Solitude or Primitive and Unconfined Recreation Quality Indicators

USFS Wilderness Character Monitoring Technical Guide

5 Solitude or Primitive and Unconfined Recreation Quality Indicators

Monitoring the Solitude or Primitive and Unconfined Recreation Quality assesses whether management of a wilderness is trending over time towards protecting outstanding opportunities for specific unique recreational experiences. Key indicators and measures monitor solitude (from activities occurring both inside and outside of wilderness), primitive recreation, and unconfined recreation. This section provides detailed guidance for monitoring the following indicators and measures:

5.2 Indicator: Remoteness from Sights and Sounds of Human Activity Inside Wilderness

5.2.1 Measure: Index of Encounters

5.2.2 Measure: Index of Recreation Sites Within Primary Use Areas

5.2.3 Measure: Acres of Wilderness Away From Access, Travel Routes and Developments Inside Wilderness

5.2.4 Measure: Miles of Unauthorized Trails

5.3 Indicator: Remoteness from Sights and Sounds of Human Activity Outside the Wilderness

5.3.1 Measure: Acres of Wilderness Away From Adjacent Travel Routes and Developments Outside the Wilderness

5.4 Indicator: Facilities That Decrease Self-Reliant Recreation

5.4.1 Measure: Index of NFS Developed Trails

5.4.2 Measure: Number of Authorized Constructed Recreation Features

5.5 Indicator: Management Restrictions on Visitor Behavior

5.5.1 Measure: Index of Visitor Management Restrictions

5.2 Indicator: Remoteness from Sights and Sounds of Human Activity Inside Wilderness

This indicator focuses on wilderness visitation and the capacity of the wilderness setting to allow for escape from the sights and sounds of human activity. There are four measures under this indicator: one required measure on encounters and three measures on other aspects of remoteness from human activity inside wilderness from which units are required to select at least one.

5.2.1 Measure: Index of Encounters

This measure monitors encounters by assessing one of the following, listed in order of preference: (1) an index evaluating traveling and camp encounters; (2) the number of traveling encounters or camp encounters (but not both); (3) the number of visitors; or (4) the trend in visitation. Local units may select the appropriate protocol option as described in step 1 below. Local data are compiled and stored in local archives. Local staff calculate the measure value. Table 2.5.1 describes key features for this measure.

Table 2.5.1—Summary of measure type, protocol options, local tasks, national tasks, and frequency of data reporting for measure "Index of Encounters."

Protocol

Step 1: Determine which protocol option is appropriate for the wilderness. The four protocol options for this measure are summarized below, listed in order of preference from highest to lowest data adequacy.

Protocol Option 1—Index of Traveling and Camp Encounters. This protocol option requires encounter data on both traveling encounters and camp encounters. The mean number of traveling encounters per hour and the mean number of camps seen from occupied sites are combined in an index to derive the measure value.

Protocol Option 2—Number of Traveling or Camp Encounters. This protocol option only requires one type of encounter data—either traveling encounters or camp encounters—but not both. The measure value is either the mean number of traveling encounters per hour or the mean number of camps seen from occupied sites.

Protocol Option 3—Number of Visitors. If direct data on encounters are not available, indirect (proxy) data on visitation may be used instead. For this protocol option, the measure value is the number of visitors.

Protocol Option 4—Trend in Visitation. If there are no direct data on encounters or data on visitation, professional judgment may be used to determine the trend in this measure. For this protocol option, the measure value is the perceived trend in visitation, reported as increasing, stable, or decreasing.

To determine which protocol option is most appropriate, local units must first identify which types of data sources (encounter data, visitation data, or professional judgment) are available for a wilderness. Given the great amount of local variability in data collection, there is no strict formula for selecting the preferred data source. If a wilderness has multiple types of data (e.g., traffic counter data for some areas, traveling encounter data for other areas), select the best source, taking into account the geographical coverage, amount of data collected (i.e., number of days of monitoring), and accuracy of the method. In some cases the local unit may choose one of the options as the required measure and then one of the other options as a locally developed measure. And in some cases, it may be that combining different data sources generates the best overall assessment; if so, this should be considered professional judgment and the Protocol Option 4–Trend in Visitation, should be selected. Additional considerations for determining the most appropriate data source are described below by data type.

