Scale & Monitoring

Asking What, Where, and For How Long: Scale Matters

Social, economic, and ecological systems reveal different characteristics both in time and space. A single tree does not look the same or play the same interactive role within the parent community at 5, 50, and 500 years; and a forest does not look or function the same at 500 or 5,000 or 50,000 acres. Simply stated, scale, both spatial and temporal, matters. Because the systems we are trying to sustain reveal different characteristics at every scale, sustainability questions change at every scale. Allen and Hoekstra (1994) note that there is “no nature-given scale at which a system is sustainable or otherwise.” Therefore, we should monitor for sustainability at a variety of scales.

Although critical in framing the issue of sustainability, scale is often neglected in sustainability discussions (McCool and Haynes 1995). Sustainability monitoring initiatives occur at a range of spatial scales. The national scale of the Montreal Process focuses on examining the state of a nation’s forests. At a regional scale, a number of states are monitoring for sustainability. Other assessment programs at this scale are defined by ecological region, such as the Mid-Atlantic, the Great Lakes, and the Sierra Nevada.

The question we ask at the local scale will be specific to the dynamics of the particular place and its resources and residents. Thus, employment may be a common factor to many scales, but the meaning associated with monitoring employment at the local scale is very different from monitoring at the national scale.

Spatial Scale

Sustainability monitoring initiatives are occurring at a range of different scales. The national scale, characterized most prominently by the C&I sets of the Montreal Process and Helsinki Accord, is focused on examining the state of a nation’s forests. Global assessments (e.g., WorldWatch Institute 2001) are also being used to track trends in the status of forests across the globe. At a regional scale a number of states (e.g., Pennsylvania and Oregon) have begun monitoring for sustainability; and other regional monitoring and assessment programs have been conducted, often defined by ecological regions such as the Mid-Atlantic region, the Great Lakes region, and the Sierra Nevada region.

Because the questions about sustainability vary at different scales, Cheng and Durst (2000) note that “nation-level sets of criteria and indicators for various regions of the world should be complemented by the development and implementation of sets of criteria and indicators defined by the forest management unit level.” The need for scale-specific C&I was also a primary finding of the CIFOR-NA study (Woodley et al. 2000).

Experience has demonstrated that when indicators developed for larger scales are used for local-level scales they often do not address the system structures and functions and questions of sustainability. And while very broad and generically worded indicators (e.g., employment) can be applied at multiple scales, they typically have to be adapted to meet the local conditions for them to have any true value. Sometimes indicators are so inherently scale-specific (e.g., measures of the contribution of forestry to the Gross Domestic Product) that they have to be significantly revised.

Although there should be linkages between sustainability monitoring initiatives at different scales so that they are complementary, monitoring initiatives at the local, or forest management unit (FMU), scale are not just an application of nation-scale material. And as social, economic, and ecological conditions vary greatly from place to place, the results from FMU-scale C&I processes vary greatly.

The FMU scale can generally be described as the scale at which management policy is actually implemented with on-the-ground activity and at which one or more ownerships decide how a land area will be affected by land and resource management activities. Total land area and ownership size might vary, but the focus on the FMU scale is based on the assumption that it is at the FMU scale that most of the decisions about management occur. The design of FMU-scale C&I is intended to help provide insight into the sustainability of the underlying social, ecological, and economic systems that function coincident with the FMU scale. There are a growing number of C&I initiatives focused at the FMU scale, but chief among them are the programs of the Center for International Forestry Research (CIFOR), the Canadian Model Forest Network, the USDA Forest Service LUCID Project, and the local-scale C&I development efforts in Mexico.

Temporal Scale

The issues of temporal scale and sustainability relate to the principles of intra- and intergenerational equity: the balanced distribution of benefits and costs not just within this generation but also for future generations. McCool and Haynes (1995) ask “Over what period of time do we judge the sustainability of resource management?” The topic of temporal scale does not end with such a straightforward question; it has its own subtle convolutions as Dixon and Fallon (1989) observe: “The shorter the time horizon [in resource management decisions], the less likely any pattern of resource use will be sustainable over long periods of time.” When temporal scale is ignored in resource management, the most common result is that the succeeding generation bears the costs of the preceding generation’s benefits.

Temporal scale considerations are central to adaptive management, yet managers often treat monitoring as a singular, one-time event dependent on project implementation, agency funding, and reporting requirements. To understand systems, it is vital to consider the temporal scale at which the elements in question interact, and then to monitor the elements accordingly. In this context, sustainability monitoring tracks a suite of indicators over time to identify the changes in the states of systems. Monitoring is an ongoing, multistage process. Periodic sustainability assessments serve to analyze and synthesize data collected at more frequent intervals. Results of the sustainability assessments can then inform management decisions.

Multiple Scales: Measurement and Data Issues

The multiplicity of scales by which sustainability can be examined immediately dismisses the notion that any particular scale or any one boundary might be the right one or the best one for monitoring. Inevitably, situations arise in which the reference values for a suite of indicators are met at one scale but not at another. For example, sustainable development at the local scale may result in contributing to unsustainable development patterns at a national level (McCool and Haynes 1995). The converse is also likely to be true, for example, when a forest does not meet the reference values set for it on a watershed-by-watershed basis while at a regional level those system structures and functions may well be within target. Thus, sustainability monitoring and assessment at multiple scales is suggested.

