lichen bands at the base of large rocks or tombstones display rate of erosion 17. Erosion Lines: Erosion and lichen bands dated at the base of tombstones and along brick or concrete walls or walkways provide basic data on soil loss and surface change 18. Fire: Fire scars on trees and rocks; charcoal in alluvial soil layers 19. Plow Lines: Plow strata (as a sign of whether cultivated or not and for how long) can be read in soil pits and transects 20. Mammals: Mammal activity in the past can be read from relict soil burrows. The burrow size and pattern will determine the mammal and thus the requisite associated habitat 21. Birds: Lines of trees (having originated from birds defecating seeds while sitting on fences) display old field borders 22. Fence Lines: Field size, erosion rates, and cultivation can be read from lower-side fence rows which serve as a soil catchment. Fence posts may be buried, preserved, and available for analysis. Seasonal layers in a cross section of this soil hump may enable the duration of cultivation to be determined. Trees in fence rows may be aged to confirm fence dates 23. Buried Flora: Soil at old dikes and land fills may cover the original flora. |
Diamond (1988)"proved" the importance of ecological history to the faunal system manager. I believe such managers will want to use history to establish for their areas statements like:
This community has persisted over ___ years. About ___ years ago there were changes that resulted in a transition from A-type community to the present X-type. We can expect this stage to appear only ___ times in 200 years. We can only expect the present community for several hundred years (your great grandchildren's environment). We saw the loss of X and Y and we cannot describe significant differences in the community before and after those losses. Species Y has lived here for ___ years, surviving every high and low of environmental factor X1 to Xn. The rates of loss (or gains) have been ______ for over 500 years and we see no reasons to suspect change in the rate for the next 50 years.
 |
| Fig. 7.1. A time line for a faunal management area displays time as distance between points. |
CAP83 provides a professional time line to aid managers in comparing the rate of change in known time to change in planned time.
The community is the result of biota invading or remaining in an area; it is what the existing forest trees or environmental components like soil and precipitation "allow" to be present. The analyses of communities usually start with the dominants. Analyses, however, need to start with what once was, then advance to the present condition. Neither approach is easy; the preferred alternative is the one that makes more ecological sense, less statistical sense (thus one usually less tractable).
The need for permanent plots and permanent picture points can hardly be stressed enough. The costs are great, vandalism and losses high, and difficulties in managing records almost insurmountable. The difficulties do not deny the needs and a planned, coordinated effort to manage such information, at least regionally, at least as a viable museum and library function, need be debated no longer (Curtis 1983, Brewer and Barrier 1984).
Wilderness or natural areas are invaluable in establishing a dynamic, cumulative record of the past. Sections of faunal management areas should be"set aside," in fact"created," as any other positive managerial action, for the purpose of historical understanding. Fenced exclosures may suffice in some areas, but areas of at least a hectare should be marked and excluded from use except for making measurements. Larger areas are of equal or more importance because within them are opportunities for holistic communities developing.
Exclosures, even though expensive to construct and maintain, have been used extensively in wildlife studies, mostly to study big game effects on plants and communities. By using wire of different size mesh and variously-sized gates for exclosures inside of exclosures, differences in the effects of different size animals may be observed. Exclosures provide decision makers with some of the same problems associated with other techniques used to understand animal space. They are small, so sampling must be limited; sampling can destroy the utility of the technique. Because small, they can only represent a part of the area, a sampling stratum. Mueggler and Bartos (1977) dug through records and found data about aspen area exclosures that were extremely useful. Trends, sorted and damped by years of adjustment on a site, can provide the knowledge managers need ... before such managers are assigned after 5 or 6 years to another work area.
Mueggler and Bartos (1977) noted an ecological cross-current: "... wavyleaf thistle (Cirsium undulatum) apparently was favored by aspen removal at Big Flat, but by a closed canopy at Grindstone Flat." They explained the difference being related to a 600 m elevational difference between the two areas that resulted in differential cooling and available moisture.
Platts (1981) studied a stream protected for over 30 years by a fence around a forest guard station. Clearly the objective of putting in the fence was not to provide a baseline condition or exclosure for comparison of the effects of sheep grazing outside the fence, but it did. It may be that similar areas (e.g., wide vegetated areas between highways or inaccessible plateaus (natural or those created by mining)) may serve for comparisons.
Site Interpretation
There are special places that need to be seen and experienced. Documenting the universal feeling that seeing or being in some places stimulates people is like documenting the need for breakfast. Such places vary widely, so presenting a list of all types will not be done, but at least people can agree that there occur unusual experiences of beauty, e.g., the feeling that comes with seeing certain managed forest stands or a tight collection of wildlife management practices. Look at that! is all that is needed to get a response, but more is needed because "notice" is rarely sufficient. Increasing numbers of people are jaded by television viewing. (They have seen 10 close-up examples of nature, far better, far more intimate, than any manager can hope to provide.) Increasing numbers of forest visitors are urbanites and cannot see what the manager sees. Interpretation is needed.
The objectives of the interpretation system are:
The inputs for the system are the range of information about the site and the specific objectives (e.g., to demonstrate early human dependence on small mammals). The time available and the reading or listening speed of target groups are critical to the input process. Use of multi-media is needed both because of diversity among people and because of effectiveness of each medium to transmit information of a particular type.
The process can range from the educational queue for visitors to the sequence of messages, but the key elements are in answering: who, what, where, when, how, why, and so what?
The evaluation should include whether people saw what was intended and if behavior (e.g., the viewer's blood pressure or littering) changed. (The past record or a before-"treatment" and after-treatment of the viewer is needed.)
Adjusting to future trends in the area as well as to the people (including the percent of people who return) is essential.
Inventories are needed to document the conditions of faunal spaces. With faunal abundance estimates, it may be possible to relate these conditions to richness and abundance. Always a danger of collecting too much or too little, it is essential that the risk be taken. The needs are for:
Forest Inventory and Analysis, formerly "Resources Evaluation" is an endeavor required under the Forest and Rangeland Renewable Resources Act of 1974 and the McSweeney-McNary Forest Research Act of 1928. Workers periodically inventory the amount and conditions of the forest of the nation. It takes 3-4 years to complete a statewide inventory and they are conducted about every 10-20 years. These inventories can provide valuable information about the forest wildlife resource such as:
The distance of forests from water (in Michigan, USA, one third of the forests are within 1 mile (5-km of open water)) related to animal water needs, to human use of areas for fishing and other wildlife resources uses, to potential impact of harvesting on water quality and the fishery, and to distribution of land by site index, thus the basis for assessing forest dynamics.
There are many ways to obtain these data. Coordination of such efforts are difficult. There are hundreds of databases (e.g., Lander et al. 1979) but they are difficult to locate and most require extensive work to achieve compatibility with hardware, software, or objectives.
McClure et al. (1979) described multiresource inventories, an expansion of the traditional timber inventory (e.g., Spencer 1983). Sheffield (1981) showed how multiresource inventories made by the U.S. Forest Service and cooperators could be used to evaluate the conditions for 10 species of birds. Dissmeyer and Cost (1984) described watershed conditions based on the inventories. Craver (1982) used the inventory to describe the distribution and amount of honeysuckle (Lonicera japonica) (an excellent food and cover plant but often a problem in forest regeneration work).
