Bird Monitoring Protocol for the Central Alaska Network (CAKN)

Standard Operating Procedure #1: Sample Designs

Please cite this document as:

Knutson, M., B. Route, and T. Gostomski. 2010. Standard operating procedure #1: Sample designs. In Gostomski, T., M. Knutson, N. P. Danz, B. Route, and T. W. Sutherland. 2010. Landbird monitoring protocol, Great Lakes Inventory and Monitoring Network. Natural Resource Report NPS/GLKN/NRR—2010/225. National Park Service, Fort Collins, Colorado.

Contents Page

Figures...... v

Tables ...... v

Introduction ...... 1 Species of Management Concern ...... 1 Target Population and Sampling Frame ...... 1

Objectives ...... 3 1. Inventory ...... 3 2. Change Over Time ...... 3 3. Community Composition ...... 4 4. Management Action ...... 4 5. Adaptive Management ...... 4 6. Habitat Associations ...... 5

Revisiting Sampling Points ...... 7

Spatial Considerations ...... 7

Sample Size ...... 9

Sampling Designs ...... 11 Design #1: Random design, for small to large land units ...... 11 Design #2: Stratified random design, stratified by access, for small to large land units ...... 12 Design #3: Stratified random design, stratified by habitat type, for small to large land units ...... 13 Design #4: Multi-stage probabilistic design, stratified by accessibility, for large land units ...... 15

Literature Cited ...... 19

iii

Figures Page

Figure 1. Diagram of macro-grid with mini-grids marked in gray. Each mini-grid contains twenty-five 300 m cells...... 15

Tables Page

Table 1. Minimum sample sizes and the number of required visits per year for sampling relatively common species. Sample sizes and number of visits correlate to different levels of alpha risk and power and to a minimum detectable trend and the number of years of sampling required to observe that trend...... 9

Table 2. Example of a sampling rotation for landbird monitoring using a multi-stage design over a 20-year period (from McIntyre et al. 2004)...... 17

Introduction

This SOP provides options for sampling designs to be used for purposes of monitoring landbirds during the breeding season. Several resources that describe basic sampling designs were used in developing this SOP (Cochran 1977, Thompson 2002, Hill et al. 2005).

The monitoring coordinator for each park should work with his/her supervisor to define management and sampling objectives and to select the appropriate sampling design to meet their needs. There may be technical resources available to assist in generating a set of points the meet the sampling design criteria.

Species of Management Concern For parks whose management objectives focus on a condition or change in the status of specific species, it will be necessary to define a subset of landbird species a priori that will serve as focal species for management. Scientific reasoning, coupled with statistical analyses of inventory or pilot data are used to identify appropriate focal species for future monitoring. These species may have one or more of the following attributes:

1. closely associated with major habitat types under management (habitat specialist or indicator species); 2. classified as a species of management concern by Partners in Flight, North American Bird Conservation Initiative (NABCI), or U.S. Fish and Wildlife Service (USFWS) Division of Migratory Birds; and 3. detected with sufficient frequency that the sampling objectives are likely to be attainable with available resources.

Target Population and Sampling Frame Important issues to consider in a monitoring plan are the target population and the sampling frame. The target population is what you are interested in learning about (i.e., which landbird birds occur in the park or in a particular management unit or habitat?). The sampling frame, or area of interest, is the population of sampling units (locations) that most closely approximates where the target population may be found and that has some possibility of being sampled (e.g., entire park, a management unit, particular habitat[s] in the park). All monitoring plans should clearly define the target population in the objectives and define the sampling frame in the sampling design so that it is clear to which area or population the summary information applies.

A fundamental rule of sampling design is that you cannot extrapolate your results to locations that had no opportunity to be sampled. For example, if large portions of your target area are inaccessible because they occur on private property, you might be very interested in the birds that breed in those locations (target population), but if those areas are not in the sampling frame (no possibility of being sampled), then you cannot attribute your results to those areas. You should be able to describe your sampling frame and create a map, blocking out any areas that were not in the sampling frame. It is rare that your sampling frame will encompass your entire population of interest; your sampling design is the best representation of the population of interest that you can achieve, given logistical and resource limitations.

Objectives

In any new monitoring effort, the first step is to clarify the management questions and objectives. Why do you want to monitor birds? How will you use the information to guide your land management decisions? A workshop may be needed for key park staff, partners, and advisors to answer these questions, set objectives, consider alternatives, and design an adaptive management framework that integrates bird monitoring with management objectives and decisions. These decisions drive your sampling design.

