Background
Wetland assessment procedures are tools in the trade of wetland science that provide a definitive procedure for identifying, characterizing, or measuring wetland functions and/or social benefits. They are used in a variety of contexts for regulatory, planning, management, and educational purposes. In the 1980s, wetland assessment was simple, as there were only a few procedures to choose from (see summaries in Lonard et al. (1981), U.S. Environmental Protection Agency (1984), World Wildlife Fund 1992)). Wetland scientists were, however, frustrated by the lack of suitable procedures for most situations. Today, there are a variety to choose from, but the process is complicated by the ever-increasing number of procedures, differing approaches, and new terms. For example, at least 31 wetland procedures have been published since 1990 and more are being produced (Bartoldus 1999a). This paper summarizes key information on 39 current procedures (Table A).
Each procedure outlines, in varying degrees, the steps to assessing wetland function. Most begin with locating/delineating the wetland and then focus on the application of models (e.g., NH Method (Ammann and Lindley Stone 1991); AREM (Adamus 1993a, 1993b); Descriptive Approach (USCOE 1995); MDE Method (Fugro East, Inc. 1995)). A few describe earlier steps such as defining the assessment objectives or screening for red flags (e.g., Synoptic Approach (Leibowitz et al. 1992); EPW (Bartoldus et al. 1994); HGM Approach (Smith et al. 1995); NEFWPBP (Hicks 1997)). The Regulatory Assessment Method (Kusler 1997) provides guidance on how to conduct a preliminary assessment of wetland functions/values and other factors (e.g., red/yellow flags, natural hazards, alternatives for landowners) before applying an established assessment procedure, but it does not give instructions on how to select a procedure.
The HGM Approach (Smith et al. 1995) takes the lead as technically the most progressive and extensive effort among the wave of recent procedures. It was developed to satisfy the requirements of the 404 Program, but is also intended for use in a wide range of assessment situations. There has been some discussion about scientific and technical issues surrounding the HGM Approach (e.g., Brinson et al. 1997, 1998; Hruby 1997, 1998). Kusler and Niering (1998) take the matter further and pose a series of questions about wetland assessment procedures in general. The underlying cause of these discussions stems from the fact that no one procedure can satisfy the diversity of assessment situations. The same limitations that render a procedure useful for one setting become the basis for questioning the scientific or technical basis in another.
The difficulty in selecting a procedure has led to a need for further guidance. That guidance has not and cannot come from the regulatory agencies for a variety of reasons. As noted before, it is impossible to identify one preferred procedure, because no one method is suitable for all assessment situations. There are numerous procedures and each has been developed for specific purposes. The suitability of a procedure for a project depends upon the assessment objectives, geographic area, wetland type, desired level of detail, availability of applicable models, and other considerations. Additionally, policy favoring any one procedure might be considered an unfair practice. It might also discourage the development of additional, perhaps improved, techniques.
Publications
At least two publications are directed at simplifying the process of selecting a wetland assessment procedure. "A Comprehensive Review of Wetland Assessment Procedures" (Bartoldus 1999a) provides detailed information about 40 current procedures to help the user determine which best meets their specific needs. The manual contains a profile on each (e.g., references, contact person, definitions of key terminology, applicable habitat types, list of function/values categories). The profile is followed by an outline that includes the procedure steps, a sample of the model format, and time requirements. Several summary tables are also provided to facilitate a comparison of the different procedures. The ASTM "Standard Guide for the Assessment of Wetland Function" (Bartoldus 1999b) is designed to assist wetland managers by prescribing a sequence of steps for defining and assessing wetland functions. It also identifies properties that must be considered in the selection of a wetland assessment procedure to determine whether it will assist in satisfying the requirements of the wetland regulatory programs and/or produce valid design criteria for planned wetlands.
While the information contained in these two publications (Bartoldus 1999a, 1999b) is useful, it is insufficient, especially for the individual who needs succinct guidance on how to select a wetland assessment procedure. The Comprehensive Review is organized as a reference manual. Comparisons among procedures are facilitated with a few summary tables; however, the report is best used for learning the details of how each procedure works. The ASTM guide outlines some of the steps and considerations for selecting a wetland assessment procedure; however, it was written over 2 years ago and much of the information is outdated. The ASTM Guide steps focus on how to define and assess wetland functions, not on how to select a wetland assessment procedure.