The beauty of having baseline data, the stuff that does not change or that is historical (like the deer harvest last year), is that it can be used. It does not wear out. It can be creatively transformed (e.g., the logarithm of x+1; CAP9072) to gain new meaning. It can be correlated and analyzed wherever there is a new hypothesis that animal A is a function of X or any variable in the database.
Grosenbaugh (1973) described how a computerized geographic location system was being developed and how results of inventories could be mapped. Fales (1969) demonstrated computer mapping potentials for a wildlife area. There have been many advances since 1969 in hardware and software but, sadly, there are few faunal system managers who have matched inventory to place, place to a hundred known factors, and factors then to planned forest harvests and largely predictable changes in faunal populations. Animals are largely a function of their spaces. To know spaces and to computer-map them is to know populations.
Baseline data are needed. They must be preserved, turned over to successors, perpetuated, used, and feedback applied to add to them factors which are needed as environments and knowledge of them changes.
Area
Area is a human concept symbolized on maps. Animals actually occupy volumes but, laying aside that concept temporarily, even in two-dimensional space they occupy areas in which it is energetically efficient to move and gain food and other life needs. There are areas of uncertainty for animals; there are zones in which the ears only provide information. There are areas to be defended (called "territory") and those areas which, if entered, a fight, not likely to the death but very costly, will likely occur. There are seen areas, heard areas (functions of sound attenuated and hearing acuity), and smelled areas. Microtus townsendii, for example, has hip glands used in area marking behavior (MacIsaac 1977). Other animals have similar glands and mark areas. Whatever the purpose of such marking (mate attraction, area ownership or control, or simply information such as "I have been here before"), such marking creates areas of faunal importance invisible to the manager. It is likely that knowledge of these areas will help managers understand the meaning of area in density estimates (animal abundance per unit area), will help understand how highways and other land use changes cause unexplained disruptions of animal activities due to changes in these unseen areas, and how managers should try to include such areas, not disrupt or partition them.
All fauna occur in an area. Management always involves area. (CAP145 provides conversion for metrics and CAP132 calculates area of a polygon from entered points along the perimeter.) "Sanctuaries", "reserves", "preserves", "refuges", "wilderness", and "parks" are all words used by different people in different areas of the world to mean areas for wildlife. They emphasize, depending on local or legal expression, degrees of protection, usually from hunting. "Wildlife management areas" may include internal refuges or protected areas. There is no known right definition of each. CAP06 provides some, but local use will prevail. Herein, a managed area could be all or part of a National Park, a National Forest, a National Wildlife Refuge, similar state-owned area, or a part of a corporate or private holding. Wildlife agencies often call areas "wildlife management areas" and manage them, or parts, as they see fit.
Management units are parts of these areas. They are often hunting units, areas with different regulations such as those limiting the animals that may be taken or the length of the hunting season. Elsewhere they are administrative and only designate areas within which staff work or have responsibility. Administrators change the name of forest area units but the progression is usually: region, forest, working circle or unit, compartment, stand, and operation (e.g., what part of a large stand will be treated in some special way).
These areas are identified in many ways. Some are administrative: e.g.,"you take half." Others are based on ownership lines, others on topographic features (e.g., mountain crests), others on roads: "everything west of Rt. 1." Some are based on watersheds. Increasingly none of these make as much managerial sense as they once did. Each was limited (which is why there are so many ways to designate areas). Now it is feasible, and I think desirable, to imagine all faunal areas as small square map cells, thus squares in the forest. See Fig. 7.2.
 |
| Fig. 7.2. A grid can be placed over any mapped area and information recorded about what is in each cell. Individual-factor maps are created such as one for the elevation at the center of a cell, the presence of a stream, or an ownership code. |
Any map with any features can be overlaid with a square grid of lines. Every square can be located by an x, y coordinate. By using computers, maps can be drawn of any feature. There is no need to decide on a fixed means for designating a wildlife area. This year the area may be the watershed, next year, only the south-facing slope of the watershed. Suboptimal designation of wildlife areas suboptimizes faunal management. A flexible process allows feedback to correct the designation.
I have learned that in developing countries where the U.S. refuge or park model is attempted, often a line is drawn on a map and an area designated. Not only is the U.S. concept for parks probably suboptimal in developing countries, but also where the boundary was drawn was also suboptimal. It is possible to state a set of criteria such as "minimum impact on villagers", "maximum species richness", "maximum access for tourists", "minimum cost of forage production ", etc. and to allow computer mapping to indicate how well each cell satisfies these criteria. Then they can be weighted, a new map produced, and then the management area (or areas) boundary can be drawn.
There is much literature on what is the optimum size for wildlife and related biological preservation and management areas. The answer begins with the question: What are you attempting to optimize? State your objective, then the constraints, then we are in business. Until these are produced, the debate will continue. Often simulation helps when objectives are difficult to state. " What if these boundaries were used, then what would the map look like?"; "what if these ...?" In some cases, intuitively, the answer for the optimum size is "as large as possible." It is extremely important that all benefits as well as all costs be considered. Faunal system managers often point to people constructing dams, highways, and similar developments through wildlife areas as not counting full costs or the "externalities" or the benefits foregone over the long run. In creating new wildlife areas, wildlife managers need to be subjected to the same rule and need not fall under their own harsh judgement. The optimum area decision will include the foregone benefits from alternative uses in its determination. There will be portions of areas lost; sometimes whole areas. That is the nature of the game (Chapter 17).
Seasonal Range
In the western U.S. and Canada (and elsewhere in the world) snow depth builds rapidly in the high mountains. Animals there hibernate or move to lower elevations. These lower areas are called "winter range." Winter range is a much more complex idea than area to which animals move. It includes variable dates of first snow, snow depth, what capabilities animals have to move and feed through snow, when animals start to move into (or out of) the lowlands, the ancestral migration routes, use to which the lowlands have been put by humans (e.g., farms, reservoirs, roads, towns), the abundance and quality of the food, and direct mortality (e.g., poaching, predation). In other areas there is an alternative area usage, a reversal of the snow and energy-loss phenomenon of the cold regions. In the arid zones, the riparian forests are intensively used by animals, strongly analogous to winter range.
When the snows melt and forage becomes available, animals return to summer range. The concept is intuitively simple but it has profound consequences. In some areas, only 10 to 30 percent of an entire management area can be classified as winter range. The amount varies each year with snow depth. The relations in the system are not linear. An additional foot of snowfall laid throughout a convoluted valley system over a large area can withdraw thousands of acres of land from foraging animals.
When I moved to Idaho from the Eastern U.S., I found the concept of winter range difficult to grasp in its entirety; I had to live with it. It included: a winter with light snows allowed a large winter range; many female deer and elk survived; many returned to the summer range. Production of young was high. If in the next year the snowfall was heavy, large numbers moved to the valleys; forage was sparse; large numbers died; the range plants were hit hard resulting in their reduced vigor for 3 to 10 years; recovery of the population could be slow, even with good reproduction on the summer ranges. The valleys are now filled with people, roads, buildings, reservoirs - all having destroyed foraging areas or blocked animal movement to foraging areas at lower elevations. Comprehending winter range (and similar phenomena related to seasonal shifts, long range as well as the above short-range migration) is difficult. It is a massive geographic phenomenon, dwarfing the ego of the boldest manager seeking to control sensitively any population.