Typically, the smaller the spatial scale the more difficult it will be to resolve conflicts between competing objectives and the more likely that all indicators will not meet their reference values. At larger spatial scales resolving some of the conflicts between competing objectives may be easier in theory, but merely changing the scale of analysis may not resolve the underlying problem or conflict. Mechanisms to resolve conflicts across spatial scales are highly recommended as part of the monitoring design (McCool and Haynes 1995).

Scale-Dependent Measures

Living systems operate in similar fashions at multiple scales in time and space. For example, landscape systems on a decomposing log can have the same structure and function characteristics of landscape systems that operate in a watershed. Often, however, the way we measure (the protocols) living systems is scale-dependent. Consequently, although an indicator of a system attribute may appear to be common across multiple scales it may require different measures and metrics. Given this scale dependency in measures, C&I measures need to be developed that are particular to the scale of interest.

The measures of system sustainability differ at regional and FMU scales (see Figure 1). Therefore, summarizing results of FMU sustainability assessments for several National Forests within a region (administrative, ecological, or economic) would not be equivalent to a sustainability assessment at a regional level and will yield neither desired nor legitimately useful results. Additionally, National Forests are but one piece of the puzzle of forest ecosystems across the United States. A sustainability assessment at a regional or national level must consider the total landbase and associated systems that operate at these scales, entailing examination of National Forest land in conjunction with private, state, and other federal lands. And because cause-and-effect parameters are scale-dependent, it is impossible to take the FMU-scale C&I data sets for all National Forests and additively summarize them to generate meaningful conclusions relevant to regional or national assessments.

Scale and Data Aggregation

Just as matching the scale of the question to the scale of data collection is important so are issues of data aggregation. There is a great temptation to aggregate data collected at various scales to answer questions concerning a particular scale; however, this results in a data intensification challenge. The obvious benefits of this are increased efficiency in the cost and effort associated with monitoring. The aggregation temptation is facilitated by commonly defined data requirements prescribed across scales as core information requirements. Indeed, data may be collected at the FMU scale to assist regional or national monitoring, but those same data may not always be useful either for addressing sustainability questions at the FMU scale or to respond to on-the-ground forest management needs. The nature of systems with respect to scale and the implications this has on assessing sustainability at multiple scales is not well understood and has resulted in frequent requests for mirror sets of C&I between scales and simple aggregation algorithms to facilitate upward reporting and sharing of data.

Considerations of scale must be designed into monitoring protocols for FMU measures. This is paramount to ensure that monitoring provides information about the systems that are being assessed. Though FMU-scale measures are not the same as those that should be applied to a regional or national set of C&I, they can provide rich narrative insight into assessments of sustainability at other scales. In turn, regional or national assessments of sustainability can describe the context in which the individual FMU sustainability assessment is conducted.

Scale issues are also complicated by whether or not systems are nested or unnested. For nested systems the issues of sampling and data aggregation are straightforward. Data are typically sampled at least one scale finer than the question of interest and are then aggregated upward. Sampling and data-aggregation in un-nested systems are more difficult because the emergent properties of systems mean that simply aggregating data will overlook the synergistic effects of systems. For example, percent soil carbon can be aggregated using weighted area values while the volume of soil carbon/cubic meter cannot be aggregated, the latter because it is a process-dependent measure, the former because it is a process-independent measure.

Conclusions

Sustainability is a multiscaled problem and consequently there is no right scale to assess or manage for sustainability. Although sustainability can be studied at multiple scales, once the components of systems are identified for monitoring, selecting the correct scale is critical. The context of the systems that we are trying to sustain change at every scale because the constraints change. Using the wrong scale to look at certain system properties could be like trying to see an elephant through a microscope. Managing for sustainability requires consideration all scales, but monitoring and assessing sustainability must be based on the recognition that different questions and different methods are appropriate for different scales.

References

Allen, T. F. H., and T. W Hoekstra. 1994. “Toward a definition of sustainability.” In Sustainable ecological systems: Implementing an ecological approach to land management. W. W. Covington and L. F. DeBano, tech. coord. Fort Collins, CO: USDA Forest Service, Rocky Mountain Range and Forest Experiment Station General Technical Report 247. 363 p.

Cheng, T. L., and P. B. Durst (2000). Development of National-level Criteria and Indicators for the Sustainable Management of Dry Forests in Asia: Background Papers. Food and Agricultural Organization of the United Nations (FAO).

Dixon, J. A., and L. A. Fallon (1989). "The concept of sustainability: Origins, extensions, and usefulness of policy." Society and Natural Resources 2: 73-84.

McCool, S. F., and R. Haynes. 1995. Sustainability. Prepared for the Interior Columbia Basin Ecosystem Management Project. Available online at http://www.icbemp.gov/science/scirpte.html or http://www.icbemp.gov/science/mccool.pdf.

Woodley, S. J., G. Alward, L. I. Gutierrez, T. W. Hoekstra, B. Holt, L. Livingston, J. Loo, A. Skibicki, C. Williams, and P. Wright. 1999. North American test of criteria and indicators of sustainable forestry. Fort Collins, CO: USDA Forest Service Inventory and Monitoring Institute Report No. 3. Available online at http://www.fs.fed.us/institute/cifor/cifor_1.html.