Ecosystem Principles and Concepts

Ecosystem Principles and their Implications for Management

Four broad principles guided the development of this framework. These principles are:

  1. Ecosystems are dynamic, evolutionary, and resilient.

  2. Ecosystems can be viewed spatially and temporally within organizational levels.

  3. Ecoystems have biophysical, economic, and social limits.

  4. Ecosystem patterns and processes are not completely predictable.

Ecosystems are dynamic, evolutionary, and resilient -- Change is inherent in ecosystems; they develop along many pathways (ONeill and others 1986, Urban and others 1987). An ecosystem is said to be resilient if when disturbed or otherwise changed, it tends to return to some developmental pathway or it is cyclic such that its state is always changing within some definable bounds (Hilborn and Walters 1992). Ecosystems are the products of their history (Barret and others 1991). Natural fires, volcanic eruptions, floods, and wind events, along with people setting fires, clearing land, and introducing new (exotic) species have been sources of ecosystem disturbance (Agee 1994, Robbins and Wolf 1994). Forest and grassland ecosystems are generally resilient to a variety of disturbances. Just as past disturbances and the actions of past human generations shaped the ecosystems of today, actions of this generation will transform ecosystems of the future. Past management decisions, combined with natural environmental disturbances and conditions have influenced future options (O'Laughlin and others 1993, Maser 1994).

Historical and potential disturbance regimes have influenced the patterns and processes on today's landscapes (Oliver and others 1994). Ecosystems are constantly changing and cannot be kept indefinitely in any given state. While ecosystem management can recognize the inherent resiliency of natural systems, it should also recognize that maintaining the status quo is difficult and not necessarily a goal. It should consider the outcomes of management activities and how they influence ecosystem development. Scientific knowledge can be used to help public and private natural resource managers make choices about dynamic ecosystems in the face of uncertainty. In addition, assessments and monitoring programs can be used to evaluate and track changes in outcomes related to biophysical, social, and economic structures and functions.

Within ecological limits, a wide range of sound management options, providing different mixes of goods and services, will exist. No one landscape condition will be "best". Selecting the desired condition is a social decision that can be made by understanding ecological limits. Fortunately, the inherent resiliency of ecosystems provides opportunities to test various management approaches, and adaptive management will allow managers to learn from experience and make appropriate changes without significant risk of irreversible environmental damage. The dynamic nature of ecosystems requires a dynamic planning process.

Ecosystems can be viewed spatially and temporally within organizational levels -- To describe the dynamic nature of ecosystems, it is useful to view them as having multiple organizational levels varying over time and space. These levels can be organized within hierarchies, in which every level has discrete ecological functions but at the same time is part of a larger whole (Allen and Starr 1982, Allen and others 1984, Koestler 1967). Higher levels usually occupy larger areas and are usually characterized by longer time frames (Delcourt and Delcourt 1988, King and others 1990, King 1993).

As applied to landscape ecology, hierarchy theory allows for the definition of ecosystems and the linkages between the different levels of ecological organization. Ecosystem descriptions (Rosen 1975) and ecosystem processes, structures, and functions are all defined by the observer (Pattee 1978). Within the vegetation component of ecosystems, trees can be nested within forests, forests can be nested within series, and series nested within formations (fig. 3a). There are a multitude of environmental constraints, vegetation patterns, human behaviors, and disturbance processes that can be described for each level, time frame, and area (Pickett and others 1989, Robbins and Wolf 1994). Spatial extent can range from a few square meters to millions of square meters, and time frames can range from less than one year to millions of years.

Social and economic components of ecosystems may be defined spatially and temporally as well, along an organizational or institutional continuum (fig. 3b). Organizational levels, time frames, and spatial extents that are significant to human decision-making often overlap and do not correspond to the same time frames and spatial extents as biophysical systems. For example, ecosystem processes (such as soil formation) that occur over long time frames (centuries or millennia) hold little meaning for political processes that operate biennially. In addition, people respond to environments symbolically, and places important to people cannot typically be defined by using biophysical hierarchies alone. To some extent, the selection of hierarchies represents a compromise among the various disciplines involved in an assessment.

Viewing ecosystems hierarchically with varying time frames and spatial extents has several implications for assessments and management. This approach provides managers with a way to organize the analysis of effects of management practices that take place over multiple time frames and spatial extents. In addition, it provides for the recognition that decisions made at one level of the hierarchy are likely linked to other levels. Viewing ecosystems as hierarchies ensures that monitoring programs (measurements, analysis, synthesis, and evaluation) track ecosystem development or change over multiple levels, spatial extents, and time frames.

