The bogs and other freshwater wetlands on theWarren Grove Air National Guard Range, located in the Pinelands of southern New Jersey, are areas of exceptional biological diversity. (Photos: Douglas Ripley)

For one tasked with the conservation of biodiversity, the idea of planning to preserve the "totality of genes, species, and ecosystems of a region" is daunting – as exhibited by the complexity of Figure 2.1. Attempting to implement the conservation of biodiversity, as defined, is an overwhelming challenge. It is far too easy to become stuck in the weeds of the details and to try to manage everything individually. A land manager will justifiably ask, "How can I hope to manage for all species on my installation? How in the world do I manage for landscape function? Where do I start?"

While it is important to keep all biological levels of organization in mind, one does not need to plan, or manage, for each. In reality, planning for conservation action leans most heavily on what is commonly called the coarse filter/fine filter approach (Noss 1987). The coarse filter approach focuses on ecological systems – ecosystem management – whereas the fine filter approach emphasizes individual species management. Successful biodiversity management relies on both. In brief, the reasoning supporting this paired approach is that most species are "captured" by the coarse filter because of their association with specific ecosystem types. Those species that are not captured in the coarse filter (e.g. wide ranging species) need then be caught by the fine filter (Groves 2003).

While the concept of the coarse and fine filters was initially conceived to be independent of spatial scale, in reality those species not captured by the coarse filter tend to be intermediate, coarse, or regional scale species as defined by Figure 2.2 (Poiani et al. 2000). These tend to be larger, wide-ranging species that are often dependent on a diversity of ecosystems during their lives.

While most resources managers, and many non-biologists, have an intrinsic understanding of these levels of biological organization, it is always a good idea to review these terms and concepts as their precise meanings are often different from the perceived gestalt.

POPULATIONS. A population is typically defined as a group of interbreeding individuals of the same species living within a defined area. The key to this definition is that individuals within a population must, at the very least, have the potential to interbreed. Thus, dispersal potential can drive the size of a population. Many wideranging species, for example migratory birds, have huge populations that can span thousands of square kilometers. More stationary species, for example, bog lemmings, will have more restricted population sizes where the entire population exists within a small peat bog.

META-POPULATIONS, NATURAL AND DERIVED. Between these two extremes, most species exist as constellations of sub-populations where most individuals interfigure act within their small group, with the rare individual dispersing over greater distances. So, these species are structured as a meta-population, or a population of sub-populations. These sub-populations are distributed across a landscape (or a military installation) as many "occurrences." Each occurrence has a low probability of persisting over the long term, in isolation from other occurrences. Most sub-populations are simply too small to be resilient to environmental variation, or demographic or genetic bottlenecks. Neighboring occurrences are constantly providing, at some low but critical rate, "new blood" into a given sub-population. These neighboring occurrences also provide sets of "founder" genes that will recolonize a vacant area.

Meta-populations can be envisioned, then, as a galaxy where each "star" is a sub-population. These "stars" are winking off and on as sub-populations disappear, and then reappear as the vacant areas are re-colonized. The space between these stars – the voids – is not suitable habitat for these creatures, and so is simply not available for colonization by this species.

SOURCES AND SINKS AND THEIR IMPORTANCE. Upon reflection, it becomes obvious that all sub-populations are not the same. Some occur on tiny patches of acceptable habitat, and never grow to more than a small number of individuals. These, of course, never really escape the consequences of being in a "population bottleneck," and many have a low probability of persisting in isolation. Others occur on large areas of acceptable habitat and, thus, tend to exist as large healthy populations. These have greater demographic and genetic resilience, and hence a greater probability of persistence. Simply because of their large size, these populations tend to be the source of most of the dispersers that colonize vacant patches, and reinvigorate the small sub-populations both by their numbers and by their genetic diversity. These are thought of as "source" sub-populations, whereas the smaller occurrences which tend to absorb migrants, but do not provide dispersers, are considered "sinks."

