This article first appeared in WildEarth in 1996 as Special Paper Number 1. All reference to this paper should be given as S.C. Trombulak, 1996, How to Design an Ecological Reserve System, WildEarth Special Paper Number 1, pgs. 1-19. Access to this paper through the Web is granted by WildEarth in order to make it easier for conservation advocates to pursue the work necessary to protect and restore biological integrity and wilderness. Copies of the original paper can be obtained by contacting WildEarth at 802-434-4077.
********************************************************************
The readers of Wild Earth know that a new wind is blowing through the conservation community. The old philosophy of conserving biological diversity one endangered species at a time is being swept away. In its place has come the belief that our true goal needs to be the protection and restoration of the natural integrity of the fabric of life, and that this can be achieved in large part by creating ecological reserve systems that allow nature to operate in its own way, in its own place, and in its own time.
The work on this continent-wide reserve system has already begun in North America. Guided by the principles laid down by Reed Noss in The Wildlands Project Special Issue of Wild Earth (Noss 1992a), countless grassroots groups have begun the task of developing ecological reserves for their region. In addition, over the past year The Wildlands Project has sponsored numerous regional meetings to stimulate and facilitate the work of local activists in developing region-wide maps of possible reserve systems.
I have been fortunate to have had the opportunity to work with a number of these groups on their reserve designs and to be involved in the work of The Wildlands Project in the Greater Laurentian Region of northeastern U.S. and southeastern Canada. The vision and commitment shown by this new movement among conservationists is inspiring.
But I have also observed a problem. Although we all agree that a system of ecological reserves is our best hope of restoring biological integrity to North America, we don't all agree on or understand how such a system should be designed. The confusing jargon, piles of data, and non-intuitive mapping techniques are formidable barriers. Some people give up and never complete their maps. Those who persevere are each put in the position of having to re-invent the wheel. Individual ingenuity is often a virtue, but with respect to our goal of a coherent reserve system, this lack of uniformity puts us all in the position of creating a collection of reserve systems based on different criteria and designed to achieve different goals. Ultimately, such a collection would fail to connect together into a unified whole.
Over the past few years, the roles of The Wildlands Project and regional groups in designing and implementing an ecological reserve system has been clarified and an outline of the steps that need to be taken has been laid out (Johns and Soule 1995), but such guidelines are still short on the details of the design process itself. This article seeks to fill that gap. Rather than trying to figure out for oneself how to design a reserve system that will protect and restore biological integrity, everyone ought to be able to follow a common plan. What follows here are some simple guidelines and conceptual tools to help you with your work. They are organized as a series of questions and answers developed through my work with The Wildlands Project and with the advice of numerous people who have pioneered this effort.
Different people will extract different things from this article. Some of you have already developed reserve systems and are looking for ways to refine and revise your work. Others are working on only parts of a system, such as a wildlife travel corridor between two parcels of public land. Others would like to design a system but don't know where to begin. The organization of this article should make it possible for all to get what they need.
Of course, a general guide like this has limitations. The details of where data are available and what data are relevant vary from place to place. In some states, data on the location of rare, threatened, and endangered species are freely available from state agencies. In others, the data are proprietary and available only after the delivery of large sums of money and promises of future favors. Also, I have my own limitations. For example, this article is more directly applicable to people working in the U.S., specifically in the east, since that is primarily where I have gained my experience. You'll need to adapt your approach to the specific conditions of your region. (I hope some folks with experience designing reserves in Canada and Mexico write follow-up articles to this one.) No cookbook is a complete substitute for imagination and finesse.
Don't be intimidated by all of this. The original wilderness movement was fueled largely by a recreational philosophy (e.g., a wilderness area ought to be large enough for a person to take a two-week pack trip without crossing their trail). We all had an intuitive feel for that philosophy. We knew what a rugged, natural landscape looked like. We knew what a good whitewater river looked like. Now the rules of the game have changed. No longer is it sufficient to have an adventurous sense of the outdoors, for this aesthetic alone will not guarantee biological integrity. Now it is necessary for conservationists to think like conservation biologists, using concepts and tools that were not used even by academic biologists until about fifteen years ago! It's bound to seem overwhelming at first. But trust me: if you are patient and persistent, you will learn these skills at least as well as the so-called professionals, and together we will succeed.
Critical Concepts and Tools
Before we can agree on how to get there, we need to agree on where we're going. Let's look first at some of the critical concepts you need to understand before you can design an ecological reserve system, and some of the basic tools you can use.
1. What is the primary goal? The primary goal of a truly comprehensive conservation strategy is the protection and restoration of ecological integrity throughout a region. Protection because we need to insure that what remains is not further eroded. Restoration because returning ecosystems to their natural condition, whenever possible, is one of the best ways to protect against further loss. This is the mantra. Write it on a piece of paper and tape it above your desk. The measure of all your work is how well your design achieves this goal.
2. What are the specific objectives of the primary goal? The primary goal can be divided into four basic objectives (Noss 1992a):
(a) represent all native ecosystem types and successional stages across their natural range of variation;
(b) maintain viable populations of all native species in natural patterns of abundance and distribution;
(c) maintain ecological and evolutionary processes;
(d) maintain responsiveness to short-term and long-term environmental change, and maintain the evolutionary potential of lineages.
3. What strategy will best achieve these objectives? The objectives of the primary goal can best be achieved through the establishment of a system of ecological reserves designed to protect and buffer large core areas of land and water and to allow for biological connections among these areas. The theoretical justification for this system, sometimes referred to as the Core/Corridor Model or the Matrix Model, is described fully by Noss (1992a).
4. What concepts do you need to understand to develop a reserve system that achieves the four objectives? Like most professions, conservation biology involves a large number of specific terms, often used without regard to whether everyone understands them. This becomes a serious problem when people are expected to use the terms correctly in their conservation work. For developing a solid understanding of conservation biology, nothing will replace reading a good book or taking a good course on the subject; but many of the specialized terms can be easily explained. What follows is a brief description of the key concepts needed to design ecological reserves to achieve the four objectives. These descriptions will be kept as simple as possible and will necessarily not tell the complete story. But they provide a good beginning. Readers ready for more comprehensive discussions of these topics should see the references mentioned below (see What should you read to help you with your work?).
Species. The species is the fundamental unit of classification of life. There are several different definitions of a species, but for conservation purposes a species is usually considered to be a group of organisms that can potentially interbreed with one another in nature. This definition does not work well for all life forms. For example, some life forms are asexual, and never interbreed at all. Therefore, biologists often consider organisms that essentially look alike to be of the same species and organisms that look noticeably different to be of different species unless information suggests otherwise.
Population. If a species is a group of individuals that can potentially interbreed with one another, a population is a subset of this group where the individuals live close enough to one another and interact with one another frequently enough to make that potential real. A species can then be considered a collection of populations. Frequently populations don't have distinct boundaries and the term is used to mean "the collection of individuals of this species in this local area."
Genetic diversity. Every organism on Earth is the way it is to a large extent because of the information coded in its DNA, complex molecules inside almost every living cell. Although the environment plays a role in influencing behavior and abilities, the information stored in DNA, genetic information, can determine how an organism looks, responds to climate and disease, reproduces, behaves toward others of its own and other species, moves, eats, and so on. Every species has a unique pool of genetic information, a genome, so that a species can often be described by its characteristic looks, behavior, and abilities. Yet the complete set of genetic information for a species is very rarely carried by any one individual. To have the entire range of genetic diversity for a species, multiple individuals are needed.