Encounter data (for Protocol Options 1 and 2). Valid traveling or camp encounter data should follow the Forest Service national minimum protocol for monitoring outstanding opportunities for solitude, or should be compiled using a locally defined protocol that provides data quality and quantity comparable to, or better than, the national minimum protocol. The national minimum protocol, available online at http://www.wilderness.net/toolboxes/documents/WC/National_Minimum_Protocol_Solitude.pdf, was used in the 10-year Wilderness Stewardship Challenge, and it provides detailed instructions for sampling, collecting data, standardizing data, and reporting results for WSP.

Use of informal encounter data collected opportunistically is not recommended unless these records include the basic data fields required in the national minimum protocol to standardize encounter rates across locations and dates. Specifically, information for each observation session must include the interval of time when data were collected and the location of data collection. If such information is not collected, it is not possible to track trends over time with confidence, and the data should not be used for this measure. If encounter data are used to evaluate this measure, local units must be confident that data from each monitoring period are comparable, meaning data come from the same location and use season.

Visitation data (Protocol Option 3). Visitation data may be collected under locally developed protocols through a variety of data sources, including permits, trailhead registers, traffic counters, or other means. To determine whether visitation data are appropriate to use for this measure, or to choose the best visitation data from multiple potential sources, consider two primary factors:

Data accuracy—Data accuracy is similar to data adequacy (see section 1.2.3 of part 2). Monitoring systems vary considerably in the accuracy of the data collected. For instance, mechanical counters (e.g., car counters, TRAFx trail counters) can be highly accurate with complete 24/7 coverage, while self-issue permits can have low and variable compliance rates, resulting in poor data accuracy. The data source with the highest accuracy should be chosen. Traffic count data are often more accurate and complete than ranger reports, so if both types of data are available, it is advisable to use the traffic count data. # Geographical coverage—Many units collect visitation data for limited locations. This is not necessarily a problem if data for the same locations are collected in subsequent years, and if the protocols provide at least as complete coverage as required in the national minimum protocol. Ideally, data from more locations would be preferable, but coverage will often be less of a concern than data accuracy.

Professional Judgment (Protocol Option 4). If data for encounters or visitation are not available, professional judgment may be used to determine the trend in visitation. Local units would not be expected to report actual estimates of encounters or visitation; instead they would report the perceived trend at 5-year intervals. If a wilderness has multiple types of data (e.g., visitation data for some areas, encounter data for others), the best decision might be to combine these sources through professional judgment.

Step 2: Document the data compilation strategy. To ensure confidence in tracking trends, data must be compiled consistently over time. Given the amount of variability in data sources and protocol options for this measure, it is essential that local units document the data compilation strategy (including the unit of measure as well as the timing, location, and intensity of data collection) for each wilderness. From year to year, local units should also document any special circumstances that may have affected data collection (e.g., equipment failures, gaps in data, or weather events that may have affected visitor use). Documentation may consist of a brief narrative or detailed instructions, and may be stored locally, on shared drives, or uploaded to the WCMD.

If encounter or visitation data are used, local units must document the unit of measure as "people," "vehicles," or "groups" (see table 2.5.2). The specific choice of unit of measure is not critical as long as each monitoring cycle uses the same units over time to assess trends. For some data sources (e.g., permits or trail registers), it is simpler to use "groups" as the unit of analysis, rather than "people," because this requires only counting the total number of entries rather than summing the number of people in each group.

Table 2.5.2—Choices for units of measure for encounters and visitation.

To be able to track change over time, data must be collected in the same places and during the same seasons each monitoring cycle. Documentation for each wilderness should include a map that clearly identifies locations of data collection (e.g., zones monitored for encounters or trailheads where traffic counters are placed). Given the extreme variation of visitor use across the year, it is advised only to collect and report data for the primary use season (additional guidance on this is provided in the national minimum protocol). The primary use season will vary depending on where a wilderness is and the type of visitor use it receives. If National Visitor Use Monitoring (NVUM) data are used (see Caveats and Cautions for concerns about using NVUM), document specific sampling sites and dates from each monitoring cycle. This information provides context for inferring whether differences in NVUM's visitor use estimates are due to actual changes in use, or whether they are an artifact of changes in sampling times or locations.