The sampling designs in this SOP can be used to address several potential management objectives, all of which employ estimates of bird abundance and occupancy. The specific objectives of a park monitoring plan should be explicitly stated before undertaking the monitoring (Johnson 2000). The objectives below are a template for expressing management objectives and the associated sampling objectives; they should be modified to match the park’s management objectives (Elzinga et al. 2001). Various resources are available to guide you in defining clear management objectives (Elzinga et al. 2001, Adamcik et al. 2004, Butler and Koontz 2005, Arkema et al. 2006, Schroeder 2006, Mahan et al. 2007).

1. Inventory Conduct an inventory of landbird species and estimate their abundance, density, or occupancy. The objective is to create a list of all landbird species in the target habitat. This objective is generally applicable when there is no baseline data for the site. For this objective, sample the largest number of points in the target habitat that you can, regardless of the size of the land unit. For the smallest land units, we simply want an inventory and some estimate of variance from which to estimate sample sizes for future monitoring. Thus, if the unit will only accommodate 10 points, we still sample the 10 points. For larger land units, the more points sampled, the higher the probability of detecting a large proportion of the species present.

Management objective: I want to create a species list for the park or management unit, and I want to estimate the abundance (birds/point), density (birds/hectare [ha]), or the probability of occupancy (proportion of sites occupied) of a set of landbird species in the years 2005-2006. I want pilot data adequate to estimate the sample sizes needed to meet future management objectives (see below).

Sampling objective: I want to be 80% confident (mean ± 20%) that the abundance/density/ occupancy estimates for species present are within 20% of their respective true values.

2. Change Over Time Detect change in abundance, density, or occupancy of landbird species. For this objective, monitoring must be conducted over long time periods (5-10 years).

Management objective: I want to be aware of a 50% decrease in the abundance (birds/point), density (birds/ha), or occupancy (proportion of sites occupied) of species X from 2000 to 2005 in the area of interest.

Sampling objective: I want to be 80% certain (α=β=20%, one-tailed test) of detecting a 50% decrease in the abundance/density/occupancy of species X from 2000 to 2005; I am willing to accept a 20% chance of inferring a change when the decrease is <50%.

3. Community Composition Detect change in community composition of landbird species.

Management objective: I want to be aware of a 20% decrease in the species richness of landbird species from 2000 to 2005 in the area of interest.

Sampling objective: I want to be 80% certain (α=β=20%, one-tailed test) of detecting a 20% decrease in species richness of landbird species from 2000 to 2005; I am willing to accept a 20% chance of inferring a change when the decrease is <20%.

4. Management Action Evaluate achievement of a species target or threshold in abundance, density, or occupancy, usually in response to management actions.

Management objective: Through management actions, I want to increase species X to an abundance of 0.4 birds/point or the probability of occupancy to 40% from 2000 to 2005 in the area of interest.

Sampling objective: I want to be 80% confident (mean ± 20%) that the abundance and/or occupancy threshold has been achieved (1-tail, t-test comparing mean with target or threshold) within 20% of their respective true values.

5. Adaptive Management Monitor bird abundance, density, or occupancy as part of an adaptive management process. This objective is used only if a specific management action or decision is framed as part of an adaptive management process (Schreiber et al. 2004). Adaptive management is a systematic approach for improving resource management by learning from management outcomes (U.S. Department of the Interior 2006). Adaptive management requires five elements: objectives, potential management actions, models of system response to management actions, measures of confidence in the models, and a monitoring program that assesses achievement of objectives and updates the models (Nichols and Williams 2006). Over time, the resource manager learns which management actions have the highest probability of achieving the stated objectives, given key system variables.

Management objective: I want to estimate the abundance (birds/point), density (birds/ha), or the occupancy of a set of landbird species in 2005 in the area of interest after applying a specific management action. These metrics will be used to update confidence measures for a set of competing models that predict bird community responses to specific management actions or decisions. These confidence measures will help me decide among alternative management actions in the future.

Sampling objective: I want to be 80% confident (mean ± 20%) that the abundance/density/ occupancy estimates are within 20% of the true value as a starting point; the minimal accuracy of my estimates need only be accurate or precise enough to distinguish among my competing predictive models. If my competing models require more accuracy or precision than I can reasonably obtain, given the size of the management units or available resources, then I should reconsider using these metrics and explore other monitoring metrics that will have the needed accuracy and precision. Another alternative is to re-define my set of competing models.

6. Habitat Associations Explore habitat or other factors that may contribute to changes in abundance, density, or the probability that a site is occupied by a species. This objective requires defining a priori a set of competing models to explain expected changes in bird abundance or occupancy and then collecting data on the environmental variables defined in the models. Expected change can be anticipated based on factors including stressors, plant succession, management actions, or climate change.

