From Information to Action: Monitoring Objectives, Benchmarks and Required Proportions
Monitoring objectives help translate information to action in natural resource management. These statements establish quantitative guidelines to help us determine if the management goals were achieved. For example, robust monitoring objectives can answer the following types of questions: Is there enough suitable sage grouse habitat on the landscape? Are watersheds functioning properly? Are management actions maintaining the health of the public lands? As outlined in the Developing Monitoring Objectives section of the Monitoring Design Worksheet, monitoring objectives at a minimum should include: 1. the focal indicator(s), 2. quantitative benchmarks or thresholds for each indicator, and 3. the proportion of the landscape that is required to meet a given benchmark.
Benchmarks are a key component of every monitoring objective. Benchmarks are indicator values, or ranges of values, that establish desired conditions and are meaningful for management. Benchmarks are used to determine if observed indicator values at assessed points (i.e., monitoring reaches or plots) are within the range of desired conditions. Conversely, failure to set benchmarks can make it difficult to interpret monitoring data. For example, achieving a plant density benchmark value following a seeding treatment may tell you that the project was successful; where failure to meet the benchmark may trigger reevaluation of the seeding methods. On the other hand, observed electrical conductivity (EC) values characterize the amount of cations and anions dissolved in stream water at a monitoring location, but without appropriate benchmarks, these observed values cannot be used to assess condition or the attainment of management objectives.
When considering an area with multiple monitoring locations, some amount of failure to achieve a benchmark is often acceptable. Natural events such as floods, droughts, fire and disease result in natural variability across a landscape. For this reason, monitoring objectives also include the proportion of the landscape that is required to meet a given benchmark. For example, achieving a benchmark density of plants on 80% of seeded acres can indicate success, even if 20% of the acres did not meet the benchmark value. If monitoring information shows that an insufficient amount of the resource has met a benchmark, then management changes will be triggered.
Benchmarks, along with associated required landscape proportions, provide a way to objectively operationalize policy statements such as: “take appropriate action” to make “significant progress toward fulfillment” of land health standards.
Example Monitoring Objectives, with Benchmarks and Required Proportions
- Soils Land Health Standard: In the grazing allotment, maintain soil aggregate stability of 4 or greater on 80% of lands with 80% confidence over 10 years.
- Watershed Function within Land Use Plan Area: Maintain bank stability of greater than or equal to 75% for 80% of perennial wadeable streams in the planning area with 95% confidence over 10 years.
- Sage Grouse Habitat within Land Use Plan Area: In all SFA and PHMA, the desired condition is to maintain all lands ecologically capable of producing sagebrush (but no less than 70%) with a minimum of 15% sagebrush cover or as consistent with specific ecological site conditions over 5 years.
Benchmark values should take the ecological potential of the landscape into account. It is important to consider how much the ecological potential varies across the study area when setting benchmarks, as areas with large differences in ecological potential may require that you establish multiple benchmarks for the same indicator. The goal is to ensure that we are comparing indicator values at assessed sites to benchmarks that are relevant to potential for those sites; i.e. that we are making an apples to apples comparison. For example, the benchmark for perennial grass cover will be set higher in loamy areas than in salt desert areas.
Setting benchmarks is a necessary but often challenging step. Benchmark values can be gleaned from a variety of sources, including existing policy (e.g., sage grouse habitat standards; state water quality standards), ecological site descriptions (USDA NRCS 1997), peer-reviewed literature, range of variability among a group of least disturbed sites (Hughes et al. 1986, Hawkins et al. 2000), paired reference sites, natural conditions predicted by ecological models (Hill et al. 2013, Olson and Hawkins 2012), reference data from other comparable monitoring efforts, and best professional judgment.Often, a combination of these resources will be used.
There are two general, valid approaches to setting benchmarks:
1) Reference or expected conditions (e.g., natural range of variability). This approach assumes that the goal of land management is to maintain or restore ecosystems to minimally or least disturbed conditions. For example, the EPA has partnered with BLM and other agencies to identify a network of “least disturbed” sites on streams throughout the West. Benchmarks are then defined in terms of departure from the indicator values found at these “reference” sites (reviewed in Hawkins et al. 2010). Similarly, ecological site descriptions describe reference plant communities, and benchmarks can be defined from the attributes of these plant communities (USDA NRCS 1997).