Minimum viable population. Viable for conservation biologists means "able to survive." A minimum viable population is the smallest number of individuals that will allow that population to survive. Since survival of populations is a goal of conservation, maintaining viable populations is obviously of great importance. Small populations are more susceptible to extinction than large populations because they are more likely to be completely wiped out by a single random event--like a hurricane or plague--or suffer genetic problems due to inbreeding or the loss of important genetic information. Therefore, the minimum viable size is really the lowest acceptable size; it is not an optimal size. Being above the minimum size adds a margin of safety. Many factors will influence what the minimum viable population size is for a species: its mating system, age at maturity, life span, the amount of existing genetic diversity within the population, the frequency and severity of environmental disturbance to which it is subject, and so on. Also important is the length of time over which you want the population to survive. No species is ever 100% safe from extinction forever, no matter how big it is. Just consider what happened to the dinosaurs. Viability estimates are usually expressed as something like, "There is an X% chance that this population will survive Y years." Everything else being equal, populations need to be bigger the longer you want them to be safe from extinction and the larger the probability you want them to survive for a given length of time.
Calculating minimum viable population sizes requires detailed ecological information and some mathematical training, and the best estimates will be different for each population of each species. In the absence of these calculations, a rule of thumb you can use is that for most species that have been studied, minimum populations for long-term viability and evolutionary potential fall within the range of 5000 to 10,000 individuals, and sometimes the number is higher.
Metapopulation. This refers to a group of populations that are loosely connected to one another by immigration and emigration. Metapopulations are of interest in conservation because it appears that for some species, for example butterflies and amphibians, individual populations depend on immigration in order to remain viable. This is one reason why habitat fragmentation is such a problem; it lessens the connections between populations, which then may result in permanent local extinction.
Subspecies. Another way to subdivide a species is by distinctiveness in appearance. A group of individuals within a species that looks noticeably different from another group within the same species is called a subspecies. The characteristics that set subspecies apart are often subtle; not large enough to put them into different species. The physical differences between subspecies imply genetic differences between them as well. Different subspecies are often found living in different environmental conditions, such as the northern and southern parts of the species' range, which may partly explain the genetic differences among them. Some evolutionary biologists think that the development of subspecies or genetically distinct populations adapted to different environmental conditions is the first step in the evolution of new species. Subspecies are of conservation concern because they each contain unique patterns and combinations of genetic information and are part of the evolutionary process.
Community. A community is a collection of interacting populations of different species. The term is also sometimes used to mean a collection of ecologically-similar populations, such as a "bird community" or "aquatic plant community."
Ecosystem. An ecosystem is the combination of the community of organisms in an area and the non-living components required for life: air, water, nutrients, and energy. Like other ecological groupings, ecosystems do not have precise boundaries, but the term is often applied to areas with common physical and biological features, like forest ecosystems and aquatic ecosystems. Use of the term in this way lets us recognize nested sets of ecosystems; for example, the beech-maple-birch forest of the Northern Appalachians is a subset of the northern hardwood forest ecosystem which is a subset of the deciduous forest ecosystem (which is a subset of the forest biome).
Biological diversity. The diversity of life at all levels of organization, including genetic, species, and ecosystem. The first two basic conservation objectives described above relate to the conservation of biological diversity. Note that distinctions are not made among species based on how cute or closely related to humans they are. Soil bacteria, trees, fungi, insects, plankton, sharks, and the ecosystems they participate in are (from an ethical standpoint) all equal members of the biosphere.
Ecosystem processes. The components of an ecosystem do not exist separate from one another but interact in various ways. Interactions between species (e.g., predation, herbivory, parasitism), movement of energy and matter from place to place within the ecosystem (e.g., water cycle, nutrient cycle), and changes in the ecosystem (e.g., succession, fire, drought) are all ecosystem processes, and are as vital to biological integrity as are the species and ecosystems themselves.
Home range. An animal's home range is the amount of area it typically uses. Its daily home range, its yearly home range, and its life-time home range may differ because of different needs over those time periods. A territory is a defended home range. To use the home range concept for conservation planning, it's best to consider life-time home range, since that is required to let the animal successfully live its full life.
Spatial scale. This just means the amount of land and water you are looking at. Some ecological questions only require information from a small spatial scale for an answer, such as "How far from its nest will an ant walk in search of food?" Other questions require larger scales, such as "What controls the weather?" The answer to a specific question may depend on the scale at which you look. For example, the answer to the question "Are the trees in this area getting older?" depends on the size of the area. If you look at only one acre, you may conclude that the stand is maturing. But if you look at a much larger area, the pattern shown by the few trees in the smaller area is then averaged with the patterns of many more trees, and the results may be different. The few trees may be getting older, but over the larger area the average age may not be increasing as some old trees die.
Conservationists now understand that it is important to consider large-scale spatial patterns in their planning, such as continental migration of birds, wide-ranging disturbances, and long-distance dispersal of individuals.
Temporal scale. The temporal scale, or time frame, over which an event occurs can range from very long (for example, the first appearance of life on Earth took about 1 billion years) to very short. In conservation, we need to be concerned about events at all temporal scales, including evolution (from several years to millions of years), climate change (hundreds to thousands of years), major range shifts of species (hundreds of years), population growth (years), habitat modification (days), and disturbance (minutes).
Succession. Ecosystems change. Species composition, environmental conditions, and physical structure all may change over time. In many cases, this change is predictable, in that a bare surface in a given place will pass through a somewhat predictable set of stages, or seres. Although the concept of succession is most commonly applied to forest ecosystems, it is relevant to all others, too.
Ideas about ecological succession have changed in recent years. First, in many ecosystems, the exact species composition achieved may not be predictable, but is instead determined by random events, such as the identity of the first individual to colonize. Second, many ecosystems do not tend toward some stable "climax" community. Rather, the rate of natural disturbance may be such that the "successional clock" is frequently reset, and the vegetation in a large region is constantly changing. Third, adding an ethical dimension, all successional stages have unique biological value and should be maintained in a landscape. Some species depend on early successional conditions, like bare ground and high light levels; others depend on later successional conditions, like high humidity and thick ground cover. Labels that imply a negative value to a successional stage (e.g., "overmature" forest) or practices that restrict natural disturbance regimes (e.g., fire suppression, flood control) should be eliminated.
Natural. A natural system is one that looks and operates as it would in the absence of the expenditure of cultural energy by humans (Anderson 1991). This definition does not conflict with the view that humans are a part of nature, since it must be possible to define natural as a subset of things in nature--otherwise we would have to consider all deaths, even those caused by human violence, to be due to "natural" causes (Trombulak 1995). Nor does it mean that any system with humans in it is unnatural; it is possible for humans to live in a landscape without altering how it would look or operate. For conservation, the terms "natural" and "unnatural" refer not to two conditions, but rather a spectrum describing how altered a system is.
Natural range of variation or variability. Nothing in nature remains constant; everything varies over both time and space. However, this does not automatically justify human manipulation of population sizes or ecosystem conditions. Biological integrity requires that the variation occur within a specific range, the range of change to which species are evolutionarily adapted.
Native. An organism that evolved in an area or migrated to an area under its own power (or by means of wind, ocean current, avian dispersal, or some other natural force) is considered to be native or indigenous.
Exotic. An organism that was introduced into an area by humans, either accidentally or on purpose, is an exotic. Such species are also called non-native, non-indigenous, or alien.