Management objective: Identify habitat or other factors associated with high vs. low abundance/density/occupancy of species X for the time period in the area of interest.

Sampling objective: Develop an appropriate and limited set of competing models and collect the associated habitat variables along with the bird data. Identify the best-fitting models that explain variation in abundance, density, or occupancy. This requires careful planning so that the habitat data are designed to meet the objectives; consultation with a statistician during the design phase is recommended.

Revisiting Sampling Points

One visit to a sampling point within a breeding season is the minimum requirement for estimating change in abundance, density, or species richness of birds in this protocol. Use one visit if you want to maximize the total number of survey points visited, if access is difficult or time-consuming, or if you cannot obtain sufficient resources for multiple visits.

Raw counts of individual birds and estimates of species richness need to be adjusted for detection probability to obtain accurate bird density estimates (Nichols and Conroy 1996, Pollock et al. 2002). The data collected using this SOP meet minimal requirements for analysis using time-depletion (removal) methods (Farnsworth et al. 2005), which, along with measuring the distance of birds from the observer, increases the accuracy of our density estimates.

If occupancy is the metric of interest, multiple visits to a point within a single breeding season are required; counts of individuals are not required (Bailey and Adams 2005, MacKenzie et al. 2006). A species need only be recorded as present or absent at a point. Keep in mind that count data can always be collapsed to presence/absence, but the inverse is not true. We suggest recording the count of individual birds even if occupancy is the desired metric. Also, recent advances in occupancy methods allow use of abundance indices (Royle 2004).

Use three or more visits if you have a small area to survey and want to increase precision, access is simple, or if you plan to use occupancy to model habitat associations. Three visits are estimated to be the minimum number needed to estimate occupancy when detection probability is >0.5; more visits are needed when detection probability is low (MacKenzie and Royle 2005, MacKenzie et al. 2006, Bailey et al. 2007).

Panel designs that allow some points to be revisited annually and other points to be revisited in rotations of >1 year apart may be advantageous for long-term monitoring because such sampling distributes the “response burden” among sites, and because split panel designs generally have higher power for estimates of both trend and status (McDonald 2003). Statistical consultation is advised before implementing a panel design. Design #4 (below) uses a panel design to define the timing of revisits.

Spatial Considerations

Ideally, sampling design employs digital (GIS) maps of the study area, using the best available land cover maps. In virtually all habitats, >99% of individuals are detected within 125 m of the observer (Ralph et al. 1993). Therefore, the minimum distance between points in wooded habitats is 250 m. Greater distances should be used in open environments, especially along roads where vehicle travel is possible; in those locations, recommended distances between points is >500 m (Ralph et al. 1993). If your objective is to count birds within a specific habitat type, avoid placing survey points at the edge (within 50 m) of the target habitat. However, park-wide surveys and surveys of species that are known to occupy edges should include edge habitats.

Sample Size

The sample size depends upon the management objectives, the desired minimum detectable trend, and the time frame. That is, how much change is meaningful from a management perspective and over what time period? Consult a statistician to answer these questions. The following guidelines illustrate the complexity of this issue and provide only general guidance.

For count data, power is more sensitive to sampling period (number of years) and among-year variability than to within-year variance (Gray and Burlew 2007). Increasing the number of visits within a year will increase power (Table 1). Over short time frames, only catastrophic trends (>25% change/yr) are likely to be detected, even for common species (Thogmartin et al. 2007). There is no requirement to monitor every year when estimating trends, but annual monitoring will help define inter-year variance, and in general, larger sample sizes will be needed to estimate trends in occupancy compared with trends in abundance (Knutson and Gray 2008).

Monitoring rare species or those difficult to detect may require alternative methods, but the required sample sizes may exceed what is feasible to complete during a field season. The following minimum sample sizes for relatively common species illustrate the trade-offs between sample size, sampling period, and other factors (Table 1). Additional factors such as over- dispersion and low abundance will increase the necessary sample size. Table 1 provides only general guidance regarding the minimum number of sampling points needed. Pilot data will be needed to make more precise estimates of sample size.

Alpha Minimum Sample Required risk detectable trend Time Type Size visits/yr (α) Power (% change/yr)1 (yr)

Inventory Maximize Count data 30 1 0.10 80% 5% 20 (Gray and Burlew 2007) 10 1 0.10 80% 5% 25 30 1 0.10 80% 3% 30 Relative abundance 80 1 0.10 90% 3% 10 (Purcell et al. 2005) 20 1 0.10 90% 3% 20 30 3 0.10 90% 3% 10 10 3 0.10 90% 3% 20 Occupancy 500 - 0.05 80% 5% change in 10 (Brian Gray, unpub. odds of species report) presence 100 - 0.05 80% 5% 15 1A change of 5%/yr means that the species will be extirpated in 20 years; if you assume a non-linear Poisson distribution for the count data, a change of 3%/yr corresponds to a 50% decrease in 20 years (Purcell et al. 2005).