Steps in selecting a function assessment procedure
The following section lists the steps and provides brief guidance for selecting a function assessment procedure. These steps are summarized in Table B. Also provided is a checklist for the user to follow and record decisions for each step (see example in Table C, blank checklist in Appendix A). The tables containing key information for each step can be accessed at appropriate points in the paper.
Step 1a: Define goals of the assessment (Table Step 1a)
The user must define exactly what he/she wants to achieve and list what the procedure must be able to do to attain these goals. Identify the overall goal(s), then identify specific requirements for the assessment (see Table Step 1a and example in Figure 1). Consider geographic area, habitat types, detail or sensitivity of the procedure, acreage in the measure of function, functions, and comparison of different habitat types.
Users may find that no one procedure can meet a broad range of goals. All procedures have constraints because they are designed for different purposes. For example, the NH Method (Ammann and Lindley Stone 1991) is used to evaluate New Hampshire freshwater wetlands in planning, education, and wetland inventory; but it is not used for detailed impact analysis on individual wetlands. It addresses 15 functional values including wetland wildlife habitat. In contrast, the Habitat Evaluation Procedure (HEP) (USFWS 1980) is used throughout the United States to document the quality and quantity of available wetland/upland habitat for selected fish and wildlife species. HEP can be used for detailed impact analysis and in other planning activities such as compensation analysis. It does not address other function categories with the current published Habitat Suitability Index (HSI) models.
Do not be concerned with the constraints of each procedure at this point. Just focus on the goals. The properties of each procedure are summarized in tables that accompany the following simple steps. Remember to leave options open to the possibility of using more than one procedure. One may be sufficient, but you may have to use or revise a combination of models from different procedures.
Step 1b. Select general objectives/applications (Table Step 1b)
Check one or more of the following set of objectives (or applications). Refer to Table Step 1b to identify procedures that can be used for the selected objectives.
q Impact analysis, i.e., procedure qualifies and/or quantifies impacts
q Compare landscape subunits (e.g., watersheds)
q Establish mitigation/compensation ratios, i.e., procedure provides numerical output that is suitable for establishing compensation ratios
q Guide to design, i.e., procedure can be used as a guide to design of planned habitats (e.g., plans for restoration, enhancement, or creation)
q Inventory or planning
q Describe biological condition or integrity
Step 2. Select preferred construct of procedure
All procedures may assess function, but the approaches and construct of these vary greatly. The procedures have been classified into the following five different constructs (Table Step 2). Select one or more. For example, if the construct does not matter, then all categories may be selected. Note that there are other differences, even among procedures that fall within the same category. It is strongly suggested that after selecting a procedure, users learn how the models (if any) have been developed, their limitations, and underlying assumptions.
q No models – User judgment. Function measure based on user judgment of site conditions as they describe level of function. Score may be qualitative (e.g., low, moderate, or high) or quantitative (e.g., scale 0 – 10).
q Model: Based upon population data. Function measure based on actual population data (e.g., species richness) for defined geographic area. Data then collected on organisms at specific site and score calculated for site as it compares to a reference.
q Model: Site data collected and score/index considers function and opportunity. Function measure based on observation/measurement of specific site conditions for one or more variables. Each variable is assigned subscores based on the site conditions. Variable subscores are then used to calculate function score (or index). Variables include those that measure functional capacity and opportunity/value (e.g., score higher if the opportunity to perform function is greater).
q Model: Site data collected and score/index only measures function. Function measure based on observation/measurement of specific site conditions for one or more variables. Each variable is assigned subscores based on the site conditions. Variable subscores are then used to calculate function score (or index). Variables only include measure of functional capacity.
q Landscape approach that uses models.
Step 3. Select applicable geographic area (Table Step 3)
Select the geographic region of interest (e.g., state, watershed, ecoregion). Refer to Table Step 3 to identify procedures used in each state. Use this as a guide to identify those that are applicable, but also realize that others may be modified for use in a particular area. The modifications may be minor or major, to ensure that the models are calibrated to the defined area.