Biological (or ecological) integrity. This refers to how close an ecosystem looks and functions to the way it would if humans were not present. Biological integrity will increase in a region as we accommodate the four basic objectives listed above.
Mapping scale. This refers to how distance on a map relates to distance on the ground. Mapping scale is expressed as a ratio of map distance to real distance. If 1 inch on a map equals 100 inches on the ground, then the scale is 1:100. Since mapping scale is a ratio, whether the scale is large or small depends on the fraction of the two numbers. For example, 1:100 is a larger scale than 1:1000 because 1/100 is a larger number than 1/1000. The smaller a map's scale, the less the detail and accuracy but the more real area shown for a given map size.
Digital and analog. Something is digital when it is expressed in numbers and analog when expressed in some other kind of symbol. Think of a clock. A digital clock displays the time as specific numbers. An analog clock displays the time as a big and a little hand pointing to numbers which symbolize the minutes and hours. Spatial data, such as the location of a stand of old growth, can also be in digital or analog form. Digital spatial data might be the latitude and longitude of the stand recorded as numbers. Analog spatial data might be a dot on a map symbolizing the location of the stand.
GIS. This acronym stands for Geographic Information System. GIS is nothing more than a class of computer software programs that, among other things, can manipulate and display digital spatial data. Individual data sets (for example, the locations of all endangered species in an area) can be plotted on a map, and multiple data sets can be mapped together to look for overlap. The mapping tasks performed by GIS can also be done by hand using transparencies on paper maps; but with GIS, data and maps can easily be edited and updated, and numerical calculations from the data (for example, calculating the percentage of all roadless areas within an existing ecological reserve) can be made. GIS technology is rapidly improving, becoming easier to use, cheaper, and available for personal computers.
Core area or reserve. A strictly protected area managed to maintain or restore its biological integrity.
Buffer zone. A region surrounding a core reserve designed to minimize stresses on the core that would decrease its biological integrity. For political reasons, some choose to call these zones transition areas, zones of cooperation, multiple-use zones, ecosystem management zones, or sustainable development zones.
Corridor. An area that provides for the natural movement of individuals among core reserves. The primary purpose of corridors is to provide biological connectivity. Therefore, they are sometimes called "connectivity zones" or "landscape linkages" to avoid the false impression given by the term "corridor" that connectivity is achieved only through long, narrow passageways.
5. What should you read to help you with your work? Literature is the fuel that makes this process run. I highly recommend each of the following as starting points.
The Wildlands Project: Plotting a North American Wilderness Recovery Strategy. Wild Earth Special Issue 1992.
Reserve Design Framework Package. This is a compilation of papers that provide a wide array of examples, guidelines, and lists of resources. It is available from The Wildlands Project by contacting Rid Mondt (P.O. Box 5365, Tuscon, AZ 85703; 520-884-0875).
Reed F. Noss and Allen Y. Cooperrider. 1994. Saving Nature's Legacy: protecting and restoring biodiversity. Island Press, Washington, D.C.
William S. Alverson, Walter Kuhlmann, and Donald M. Waller. 1994. Wild Forests: conservation biology and public policy. Island Press, Washington, D.C.
Gary K. Meffe and C. Ronald Carroll. 1994. Principles of Conservation Biology. Sinauer Associates, Inc., Sunderland, MA. (Or any other basic text on conservation biology.)
Robert E. Ricklefs. 1993. The Economy of Nature. 3rd edition. W.H. Freeman, New York. (Or any other basic text on ecology.)
Conservation journals, particularly Conservation Biology (Blackwell Scientific Publications, Inc., 238 Main Street, Cambridge, MA 02142), The Natural Areas Journal (108 Fox Street, Mukwonago, WI 53149), and Wild Earth (P.O. Box 455, Richmond, VT 05477).
The Conservation Directory. National Wildlife Federation, 1400 16th Street, N.W., Washington, D.C. 20036-2266. Updated every year, this book has listings of all state, provincial, and federal offices in the U.S. and Canada that have anything to do with conservation or natural resources, as well as of most private conservation groups, colleges, universities, and environmental on-line databases. It isn't a book to read, but a book to use when you need a phone number, address, or a description of what a particular agency or group does. This will come in handy as you track down specific sources of data in your area.
6. How do you perform a literature search? In your work you may find it necessary to read extensively on a particular subject, such as what is the home range of a Grizzly Bear, what kind of habitat does a Wolverine need, or how did the pine forests move northward through the Rocky Mountains after the last continental glacier retreated. Guessing won't do, and looking it up in the encyclopedia only worked in high school.
You need to find and read as much previously published information on the subject as you can. This actually has two advantages. One is that you get the right answer and your reserve design does a better job at achieving the objectives. The other is that you will be better informed than those who will oppose your design when you try to implement it. Reading alone won't transform you into a professional biologist, but you will gain a great deal of credibility when you can cite the numerous scientific studies that support your plan.
The best way to find all this information is to perform a literature search. Many environmental on-line databases are listed in the Conservation Directory. Literature searches are quick and easy to do yourself if you have access to a college or university library. Make a list of the types of sources for information on your question. For scientific literature, restrict yourself to books, journal articles, and perhaps government documents. Magazines and newspaper articles usually only report the results of other studies, and are not authoritative by themselves. Make a list of the keywords associated with your question. The more specific your keywords, the more helpful the references you find will be. The more general the keywords, the more references you will find. You'll have to experiment to find the best compromise.
Most academic libraries have their "card" catalogs on computer, which makes it easy to search for books on keywords. Libraries that are repositories for government documents have them cataloged on a CD-ROM provided by the government, which also allows you to search on keywords. You will find it hardest to search for journal articles. Some journal databases are on-line and accessible through the same system as the book catalog. That is unusual, however, and you will probably need to search on your keywords through indexes by hand. For articles about conservation, the two most likely indexes are Biological Abstracts and Zoological Record, which are available in most academic libraries.
7. Why should you develop a cooperative relationship with the academic community and how can you do it? (This question is for those who aren't already a part of the academic community or collaborating with someone who is.) Academic institutions have many resources, including people, equipment, and information that will make your job easier and your final reserve system more successful at achieving biological integrity.
How can you develop such a relationship? There is no guaranteed method, but here are some steps that will improve your chances. First, make a list of all the academic institutions in or close to your area. Don't limit yourself to big research universities. Often small colleges have good resources and interested people, and differ from universities only in that they don't have graduate students. The Wildlands Project can also advise you about the academic community in your region.
Next, look in the catalogs of those institutions (available by calling the Admissions Office and asking for a copy) and list those that have the kinds of resources and people that match your interests. You are looking for people who teach courses like ecology, conservation biology, regional planning, geography, GIS, environmental studies, remote sensing, or wildlife biology,. This will only give you information on faculty, but they can often help you make connections with students. Also, talk to other conservationists in the area and ask whom they have worked with in the past and whom they recommend you approach.
Once you have decided whom you would like to work with, make your approach. It's like selling encyclopedias door to door: there's no sure recipe for success, but you stand a better chance if you understand the person on the other side of the door. Like everyone, faculty are busy and don't feel they have time for anything else in their lives. So you won't have much success if you walk into someone's office and say, "Hi, I want you to collaborate on the design of a reserve system for this state."