Sampling Designs

This protocol focuses on sampling at the local spatial scale (i.e., park or management unit). Examples of some sampling designs are described below. Statistical consultation is strongly recommended when planning a new monitoring effort. Sampling designs often need to be tailored to meet specific objectives, especially when evaluation of management actions is an objective. No set of omnibus sampling designs will serve all purposes.

For all but the largest management units (>10,000 ha) the protocol relies on random or stratified random sampling. For very large land units, Generalized Random Tessellation Stratified Design (GRTS) and its variations have the advantage of ensuring the spread of samples across the target area (spatially balanced) and simplifying replacement of unusable sample points (Stevens and Olsen 2004, Theobald et al. 2007, Lister and Scott 2008). GRTS is preferred for multi-stage sampling designs across ecoregions, regions, or other large areas.

Design #1: Random design, for small to large land units

Design 1. Impose a randomly-placed 300 m grid over land unit (entire park or management area).

2. Select all grid cells with >60% of the target landbird habitat (define land cover classes to be included).

3. Randomly select 50-100 of these points to sample (more is better). If the target is 50 cells, select 50 primary and 20 replacement cells to sample (over-sampling).

4. Locate the centroid of these cells and generate a set of points for mapping and to upload into a hand-held GPS unit.

5. Overlay roads, trails, and other elements that allow access to target land cover class (define for each park). This information is used to determine access points.

6. Navigate to the primary sampling cells (points) (e.g., cells 1 to 50); if a cell cannot be reached for logistical reasons, replace it with the next cell on the list (e.g., cell 51); repeat as necessary until the full complement of cells are sampled.

7. If edge habitat is not part of the target habitat, you can remove points that include edges. Depending upon your objectives, you may want to keep edge habitats in your sample.

To remove edges: If a sampling point does not contain at least a 50 m radius circle of the target land cover class surrounding the point, the point is sampled and is labeled “edge” on the data

sheet. A replacement point is then used for the next point (see above) to acquire data for at least 50 points that are not “edges.” Theoretically, edge points could be screened out using the digital map by employing a rule that the entire 300 m cell must fall within the land cover class boundary; however, many maps are not accurate enough to do this without field reconnaissance. It will save survey time if all the points can be ground-truthed and the edge points eliminated, if necessary, before the field season, especially if the points are permanent and will be sampled regularly.

The sampling frame is the set of 300 m grid cells containing >60% of the target land cover class and this area can be explicitly mapped.

Design #2: Stratified random design, stratified by access, for small to large land units

Objectives This sampling design addresses all six potential objectives discussed above (see Objectives), and is suitable for small to large land units (<5,000 ha) when large areas of the target sampling frame are very difficult to access and linear features (roads, trails) characterize all accessible areas. Stratification is by access. If parts of the target sampling frame are permanently inaccessible for safety or other reasons (unexploded ordnance, technical climbing required, frequent illegal activity), remove those cells from the sampling frame entirely. Use this design (#2) if roads or trails characterize the accessible areas. Use Design #1, with inaccessible cells removed, if access is not limited to roads or trails. Caution: Once you have defined the strata, they are fixed and cannot be changed without introducing bias.

Design 1. Impose a randomly-placed 300 m grid over entire park or management area.

2. Select grid cells with >60% of the target landbird land cover classes.

3. Overlay roads, trails, and other elements that allow access to target land cover classes (define for each park).

4. Classify 300 m cells as accessible (Class 1) if they intersect any of the above features or difficult to access (Class 2) if they do not.

5. Select Class 1 sampling units: Extract a line coverage that represents all accessible paths (roads, trails); employ a sampling program that defines a set of points along the line coverage, with all points at least 300 m apart in all directions. Randomly select 50-100 of these points to sample. If all sampling points are on roads, distances between sampling points should be increased to 1 km (0.5 mi.) to cover more area (Ralph et al. 1995).

6. Select Class 2 sampling units: Buffer the accessible line coverage 300 m, select all Class 2 cells that do not intersect the buffer. Locate the centroid of these cells and randomly select 10 primary and 5 replacement Class 2 cells to sample. These are numbered 1-15.

7. Navigate to the primary (#1-10) Class 2 sampling cells; if a primary cell is inaccessible for logistical reasons, replace it with #11, and repeat as necessary until ten Class 2 cells are sampled.