The generalization that small populations tend to be sinks, and large populations sources, is, like all generalizations, only true to a point. The key, which is often difficult to measure, is whether the population produces significant numbers of emigrants or not. Source populations do, sinks do not. In general, "sink" sub-populations will not persist without continual immigration from "sources." Thus, the destruction of a single "source" sub-population can result in the extirpation of many surrounding "sinks" even if they are not directly impacted.

The Sikes Act and the Endangered Species Act require military installations to prevent the loss of threatened and endangered species found within their boundaries. Understanding the ecology of those species, and how their populations and sub-populations are distributed, is key to meeting this requirement. Conserving a wide-ranging species like the bald eagle might be accomplished simply by protecting a limited number of nesting sites – as only a small piece of a much larger population exists on site. The Karner Blue butterfly, in contrast, exists as a metapopulation where sub-populations exist in ephemeral patches of host plants. Conserving this species requires an understanding of the disturbance dynamics creating these patches of host plants, and the dispersal capabilities of the butterfly, so as to manage the entire meta-population and not just a few occurrences, each with a low probability of persistence in isolation. Understanding and managing a meta-population often requires looking beyond an installation's borders to subpopulations on neighboring lands.

COMMUNITIES AND ECOSYSTEMS

How are communities and ecosystems different than "habitat"? Managing for species invariably means managing habitat. Habitat (which is Latin for "it inhabits") is the place where a particular species lives and grows. It is essentially the environment – at least the physical environment – that surrounds (influences and is utilized by) the species population. The term was originally defined as the physical conditions that surround a species population, or an assemblage of species (Clements and Shelford 1939). Wildlife managers, in particular, tend to focus on habitat management – identifying and manipulating those environmental factors limiting a targeted population's size (Leopold 1933, Yoakum and Dasmann 1971). Scientists often expand the concept of habitat to include an assemblage of many species, living together in the same place. Thus, for example, wildlife managers often work to improve shorebird habitat. The U.S. Fish and Wildlife Service (USFWS) has spent many millions of dollars managing for breeding habitat for migratory waterfowl in the prairie pothole region of North America. Ecologists regard the habitat shared by many species to be a biotope – a place where a community of species lives.

The concept of habitat is not synonymous with that of the natural community or ecosystem. A natural community is the assemblage of plants and animals sharing the same habitat and interacting with each other. When one speaks of a natural community, the focus is on the species and their interactions. The habitat, or biotope, is the biophysical stage on which these species and their interactions occur. Communities typically reoccur across a landscape as they track habitat conditions. As such, communities do not occur at a single, specific spatial scale. Vegetation communities are often perceived as the classic community, but one can also describe the smaller community existing within a fallen log, or ephemeral community within a vernal pool.

An ecosystem, then, can be thought of as the whole picture; the combination of a natural community and its habitat (or biotope). As such, an ecosystem can extend far beyond even a large military installation. But ecosystems are more than just a community in its habitat. The concept of the ecosystem includes dynamic ecological processes (see below) and the recognition that species composition (i.e. the community) will change over time as well as over space. Every species within a community responds to the environment differently from the others. Similarly, each species interacts with different suites of other species. As conditions change, as they certainly do within military installations as in other environmental settings, some species become more abundant, while others become rarer.

Natural disturbances, ranging in size from gaps caused by fallen trees to massive wildfires, all affect species abundance and distribution differently (Picket and White 1986). Thus, ecosystems are neither static nor homogeneous. Rather, they are composed of "patches" of various sizes and ages, and the relative abundance and distribution of these patches is crucial to maintain the full suite of biodiversity within an area. Maintaining ecological processes, such as fires, floods, and periodic disease epidemics, is the keystone of successful ecosystem conservation. Indeed, the core of the ecosystem-based management approach is the understanding that the persistence of all biodiversity within an area is contingent on the persistence of this crazy-quilt pattern of disturbed and recovering patches. Management, then, needs to focus on the dynamic processes creating this pattern and not on maintaining a static structure and condition. Military activities can mimic some natural disturbances, and thus can often be integrated into a biodiversity management plan.