You will have more success in the long run if you start small. For example, ask if they would be willing to evaluate an early stage of your design for scientific accuracy. Ask a specific question that connects their area of expertise to your project. Ask if they know of other faculty or students working on such questions. If they teach an advanced seminar or discussion group related to conservation, ask if you can sit in and listen to help you with your work. Make a list of small projects associated with your work that you think would be appropriate for undergraduate or graduate students. These might include literature searches, initial maps of single data sets, ground truthing, or small GIS projects. Offer to sponsor a student intern to work with you on your project. Volunteer to give a guest lecture on your project or region.
In other words, make yourself a resource rather than a person who just wants something for free. If the grounds for a long-term and serious collaboration exist, the rest will fall into place.
Getting Started
1. What area are you working in? Every conservationist wants to save the world, but it's a pretty big place to deal with as a single project. You need to be able to describe your area from the beginning of your work so you don't make any major planning mistakes. There is an important trade-off to keep in mind: the larger your area, the more data and maps you will need to acquire and the longer it will take to do your work, but the easier it will be for you to include an entire ecologically-relevant region. How do you balance these two concerns? Start with an entire area that you can describe in geographical terms; for example, a mountain range, valley, watershed, plateau, or vegetational province. Determine how many states it covers in whole or part. The more states it covers, the more sources you will have to go to for data). If it covers more than one state, ask whether you make the region ecologically irrelevant by reducing the area. That is, can you make an ecological argument for working on only a subset of the area? For example, the Connecticut River watershed includes parts of one province and four states, but it can be subdivided into northern and southern parts based on vegetational types. The deserts of the Southwest can be subdivided by climate. The Pacific Northwest can be subdivided by mountain ranges.
There is no good rule of thumb on how long it takes to map a reserve system for a given amount of area. It will depend on your experience, the amount of time you can devote, and the ease with which you find the data. If it is going slower than you like, subdivide your area. If it is going faster than you predicted, finish your work and then start on an adjacent area.
Although it is best to define your area based on ecological or natural geographic features, an alternative worth considering is to choose an entire state. Since you will get most of the data you need from state agencies, it is often just as easy to get the data for the entire state (ecologically relevant or not) as it is to get the data for only a portion of it. Wildlands proponents have already produced good preliminary reserve maps of both Minnesota and Pennsylvania, even though both of these states include multiple distinct bioregions.
2. Over what time frame do you plan to have your system implemented? The answer to this question will influence the decisions you make on how to deal with the data you collect. For example, if you hope to have your system implemented by next year, then there is little chance that any roads will be closed, land purchased, or management plans changed, and your reserve design will need to reflect those limitations. However, such a system will certainly not completely achieve the four basic objectives. If, on the other hand, you hope to have it implemented in the next 200 years, then almost anything made by humans could be gone within your planning time frame, including major cities and interstate highways, and you should completely ignore their current locations. The resulting system would almost certainly meet the four basic objectives, but at a time so far in the future that it may be of little real value to the crisis we now face.
There is no one right answer. Some have argued for a 10-year time frame (hoping to promote rapid progress toward the objectives) and others for a 75-year time frame (hoping to develop a system that will be more effective at achieving the objectives). The Wildlands Project recommends an incremental plan (10, 25, 50, 100 years) recognizing that full biological recovery may still take 200-500 years.
3. Who else has worked or is working on a reserve system in your area? Chances are good that the area you want to work in has not been looked at in this way before, but why risk re-inventing the wheel. Perhaps someone else has just finished doing all this work and needs help revising the plan and getting it implemented. Perhaps they collected some of the data but stopped because they needed help. Check around to see. Here are some places to ask whether anyone is developing a plan for a system of ecological reserves in your area:
(a) Wild Earth, P.O. Box 455, Richmond, VT 05477 (802-434-4077). Look in back issues for articles about your area or articles written by people who live in your area. Ask the Wild Earthlings whether they know anyone working in your area. Send them a letter to publish that asks for anyone who has done work in your area to contact you.
(b) The Wildlands Project, P.O. Box 5365, Tucson, AZ 85703. Write and ask whom they know working in your area.
(c) Planet Drum, P.O. Box 31251, San Francisco, CA 94131. Planet Drum acts as a clearinghouse for information on bioregional groups throughout North America (and elsewhere). Find out if any bioregionalists are active in your area and ask them what they know about any work being done on regional ecological reserves.
(d) Sierra Club, 730 Polk Street, San Francisco, CA 94109. The Sierra Club, through their ecoregions campaign, has taken a great interest in the subject of ecological reserves. Tap in to their network.
(e) Your local and/or state chapter of National Audubon Society, The Nature Conservancy, and Sierra Club. These groups themselves are probably not doing this work, but often the individuals in these groups know who in the regional activist community is doing what.
(f) Conservation organizations specific to your region, such as the Friends of the X or the Alliance to Save the Y. You can locate these groups through the informal network of grassroots wilderness activists often called the New Conservation Movement, or by checking the state listings in the Conservation Directory.
(g) The Information Superhighway. OK, so technology is the root of all evil, but the fact is some routes on the superhighway give you access to a wider community than you might reach otherwise. Depending on your access, try posting a message on the Internet, Compuserve, America Online, or Prodigy. These services are growing so quickly that the appropriate lists and newsgroups will surely have changed before this is published. Search for them with relevant key words like wildlands, wilderness, and conservation.
4. What maps do you need and where can you get them? Once you've decided where you are working, you need to start acquiring maps. Unless you plan to work exclusively with GIS on a computer (see Do you need to know or use GIS technology?), you'll need at least two basic types of maps: base maps (used to draw your borders and geographic features) and data maps (used to obtain data for your reserve planning).
For base maps, I recommend using US Geological Survey topographic maps. These are available at most recreational stores (at least for the local area) and directly from the USGS (call 1-800-USA-MAPS). USGS maps come in several scales and types. For accuracy, it is necessary to map at a scale of 1:100,000 or larger. This means using USGS maps of one of the following types: 7.5-minute (approximately 1:25,000), 15-minute (approximately 1:50,000), or 30 x 60-minute (1:100,000). The larger the scale you use, the more accuracy you get, but the more maps you need to cover a given area. A 30 x 60-minute map covers the same area as eight 15-minute or thirty-two 7.5-minute maps.
USGS also produces maps at smaller scales, like state maps (usually 1:500,000) and regional maps (1:1,000,000). These are useful for keeping the big picture in mind, but they cover too much area with too little detail to be useful for the mapping itself.
If your area is entirely within a single state, you may be able to use maps prepared by a state agency. For example, in his reserve design for the Oregon Coast Range, Noss (1992b) used Forest Protection District maps from the Oregon State Department of Forestry.
It is also necessary to be consistent in the scale you use. You shouldn't map part of your area at one scale and another part at another scale.
If you are working with several maps, code them so they are easy to catalog and match with data maps. Keeping track of all the maps you use may turn out to be the most difficult part of your task. I recommend giving each map a row and column code. For example, if your region is covered by nine maps arranged in three rows and three columns, label the rows A through C and the columns 1 through 3. Each map then gets its own code, such as A1 or C3.
Getting the Data
1. What data sets are of critical importance and where do you get them? In general, you will get most of your data from state agencies, even if you are working in only a subset of a state. Many data sets originally developed by federal agencies and private conservation organizations are available from state agencies (e.g., wetlands inventories, Natural Heritage element occurrences), making them the first place to look. Consider both the agency responsible for the type of data you are interested in (e.g., fish and wildlife, transportation, forestry) and the agency responsible for GIS databases regardless of their type. Because different states have different agencies, priorities, and programs, I can only give a rough guideline to sources of data. You will need to refer to the Conservation Directory to make sure you have the right office in your state and ask numerous questions of the people who work there to learn of all that they have available.