Documentation for this measure should also include information on the sampling intensity and data accuracy. Because visitor use is highly dependent on weather, fire, publicity about specific locations, and other factors, measures based on a small sample of dates in any given year may be a poor indicator of overall visitor use or encounter rates. The more dates included in sampling, the greater the likelihood of drawing correct interpretations about trends over time. Similarly, including assessments of data accuracy (e.g., documenting compliance rates for self-issue permits) allows for more confidence in interpreting trends over time. If traffic counters are used, local units also need to perform calibration tests to ensure the accuracy of the data. TRAFx provides links to various studies and documents to help local units design calibration studies (https://www.trafx.net/counting_methodology.htm).

If professional judgment is used to assess trends in visitation, local units must document who made the assessment and the basis for their determination. For example, if informal encounter data were combined with self-issue permit data to derive the trend in visitation, those data sources and the data adequacy of each should be recorded.

Step 3: Compile and process the data. Data may be collected over a span of multiple years within the 5-year reporting period for this measure. For example, encounter data may be collected in different locations in different years, with a full cycle of monitoring (all identified monitoring areas) completed after 5 years. Or, partial data may be collected for a single area across multiple years. Data are considered complete when a sufficient amount of data (per the national minimum protocol) have been collected for all monitoring areas. Ideally, it would be better to collect encounter or visitation data annually (i.e., more frequently than the minimum 5-year frequency required for this measure) because wilderness visitation can be quite variable and can depend on many factors, such as snowpack, weather conditions, fire events, and economic conditions. It is recognized, however, that most wildernesses will not be able to collect complete data for all locations every year.

Once all data have been compiled, the data are processed to derive the measure value. Data processing requirements are described below for each protocol option.

Protocol Option 1—Index of Traveling and Camp Encounters. Traveling encounters are generally reported as the number of encounters (people or groups) per hour, while camp encounters are generally reported as the number of camps (i.e., the number of groups) seen from occupied sites. While traveling and camp encounter data may be processed in a variety of ways, the national minimum protocol suggests recording the mean number of encounters per hour (i.e., the mean encounter rate) separately for each monitoring area. For this protocol option, local units must compute the grand mean (i.e., the mean of means) for a wilderness for both traveling and camp encounters by averaging the mean encounter rate across all monitoring areas within a wilderness, as illustrated in tables 2.5.3 and 2.5.4.

Table 2.5.3—Example of computing the grand mean number of traveling encounters per hour based on data collected using the national minimum protocol.

Table 2.5.4—Example of computing the grand mean number of camp encounters based on data collected using the national minimum protocol.

Local units then combine the grand means for traveling encounters and camp encounters into an index using the following formula:

Traveling encounters + (2 × Camp encounters) = Index of encounters

For example, using the grand means from tables 2.5.3 and 2.5.4, the calculation for the index of encounters would be:

10 + (2 × 2) = 10 + 4 = 14

This index weights camp encounters twice as heavily as traveling encounters because research has shown that seeing or hearing other campers is substantially more impactful on visitors' experiences than encountering people on the trail. The measure value reported for this protocol option is the index value.

Protocol Option 2—Number of Traveling or Camp Encounters. This protocol option follows the same initial steps described above for the Index of Traveling and Camp Encounters protocol option using whichever encounter data are available for a wilderness. Once the grand mean of either traveling or camp encounters has been calculated, however, no further data processing is required. The measure value reported for this protocol option is either the grand mean of traveling encounters or the grand mean of camp encounters.

Protocol Option 3—Number of Visitors. If visitation data are used, sum the total number of people, groups, or vehicles across all trailheads or access points monitored for a wilderness. If traffic counters are used, the data may need to be corrected to account for entries and exits and ensure visitors are only counted once. This would be the case whenever vehicles must travel both in and out over the sensor. (If the site layout and counters are arranged so that each vehicle is only counted once, correction is not needed.) To correct the data, divide traffic counter totals by two. The measure value reported for this protocol option is the total number of people, groups, or vehicles. Table 2.5.5 shows variations of the measure value based on the data source.