Step 4. Select applicable general habitat type(s) (Table Step 4)
Select the general habitat type(s) to be assessed from the following: upland, non-tidal wetlands, tidal wetlands, and aquatic. Then, refer to Table Step 4 that identifies both general and specific habitat types that are applicable to each procedure. Note that with regard to the specific habitat types, there is no standard habitat classification used in assessment procedures. For example, wetland types may follow the USFWS classification (Cowardin et al. 1979), the HGM classification (Brinson 1993), or local terms (e.g., fresh/intermediate marsh). The entire table should be reviewed to identify applicable procedures. Please note that one procedure may not be sufficient when you are dealing with more than one habitat type. Also, realize that other procedures may be modified for use on other habitat type(s).
Step 5. Define the desired level of detail and sensitivity. Consider time, resources, and cost. Then select maximum time to be allotted for the assessment of each site (Table Step 5)
As with any study, more information will be obtained with an increase in effort, time, and money spent. Assessment procedures represent an alternative to extensive data collection and scientific analyses. While these procedures are more rapid, there is still a broad range of time requirements among them. Some are brief (e.g., 0.5 hr per assessment area), most are relatively rapid (1-2 hrs), and others can be comprehensive (e.g., 136 hrs). In general, procedures that are most time-consuming are also the most detailed and sensitive. Before selecting one, decide how much time, money, and resources are available. Estimate the number of assessment areas and determine how much time can be spent on each. Unfortunately, there is no standard approach to deciding the number of assessment areas and very few procedures provide guidance. The HGM Approach (Smith et al. 1995) identifies some criteria for identifying wetland assessment areas (e.g., same regional wetland subclass, physical separation, different level of impact). But, ultimately assessment goals and resources determine the number needed. Sometimes the answer is simple. For example, one is sufficient when evaluating a small homogenous wetland for impact analysis. If a proposed highway project will impact 25 wetland areas, each could be assessed separately or information could be collected on a few representative sites.
Once you have decided how much time can be allotted to each area, then identify the procedures that meet your time constraints (refer to Table Step 5). These data are from the authors of the individual procedures and reflect their perspective. There are two tables. The first identifies time estimates and assumes that models are available for a given habitat. The second identifies time estimates for those procedures that may require the development of new models. Note that the procedures are arranged by the maximum time needed for data collection and analysis. Preparation (e.g., habitat delineation, map preparation) was not included because it was difficult to directly compare the estimates provided by the authors. Some authors did not include preparation time in their response to the survey. Others did, but it's not always clear what was included.
Step 6. Determine whether there is a need to generate results that include the size of the habitat in the measure of function, i.e., do you need to distinguish between a I-acre and 100-acre site for function and size? (Table Step 6/8)
Although most procedures consider size of the habitat, it can be addressed in two very different ways. First, size may be one of several variables used in a model to define the level of function. A difference in size may increase or decrease the output (e.g., index, rating), but the output will always remain within set limits. For example, the HSI model used in HEP for breeding blue-winged teal uses "area of wetlands per 259 ha" as one of seven variables to estimate habitat suitability (Sousa 1985). The optimum area is 64.8 ha. Anything smaller lowers the HSI score. The size may be greater than 64.8 ha, but it will not change the HSI, which has limit of 1.0. HEP is also useful for describing the other way in which size is addressed in assessment procedures. Size (acreage) may be used as a multiplying factor to generate a separate unit of measure. For HEP, this alternative measure is the habitat unit (HU) and is derived by multiplying the HSI by the area of available habitat.
Outputs from the assessment procedures vary. Table D lists a few of the commonly used units of measure along with an explanation on how or if size is incorporated.
Some situations may not require a unit of measure that accounts for habitat size. It may be critical in other situations if the assessment results are used to make decisions with regard to habitat function and acreage. For example, size is important in permit actions where there is a need to describe not only the functional capacity of the mitigation wetland, but also the acreage required to compensate for wetland impacts. Once the size is defined, then the site selection process can begin.