I find it useful to think of data sets as falling into three distinct categories: critical, useful, and unimportant. The following are critical data sets:
Locations of rare, threatened, and endangered species, populations, and communities. These are the biological "elements" that some government agency, either state or federal, has identified as being of special concern for conservation. That doesn't mean they are the only elements of concern. Numerous species are so poorly known and the listing process is so slow that many additional species that probably should be classified as being of special concern are not. Nonetheless, the officially-designated elements are a good place to start because data on their locations are often easy to obtain.
The best source of these data is the state agency that houses the Natural Heritage Program. This program, started by The Nature Conservancy, seeks to inventory all biological elements within each state. Although these inventories are generally still incomplete, they are used to help determine the conservation status of species, populations, and communities. A standard classification scheme involves ranking each element according to its global (G) and state (S) abundance. Each category contains five abundance ranks, ranging from 1 (less than six viable populations or less than 1000 individuals total or a geographic distribution of less than 2000 acres) to 5 (numerous populations or individuals, or widespread geographic distribution).
Obtain the locations of all G1 though G3 and S1 through S3 elements in your region, which you can think of collectively as the rare, threatened, and endangered (RTE) elements. The data may be available both digitally and on maps. Data are usually "buffered," which means that the exact location of an element is not given, but rather an area of several thousand square meters that includes the element somewhere within it. This is done to protect the element from potential poachers, but is not a problem for you. Since you will be designing reserves on the order of square miles, data buffered on the order of meters is just fine.
A category of rare, threatened, and endangered communities that may not be completely reported by the Natural Heritage Program consists of unmodified ecosystems or ecosystems so rare or imperiled that all remaining examples should be protected. These must be individually defined in your area, but may include old-growth forest, native grasslands, wetlands, and spawning grounds. This category of data is the hardest to give general advice about because good data are so few and potential types are so many. In general, start with state agencies and The Nature Conservancy to see whether they have inventories of native ecosystem patches and maps of their locations, and include these in your data set on RTE elements whether or not they have been given a G1-3 or S1-3 rank.
Tracts of old growth are generally poorly mapped, and you will likely only find approximate locations. In the eastern U.S., the inventory of old growth by Davis (1993) is the best initial source of information.
Wetlands were mapped in the late 1970s for the National Wetlands Inventory. These data are available from state agencies, many of which have updated the data set with their own inventories. However, the inventories may be incomplete, especially for wetlands below three acres in size.
Natural ecosystems. Knowing the distribution of ecosystem types in your area is key to meeting the first basic conservation objective. Unfortunately, standardized systems for classifying ecosystems in detail do not yet exist. Ecosystem maps based on state-specific schemes may be available from state agencies of natural resources or planning.
A hierarchical system of ecoregion classification based primarily on vegetational type, known as the Bailey-Kuchler system, has been developed for North America. This system continues to be refined to some extent, especially in the U.S. by the Forest Service. You can get access to the most recent classifications in your area by calling the nearest U.S. Forest Service Experiment Station and talking to someone involved with regional ecosystem mapping.
The U.S. Forest Service also has digitized data available on forest types at a 1 km2 scale throughout the U.S. (contact the Southern Forest Experiment Station at 601-338-3110 for information about availability). If your area is predominantly forested and you are working with GIS, this data set is indispensable.
Existing public land. The purpose of this data layer is to identify areas owned by the public that may already operate as a core reserves, buffers zones, or corridors, or could do so with increased size or changes in land-use plans. Keep this purpose in mind when you decide on what kinds of public land to include and exclude. For your purposes, a forest is more useful than a county courthouse.
Public land can be roughly divided into three categories: federal, state, and municipal. Federal lands are the easiest to map since their borders are well known and appear on almost every map you will find of your area, including state atlases and USGS topographic maps. Digitized boundaries of federal public land may be available from your state GIS offices. The categories of federal land of most interest are National Forests, Parks, Grasslands, Wildlife Refuges, Seashores, Marine Sanctuaries, Monuments, Preserves, Parkways, Lakeshores, Reserves, Wild and Scenic Rivers, and Recreation Areas, as well as Wilderness Areas, Bureau of Land Management lands (especially Areas of Critical Environmental Concern), Research Natural Areas, and Estuarine Research Reserves.
Accurate data on state public land is also generally available either from the state agencies responsible for managing the land or from the state GIS offices. When talking with them about getting maps, be sure to indicate that you want maps with accurate boundaries at a scale of 1:100,000 scale or larger. If you don't, they will probably give you the standard tourist map, which is not accurate for your purposes. As with federal land, there are many types of state public land, including forests, parks, recreation areas, wildlife management areas, and wilderness areas. You may need to go to more than one state agency to get the maps. Keep a list of the types of land you have information on, and ask each person you talk to whether there are other types you should consider and who administers them.
Municipal land, which I use to mean everything administered by a government at a level below that of the state, is much harder to get accurate information on. Counties and towns may own forests and parks but not know exactly where the boundaries are. If you cannot get accurate information on municipal public lands from a state map, compile a list of all the municipal lands you are aware of from all your sources, such as state atlases, town plans, conversations, and newspaper articles. For each town in your area, ask the town clerk for a copy of the survey map for each parcel. These will probably be available for only some of the parcels. For the others, get a map of the general area where the parcel is, find out who would be the best person to estimate where the boundaries are, and get them to draw these on your map. When developing your final data set, parcels with estimated boundaries should be distinguished from those with exact boundaries.
Existing private conservation land. The best example of privately owned land specifically managed for conservation goals is the holdings of The Nature Conservancy. Contact the office in your state for information on the availability of accurate maps of the boundaries of their land.
Regional conservation groups, Trust for Public Land, local Audubon groups, local land trusts, and educational institutions may also own land. Ask them.
Roads. Data on roads are available from state departments of transportation, USGS and BLM maps, state atlases, and basic road maps. Digitized data are also available from GIS offices. It is useful if each road is categorized by type, such as primary, secondary, and trail. This allows you to map them as separate data layers to experiment with the size of core reserves you could design by closing roads of varying categories. Note that data on roads allows you to determine the location of roadless areas. Maps of the locations of most large roadless areas are available from the BLM and Forest Service.
The preceding five data sets are critical to your work. Another six data sets will also be useful if you can acquire them for your area. Useful data sets include everything else of any biological or ecological relevance, plus data sets relevant for implementation of the system, such as:
Soil types. Ecologists generally know little about biological diversity within soil, especially how it varies from one ecosystem to another, but using soil type as an indicator of soil biological diversity may be a reasonable alternative to an incomplete or non-existent inventory. These data are available at different scales for different regions from the Soil Conservation Service or state GIS offices. Because soils are described as a series of nested categories, you may find that you can get more data on soil type than you actually need. Try to get the most detailed data you can for each part of your area and use only the level of detail necessary to meet your conservation objectives (see What do you do when you have too much data?)
Distribution of all biological diversity. Ideally, your reserve design would protect all species, no matter what their status, not just rare, threatened, or endangered species. This would require, however, information on exactly where every species in your area is found (or the proposal of your whole area as a core reserve, which would generally not be taken seriously). Such data exist at relatively course spatial scales for a few well-studied taxonomic groups like birds, but for most groups are completely unavailable. I mention this data type primarily as a reminder of your true objectives and in recognition that one day such data may become available and will need to be incorporated into reserve designs.