The controls are limited, but managers can resist developments that block migration routes, stay attuned to annual snow depths and range conditions, maintain flexible harvests to reduce range pressures, and restore the areas heavily "beat down" by winter-stressed groups.
It is evident that animal movement is related to snow and temperature. I believe animals as "energy balancers" engage in similar subtle movement in response to other factors. (See Moen 1973.) They shift time spent between elevations and aspects, between wind-swept ridges where there are few valleys, between conifer clumps and open stands. In the northern states, deer move to "yards" where winter conditions are favorable.
Relevant Areas
Portions of some natural resource or management areas are for the animals and very specific users such as those who do baseline research studies and monitoring. Other parts of these areas are for intensive user such as anglers, hunters, and animal observers. Except for hikers who will go almost anywhere that there is a trail, and mountain climbers, these users frequent very limited areas. Rather than comparing total areas or allocating funds based on total area, it is appropriate to estimate the area that 9 out of 10 people will probably (0.90) use and also the probable days of use over a 10 year period. Using these criteria usually reduces the area for management significantly. The cut can focus attention, clarify work responsibilities, and improve budget allocations. The "area sword" has two cutting edges. For example, U.S. Forest Service land contains 90 million of the 500 million acres of U.S. forest land capable of growing more than 20 cubic feet of wood annually. The statistic can be used to boast of control or to absolve responsibility; the need suggested is for understanding the managerially relevant area.
The relevant area is usually a zone around access points (e.g., boat landings) and roads. The potentially useable area needs to be estimated based on reports of hunters and other users, observed kills, radio tracking of hunters and recreationists, viewscape maps, trail sensors, and other methods. Simple time-rate-distance relations can provide a first estimate of the width of the potentially useable area. This may be the area where most faunal system management should be conducted.
The Unique Square-Meter
Computer technology has allowed managers to think in new ways about wildlife area. Once it was essential to aggregate land into large conceptual units like regions or watersheds. Now it is possible to think about very small units, mentally working in the reverse order. Satellite images are made from pixels or picture elements. These cellular images of Earth's radiations are only one factor among hundreds that are mappable. Commercial computer programs are now available for manipulating such maps. The size of each map or land cell is an important consideration. There is no optimum. Size depends on objectives, available budgets and maps, and required accuracy.
Animal-to-space relations are needed but not as much as some people think for reasons already discussed abundantly in this book. It would be very desirable to know the relationship of animals to many major variables, such as in:
Nt+1 = f (at, bt ... nt)
Predictability is only possible in the sense of potential populations or standard conditions. Animals are influenced by people, even on wilderness areas! Animals are hunted, subsidized, protected! Every site is unique! There is equifinality (Angermeier and Schlosser 1989:1460), many pathways to the same population tomorrow! Non-managers, laboratory-bound scientists, readily conceive of one precise relationship. The alternatives needed in the field are pathway models, expert systems, and models that account predictable change and variety over time. Emphasis needs to be on unique entities of a population, not on mean conditions for which there exists only one unique entity. A technique that evaluated a site today yields wrong results tomorrow. The needs are rarely for estimating of the number present in a habitat, but for the potential for that habitat. Once that is decided as the objective, then the analyses shift to relative differences between faunal areas. Is this area better than that? Based on the conditions present and observable by analysts, conditions positive and negative, conditions stable or with a high probability of change from a one balance to the other, these can be determined at a very high level of confidence by a student of a species or life group. That the well-estimated population of X in many areas does not regress well on some index of current habitat goodness or suitability is of no surprise. Maximums (potentials) should provide a good regression. Pure chance may match up a population with a treatment. A brief, expensive test to determine this would be audacious given the number of species that must be mastered in order to bring the faunal system under semblance of control.
In geographic information system work, the cell-size question is one of precision. How precise will you make your measurements? Will a cell be 1 square kilometer or 0.1 hectare? It depends. Most people want very small cells because they realize how variable land is. They are familiar with their house and yards. They know that the fountain is small, just as is a corner of flowers, a walkway, and a spot that for unknown reasons will grow nothing. All of this is in a small area 10m by 10m. How then can realistic work be done with larger cells? Even with computers and their inconceivable speed, cell size has to be limited for most problems to be solved. In Virginia there were about 1.1 million cells representing the entire state in a computer data base. Each cell was 1/9 square kilometers (about 27 acres). There were factors known for each cell such as average slope (CAP113). All outdoors people know that slope changes quickly and is varied (except on wetlands). Standing on a large rock or stepping into a stream channel can produce grave doubts about the meaning of average slope. Yet slope estimates for cells can be used, have been, and can be very expressive about average conditions ... and these estimates are more accurate than a map with no slope information, one requiring mental interpretation of the denseness of contour lines, or one with large areas painted one color indicating slope changes of 10 percent. The tension that the systems person feels is that between working with nearly complete ignorance one day and great knowledge the next, between being praised for precision used in small areas and criticized for unusual skepticism about large-area observations. The computer mapping environment is dynamic and its beauty is in the advances made, almost self-generated, by the technology - the hardware-software-staff system.
Every square meter of a forest is unique ... absolutely. There will be similarities, but like egocentric people who know they are unique, each plot has combinations unlike any other. Pielou (1969) observed "not only does each species differ from every other, but also all the individuals within anyone species are unique. ...It can truly be said that no two of the individual units making up a community are alike and that each of them, throughout their lifetime, varies continuously in a manner peculiar to itself. ...The components of a community are never the same at two successive occasions." Facing the uniqueness of areas realistically, the forest faunal system person needs a geographic information system with 1 square meter cells. (Smaller, perhaps, but enough is enough!) For each cell in large faunal management areas I could potentially have 50 factors or more, and then I would have conceptual control, be able to understand what goes on in every place on my area, explain differences, predict likely future faunal systems, estimate impacts, etc. This is now conceivable and almost executable but even with computers, the data storage becomes enormous (how many meter-sized cells are there in a moderate 7000-hectare wildlife management area?), the hardware is beyond personal computer limits, the allocated costs of computation, time allowed for certain types of emergency decision making (e.g., forest fire fighting and lost-person rescue work). One meter may be too small for work in forest faunal resource management which always involves animals with fairly expansive and variable home ranges, populations of animals (not individuals), and with a seasonal as well as a spatial dynamic (e.g., in rotating forest harvests over an area). It will soon be the proper size.
I cannot provide an answer to the optimum map cell size. In urban yards-for-wildlife, I would work with a 1-meter cell and define "the yard" as the owner's and all contiguous yards. This would be the managed unit. I have seen powerline corridor analyses done with 1 square kilometer cells. This, it seems to me, is too gross for any purpose, but in very large areas that are not variable, where data are scarce, and where objectives are to avoid bad "spots" or cells, then such maps may be of service. I believe the 1/9th square kilometer represents a reasonable compromise for many uses in natural resource planning. It is too gross for centerline location of roads or trails, but too precise for printing planning maps, boundaries, and riparian vegetation areas. I suggest that most applications of geographic information systems can be handled as a 3-level operation. A gross system can be created, for large areas (e.g., regions of 20-25,000 sq. km.), perhaps with one-square-kilometer cells. Then at an intermediate level, cells can be developed (perhaps 1/3 km on a side) for areas of special concern and intense planning. The third level would be at the 10 meter x 10 meter cell size (called Alpha Units by Giles, ms)(with images of cover now available from satellites) . This size allows fair discrimination, requires abundant data collection, entry, and verification, but is feasible for many areas with proper scaling of work, equipment, space, and staff. Past efforts to do extremely detailed, large-area work have often failed. Improved technology and programs are reported regularly so the problems are diminishing. The warning needs to be sounded. Start at level 3, with very precise cells and data but with a very small area that can be expanded easily into larger cells of the same metric. Otherwise, start with large cells for a large area, then, using it, locate parts of the system that need greater precision and more data collected and processed.