To remove edges: If a sampling point does not contain at least a 50 m radius circle of the target land cover class surrounding the point, the point is sampled and is labeled “edge” on the data sheet. A replacement point is then used for the next point (see above) to acquire data for at least 50 points that are not “edges.” Theoretically, edge points could be screened out using the digital map by employing a rule that the entire 300 m cell must fall within the land cover class boundary; however, many maps are not accurate enough to do this without field reconnaissance. It will save survey time if all the points can be ground-truthed and the edge points eliminated, if necessary, before the field season.

The sampling frame is the set of Class 1 and Class 2 grid cells containing >60% of the target landbird land cover class and this area can be explicitly mapped.

A statistician can use the Class 1 and Class 2 designations to check for bias in the Class 1 data set and to allow extrapolation to the sampling frame. Thus, the sampling frame at the local scale is the set of Class 1 and Class 2 cells across the park or management area and this area can be explicitly mapped.

Design #3: Stratified random design, stratified by habitat type, for small to large land units

Objectives This sampling design addresses all six potential objectives discussed above (see Objectives), and is suitable for small to large land units (<5,000 ha) when most of the target sampling frame is accessible. In this design, the objective is to stratify by a factor relevant to your management objective, such as habitat type (land cover class) when you are concerned that a simple random sample might miss or under-sample a sub-set of that factor (e.g., rare habitat type). Sometimes other factors, such as land ownership, may be of stronger interest than habitat type. With a simple random sample, it is likely that habitats that are small in area relative to other habitats may, by chance, have too few samples or none at all. Stratification solves this problem.

If the objective is to obtain habitat-specific estimates of abundance or occupancy, then equal sample sizes in each target habitat are recommended, or you can design the number of samples to meet your sampling objectives (confidence intervals). For example, if you have a stronger interest in public land versus private land, then you would sample public land more intensively than private land.

Design 1. Impose a randomly-placed 300 m grid over a map of the park or management area.

2. Select grid cells with >60% cover of the target landbird habitat (define land cover classes).

3. Classify 300 m cells by the dominant target land cover classes (Habitat 1, Habitat 2, etc.).

4. Select replacement cells if primary cells cannot be accessed or if they occur on a habitat edge. If 50 cells is the target, select 50 primary and 20 replacement cells to sample in Habitat 1. Do the same for the other target land cover classes. You can choose to allocate the samples proportional to the size of the land cover class or you could sample equal numbers of points in each land cover class.

5. Independently, randomly select some number of points (more is better) in each land cover class. Locate the centroid of the selected cells and generate a set of points for mapping and to upload into hand-held GPS unit.

6. Overlay roads, trails, and other elements that allow access to target land cover class (define for each park). This information is used to determine the best access points.

7. Navigate to the primary (#1-50) Habitat 1 sampling cells. If a primary cell cannot be reached for logistical reasons, or if the habitat at the point does not match what is indicated on map, use a replacement cell (#51) within the same habitat. Repeat as necessary until 50 cells are sampled.

8. Repeat steps 5-7 for remaining land cover class until all the target land cover classes are sampled.

To remove edges: If a sampling point does not contain at least a 50 m radius circle of the target land cover class surrounding the point, the point is sampled and is labeled “edge” on the data sheet. A replacement point is then used for the next point (see above) to acquire data for at least 50 points that are not “edges.” Theoretically, edge points could be screened out using the digital map by employing a rule that the entire 300 m cell must fall within the land cover class boundary. However, many maps are not accurate enough to do this without field reconnaissance. It will save survey time if all the points can be ground-truthed and the edge points eliminated, if necessary, before the field season.

The sampling frame is the sum total of all Habitat cells (Habitat 1 + Habitat 2 +…) across the park or management area and this area can be explicitly mapped.

Design #4: Multi-stage probabilistic design, stratified by accessibility, for large land units

Objectives This sampling design addresses Objectives 2, 3, and 6 above, and is limited to very large land units (>10,000 ha), where large areas of the target sampling frame are very difficult to access, and the target population is the entire park.

Design 1. This sampling design is adapted from Roland et al. (2003) and was used in Denali National Park (McIntyre et al. 2004).

2. Impose a randomly-placed grid (macro-grid) over the entire park or management area (Stage 1). (The size of the macro-grid can be scaled to the size of the overall management unit; many units may be well served by a grid with cells measuring 10 km on a side.)

3. In the southeast corner of each macro-grid, generate a mini-grid consisting of a 5x5 lattice of twenty-five 300-m cells (Stage 2; Figure 1). The mini-grids are your target sampling units.

Figure 1. Diagram of macro-grid with mini-grids marked in gray. Each mini-grid contains twenty-five 300 m cells.

4. On each mini-grid, overlay roads, trails, and other elements that allow access to sample points (define for each park).