2) Desired conditions (e.g., habitat criteria). This approach focuses on the ecosystem goods and services that society desires the land to produce, and what range of values for a given indicator will sustain them. For example, the BLM has set a number of benchmarks for sagebrush cover and other vegetation characteristics in order to maintain habitat for the Greater sage-grouse (e.g., Stiver et al. 2015).
The key with setting benchmarks and the required landscape proportions is to clearly justify and clearly document the approach that you took. Likewise, the required proportions may follow from a variety of sources including ecological and policy considerations.
For more information, see Chapter 4: Management Objectives in Measuring and Monitoring Plant Populations (Elzinga et al. 2003); Stoddard et al. 2006; and Hawkins et al. 2010.
Reporting Monitoring Results
Using benchmarks to interpret monitoring information is not a new concept for land managers. However, applying benchmarks to estimate the proportion of a landscape that achieves benchmark conditions is new. To accomplish this, first, monitoring locations are assigned a condition class (e.g., meeting vs. not meeting the benchmark) based on the departure of observed indicator value(s) from the benchmark(s). Importantly, benchmarks can vary for different monitoring locations according to biophysical characteristics and ecological potential (see Setting Benchmarks). The use of benchmarks and related condition classes enables you to report the proportion of the landscape meeting objectives even across variable landscapes, simplifying stratification and reducing sample size requirements. After condition classes are assigned, with a probabilistic monitoring design, you can combine site-based results to estimate the proportion of the landscape that is meeting or not meeting objectives. An example monitoring result from a benchmark approach could be: 90% of the grazing allotment has sagebrush height of 30 cm or higher. See more detailed examples below.
Monitoring Examples Utilizing Benchmarks
Terrestrial Monitoring Example
Fig. 1. Proportion of early brood-rearing sage-grouse habitat that is meeting the benchmark of 15-25% sagebrush cover. The objective was for sagebrush cover to meet this benchmark across 80% of the habitat area, but it was not achieved.
For nesting and early brood-rearing sage grouse habitats, one objective is for sagebrush cover to be greater than 15% and less than 25% across 80% of the habitat area. The benchmark in this case is greater than 15% and less than 25% sagebrush cover, and it is set in policy based on sage grouse research (e.g., Stiver et al. 2015). The proportion of the habitat area required to meet the benchmark is 80%.
We can estimate the proportion of habitat area meeting the sagebrush cover benchmark based on the number of monitoring sites achieving benchmarks. Sagebrush cover values at 19 monitoring sites in early brood-rearing habitat were compared against this benchmark (Table 1). Overall, 33% of early brood-rearing habitat met this benchmark (Fig. 1). Given that the objective for sagebrush cover was to meet the benchmark across 80% of the habitat area, the objective was not achieved.
Table 1. Example sagebrush cover data from monitoring sites and how it relates to the benchmark in the objective. Site 1 achieves the benchmark of 15-25% sagebrush cover set forth in the objective whereas Sites 2 and 3 do not.
Note that sagebrush cover is only one indicator used to assess sage grouse habitat. To complete site-scale habitat suitability ratings for a HAF assessment (Stiver et al. 2015), the ID team would take into account multiple indicators and the proportion of the assessment area meeting benchmarks.
Aquatic Monitoring Example
Figure 2. Distribution of bank stability values observed among a network of 30 least disturbed reference sites. Bank stability decreases in response to stress; thus, the lower 25th and 5th quantiles of the reference distribution were used to define benchmarks to differentiate ‘Minimal’, ‘Moderate’, or ‘Significant’ departure from reference condition. These quantiles correspond to bank stability values of 87% and 75%, respectively. The objective is to maintain “Minimal” or “Moderate” departure from reference conditions, or bank stability of greater than 75%.