Giles (1988) presented warnings about creating large geographic data bases without appropriate plans for their maintenance and use. One beauty of data base systems is that with each use, the per unit cost goes down. The data are not used up! Once such data are available, then the creative juices of managers invariably flow. The systems can be used for many faunal system purposes, then uses become readily apparent in planning, for optimum site location, for corridor location (e.g., roads, utility right-of-way (Giles et al. 1976), trails), for points (e.g., airports (Koeln 1980) and solid waste processing sites). Once the wildlife agency has such a system, then "knowledge is power" and it can use it in an active, progressive, cooperative mode to improve all resource management in an area and thus wildlife within that context. Otherwise, it can be used defensively, assuming land use of most types is likely to be harmful to some fauna and therefore to be stopped, delayed, or diverted. I favor the former.
There is a debate over what methods to use to enter map data into a computer and how to use it. The major options are the square or hexagonal cell and the polygon (including the variable-size triangle). The debate is extensive and cannot be reviewed here.
The manager needs to be able to use the data and (1) to select from the entire area a relatively narrow line or corridor that meets at least 40 criteria of goodness and that has ability to allow the factors that meet those criteria to change over time (e.g., trees to grow, soil to erode); (2) to select a point that meets similar criteria (e.g., for a water hole, a tourist center, a hunter camp); (3) to select an area that meets such criteria (e.g., pest outbreaks, viewed areas, fire hazard areas, taxation impacts); (4) to select a volume that meets criteria (e.g., a forest layer, a ground water volume); and (5) to transfer effects between cells and layers (e.g., runoff, groundwater recharge, wind in forest layers). The emphasis I wish to make is that maps need to be drawn by computer but they also need to be decision aids; they must participate, by design, in optimization. To my knowledge, only cellular systems (or polygon systems converted to cellular systems) can perform consistently in this manner. Overlays of polygons can be made and used in making decisions, of course. Cell-based systems, however, allow impacts or costs per unit area to be accounted, each incremental change to be readily accounted. Fig. 7.4 shows a computer-produced map of a power line corridor selected from thousands of possible routes between two specified points.
 |
| Fig.7.4 A computer-produced map resulting from analyzing alternative routes for a powerline. The analysis included 12 weighted dimensions of impact, 40 variables including wildlife, costs of construction and maintenance, and successional changes over 30 years. Each map cell is 27-acres. Two lines are shown for a vast area - one proposed by a power company and one computer-selected. Shades of gray in each map cell indicate likely long-term total costs. (Map developed by A.B. Jones III.) |
The company-proposed corridor can be compared to such selected routes. One use is in court hearings in which a company-proposed route(s) is compared to routes suggested by opponents or a public body. As in other types of developments and land use change, there is no "best place" (the typical question). All development modifies faunal space and thus is good for some animals or resources in that space, bad for others. The impact question needs to be formulated, usually, as "where is the least bad place to put X?" or "where will net positive effects of X be maximized?" Once done, the solution method can usually be found among those labeled cost, risk, or loss minimization. The literature of economics and operations research (e.g., Taha 1971) are full of such methods. Overlay maps (without computers) are useful. After manually placing 6 or 8 overlays, no further discrimination is possible. Wildlife decisions usually involve more than 8 factors but using 8 usually produces an improvement in decisions.
The dominant vegetation can usually be mapped with the aid of low-altitude aerial photos or satellite images. Fies (1983) used satellite images to identify hardwood and coniferous forests (about the only level of discrimination consistently possible from these images), then added data on slope, aspect, and elevation and was able then to specify potential forest type in each cell. Imagine the power this gives to the faunal manager able to relate fauna to forest type and age!
A Question of Managerial Control
The manager seeks control over the benefit-production-and-cost system of the forest. The faunal spaces are one part of the production system. Much control is possible, but there are limits. Can precipitation be controlled? Politicians? Poachers? The answer is yes, but it requires a particular perspective about control. When knowledge is gained about something, that thing begins to be brought under control. Models are knowledge units. With them, risks are reduced, predictability is gained. Spies help bring an enemy under control. Models help bring climatic factors under control. To name something begins to bring it under control, for then it can be studied, information shared, the entropy of undirected study reduced. There are many things that cannot be brought under full control. A 100-percent efficient machine is known from elementary physics not to exist. Yet the concept remains. Control, like efficiency, is sought even though it may never be obtained.
To know that one forested slope receives more isolation than another slope throughout the growing season puts the manager in control. The evapotranspiration and energy input differences can be noted. These help explain differences in forage on a site, in animal population behavior relative to it. No longer responding to the land as if it were a whimsical earth-god, the manager gains partial control, at least some potential power, for now he or she may decide to do more (or less) on this (or that) area, now, because of knowledge.
In a manager-as-driver-of-a-car analogy, the driver is clearly in control but as every driver knows, control is relative. There are wind gusts, irregular pavements, breakdowns, forgetfulness (as in filling the gas tank), and other aspects of a purposefully controlled, high-tech activity, that deny "control" as a 100 percent, finite condition.
The manager seeks control; the dynamics of this two-word phrase is very important. The manager who is statistically versed knows that experimental designs are largely made to control statistical variance. To know several factors and to have data about them allows the researcher- manager to control or reduce the measure of variability in the relations of a factor (e.g., animal density) to environmental factors (e.g., forage). Measuring and relating forage and density may give a gross predictive or explanatory relationship in a model or equation, but knowing insolation, soil texture (bulk density), and slope may give very great predictive power (e.g., may move the R2 index to the goodness of the model from 0.3 to 0.8). This is gaining control - conceptual power - over the system. The manager may not be able to change or manipulate land slope, but he or she may select study sites or work areas differently because of such knowledge. He or she may not be able to change soil bulk density, but will know the effects of hunter-vehicles or logging equipment compacting soil and thus be able to explain changes in forage and animals responding to it. Change insolation!? Perhaps not, but changes in cloud cover or particulate air pollution might be projected or hypothesized and thus their effects on wildlife and other phenomena known at a high level of confidence. To know these consequences or even to be able to begin to estimate them with preliminary tools (not just good guesses) brings the forest faunal manager into control of the system.
In rough terrain in the early morning, the hunter or woods worker has surely experienced the sun "coming up over the ridge." It may be very late in the morning before plants on a site experience direct isolation. To know this primary, driving, ecosystem force on every forest site (perhaps map cell) seems to me almost essential for managerial control. No (or little) field sampling is needed. The path of the earth is fairly well known as is the trigonometry of light angles. The wildlifer may not be in control of the sun but ability to estimate the rate of tree height growth on a mountain ridge provides him or her power to estimate changing solar radiation on an adjacent slope and thus provides insights into patterns and differences in animal density, behavior, forage quality and quantity ... the stuff of the faunal manager's decision system. Knowledge of faunal spaces, particularly the abiotic factors of such spaces, gives decision power and control.