5. Extract a line coverage that represents all accessible paths (roads, trails); buffer the accessible line coverage 300 m.

6. Classify mini-grids as accessible (Class 1) if they intersect the buffer, or remote (Class 2) if they do not intersect the buffer.

7. Use software to generate a spatially balanced sample of 50 primary Class 1 mini-grids and 10 secondary mini-grids to sample (Stevens and Olsen 2004, Theobald et al. 2007). These are numbered 1-50 (primary) and 51-61 (secondary).

8. Locate the centroid of each cell in the primary mini-grids (25 points) and conduct the point counts. If one or more points in the mini-grid are inaccessible, mark that on the data sheet and conduct as many as possible. If more than 5 points within a mini-grid are inaccessible, replace it with a secondary mini-grid (# 51). Continue until you have sampled fifty Class 1 mini-grids, each with 20 or more accessible points.

9. Randomly select 10 primary and 5 replacement Class 2 mini-grids. These are numbered 1-15.

10. Navigate to the primary (#1-10) Class 2 mini-grids; if primary mini-grid cannot be accessed for logistical or safety reasons, replace it with #11. Repeat as necessary until ten Class 2 mini-grids, each with 20 or more accessible points are sampled.

If possible, it is best to ground-truth the sampling points before the field season to finalize the set of Class 1 and Class 2 mini-grids.

Long-term Monitoring: For long-term monitoring (>10 years), a panel design should be used (Table 2).

The sampling frame is the full set of Class 1 and Class 2 mini-grids, and these can be explicitly mapped.

The sampled set contains fifty Class 1 and ten Class 2 mini-grids. Each panel contains two Class 1 mini-grids. Panel 1 is surveyed every year, Panels 2-7 are surveyed in two consecutive years every four years, and Panels 8- 20 are surveyed once every 12 years. The ten Class 2 mini-grids are sampled as in Panel 8 (the first year and every 12 years thereafter).

A statistician can use the Class 1 and Class 2 designations to check for bias in the Class 1 data set and to allow extrapolation of sampling results to the sampling frame. Thus, the sampling frame at the local scale is the set of Class 1 and Class 2 cells across the park or management area and this area can be explicitly mapped. Panel designs will require statistical consultation during the planning and analysis phases of the survey.

Table 2. Example of a sampling rotation for landbird monitoring using a multi-stage design over a 20-year period (from McIntyre et al. 2004).

Sampling Occasion (year) Panel 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 1 X X X X X X X X X X X X X X X X X X X X 2 X X X X X X X X 3 X X X X X X X 4 X X X X X X 5 X X X X X X 6 X X X X X X 7 X X X X X X X 8 X X 9 X X 10 X X 11 X X 12 X X 13 X X 14 X X 15 X 16 X 17 X 18 X 19 X 20 X

Literature Cited

Adamcik, R. S., E. S. Ballantoni, D. C. DeLong, Jr., J. H. Schomaker, D. B. Hamilton, M. K. Laubhan, and R. L. Schroeder. 2004. Writing refuge management goals and objectives: A handbook. U.S. Fish and Wildlife Service, Arlington, VA.

Arkema, K. K., S. C. Abramson, and B. M. Dewsbury. 2006. Marine ecosystem-based management: From characterization to implementation. Frontiers in Ecology and the Environment 4:525-532.

Bailey, L. L., and M. Adams. 2005. Occupancy models to study wildlife. U.S. Geological Survey Fact Sheet 2005-3096. U.S. Geological Survey, Laurel, MD.

Bailey, L. L., J. E. Hines, J. D. Nichols, and D. I. MacKenzie. 2007. Sampling design trade-offs in occupancy studies with imperfect detection: Examples and software. Ecological Applications 17:281-290.

Butler, K. F., and T. M. Koontz. 2005. Theory into practice: Implementing ecosystem management objectives in the USDA Forest Service. Environmental Management 35:138-150.

Cochran, W. G. 1977. Sampling Techniques, 3rd ed. John Wiley & Sons, New York, NY.

Elzinga, C. L., D. W. Salzer, J. W. Willoughby, and J. P. Gibbs. 2001. Monitoring Plant and Animal Populations. Blackwell Science, Malden, MA.

Farnsworth, G. L., J. D. Nichols, J. R. Sauer, S. G. Fancy, K. H. Pollock, S. A. Shriner, and T. R. Simons. 2005. Statistical approaches to the analysis of point count data: A little extra information can go a long way. Pages 736-743 in C. J. Ralph and T. D. Rich, editors. Bird conservation implementation and integration in the Americas: Proceedings of the Third International Partners in Flight Conference, Volume 2. Gen. Tech Rep. PSW-GTR- 191. USDA Forest Service, Albany, California.