Dr. Walter Bien, Professor of Biology at Drexel University, Philadelphia, explaining his field work to graduate students and the natural resources staff at the Warren Grove Air National Guard Range, New Jersey. Research by university and environmental organization scientists has contributed significantly to the DoD's understanding of ecological processes on its lands. (Photo: Douglas Ripley)

ECOLOGICAL PROCESSES

In most human-dominated landscapes, including most military installations, native ecosystems have been fragmented and now occur as islands in seas of intensively impacted and managed lands. As mentioned above, this fragmentation harms species populations by restricting the movement of those pioneering individuals necessary to found new sub-populations and reinvigorate population sinks. Similarly, fragmentation changes how natural disturbance plays out on the landscape. Fires, for example, may be prevented from running across the landscape by the cutting of firebreaks. Thus, vegetation patches may persist for greater periods of time between fires, resulting in greater fuel accumulation, and subsequently more severe fires when they do occur.

The intensity and impact of any ecosystem process varies over time. Species and ecosystems respond to, and are organized around, these natural ranges of variation within these ecological processes. Thus, fires returning every five years will result in a very different community than when they return every hundred years. This is exemplified by both the longleaf pine forests of the southeast and ponderosa pine forests of the Rocky Mountain west. While there was, of course, variation in the frequency of naturally-ignited fires, typically any given patch would burn every ten years or so. This resulted in open forests, with relatively few large trees in a matrix of grasses and forbs. Both long-leaf and ponderosa pines have thick, fire-resistant bark and so the adult trees are not damaged by low-intensity ground fires. Active fire suppression over the past several decades has decreased the fire frequency and allowed other, less fire tolerant, species to get toeholds. Now, when fires do occur, the fire climbs into the canopy and the results are conflagrations that consume everything rather then the historically less intense ground fires that did not impact the trees.

Ecological processes that are impacted by military land uses include:

• Fire, both in terms of frequency, seasonality, and intensity
• Flooding, including frequency, sediment movement
• Disturbance of turf in prairie systems
• Sheet flow, and other water movement patterns in desert systems

Active ecosystem management by humans can mimic historic ecological processes and their effects; conservation managers can achieve both their conservation goals and meet the needs of the military. However, management with an eye toward variation is more challenging than managing for consistency. A large forest ecosystem will be very different if every management unit is burned on a 10-year cycle than if units were burned randomly on a 5- to 30-year pattern. The former is easier to plan and to implement, as managers can anticipate needs many years in advance. The latter is more complex structurally, and hence, harbors greater biological diversity.

NATURAL LANDSCAPES

From a biological perspective, a military installation is not an island, existing in isolation. It lives within a larger landscape comprising both natural and anthropogenic systems. A natural landscape can be thought of as the spatial scale at which ecosystems reoccur (Forman 1995). Meta-populations often function at this scale, with sub-populations occurring in ecosystem patches scattered throughout the landscape. Many wide-ranging species are very sensitive to the landscape pattern. These species often use, and require, two or more ecosystems for survival. These ecosystems may often not be congruent, and the species must travel through the landscape. Smaller installations may encompass only a small portion of the landscape mosaic and, as a result, critical habitats and ecosystems may only occur off-site. In these circumstances, it is very important to look beyond the installations boundaries.

Alternatively, a large military installation can often be fruitfully managed as a landscape unto itself – or sometimes as a microcosm of a much larger landscape. Natural buffer zones, impact areas, training areas, and other developed lands together join to form a landscape mosaic. There is great opportunity to build upon this existing mosaic, creating missing patches or systems, and enhancing others to effect significant conservation results.

Proceed to Next Section: Learning to Think Like a Mountain: Tools for Conservation Practitioners