The previous section was about "area." Area perhaps has more influence on faunal populations than any other factors. The population is a function of density and is clearly and simply number of animals per unit area. The magnitude of hectares typically overwhelms any other descriptive conditions about the hectares themselves. The manager needs to be concerned about energy, edges, and many other factors. Area is the place to begin for faunal resource system control.
Global positioning satellite (GPS) technology can help locate corners and improve area estimates. Wild area surveys across the U.S. are surprisingly inaccurate, resulting in faulty estimates of area. The problems of estimating population density begin with the estimates of area, not the animals.
Classification
There seems to be a need for a common system for classifying, inventorying, and analyzing natural systems in which faunal resources exist and progress has been made (Marmelstein 1978). They are needed for national and regional inventories and their analyses. Selecting the most useful for a stated period is a desirable task, and, with feedback, can be improved. There is no need to avoid committing to a common classification if (a) it is reasonable and the best seen (by a majority); (b) it will be held for a few years (avoiding costly re-mapping, data transformations, and re - analyses); and (c) if it is known to be under study and that change is possible at a designated time. There is good reason, however, to consider an alternative.
To classify land means to name homogenous groups. This furthers the illusion that a thing named is a thing explained (cf Beadle 1974:307). The name is thus a code, that, if well-selected, will communicate many things about an area to others who understand the code. Anyone who has been to the Pacific Northwest understands "mature Douglas fir." It is not simply a statement about the dominant tree but about soils, moisture, understory vegetation, and wildlife. Such a "class" provides information for many users. Does it provide information for most users? For a few users with costly and risk-filled decisions? How many classes of land must be processed for a person to be able to map the range of a salamander that frequents old Douglas fir stands but also several others? Dominant vegetation is one basis for forming classes. It is good; it is predominant. It was developed bc, before computers, and is influenced by conventional map making.
Perhaps what humans readily see is not "seen" by animals; perhaps faunal responses are to infrared or other electromagnetic stimuli, and to olefaction. Continuing to assume that animals react as humans may prevent achieving the promise of "habitat classification", said by Steele et al. (1983:83) to be:
Lotspeich and Platts (1982) observed that resource inventories without classification are unorganized lists and that planning efforts require inventories. Their approach attempted to unify land and aquatic system categories; emphasized natural attributes (not present or projected land uses); and progressed from the first-order watershed as the smallest unit, the "basic ecosystem." The system suggested was hierarchical, namely: 1. Domain, 2. Province, 3. Section, 4. Region, 5. Land-Type Association, 6. Land Type, 7. Land Type Phase, and 8. Land Site. Streams are classed at level 6 because they are continua with a variety of land types. Bailey (1978) had a very similar taxonomy (Table 7.2). The system is a reasonable basis for inventory and for deductive work in unknown areas. As in other systems, difficulties exist due to scale, and in separating (if desired) fire, grazing, trapping, hunting and logging and dam affects from the "natural" attributes of an area.
| Table 2. Bailey's ecoregion element hierarchy based on that presented in Fink and Elder (1982) as adapted from Crowley (1967) and Wertz and Arnold (1972). | ||
|---|---|---|
| Level | Element Name | General Characteristics and Comments |
| 1 | Domain | Subcontinental area of similar climates |
| 2 | Division | Single regional climate at the level of Koppen's types (Trewartha 1943) |
| 3 | Province | Broad vegetation region with the same types of zonal soils |
| 4 | Section | Climatic climax at the level of Kuchler's (1967) potential vegetation types |
| 5 | District | Part of a section having uniform geomorphology at the level of Hammond's (164) land surface form regions |
| 6 | Land Type Association | Group of neighboring land types with recurring patterns of landform, lithology, soils, and vegetation associations |
| 7 | Land Type | Group of neighboring phases with similar soil series or families with similar plant communities at the level of Daubenmire (1968) habitat types |
| 8 | Land Type Phase | Group of neighboring sites belonging to the same soil series with closely related habitat types |
| 9 | Site | Single soil type or phase and single habitat type or phase |
Pennak (1978:65) said about the running stream faunal volume: "It is obvious that we are dealing with an enormous variety of running waters where there are wide ranges of chemistry, physical qualities, and biological details, and where single-criteria and several-criteria systems are of limited coast-to-coast value." He suggested 13 criteria. [If there are only 2 states of each, the options are merely 8,192 types of streams!] Those were: width, temporary or permanent flow, current speed, substrate, winter and summer temperatures, turbidity, total dissolved inorganic matter, total dissolved organic matter, water hardness, dissolved oxygen, rooted aquatic vegetation, and streamside vegetation.
Classification is an effort to aggregate knowledge, to lump things in ways that have meaning or utility. The ways must be few and mappable. Over-aggregation is almost assured, no matter what the practical problem. (General interest, breadth of curiosity, artistic endeavor, or public education are not the topics.) The result is likely to be suboptimal decisions and thus resource use.
If the greatest possible refinement of data is used, when new problems and new needs for locating areas with a special set of characteristics arise, a unique class can be defined for those new situations. A new map can be created, perhaps one never seen before ... and perhaps useless tomorrow. This is called dynamic classification (Williamson 1981). After the map is discarded, the system that produced it remains. The map becomes more like a newspaper than a thing to be preserved in a library. One day it need be only an image produced or controlled by a computer.
On a conventional map, more than 20 colors are rarely displayed. Few people can easily discriminate among 3 shades of green or other colors. In most faunal space work, it is easy to conceive of areas being needed to be analyzed by: (1) high-low, (2) north-south facing; (3) near-distant from water; (4) vegetation groups 1, 2, 3, or 4; (5) near-distant from roads. There are 64 categories suitable for mapping and needing color codes in this simple list of 5 items. Faunal space information cannot be mapped. The above seem at odds with expressed needs for maps serving the layperson, being simple, focusing on priority habitats, and being standardized (cf Kusler 1978). They may be, but not necessarily. There are needs to be met that do not require the kind of creative involvement suggested above. Where are the "critical areas?" Make a map of the areas. The criteria for critical areas varies by states. A map could be prepared using alternative criteria. "Make an endangered species map" may be a reasonable demand. What is to be mapped - counties in which they occur, or potential habitats, or cells, or last known record, etc.? It is difficult to get people who make what appear to be reasonable requests to be specific enough so that a systems output, once delivered, could be judged useful.
Steele et al. (1983) briefly reviewed the debate over whether forest types of similar classes exist, whether all plant and animal assemblages vary and occur along a continuum, or whether there is some intermediate condition (Collier et al. 1973). They were fairly pointed in suggesting "get on with it."
Although this debate may still be of some academic interest, it need not preoccupy natural resource managers and field biologists who need a logical, ecologically based environmental classification with which to work. We acknowledge that continua may exist in the landscape; nevertheless, our objective is to develop a logical site classification based on the natural patterns of potential climax vegetation. Local conditions that deviate from this classification can still be described in terms of how they differ from the typal descriptions presented here.
Cole (1978) reviewed lake classification and observed that seeking to describe lake typology as an objective has not been without value. Out of the effort has emerged patterns that have advanced limnology and ecology. He said "continuing the search for pigeonholes into which lakes can be fitted naturally on regional bases can lead to understanding of local aquatic habitats and their potentials. There is, however, little chance that a couple of adjectives can explain the complexities and dynamics of a lentic habitat."