Determine whether you need to generate results that include size in the unit of measure. If so, refer to Table Step 6/8 to identify procedures that provide a unit of measure that factors in size (acreage).
Step 7. Identify the desired function/social categories (Table Step 7)
Review the list of functions and refer to Table Step 7 to identify the general categories (e.g., hydrology, water quality, wildlife, and social) and specific functions for each procedure. Be aware that this table only includes the function titles. At this point, the definition can be assumed, but at some point it would be best to obtain and review the complete definitions to insure that the models are compatible with the functions that have been identified. Subtle differences in definitions can be important. For example, in some models the shoreline stabilization function is defined as the capacity of a wetland to bind sediments/soil. The sediment stabilization function as defined in WET (Adamus et al. 1991) is broader and refers to the effectiveness for binding soil and dissipation of erosive forces. Other procedures may consider the capacity to dissipate erosive forces as a separate function. Some procedures focus on one category (e.g., HEP measures habitat suitability for fish and wildlife species), while others look at a variety of functions and social categories.
Review the entire table to identify applicable procedures. Please note that one procedure may not be sufficient when dealing with more than one function. For example, HEP has often been used to assess fish and wildlife on larger projects. Models from other procedures are then used to assess other functions. If there are no models for some or all functions of the habitat types being evaluated, consider modifying existing models or developing new models.
After identifying procedures or individual models, make sure that they apply to your habitat type. Clearly, models assessing water quality in a riverine or a tidal wetland would differ in the specific definition of the function and the variables measured.
At this point and throughout the remaining steps, you may want to reconsider procedures that were initially considered not applicable (e.g., Steps 3-6). Some could possibly be modified for use on your project.
Step 8. Determine whether you want a separate unit of measure for each category or a measure that combines all function/social categories (Table Step 6/8)
Some procedures assess one function. Others may assess several functions and produce (a) one unit of measure for each function, or (b) a unit of measure that combines all functions into one. Decide which approach bests meets your needs. Refer to Table Step 6/8 to determine which procedures are applicable.
Step 9. Determine whether there is a need to compare different habitat types and/or habitats from different geographic regions (Table Step 9)
Many procedures are written for broad application and allow comparisons across a variety of habitat types and regions. Other procedures have a narrow focus that increases sensitivity and the ability to detect differences in functional capacity, but also limits comparisons that can be made. When models are calibrated to describe a particular habitat type, they can only be used to compare habitat areas of that same type. The same is true for models calibrated to describe habitats within a specified geographic area (e.g., watershed, Colorado Plateau region). It is inappropriate and meaningless to compare habitats from different habitat types and/or regions. When comparing similar habitats (same type/region), proceed to Step 9. If the habitat areas are dissimilar, identify procedures that will facilitate a comparison (Table Step 9), or decide on how comparisons can be made.
Step 10. Determine whether the procedure can be used to meet the specific goals of the assessment.
Assessment procedures are used within a variety of contexts (e.g., watershed planning, inventory, regulatory permit action, wildlife management, a guide to wetland design). As a result, the goals of the assessment often differ and there are special considerations. Examples of specific considerations include:
o Will the procedure provide an output that is suitable for monitoring? (Note that this will depend upon the monitoring protocol)
o Is the procedure suitable for assessing losses within an entire watershed (e.g., reservoir project)?
Concluding Thoughts
The growth of assessment procedures in recent years presents both opportunities and challenges to those needing to quantify and compare functions performed by wetlands. Wetland function assessment is not a precise activity. Too much variability exists in wetland types and conditions for ready-made approaches. Although some of the procedures are relatively simple to apply, none of them are "plug and play" but require input of data and common sense. We expect to see more development in the future, building on past approaches. Work with procedures such as HGM, HEP, and IBI will likely make the most advances.
Availability of Procedures
Only a few of the 39 procedures are available on the Internet to our knowledge. Those few are indicated in the Literature Cited section. An overview of each procedure and Point of Contact is found [here]. Additional information on the structure of the 39 is found in Bartoldus (1999a), which can be obtained from Environmental Concern, Inc.