Vegetation Hierarchy

Figure 3a

Social Hierarchy

Figure 3b


Figure 3 -- Ecosystem organization can be viewed as a hierarchy. Each level of the hierarchy has both time frames and spatial extents. A vegetation hirarchy is shown in 3a and a social hierarchy is shown in 3b.

Ecosystems have biophysical, economic, and social limits -- In all ecosystems, there are limits to the rate of production and accumulation of biomass (plant, animal, and human) (Kay 1991, McCune and Allen 1985). In addition, the environment is constantly in a state of flux, causing ecosystems to change. Given this, human populations need to recognize that the ability of an ecosystem to provide goods and services has limitations. Unfortunately, people often make demands on ecosystems that exceed their biological or physical capabilities (Robbins 1982, Young and Sparks 1985).

Science provides information about ecosystem limits; land managers use this information as they develop ways to allocate finite resources. Society uses this shared information to make choices about its behavior. People can choose to modify their behavior and organize their institutions to be consistent with the capabilities of ecosystems, or they can pursue actions inconsistent with the capabilities of ecosystems. People can also improve ecosystem productivity on some sites through investments in management practices. Societal choices regarding the use and allocation of resources have implications for inter-generational equity and trade-offs. For example, investments in ecosystem restoration made by this generation will provide benefits and options for future generations.

Ecosystem patterns and processes are not completely predictable -- The events that influence ecosystem patterns and processes usually are unpredictable (Holling 1986). Predictability varies over temporal and spatial organizational levels (Bourgeron and Jensen 1994). For example, from year to year wildfire occurrences are associated with particular seasons and environmental conditions, but a fire may occur in any season and under different environmental conditions. Similarly, eruptions of volcanos in the Cascade Mountain Range have occurred, on average, twice each century for the past 4,000 years (Dzurisin and others 1994); however, neither when the next eruption will occur, nor its size and effects can be predicted. In the social dimension, it is possible to predict crime rates at the regional or community level, but it is much more difficult to predict the occurrence of a crime at a particular household.

While people generally prefer predictability, adept ecosystem managers acknowledge and prepare for surprise (Kay 1991). The limited predictability of ecosystem outcomes has several important implications. Land management policies and practices should provide sufficient flexibility for managers to respond effectively to any unanticipated effects of previous decisions. For example, knowledge gained from adaptive management and monitoring programs may help managers prepare for and respond to surprise events. As knowledge increases, managers are better able to predict outcomes. Yet, long-term yields of goods and services may remain unpredictable. Although models are simplistic representations of real world systems, they may improve the predictability of outcomes. Models are never error free, but through adaptive management their generality, accuracy, and realism can be improved (Slocombe 1993).

Ecosystem Management Concepts

Boundaries -- Delineation of assessment boundaries should be based on a combination of biophysical, economic, and social attributes. On Federal lands these boundaries should be delineated to facilitate national strategic planning as well as to set the context of management areas for local planning and decision-making. Assessment boundaries should strive to delineate specific measures of homogeneity. These measures and the associated boundaries will vary depending on the perspectives of those defining them. For example, areas delineated based on biophysical patterns and processes may not coincide with those based on hydrologic and aquatic processes, economic trade areas, or social settings.

Delineation of boundaries internal to an assessment area involves multiple approaches. Economists tend to view an area in terms of economic regions or counties, hydrologists view an area in term of watersheds, while ecologists may view an area in terms of patches of vegetation. These different approaches result in internal boundaries that challenge the integration process. It is important to recognize multiple boundaries, but also to compromise on a common internal boundary set for analysis and description.

Scales -- There are different notions of what scale means in the literature and often there is confusion between geographic extent and data resolution. An approach clarifying the use of scale is shown in tables 1, 2, and 3, where geographic extent refers to the area assessed and resolution describes the amount of detail incorporated in the data. Assessments can be described by two-part names designating both the geographic extent and the resolution of the data.

In regional and sub-regional assessments, some ecosystem components cannot be adequately addressed using broad resolution data. For example, habitat conditions for species with small home ranges cannot be adequately assessed with broad resolution data (O'Neill and others 1986). Similarly, assessments of economic patterns in rural communities may be more appropriate at the landscape rather than the regional extent.