Hoffman and Alexander (1980) used "plant association" to mean forest stands that have the same overstory and understory, and "plant series" to mean those having only the same overstory. Within-series classes were based on differences in understory. They proceeded to analyze a forest on the basis of potential vegetation, not that which resulted from extensive past disturbance.
The capabilities now exist within geographic information systems for any manager to ask for a unique map to meet a specific need. For example: Show me all sites above 900 meters with conifers adjacent to water on north-facing slopes with steepness less than 15 percent. A manager might be able to use conventional maps and create such a map but at great costs of time and effort, and the results would be of questionable accuracy. Often, time for making a response is the resource in shortest supply. The map is needed tomorrow in a critical meeting! It is to be used for a specific purpose for one life-form in one managerial context. It must communicate well, given the people likely to be at the meeting.
Just as a person uses a plant key, working through taxonomic criteria until a specimen is classified, so can a manager specify criteria, then in reverse order find the "specimen map cells" that meet those criteria. The criteria may be dynamic (as they are in so many managerial situations); the data may be revised or new data added and a new map produced. The new map is a tactic for resisting suboptimization (Williamson 1981). Logistic regression has been used by McCombs (1998) to find the probable sites where the northern flying squirrel will be found.
Because of the capability of computer-based dynamic classification, the extreme variability in types already named, the occurrence of identical dominant vegetation with extremely different conditions (expressed equifinality), extreme overlaps in type occupancy by fauna, faunal differences greater in forest age classes than in forest types, limited utility among resource managers (though within-a-resource, use may be high), a visual limit to the number of types used that is imposed by cartography (about 20), and the broad transition bands that occur between some types, I suggest that faunal system managers be classification aversive. Shelford (1963:238) included the grizzly bear in a classification system and used the "spruce-moose" type. Even this will not overcome the above difficulties. Typing areas may be necessary but, for the long run, since change now seems to be exponential, eclectic emphasis on faunally--influential factors now seem appropriate.
Primeness
Giles and Koeln (1983) argued for using a concept of agricultural land "primeness" and demonstrated how it could be calculated and used. Rather than using a simple concept of "prime farm land" (two options, prime or not-prime) they suggested that primeness, a continuous variable, could be meaningful. Somewhat like "suitability", "prime" denotes a maximum and implies that"less-than" can be discussed, even to zero primeness. "Suitable" however connotes a yes-or-no condition, does not suggest a maximum or minimum condition as a basis for comparison, and does not contain a time concept, one related to forest succession in an area or to the suitability "next year." Habitat "suitability" models and indexes (U.S. Fish and Wildlife Service 1980, Collotzi 1980) are now abundantly used so "suitability" will probably remain in use. Lorio et al. (1982) described a stand risk rating for southern pine beetles. Although the concept of risk is associated with beetle damage, the rating is an insect habitat primeness concept. The insect occurrence (frequency of infestation among stands) cannot be well predicted, a situation not unlike that for other animals. The utility of primeness models and their results is that land (each cell) can be evaluated for the probability of how well it meets the present as well as future needs of a species or life group. Especially important in land-use-decision-making and impact analyses, areas of lower primeness may be selected first for use, saving those areas of greatest primeness for later. This results in preservation of future options. Similarly, tradeoffs can be made among cells, estimating the long term species- or group-specific benefits from cellular areas and their combinations. It may be that 2 areas with primeness 0.45 will substitute for 1 with value 0.9 ... but then cumulative scores over many years need to be studied.
Quality Primeness, suitability, or other terms express area quality. Quality is a species- or life-group specific concept, varies seasonally, and over time. It is essential that managers view area size (quantity), as if through a stereoscope, with area quality. One-thousand hectares for an animal life group is, if of a quality index of 0.4, equivalent to 400 hectares of index quality 1.0. Agencies may think they impress the public with statistics about areas held or treated annually. Some people want to know which areas; of what quality; and what was the change produced? Land acquisition and set-aside programs need to be based on area and quality. Rayburn and Giles (1975) compared areas for acquisition for deer management purposes based on the net energy that these areas would likely supply over a long period. Rough topography, temperatures, and snow may function to cause energy drains to deer. Available forage produced and its metabolizable energy content may contribute to the population. Areas of the same size have very different potentials for supporting wildlife or providing benefits for users.
Because quality is so important, also important along with area (for what is the significance of zero quality?), then questions of minimum areas become almost unanswerable. I suspect an expert system (CAP54) will be developed to help determine minimum sizes for faunal management areas. The species-area curve (see Chapter 4; CAP2041) has been discussed. Where species richness is an objective, the curve may provide insights into the minimum area size, but it will not free the manager or the public, represented by managers, from making the decision about the proportion of the area under the curve that will be acceptable.
Total area is important to some species. In general, areas that are at least 50 home ranges in size are needed. This assures genetic diversity sufficient to reduce inbreeding and to assure a viable population. In some areas with large animals this cannot be achieved and this presents the problem experienced with other animals, namely, what is the smallest patch size? A patch is a small community like a clump of trees in a grassland. There is evidence that each patch has to be larger than a particular size for each species or it will not contain that species. If an animal requires 5-hectare patches and there are 50 patches saved for it, each 4 hectares in size, then the species will probably not exist in the area. It may be asserted that 200 hectares have been preserved for it! The intention was good; the execution flawed. Each species needs to be studied by observing areas of many sizes and plotting the size of patches and species abundance, at least presence-absence, and proximity of other patches. "Corridors" or connecting vegetation units are over-emphasized.
Lynch and Whitcomb (1978) concluded from their studies that breeding birds in the eastern U.S. deciduous forests have declined in richness and abundance due to decreases in forest patch size. Birds in large forests are not so affected. Birds most affected are forest-interior species that migrate out of the areas in winter. They thought that small forest parks are ineffective as reserves for avifauna because they are too small, too far away from generator populations, and have too much human disturbance. They were concerned about the likely extinction of many forest bird species and called for a halt in the trend toward insularization of the remaining forests of the Atlantic Coastal plain. Harris (1984) called for the same in the Pacific Northwest.
We can return to the questions of how much area should be used for forest fauna and what is the optimum size to be designated for such management. Those are complex questions and there is much debate over what are the proper policies to provide answers. The questions are leading. Refuges and reserves, dedicated areas, are one way to manage wild animals. Setting aside areas is not necessarily best and rarely necessary in light of the context and other options available to achieve objectives. Dedication of land is a conspicuous act easily done by politicos. The hard work begins afterwards and not by those signing the grand documents. Perhaps dedication and public notice are necessary precursors, but the effects on wildlife that seem to come from "establishing a park" are really from altered habitat destruction, stopping poaching or other critical activity, and actively managing spaces. The idea of a refuge as a center from which most animals disperse seems largely unsupported. In protected areas, animals behave differently to humans than do animals outside. The richness and densities desired (stated in objectives) can often be achieved without reserves. In some countries the area of complete protection causes social as well as ecological problems resulting in negative benefits at high costs - hardly the stuff of modern wildlife management. Without predators, for example, ungulates on protected areas often increase to numbers that destroy native plant communities. Controlling hunting or other animal removals as well as other behavioral controls are needed to protect wild flowers, butterfly and songbird habitat, forest regeneration, soil on critical sites, and water quality.