Table 2.5.5—Values to report for various data sources used for indirect measures for the index of encounters

Protocol Option 4—Trend in Visitation. If professional judgment is used, consult with individuals with the best knowledge of on-the-ground conditions (e.g., lead wilderness rangers) to assign an applicable trend category from the following options:

Decreasing visitation—visitation levels appear to be trending over time towards fewer visitors.
Stable visitation—visitation levels appear to be remaining about the same.
Increasing visitation—visitation levels appear to be trending over time towards more visitors.

For the measure baseline year, the stable visitation category should be selected. In subsequent monitoring cycles, trends in visitation should be assessed by comparing current perceptions of visitation levels with perceptions from the measure baseline year. Given the subjective nature of professional judgments, it is important to include additional documentation (e.g., a brief narrative) for each monitoring period that explains who assigned the trend category and the basis for their determination. The measure value reported for this protocol option is the selected trend category.

Step 4: Enter data in the WCMD. Enter the appropriate measure value for the selected protocol option in the WCMD. The measure value is either the index value, the grand mean of encounters, the number of visitors, or the trend category for visitation.

As described above, it would be ideal if complete encounter or visitation data were collected annually for this measure, although it is recognized that most units will be unable to do this. If a wilderness using protocol options 1, 2, or 3 (but not protocol option 4) does have complete data collection each year, however, the measure value is instead calculated as a 3-year rolling average. With annual (complete) data collection, local units must still enter the values described above in the WCMD, but the WCMD then automatically calculates 3-year rolling averages from those data.

Caveats and Cautions

Encounter and visitation monitoring protocols tend to change frequently. For example, local unit managers decide to monitor different areas, collect different data, or implement (or discontinue) self-registration systems. Because of such changes, units should not simply assume that differences over time reflect real change. Before drawing conclusions about meaningful change, verify that the data collected in different time periods are in fact comparable.

The suitability of the NVUM program (https://www.fs.fed.us/recreation/programs/nvum/) to generate wilderness visitation data should be addressed. NVUM provides a measure of wilderness visits for each national forest, along with the 90-percent confidence interval. Because NVUM was designed to generate estimates for the NFS as a whole, some features of the methodology make it problematic to use for estimating use of a specific wilderness. Using NVUM as a proxy for the number of encounters in a wilderness is not recommended for the following reasons:

All wildernesses in a single Forest Service unit are combined in a single sampling stratum so only a few sites may represent data for any given wilderness. Professional judgment would have to be used to apportion visits across the different wildernesses.
Sampling intensity is low, with as few as eight sample days for wilderness for any given Forest Service unit.
Confidence intervals for visitation tend to be wide.
If a wilderness is shared across units, it is not possible to assemble data for an individual wilderness.

The current NVUM report (USDA Forest Service 2013b) (http://www.fs.fed.us/recreation/programs/nvum/2012%20National_Summary_Report_061413.pdf) states that data "currently cannot be used to identify trends or make assumptions about changing use patterns." Changes have been made in the NVUM protocols concerning reclassification of wilderness sampling sites as high, medium, or low use. This means that data in some cases will not be comparable, or will produce questionable trends. For example, table 2.5.6 shows data for two rounds or cycles of data collection for the Deschutes, Mt. Hood, and Willamette National Forests in Oregon. The wide confidence intervals around the estimated number of visits (± 50 to 60%) make it extremely difficult to detect trends. Moreover, the data suggest that wilderness use in the Deschutes and Willamette National Forests decreased during a time in which the local population grew considerably and local permit data show increased use, calling into question the validity of the NVUM data.

Table 2.5.6—NVUM estimates of wilderness visits for three Region 6 National Forests (Deschutes, Mt. Hood, and Willamette) for years 2005–2009 and 2010–2014.

Despite these limitations, there may be a few situations when NVUM is appropriate for documenting visitation, such as if a local unit has only one wilderness or if a forest conducted additional data collection in conjunction with the standard NVUM surveying. If NVUM data are used the measure value is the total annual number of wilderness visits.