Regional assessments (table 1) show trends and describe general conditions for biophysical, economic, and social components. These assessments describe social characteristics such as state and county trends in human populations and urban versus rural economic growth. They usually contain broad resolution information on spatial patterns of resources and associated risks to resource values. Sub-regional assessments (table 2) typically rely on mid-resolution data to provide information on patterns of vegetation composition and structure, trends in social well-being for human communities of interest, and trends in basic conditions of communities (places). Assessments of landscapes or specific sites provide the greatest detail (table 3). These assessments may cover landscapes, watersheds, or individual project sites and specific human communities. They typically rely on fine-resolution data regarding vegetation patches, stands, meadows, streams, and social and economic data. Landscape assessments, as described here, are essentially the same as "ecosystem analysis at the watershed scale" as used in implementing the Northwest Forest Plan (USDA 1994a).

No single assessment will adequately address the complex issues facing resource managers today. Higher level assessments set context while lower level assessments help understand processes. Assessments of landscapes or sites cannot adequately address broad patterns and processes, such as habitat conditions for wide-ranging species or global climatic processes. Regional and sub-regional assessments provide a necessary context for landscape and site assessments. Together, multiple assessments (site specific to global) provide a comprehensive basis for land management decision-making.

Conducting assessments at different geographic extents also can promote more effective stakeholder participation and learning. Many people see their interests affected primarily at the local level. They may choose not to participate in sub-regional or larger assessments because they might feel their local concerns will be diluted or unnoticed. Without sub-regional and regional assessments, stakeholders and decision-makers may have difficulty assimilating the cumulative magnitude and complexity of many highly detailed, or localized, landscape and site-specific assessments. Conversely, stakeholders whose interests are national or regional may find it difficult to participate effectively in numerous landscape assessments based on fine-resolution data.

Undertaking assessments at multiple geographic extents promotes the inclusion of more interests into the assessment process. It also serves to provide decision-makers with the appropriate information for particular levels of decision-making. Therefore, depending on the issues and policies being addressed, the type of assessment (data resolution and geographic extent) to conduct can be specified (see tables 1, 2, and 3).

Table 1. Attributes and characteristics typically associated with broad resolution, regional assessments. 1
Attributes Landscape Ecology Terrestrial Aquatic Social/Economic
Geographic extent River basin River basin River basin States
Data Resolution2 >100 ha >100 ha >400,000 ha Sub-basins State, County
Organizational hierarchy Multiple watersheds Community & species associations Watersheds, communities of species State, County
Mapscale >1:100,000 1:2,000,000
1:1,000,000
1:100,000 1:1,000,000
Time period3
.....Short term
.....Long term

1-10 years
10-300 years

1-10 years
10-100 years

1-10 years
10-100 years

1-5 years
5-50 years
1 The general size of these assessments is millions to billions of km2 and the general use is for national and regional planning and policy-making.
2 Defining vegetation components is typically on a resolution of 100 ha while the aquatic components are defined by river systems (>400,000 ha).
3 Short- and long-term time periods for historical and projected patterns and processes differ between types of assessments.




Table 2. Attributes and characteristics typically associated with mid-resolution, sub-regional assessments. 1
Attributes Landscape Ecology Terrestrial Aquatic Social/Economic
Geographic extent Multiple watersheds Province Multiple County watersheds
Data Resolution <100 ha 1-5 ha watershed 15,000 ha County
Organizational hierarchy Watershed Species groups Species groups County
Mapscale 1:100,000 1:24,000 1:100,000 1:24,000 1:100,000 1:24,000 1:100,000
Time period2
.....Short term
.....Long term

1-10 years 10-300 years

1-10 years 10-100 years

1-10 years 10-100 years

1-5 years
5-50 years
1 The general size of these assessments is thousands to millions of km2 and the general use is for state, regional, and local planning and policy-making.
2 Short- and long-term time periods for historical and projected patterns and processes differ between types of assessments.




Table 3. Attributes and characteristics typically associated with fine resolution, landscape assessments. 1
Attributes Landscape Ecology Terrestrial Aquatic Social/Economic
Geographic extent Watershed Watershed Watershed Household
Data Resolution <25 ha 1-5 ha Streams Household
Organizational hierarchy Streamsand vegetation patterns Species Species Household
Mapscale 1:24,000 1:24,000 1:24,000 1:100,000
Time period2
.....Short term
.....Long term

1-10 years
10-100 years

1-10 years

1-10 years

Months-5 years
1 The general size of these assessments is tens to hundreds of km2 and the general use is for multi-forest/district, forest/district, or area planning and policy-making.
2 Short- and long-term time periods for historical and projected patterns and processes differ between types of assessments.