The needs for areas can be related to the local expression of a system performance measure, Q* - benefits from all species at low costs over at least 50 years. To meet demands for each species, area can be calculated. Then they need to be listed from largest to smallest. This may suggest the need for a large area to meet the needs of the most wide-ranging and lowest-density animal. This may not be sufficient, for there may be habitat needs for other species that do not overlap the range of the wide-ranging creature. Iteratively, the questions need to be phrased, high to low, "sufficient size for desired benefits from species 1? 1 and 2? 1, 2, and 3? ... etc." In some environments, meeting the area size requirements for species 1 may be sufficient. Meeting that need meets needs for all other species. Costs and available areas come into play. The manager may work up from the bottom of the ranked cumulative list, deciding on the greatest area that can be afforded that meets the needs of the most importance-weighted species in demand.
Generally, the smaller the perimeter-to-area ratio of a managed area, the easier the area is to manage. Irregular or long areas, however, usually enclose more species ranges. See CAP2011, CAP9062, and CAP9065. The more proximal are similar small areas of the same habitat, the better. There needs to be maximum opportunities for animals to find enough suitable conditions - whether these be nesting sites, food, water, escape areas, or escape from inbreeding. The criteria for parks or refuges are not of areas but of animals, all of these specifically quantified in the objective.
I find appealing the idea of identifying the species in an area of interest (or the species of interest) having the maximum home range (h) (Chapter 8). Then assuming an area is needed for at least 5 to 10 (g) of these groups, and there is some overlap (z) an averaged proportion of the home range size, then the area needed is:
A = hg - hz (g - 1.0)
CAP2000 allows the manager to play with a variety of assumptions or hypotheses about each of these variables. In some cases, detailed studies on one or more may have been done or are in progress. Often a species being protected for the first time is almost unknown. Deduction based on similar species and spaces is all that is possible. Like engineers, faunal system managers need to make estimates as accurately and precisely as possible ... then add a bit for safety. How to determine the "bit" is another question, but it can be added to the force-field in which the manager lives - accuracy, precision, confidence. The manager adds a "bit," a measured practical attempt at reducing the probability that an undesirable limit will be passed.
Wildlife is produced on many areas. All good animals are not assigned only to refuges or special faunal areas. Fauna are to be counted as satisfying objectives wherever they are found. There may be horror stories about habitat losses but few people will express how many extra animals or benefit units can be produced by more intensive management on the areas remaining. In some cases (most I wager), animal abundance losses can be more than overcome by intensive management elsewhere. Wildlife potentials need to be estimated. Loss of 100 hectares producing at 0.10 of its natural potential or producing animals for which there is no current or projected demand, may be overcome by managing 10 hectares producing nearly at their potential. Mitigation is a concept implying replacement in like kind lands lost to various projects such as dams. Acquisition of wildlife lands for the public to replace wildlife lands lost is often a desirable action but rarely can lands of equal value or function be acquired and long term maintenance and management costs be the same or less. If faunal system management "works", it may be better to require trust funds for management of residual areas replacing the animals lost, for the long run, not the area lost. The reality of lands needed is blurred by whether land for fauna should be studied as if it is producing naturally or producing as it might under sophisticated management. The latter brings into discussions the questions of cost. There is at least one book to be written on this complex topic and I cannot resolve it briefly here. I emphasize: (1) the manager can only count the additional animals produced as a result of investments; (2) benefits, not animals, are to be produced, seeking to maximize Q*; (3) without a view of a need for or the possibility of increasing B in Q* = B/C, there is no loss in the wildlife resource when habitat is destroyed or lost; (4) areas lost or gained (as when parks or reserves are dedicated) are meaningful only in the context of Q* where benefits can be enhanced from different sizes and arrangements of areas depending upon or viewed as a function of cost. Complete enumeration of the options, while not yielding a global optimum, will probably show preferred solutions and even several nearly equal "best solutions" over which the public, politicos, and pundits can squabble for years. The manager may choose sides in such situations, then laugh in defeat, knowing that the lost cause was equally as good as the one overtly supported.
Confidence
Readers may review the concept of "confidence" in any statistics book. See Fig. 7.3. Here I wish the reader to see that the level of confidence at which the manager works is a managerial problem, not a dictate of a journal, a professor, or a concept "acceptable throughout all of agriculture." When a manager uses statistical tests to determine if a fertilizer was better than another one, if these users were more satisfied than those, or if production averaged J kilograms per hectare, then a confidence level is needed. The manager will want to say "this is the case, based on my limited studies, and I'll stand by it at least 9 times out of 10." This simply acknowledges that he or she has not studied all possible situations, that they may have gotten a peculiar sample, and that they cannot be certain. Being wrong one time in 10 may give people the shivers when dealing with a threatened species or an advanced-age timber stand, but for most wildlife managerial decisions, the manager of the long term resource will not suffer much if decisions are made at the 0.80 level - the chance of being wrong only 1 time in 5! CAP222 and CAP121 allow the reader to look at the consequences of changing the confidence level in sampling. The manager realizes that experience, tests, preliminary studies, etc. all lead to knowledge about a population. They rarely have time, money, resources or enough diverse situations in which perfectly designed studies can be done. Observing in the field is sampling. Knowing how many samples are really needed to draw a conclusion at a stated level of confidence and within acceptable limits of accuracy can have a very conservative influence on an observer. It can produce anomie; the manager may never be able to make enough observations to be confident enough to make a decision and act on it!
Take the situation in which the sampling equation is used:
n = s2t2 / d2
and n is the sample size needed; s2 the variance in observations based on quick preliminary analysis or reports from other observers; t2 the Student's t-test statistic taken from a table. Within t is expressed the level of confidence. Assuming at least 20 samples will be taken, then the value for t, when the confidence level is about 95 percent, is 2.08. When 80 percent, then t is 1.33. Look! The value 2.08 squared is 4.3 and 1.33 squared is 1.8; the one is 2.4 times larger than the other ... and so must be the value of n, the sample size. Nothing physical changed in moving from one value of t to another, only the manager's decision about the confidence level. The value of n increased and so did time, costs, processing, data analysis, data entry, etc. As suggested in Chapter 3 and 17, perhaps winning 51 percent of the time is all that is really required of the successful manager. At the 50 percent level of confidence, the value of t is only 0.69 and t2 is 0.47, about 10 times less work required in a managerial study than required if a level of 95 percent is insisted upon. When drugs are to be administered to humans, and we might be one to whom they are given, then we tend to think that a level of 99 percent or higher may be a good idea. When setting a squirrel hunting season or applying water to a wildlife food plant nursery, must the confidence be so great? If you personally are paying for the study?