Data Adequacy

Data adequacy must be assessed for each wilderness individually based on the quality and quantity of local data. A general overview of data adequacy for various data sources is provided below as a starting point for local data adequacy assessments.

Data on the number of encounters collected in accordance with the national minimum protocol should be of good quality, but in most cases will only be available for selected monitoring zones, so data quantity will usually be partial. In some wildernesses, such as a small wilderness with only two or three primary use areas, data from the national minimum protocol will capture overall encounter conditions and can confidently be used for assessing trends. However, in large wildernesses with variable use across different locations, the data only reflect conditions within the areas monitored, which may or may not be representative of a wilderness as a whole. Even if not representative, trends in those selected areas could be very informative in terms of suggesting the need for more intensive monitoring. Thus, the overall data adequacy is medium.

Data on visitation collected by traffic or trail counters are generally of good data quality, and depending on coverage will range from insufficient to complete data quantity. Data are usually complete for each location because counters record continuously, but only a few locations may have counters installed. Overall data adequacy is therefore medium to high.

Data on visitation collected by trail registers or self-issue permits can have significant limitations. For example, compliance rates with self-issue permits tend to be low and variable (Cole and Hall 2008), and if observational data are not collected to calibrate the counts generated, accuracy may be low. Moreover, self-issue stations may run out of forms or pens, leading to gaps in data, particularly during high use times. However, mandatory registration systems can provide good coverage, and public contact reports or other observations can generate correction factors to adjust for non-compliance and improve confidence in data quality. Thus, data adequacy can range from medium to high.

The other widely available data source, NVUM, is not designed to monitor individual wildernesses separately, and therefore should be used cautiously, if at all. While consistent protocols and a trained workforce ensure moderate to good quality data, representation of individual wildernesses is poor, making data quantity insufficient unless additional sampling has been conducted. NVUM reports provide a 90-percent confidence interval for wilderness visits, which gives local units information about the precision of the estimates. Data accuracy is usually low to medium.

Professional judgment about trends in visitor use can be good in some cases, if it is based on multiple years of on-the-ground experience. However, professional judgment is not acceptable for estimating actual encounter rates or visitation levels.

Additionally, natural human tendencies to focus on anomalies may call into question the validity of professional judgment. Thus, data adequacy can range from low to high depending on the sources of data used.

Frequency

Every 5 years, data on encounters or visitation are compiled and the index of encounters, number of encounters, number of visitors, or trend category is then entered in the WCMD. Field data collection may span multiple years within the 5-year reporting interval. Although the minimum frequency for this measure is every 5 years, data compilation, analysis, and entry may occur more frequently if so desired for the Index of Traveling and Camp Encounters, Number of Traveling or Camp Encounters, or Number of Visitors protocol options.

Threshold for Change

The threshold for meaningful change differs depending on the protocol option used. If Protocol Option 1—Index of Traveling and Camp Encounters, Protocol Option 2—Number of Traveling or Camp Encounters, or Protocol Option 3— Number of Visitors are used, the threshold is a 10-percent change in the measure value or number of encounters or visitors. Once there are five measure values, the threshold for meaningful change will switch to regression analysis for these three protocol options. (If a wilderness has annual data collection for any of these protocol options, the threshold would be a 10-percent change in the 3-year rolling average or, once there are five measure values, switching to regression analysis of the rolling averages). If Protocol Option 4—Trend in Visitation is used, the threshold is any change in categories. A decrease in the encounter or visitation measure value beyond the threshold for meaningful change, or a change in categories towards decreasing visitation, results in an improving trend in the measure.

5.2.2 Measure: Index of Recreation Sites Within Primary Use Areas

This measure is an index that assesses the number of recreation sites and their condition, based on the national minimum protocol for recreation site monitoring. Local data are compiled and stored in local archives. Local staff calculate the measure value. Table 2.5.7 describes key features for this measure.

Table 2.5.7—Summary of measure type, protocol options, local tasks, national tasks, and frequency of data reporting for measure "Index of Recreation Sites within Primary Use Areas."