Gray, B. R., and M. M. Burlew. 2007. Estimating trend precision and power to detect trends across grouped count data. Ecology 88:2364-2372.

Hill, D., M. Fasham, G. Tucker, M. C. Shewry, and P. Shaw, editors. 2005. Handbook of Biodiversity Methods: Survey, Evaluation, and Monitoring. Cambridge University Press, Cambridge, UK.

Johnson, D. H. 2000. Statistical considerations in monitoring birds over large areas. Pages 115- 120 in R. Bonney, D. N. Pashley, R. J. Cooper, and L. J. Niles, editors. Strategies of bird conservation: The Partners in Flight planning process. Proceedings of the third Partners in Flight Workshop. Proceedings RMRS-P-16. USDA Forest Service, Ogden, Utah.

Knutson, M., and B. Gray. 2008. Standard operating procedure (SOP) #1: Sampling designs. In M. G. Knutson, N. P. Danz, T. W. Sutherland, and B. R. Gray. 2008. Landbird Monitoring Protocol for the U.S. Fish and Wildlife Service, Midwest and Northeast Regions, Version 1. Biological Monitoring Team Technical Report BMT-2008-01. U.S. Fish and Wildlife Service, La Crosse, Wisconsin.

Lister, A. J., and C. T. Scott. 2008. Use of space-filling curves to select sample locations in natural resource monitoring studies. Environmental Monitoring and Assessment (Online). http://nrs.fs.fed.us/pubs/jrnl/2008/nrs_2008_lister_001.pdf.

MacKenzie, D. I., J. D. Nichols, J. A. Royle, K. H. Pollock, L. L. Bailey, and J. E. Hines. 2006. Occupancy estimation and modeling: Inferring patterns and dynamics of species occurrence. Academic Press (Elsevier).

MacKenzie, D. I., and J. A. Royle. 2005. Methodological insights: Designing occupancy studies – general advice and allocating survey effort. Journal of Applied Ecology 42:1105-1114.

Mahan, C. G., D. R. Diefenbach, and W. B. Cass. 2007. Evaluating and revising a long-term monitoring program for vascular plants: Lessons from Shenandoah National Park. Natural Areas Journal 27:16-24.

McDonald, T. L. 2003. Review of environmental monitoring methods: Survey designs. Environmental Monitoring and Assessment 85:277-92.

McIntyre, C. L., R. Drum, K. L. Oakley, E. Debevec, T. L. McDonald, and N. Guldager. 2004. Passerine bird monitoring protocol for the Central Alaska Monitoring Network: Denali National Park and Preserve, Wrangel-St. Elias National Park and Preserve, and Yukon- Charley Rivers National Preserve, Alaska. U.S. Department of the Interior, National Park Service, Denali Park, AK.

Nichols, J. D., and M. J. Conroy. 1996. Estimation of species richness. Pages 226-234 in D. Wilson, R. Cole, J. Nichols, R. Rudran, and M. Foster, editors. Measuring and monitoring biodiversity: Standard methods for mammals. Smithsonian Institution Press, Washington, D.C.

Nichols, J. D., and B. K. Williams. 2006. Monitoring for conservation. Trends in Ecology and Evolution 21:668-673.

Pollock, K. H., J. D. Nichols, T. R. Simons, G. L. Farnsworth, L. L. Bailey, and J. R. Sauer. 2002. Large scale wildlife monitoring studies: Statistical methods for design and analysis. Environmetrics 13:105-119.

Purcell, K. L., S. R. Mori, and M. K. Chase. 2005. Design considerations for examining trends in avian abundance using point counts: Examples from oak woodlands. Condor 107:305- 320.

Ralph, C.J., G. R. Geupel, P. Pyle, T. E. Martin, and D. F. DeSante. 1993. Handbook of field methods for monitoring landbirds. General Technical Report PSW-GTR-144. USDA Forest Service, Albany, CA.

Ralph, C. J., S. Droege, and J. R. Sauer. 1995. Managing and monitoring birds using point counts: Standards and applications. Pages 161-175 in C. J. Ralph, J. R. Sauer, and S. Droege, editors. Monitoring bird populations by point counts. General Technical Report PSW-GTR-149. USDA Forest Service, Albany, California.

Roland, C., K. Oakley, and C. McIntyre. 2003. Long-term ecological monitoring program: Evaluation of a study design for detecting ecological change in Denali National Park and Preserve at Multiple Scales. Volumes I and II (Internal draft). Denali National Park and Preserve, Denali Park, Alaska.

Royle, J. A. 2004. Modeling abundance index data from anuran calling surveys. Conservation Biology 18:1378-1385.