Unfortunately, even when the National Biological Service's gap analysis projects are completed across the U.S., information on what species live in each square kilometer of land will not be available, as these projects predict species distributions at coarse scales based on existing distributional data, vegetation maps, and simple wildlife-habitat association models. To find out more about the status of the gap analysis in your state, contact the National Biological Service in Washington, D.C. and ask for the name of your state contact person (United States Department of Interior, National Biological Service, 1849 C Street NW, Washington, D.C. 20240).
Travel routes of species. Individuals travel, which is one of the main reasons the core reserves need to be connected to one another biologically. Although we commonly think of movement as being just by animals that move from one area to another as part of their natural patterns of daily or seasonal movement, over long time scales plants move as well. Rather than blindly designate a corridor at random and hope the plants and animals use it, research where they traveled in the past or try to travel now and designate your corridors there.
Sadly, the best source of data on animals' travel routes is from road kills, data on which may be available from state departments of transportation or fish and wildlife. If available at all, however, road-kill data are likely to exist only for larger species on major roads. Recording road kills of less noticeable species, like salamanders and turtles, and on other roads is a good project for volunteers or interns. General patterns of movement for some species can be inferred from information on their seasonal ranges. For example, state departments of fish and wildlife have data on summer and winter ranges of large herbivores, and the U.S. Fish and Wildlife Service has similar data on birds.
Search for information on patterns of movement by plants through your area in the scientific literature. Keywords include vegetation, paleontology, Holocene or Neogene (two names for the present geological period), and the name of your region. You may learn of important references by asking botanists at the Natural Heritage Program or a local college.
Watersheds. Because many aquatic species are restricted to particular watersheds, knowledge of watershed boundaries may help you deal with the problems of incomplete information on the species themselves. If you include representation of all watersheds in your reserve design, you increase your chance of including all species. Data on watershed boundaries may be available from state geology departments or GIS offices. If not, draw your own watershed map with a topographic map. Like ecosystems and soils, watersheds are classified as a nested set of categories. First, identify the primary watersheds in your region, those with a single drainage to the ocean. Locate each river that has a separate drainage into the ocean and draw lines along the high ground that separates each drainage basin. Then, for each of the primary watersheds continue this process as you group sets of subregions: the major tributaries (secondary watersheds), the tributaries of the major tributaries (tertiary watersheds), and so on.
There is no one right answer on how many subcategories you need to map. It is critical that a core reserve be located in each primary watershed and that each core include an entire secondary or tertiary watershed; but we know too little about watersheds in relation to floral and faunal distribution patterns and natural disturbance regimes to confidently prescribe more than these rough guidelines (see What do you do when you have too much data?).
Historical distributions of locally-extirpated species. Studies have shown that reintroductions of extirpated species are more successful than introductions of the same species outside their historic ranges (Griffith et al. 1989). Thus, to restore biological integrity, habitats within the historic ranges of species that need to be reintroduced should be included in the reserve system.
Compile a list of species known or presumed to be extinct in the area. These should be available from state fish and wildlife agencies and the Natural Heritage Program. Determine which of them still exist elsewhere and could provide a source pool for reintroduction. (You need not map the historic range of the Passenger Pigeon, since it is globally extinct.) Then determine where these species were found historically. This may be impossible for species whose historic ranges were reduced before good ecological surveys were performed; but for species that are of popular interest (like birds), were hunted, or have paleontological records, such information may be available in the scientific literature or in older field guides.
Wide-ranging species. At some point you will need to evaluate your reserve design with respect to its ability to provide habitat for species whose individuals range over a wide area. Make a list of these species in your area (including ones locally extirpated) and find what is known about the individual or group (e.g., wolf pack) home range size. For terrestrial areas, this list will consist mostly of large mammalian carnivores and herbivores and raptors, and you can probably get adequate information on them from guides to the mammals and birds in your area. Whether these species are solitary or social will in part determine how much space is required for a minimum viable population.
Irrelevant data sets are those of no direct biological or ecological value. By and large, it is not necessary to collect data on the economic importance or "human-use potential" of particular parcels of land. Reserve systems should be designed neither to necessarily avoid nor necessarily include economically valuable land. Grazing potential, frost-free growing days, number of recreational visitors per year--data sets on these won't help. Also, though the implementation of a reserve system may be facilitated by knowing who owns every piece of land in a region, it is generally not necessary to know that in order to design the reserve system.
This is not to say that you should ignore the realities of where centers of human occupancy are. Some areas have been so heavily modified that, no matter how great their original biological value, they cannot reasonably be included within a reserve system anytime within the next century. Therefore, your proposal must recognize, at some level, political reality. Again, though, the biological features of the region, not the cultural ones, are the most important as you begin to develop your reserve design. With an incremental approach to implementation (see Over what time frame do you plan to have your system implemented?), areas of potential biological value that are currently under human development can be "phased in" over long periods of time as opportunities arise.
Dealing With The Data
1. Do you need to know or use GIS technology? No. GIS technology is a tool that will increase your ability to deal with data, make maps, and answer quantitative questions, but it is not a requirement. Only you can decide whether or not you want to work with GIS. You'll need to consider whether you have access to the computer equipment and whether you want to learn how to use it. Also, some regions of the continent, particularly in the north and west, still contain such large expanses of wild or public land that detailed mapping by computer may be a low priority. (For more information about GIS in the service of conservation, see Coveny 1992.)
If you choose to use GIS technology, you might consider developing a relationship with people who already have access to and knowledge of GIS. Many colleges and universities have these facilities, staffed with folks who know how to use them and who are interested in many of the same questions and issues you are.
2. How do you draw the lines? This is really the heart of the matter for most people. Now that you have the data, what do you do with it so that in the end you have a single map that shows where the proposed core reserves, buffer zones, and corridors ought to be. Let's break this down into parts. If you are using GIS, parts A and B (mapping and combining data) simply involve following the protocols for the software you use, so I won't say anything further about it. The data can also be processed by hand, which I will describe.
(a) Mapping the data into layers. If you plan to do all your mapping by hand, you need a flat work space large enough for your largest base map, and mylar, a type of transparent plastic sheet that can be laid on top of a base map and used to map individual data layers. I buy mylar in bulk from B.L. Makepiece, Inc. (800-835-0194 or 617-782-3800). I prefer the Herculene film matte both sides type (#191253) in roles 54 inches by 20 yards. I recommend writing everything on mylar in pencil, since mistakes and revisions are inevitable and pencil can be erased. Cut several mylar sheets to the size necessary to fit over your base maps. You can work with each base map separately at this stage, combining them only at the end for your entire reserve map. Assign one mylar sheet to each major data set rather than combining them all on one sheet. That gives you the flexibility to add and subtract data sets as you develop and review your system. For the data sets I described there would be nine layers: RTE elements, natural ecosystems, public and private conservation land, roads, soils, biodiversity patterns, travel routes, watersheds, and historical distributions.
Subcategories of each of these data sets can be mapped with different symbols or shades, but they should be clearly distinct from one another since you may eventually need to interpret them through more than one layer of mylar. Be sure to include a key to the symbols you use on the mylar but off of the area that includes the mapped area.
Tape the base map flat to the work surface and the mylar sheet to the base map. Label the mylar with the base map code (see What maps do you need and where can you get them?), the name of the data set it records, and the source of the data. Mark four or five reference locations on the mylar so you can easily line up the mylar with the base map in the future. If the data set is actually on the base map, simply trace the data onto the mylar. If the data are on another map at a different scale than the base map, draw the data on the mylar not by tracing (which would result in the data being at a different scale than the base map) but by eye, using the base map under the mylar to position the data in the correct location and to scale.