The probability of being in error about a conclusion, this level of confidence (variously expressed as 90 percent or 0.10), is only one among the decisions to be made by the manager using the above equation. The other is the level of accuracy desired for the estimate that is being made. It is expressed as d2 and is "the mean plus or minus the product of some percentage."
d2 = ( Mean x a)2
In this case, a means a percentage divided by 100. If the mean value from plots is 200 grams and you are looking for very subtle differences, then plus or minus 5 grams may be reasonable (about 2.5 percent) (thus a = 0.025 and d = 200 x 0.025 and d2 = 25). However, if you are not sure about the wet-weight or dry-weight relations, know that the spring scale is of limited accuracy, that the clipping is to be done with 3 kinds of clippers, that it may be windy on the sampling day, and you are not sure of the care that will be used by all of your field staff, then plus or minus 20 percent may be reasonable (i.e., between 160 and 240 grams). Using this sample, what effect does a difference in alpha between 0.20 and 0.05 have on the number of samples needed. The relations: (200 x 0.20)2 = 1600 vs (200 x 0.05)2 = 100, or an increase of 0.15 in desired accuracy caused the sample size (whatever it was based on) to increase by 16 times! In using this equation for managerial decision-making, I encourage reducing the variance estimate, s2, by dropping outlier values if there are any, and dropping those suggesting a highly skewed distribution, using the 80 percent or less confidence level (t = 1.1), and using accuracy bounds of about 20 percent. The manager needs numbers expressive of the values near the expected mean, usually the median. The costs of thousands of samples dictated by such an equation with unreasonable confidence and accuracy limits should rarely be borne by the manager or those for whom they work. Statistical help needs to be sought and non-parametric techniques sought along with stratification and other techniques (Ackoff 1962).
Ownership and Access
 |
| Access can be combined with wildlife clearings, powerline rights of way, and alternative ownerships. |
Wright (1988) briefly reported on his study with Cordell, Rowell, and Brown on recreational access policies of individual landowners in the U.S. Average tenure of ownership was 23 years but only 38 percent resided on their property. The average ownership was 183 acres. If representative responses were gained, then 3.3 percent of the private, nonindustrial acreage in the U.S. is closed to recreation, 25.4 percent held for the exclusive use of the owners, and 46.8 percent open (with restriction) to certain others beyond the family. About 20 percent is open to the public.
Leasing and fee recreation have not increased; only 5 percent of owners lease any portion of their land for recreation (4.5 percent of the area). Forty-seven percent of the owners leased land for hunting, primarily (60 percent) for big game.
The relevant variables useful for study (as, for example, for entering into a multiple regression analysis (CAP71 and many commercial programs are available)) to discover the influence or role of owners in population size, rates of animal production, or reported user satisfaction are:
A somewhat rectangular management area of about 7000 acres (2800 hectares) has a perimeter of 111,420 feet (21.1 miles or 34 km):
There are other similar variables for studying ownership such as percent of land rented, with resident owner, corporate or family ownership, etc. The edge adjacent to the wildlife area relative to total area owned by a neighbor also suggests need for an expression of ease of working with neighbors, potentials for poaching, and the roles of other intensive beneficial uses.
Ownership of land largely determines whether wildlife will be managed or not; and, if so, then to what extent. Corporate owners and owners of large areas with substantial taxes may be easily convinced to engage in guild-like activity (Chapter 15). Other public lands (an enormous area in institutional land, rights of way, military, educational, unclaimed, abandoned mines, and other) are fertile areas for intensive forest faunal management. The private owner of a small amount of land (a majority in the southern U.S. being only 10 acres (4 hectares) each) is a very different problem. These people often take great pride in ownership, do not want intervention or second-party involvement in their use of land, do not want anyone "messing with their stuff", and know that decisions made for wildlife foreclose some other decisions later. They typically enjoy the freedom and knowledge of unspecified future options. Such freedom for future notions is a thing not quickly given away. The faunal resource manager needs to consider where to spend his or her personal life "coin." If influence on the resource is a criterion, then it will probably be spent in influencing a few individuals or corporations who own or control large areas of high quality land with minimum owners per unit of adjusted boundary (06). See Smith (1988).
On private as well as public lands, access means not only physical ability to get to and use resources but also legal and social ability. "Off limits" is well understood. For years hardwood supplies and growth have exceeded removal rates. There are many reasons for this. One notable one is that many owners of hardwood timberland use forests for recreation or other non-timber objectives (Josephson and Hair 1974:5). Denying such use or disagreeing with its propriety does not change the situation. Whether loggers want to cut trees for profits or faunal system people cut them for influencing certain animal species, makes little difference. The landowner holds decision power. They are likely to produce a tighter (than looser) timber supply in the future. The limits, the faunal system manager's constraints, will tighten.
Every tract of land acquired for or preserved for wildlife needs the rational scrutiny of questions such as:
Often the answer to questions 3 and 4 is yes, because it is possible to develop partial-area uses, to gain use during select periods, and to achieve land use control (Haigis and Young 1983) through:
Direct acquisition of land by a faunally-related public agency is one way to preserve habitat. This is expensive and may be prohibitively so or opportunities for purchase may be untimely. The Nature Conservancy, for example, a non-government organization, will occasionally purchase land for a government agency, then sell it to them, thereby jogging past the ponderous fiscal gait of such agencies that prevents them from acquiring exceptional land when it is offered.
The above list suggests options where land purchase is not feasible, where staff and acquisition resources are limited, and where social conditions prevent purchase.
In the U.S. there are at least 800 million acres of public lands that may be viewed as potentially manageable for faunal resource benefits. There are millions of other acres in military lands. Each state holds enormous area in rights-of-way, unclaimed and gift lands, educational and other institutional lands, conservation easements, and there are many others. Rights-of-way of utility and railroad companies represent vast potential areas on which the wildlife resource can be changed to produce significant net benefits.
The story about animal space becomes more complex because ownership can influence access to public or private forested land where benefits can be derived. A large, high quality public area with little or no access has low potentials for producing resource benefits. Wildlife, yes; benefit no. The Wildlife Management Institute (1984) reported hunter access to private land decreasing and since over 58 percent of all types of hunting is done on private land, the trend is of concern to hunters and agencies. The institute as others has listed reasons for posting being (1) bad experience of owners with resource users, particularly hunters; (2) apprehension about liability; (3) lack of economic incentives; (4) personal interest in property use for family and invitees, (5) trespassing and invasion of privacy, (6) littering, vandalism, and noise; and (7) fear of personal injury. Solutions advanced: (1) educating owners and users; (2) providing incentives (public leasing, tax payments, direct purchase of rights, etc.); (3) limiting or removing liability through insurance and state-level legislation (a model act is available); (4) providing cleanup; and (5) providing enforcement, especially rapid access to agents of the courts.
Forest survey data may be used to study the amount of commercial land that might be hunted intensively by making a plot of distance away from roads (Fig. 7.5).
 |
| Fig. 7.5. The distance of stands from maintained roads can be an indicator of potential human use. Forest or other wildlife roads have a zone of influence. Roads "cover" areas with potential use zones for hunting, observing wildlife, poaching, and eroding sediments. The green area at W must be removed as a contribution of the road; the area at the end (T) is an extra contribution of area covered by the road zone. Data from Michigan show how 85 percent of the forests are within 1 mile of a road (Spencer 1983). |
|---|
Preparing an area-accessible map and analysis and allocating use rates and costs to this area may provide improved insight into the results of management. Change in total available area resulting from bridges built, fords created, or land acquired may be a good way to express the change per unit invested. Often active management cannot be done because working equipment cannot be gotten to the area.
Roads, like powerlines and other major developments, need to go through a two-stage decision process. First, there must be a decision based on demand. Is a road really needed? Once the decision is reached, then stage 2 makes sense. There is no best place for a road. All have impacts. The solution must proceed to evaluate "least bad" road location and design (Rasmussen et al. 1980 and Jones et al. 1986). They can be co