To assess whether stream channels are maintaining proper form and function, bank stability is a common indicator. For example, managers might seek to maintain bank stability greater than 75%, depending on stream type, for 90% of stream kilometers with 90% confidence over 10 years. Note that the benchmark in this objective is for bank stability to be greater than 75%, depending on stream type. The degree of allowable departure from this benchmark is 20% with 90% confidence. This benchmark was derived based on the natural range of variability for bank stability across a network of least disturbed sites (Fig. 2). In contrast, the benchmark in the terrestrial example was set in policy based on research.
To evaluate whether the bank stability objective was achieved, first we can look at the distribution of bank stability at monitoring sites in relation to least-disturbed reference sites. Using the above condition classes (Fig. 2), we can classify each of these 20 sites as having ‘Minimal’, ‘Moderate’, or ‘Significant’ departure from reference depending on their departure from the range of reference conditions (Table 2; Fig. 3).
Table 2. Example bank stability data from monitoring sites, along with condition classes based on least disturbed reference sites. Sites 2 and 3 achieve the benchmark of 75% bank stability set forth in the objective.
Figure 3. Bank stability compared between 20 monitoring sites and 30 least disturbed ‘reference’ sites. Monitoring sites with bank stability values falling below the 25th quantile of reference (87% bank stability) are considered to have ‘moderate’ departure from reference condition ratings. Monitoring sites with values below the 5th quantile (75% bank stability) are considered to have ‘significant’ departure
Finally, we can estimate the proportion of stream kilometers in each of the condition ratings based on the number of sites achieving benchmarks. The objective was for 90% of stream kilometers to have greater than 75% bank stability (i.e., scoring “Minimal” or “Moderate”). A total of 91.7% of stream kilometers met this benchmark, with 83.4% of stream km achieving minimal departure from reference conditions and 8.3% achieving moderate departure from reference conditions (Fig. 4). Therefore the objective was achieved.
Figure 4. Proportion of stream kilometers having minimal, moderate, or significant departure from least disturbed reference conditions. The objective was for 90% of stream kilometers to have greater than 75% bank stability (i.e., scoring “Minimal” or “Moderate”; green and yellow bars), and this objective was achieved.
Elzinga et al. 2003. Chapter 4: Management Objectives in Measuring and Monitoring Plant Populations
Hill, R.A., C.P. Hawkins, and D.M. Carlisle. 2013. Predicting thermal reference conditions for USA streams and rivers. Freshwater Science 32: 39–55.
Hughes, R.M., D.P. Larsen, and J.M. Omernik. 1986. Regional Reference Sites: A Method for Assessing Stream Potentials. Environmental Management 10 (5): 629–635.
Hawkins, C.P., R.H. Norris, J. Gerritsen, R.M. Hughes, S.K. Jackson, R.K. Johnson, R.J. Stevenson. 2000. Evaluation of the use of landscape classifications for the prediction of freshwater biota: synthesis and recommendations. Journal of the North American Benthological Society 19:541-556.
Hawkins, C. P., J. R. Olson, and R. A. Hill. 2010. The reference condition: predicting baselines for ecological and water-quality assessments. Journal of the North American Benthological Society 29:312-358
Olson, J.R., and C.P. Hawkins. 2012. Predicting natural base-flow stream water chemistry in the western United States. Water Resources Research 48 (2): W02504.
Stiver, S.J., E.T. Rinkes, D.E. Naugle, P.D. Makela, D.A. Nance, and J.W. Karl, eds. 2015. Sage-Grouse Habitat Assessment Framework: A Multiscale Assessment Tool. Technical Reference 6710-1. Bureau of Land Management and Western Association of Fish and Wildlife Agencies, Denver, Colorado.
Stoddard, J.L., P. Larsen, C.P. Hawkins, R.K. Johnson, and R.H. Norris. 2006. Setting expectations for the ecological conditions of running waters: the concept of reference conditions. Ecological Applications 16:1267-1276.
USDA NRCS. 1997. Chapter 3: Ecological Sites and Forage Suitability Groups. National Range and Pasture Handbook. Rev. 1, 2003.