Mylar can be stored either flat or rolled up, but should never be folded. If you are working with a large area and have many base maps, develop a filing system that will allow you to find a particular mylar sheet without a great deal of searching.
(b) Delineating the core reserves. For the initial design of the core reserves, use the data sets for public and private conservation land, RTE elements, natural ecosystems, and roads. Cut a fresh piece of mylar on which to draw the core reserves. The design of the core reserves occurs as a series of steps (Figures 1-4).
Figure 1. A hypothetical example to show the development of core reserves based on the critical data sets. A. Map of the public and private conservation land (cross-hatched areas). B. The boundaries of the large parcel are straightened out to decrease the edge and some smaller parcels are grouped together.
Figure 2. A. Distribution of RTE elements throughout the area (dots), some of which are already within core reserves. B. Core reserve boundaries are expanded to encompass areas with high concentrations of RTE elements.
Figure 3. A. The distribution of five different ecosystems within the area (bounded by narrow lines). B. The addition of a new core reserve to enclose an area in an unrepresented ecosystem that has a low concentration of RTE elements.
Figure 4. A. The distribution of roads within the area (thick lines). B. The addition of a new core reserve in a roadless area.
First, put the public and private conservation land data set on top of the base map and the fresh sheet of mylar on top of that. Redraw these parcels on the top layer in such as way as to (a) straighten the boundaries of large parcels of land to give them less edge for their area, and (b) make single, larger parcels out groups of small parcels that are close to each other.
Second, replace the public and private conservation land data layer with the one on RTE elements. On the top layer, expand the boundaries of the core reserves you just developed to enclose nearby areas that have high concentrations of RTE elements. Then draw boundaries around new areas that are not near the core reserves you have already delineated but that have high concentrations of RTE elements.
Third, add the natural ecosystem data layer to the pile under the top layer. Draw boundaries around areas of low concentrations of RTE elements that are in ecosystem types not represented in the core reserves you've already created.
Fourth, replace the RTE elements and natural ecosystem data layers with the roads data layer. Draw boundaries around large roadless areas that are not already included in a core reserve.
One of the hardest concepts for reserve designers to deal with is the difference between aquatic and terrestrial communities. We think primarily like a terrestrial species: undisturbed blocks of land are good places to live, bodies of water are barriers, and river valleys are good travel corridors. To a fish or a clam, however, a river or lake is required habitat and land is a barrier. In other words, a reserve design that has all of the cores in terrestrial roadless areas is biased toward terrestrial species, and may not be complete. As part of the design process, make a special effort to consider the needs of the aquatic species and ecosystems in your area. This may mean placing large stretches of river valleys in core reserves even though they are long and thin and are the sites of major roads. In coastal regions, large areas of coastline and ocean should be encompassed within core reserves.
In drawing your boundaries, try to avoid areas with large concentrations of humans, though this may not always be possible, especially when considering rivers and coasts. How large the population density an area must have to be avoided depends on the time frame you have set for implementing your design. The shorter the time frame, the smaller the settlements that need to be avoided. I use the guideline, "Is this settlement small enough that all the land could possibly be acquired by fee-simple purchases within my time frame?" If the answer is no, then I avoid including it within a core. It is important to use political judgment in this part of the design process. This constraint will inevitably mean that some RTE elements can't be protected in large core reserves, but will be part of small cores. If your overall system includes some large core reserves, however, it may still capture all known elements of biodiversity.
You now have a map that has core reserves delineated based on one of six criteria: (1) large parcels of public and private conservation land expanded to enclose nearby concentrations of RTE elements, (2) groups of two or more smaller parcels of public and private conservation land expanded to enclose nearby concentrations of RTE elements, (3) areas not in or near public or private conservation land but with high concentrations of RTE elements, (4) areas with low concentrations of RTE elements in ecosystems not represented in the core reserves based on Criteria 1-3, (5) large roadless areas not included in the core reserves based on Criteria 1-4, and (6) small reserves near human settlements.
As you look at your map, ask yourself several questions. How big should the core reserves be? Not every reserve needs to be large. Some primarily serve to protect populations of small, sedentary species, and don't need to be very big. However, some of the core reserves need to be extremely large to provide sufficient area for wide-ranging species. How large depends on your area and how many core reserves you have. All other things being equal, the more cores you have, the smaller the largest cores need to be. The more unnatural the surrounding landscape and the more fire-prone the vegetation, the larger the largest core reserves need to be. Noss (1992a) suggests that, depending on the context, large reserves can range in size anywhere from 10,000 to 25,000,000 acres.
How much of the total area should be in core reserves? That depends on the region. The more biologically diverse as well as the more unnatural and disturbed the region is, the greater the area that needs to be in core reserves. Based on the maps I have seen so far, I would say that reserve system proposals within the conterminous U.S. comprised of less than 25% or more than 75% core should be reconsidered.
If these core reserves were the only aspect of your reserve design to be implemented, you would have a system that would go a long way to achieving the first three basic objectives. You would not, however, fulfill all four objectives.
(c) Delineating the buffer zones. Primary functions of buffer zones are to protect core reserves from outside stresses and to develop more harmonious human-nature interactions and economies, in addition to serving as viable wildlife habitat themselves. These are difficult to design in that conservationists expect them to provide both habitat for all native species and resources for humans. A good tactic is to situate buffer zones where there are low concentrations of RTE elements to help provide them additional protection. Put the RTE element data set under the map of the core reserve system and draw boundaries that enclose areas with RTE elements in low concentrations. In general, these areas should surround core reserves and in some cases will actually connect core reserves together.
(d) Delineating the corridors. The design of corridors is still the most controversial and least understood aspect of the entire Matrix Model. Yet the need for connectivity of some type is obvious if the system is to maximize biological integrity. The system of core reserves and buffer zones that you have delineated may provide a great deal of biological connectivity within your region. Put the travel routes data set under the map of your reserve system. Note where large numbers of road kills or presumed plant migration paths occur outside of core reserves or buffer zones. These represent points along migration corridors that are not currently protected in your design. Draw boundaries bracketing these points. The boundaries should enclose areas that connect cores or buffers to each other. In general, the longer the corridors, the wider they should be, with a minimum width of about one-half mile. However, be prepared to revisit these corridors as more data become available that indicate how successful different types of connectivity zones actually are in nature.
Not all core reserves need to be connected by corridors; but within your total reserve design, your system of corridors should allow for movement (a) among large core reserves and some small ones, (b) along elevational gradients and ridge lines, and (c) along east-west and north-south axes. You may need to revise your corridors later as you match up your design with designs produced in neighboring areas.
(e) Evaluating your reserve design: additional data sets. I recommended that you draw the first draft of your reserve system based on only a few of the data sets you collected--in order to simplify your task. You should now check your design for completeness by testing it against the other data sets. Does your design have core reserves in each of the primary watersheds? Do the larger reserves include entire secondary or tertiary watersheds? Does your reserve design include representatives of all soil types? Does it include reserves within the historical range of locally-extirpated species? Does it protect known hot spots of biodiversity? Does it provide large enough areas (either in individual core reserves or in networks of cores and corridors) to include viable populations of wide-ranging species? Adjust your design accordingly.
(f) Evaluating your reserve design: review questions. After you finish your first draft, ask yourself some specific questions as a way of further evaluating your work.
Does the reserve system include representatives of all ecosystem types?
Are the core reserves large enough to maintain viable populations of all native species and to allow for ecological processes? Think especially about the needs of the large mammalian carnivores and herbivores, including those that might eventually be re-introduced into the area. In many cases, several interlinked regions may be necessary to assure long-term viability. Think also about wide-scale natural disturbances, such as wildfires and hurricanes, which your system needs to be able to accommodate without loss of biological integrity.
Are the core reserves buffered from outside influences?
Is connectivity adequate to allow for biological movement of all sorts (e.g., seasonal migration, dispersal, range shifts in response to climate change)?
Are aquatic ecosystems given weight equal to terrestrial ecosystems?
Do the core reserves protect regions within the historical ranges of locally-extirpated species?
If the answer to any of these questions is "no," then you should make the necessary adjustments.
After you have a reserve design that meets your own approval, send it out for peer review. Ask your conservation-minded friends to look at it for you, after reminding them of what its purpose is (otherwise they'll just tell you that it isn't "politically realistic"). One of the many roles that The Wildlands Project would like to play in facilitating the establishment of a continental ecological reserve system is as a source of scientific reviewers of proposed designs. Call The Wildlands Project and ask whom you should send it to for review.
3. What do you do when you have too much data? If you acquire data from a source that mapped information in much greater detail than you need, you might end up with more data of a particular type than you can handle. For example, you might get a map of an old-growth stand that shows to the nearest meter where every tree is located, when all you want to know is where the borders of the stand are. The data sets you are most likely to find excessive are ecosystems, plant communities, soils, and watersheds.
To facilitate mapping, you need to group the data from each set into just a few categories. Let your common sense and experience guide you (Figure 5).
Figure 5. How much detail is needed? A. An area classified only as forest. Representation is achieved with only one core reserve. B. Subdividing the forest into two types indicates that at least two reserves are required for representation. Note that the area itself has not changed, but rather the detail used to describe it. C. Subdividing the forests further, such as into three subcategories of deciduous forest, indicates that at least one more reserve is needed for representation. D. Subdividing the forest further in this example does not require additional reserves for representation.
Instead of showing the data separately for, say, forests of all different types, first show all of the forest as one type regardless of the kinds of trees. Then do the exercise again with the data subdivided into a few logical subgroups and see if your map looks substantially different. It might, because one of our conservation objectives is representation of all ecosystem types. The more types we recognize, the more likely it is that you'll need new core reserves in order to include a representative of a newly recognized ecosystem.
Continue this exercise of splitting the data set into more and more groups until it no longer results in a map different from that with fewer data sets. At this point, you know that adding more detail doesn't result in a different conservation plan, and therefore the extra detail is not needed.
4. How and why do you ground truth? You probably got your data from another source, rather than collecting them yourself. Since your reserve system proposal is only as good as the data you used to design it (and by extension, your knowledge of the region), you need to check the data for accuracy. It isn't practical to check all the data, but you can check the most important. For example, visit each of the critical ecosystems and see if they are where your map says they are. Check whether new roads have been built in roadless areas. Spot-check the accuracy of local ecosystem classifications. Ground truthing is an ideal task to share with others.
5. How do you revise your map? Despite what the "Wise Use" movement says, conservation biology is a science, not mysticism, because above all else it has the one feature that separates science from religion: an ability to adjust its statements based on more and better data. No map that you draw from data should be carved in stone and never revised. As more and better data become available, as the data themselves change, and as you rethink your work, you will need to revise your map. Keep careful track of your data files, including information on their source and date, so that you can replace old data sets with newer versions as they become available. You might even plan to revisit your design on a regular basis, say once every three years.
Final Thoughts
No matter what, all attempts to make the process of designing an ecological reserve system mechanical and foolproof will fail. At some point in the design process you still need to use your intuition and imagination about what is right, what is important, and what is irrelevant. Don't be afraid to use your intuition. You probably know your area and its needs better than almost anyone else. Over time, as you acquire more data and gain more experience, your design will get better and better.
On the other hand, don't let inexperience allow you to lose sight of what you are trying to achieve: a system of reserves, based on the best available data, that represents all ecosystems, protects viable populations of all native species, permits ecosystem processes, and allows for change.
A good reserve system cannot be designed overnight; don't be surprised if the job takes longer than you first thought. You have to find sources for the data you need, make requests for data to be copied and sent to you, buy maps, plot data, read, ground truth, revise, and ultimately ponder every line you draw. On the other hand, it need not be a task with no end in sight. If you identified a meaningful area, if you follow these guidelines and keep the goals in mind, you will soon have a design that will provide the framework for meaningful conservation in your area, and will contribute to the continental vision.
Of course, simply designing a map based on sound data and the best available scientific theories will not by itself put the design into place. That will come over time with, as Brock Evans says, endless pressure endlessly applied. Good public relations and a well-planned strategy to bring about implementation of each regional plan will be required. Expect that as more plans are developed and we, as a community, gain more experience in implementing our visions, Wild Earth will publish a companion article to this, exploring how to gain real protection for the reserves we propose.
The work of designing a continental system of ecological reserves has just begun. As more people become involved and we collectively gain more experience, even this article on the basics will need to be revised. Just as this article benefited from the work that others did before, future versions will benefit from the experience of those who use this article to guide their work now. Please write and tell me of your experiences, the unique challenges you faced in your project, the questions you had that I didn't answer, and the solutions you found. What we are trying to do isn't easy but, like Frodo and Sam in The Lord of the Rings, we can defeat even the Lord of Darkness if we keep on doing what needs to be done one step at a time.
ACKNOWLEDGMENTS
As I said, this article grew out of conversations and work with numerous people who are involved in this process. I'm especially thankful for the opportunity to be able to work with Mike Biltonen, Marcia Cary, John Davis, Kathleen Fitzgerald, Dave Foreman, John Gartner, David Johns, Kraig Klungness, Chris McGrory Klyza, Jean McKendry, Brad Meiklejohn, Jerry Mueller, Reed Noss, Dave Publicover, Charlie Restino, Jamie Sayen, Mike Soule, Jim Strittholt, Scott Thiele, George Tukel, and Bo Wilmer.
LITERATURE CITED
Anderson, J.E. 1991. A conceptual framework for evaluating and quantifying naturalness. Conservation Biology 5: 347-352.
Coveny, S. 1992. Technology isn't entirely evil. Wild Earth Special Issue: 81-83.
Davis, M.B. 1993. Eastern Old Growth, A Survey. Cenozoic Society, Richmond, VT.
Griffith, B., J.M. Scott, J.W. Carpenter, and C. Reed. 1989. Translocation as a species conservation tool: status and strategy. Science 245: 477-480.
Johns, D., and M. Soule. 1995. Getting from here to there: an outline of the Wildlands reserve design process. Wild Earth Winter 1995/96: 32-36.
Noss, R. F. 1992a. The Wildlands Project: land conservation strategy. Wild Earth Special Issue: 10-25.
Noss, R.F. 1992b. Conserving Oregon's Coast Range Biodiversity: A conservation and restoration plan. Coast Range Association, Newport, OR.
Trombulak, S.C. 1995. Ecological health and the Northern Forest. Vermont Law Review 19: 283-333.
Steve Trombulak is a professor of biology and environmental studies at Middlebury College in Vermont. He is also the regional science director for The Wildlands Project in the Greater Laurentian Region of North America.