Schreiber, E. G., A. R. Bearlin, S. J. Nicol, and C. R. Todd. 2004. Review article: Adaptive management: A synthesis of current understanding and effective application. Ecological Management and Restoration 5:177-182.

Schroeder, R. L. 2006. A system to evaluate the scientific quality of biological and restoration objectives using National Wildlife Refuge Comprehensive Conservation Plans as a case study. Journal for Nature Conservation 14:200-206.

Stevens, D. L., and A. R. Olsen. 2004. Spatially balanced sampling of natural resources in the presence of frame imperfections. Journal of the American Statistical Association 99:262- 278.

Theobald, D. M., D. L. Stevens, D. White, N. S. Urquhart, A. R. Olsen, and J. B. Norman. 2007. Using GIS to generate spatially balanced random survey designs for natural resource applications. Environmental Management 40:134-146.

Thogmartin, W. E., B. R. Gray, M. Gallagher, N. Young, J. J. Rohweder, and M. G. Knutson. 2007. Power to detect trend in short-term time series of bird abundance. Condor 109:943–948.

Thompson, S. K. 2002. Sampling, 2nd edition. John Wiley and Sons, Inc., New York, NY.

U.S. Department of the Interior. 2006. DOI adaptive management guidebook (draft). Office of the Assistant Secretary, Policy, Management and Budget, Washington, D.C.

Verner, J. 1988. Optimizing the duration of point counts for monitoring trends in bird populations. Research Note PSW-395. U.S. Forest Service, Berkeley, CA.

Standard Operating Procedure #2: Before the Field Season

Please cite this document as:

Knutson, M., T. Sutherland, B. Route, and T. Gostomski. 2010. Standard operating procedure #2: Before the field season. In Gostomski, T., M. Knutson, N. P. Danz, B. Route, and T. W. Sutherland. 2010. Landbird monitoring protocol, Great Lakes Inventory and Monitoring Network. Natural Resource Report NPS/GLKN/NRR—2010/225. National Park Service, Fort Collins, Colorado.

Contents Page

Tables ...... v

Introduction ...... 1

Procedures ...... 1 General Preparation and Review ...... 1 Scheduling Field Work ...... 2 Organizing Supplies and Equipment ...... 2

iii

Tables Page

Table 1. List of field equipment for point counts...... 3

Introduction

At least one week prior to beginning the surveys, the park’s monitoring coordinator (Park Coordinator, or PC) should review this entire protocol, including all the SOPs, with the staff/volunteers who will conduct the point counts (observers and recorders). All observers and recorders should receive a copy of this protocol during pre-survey training. Several weeks may be needed for crew leaders to prepare for the training session and organize field gear, especially in the first year of a new monitoring effort.

The PC will pay particular attention to training observers for this project (see SOP #3). Bird identification and distance estimation is extremely important; accurate identification of birds by sight and sound is essential. The training sessions will prepare observers to accurately estimate distances to all types of bird detections.

Procedures

General Preparation and Review 1. Review the protocol narrative and all SOPs. The PC should understand the goals and procedures of the monitoring program and begin preparations for the training period and field season.

2. Ground-truth point locations if time allows. Replace missing point markers (if used). Are all the sites still accessible? Have any habitats been changed by some disturbance?

3. Discuss the season’s plans with supervisor. Travel logistics, meeting times and places, work schedules, and camping requirements are a few of the topics to work out ahead of time.

4. Review field notes and trip reports from past sampling. Identify any unique species or conditions (e.g., hazardous routes, missing markers) that may be encountered.

5. Print and review the species list in Appendix A of SOP #5. Review the species most likely to be encountered. Crew members may want to carry copies of the species list in the field to use as a quick reference for identification.

6. Generate a list of point coordinates for all points to be surveyed (see SOP #1 and SOP #4). Include GIS-generated points for new points and a list of the actual marker coordinates for previously sampled points. Be sure instructions for navigating to the points also include information such as compass bearing, approximate distance, approximate travel time, and landmarks to watch for. Assume that the GPS unit will break down or not work under the forest canopy, and plan accordingly!

7. Print and laminate maps of point locations for each crew member.

8. Print and copy the field data forms. Data forms are in SOP #5, Appendix B (double- sided). Print these onto Rite-in-the-Rain™ paper if available. We suggest creating data sheets with the sampling points and GPS coordinates pre-printed to prevent mistakes in recording GPS numbers in the field.

9. Upload waypoints onto the GPS units. Waypoints (the latitude and longitude coordinates for each survey point) may be uploaded via computer into all GPS units before the start of the field season (see SOP #4).

Scheduling Field Work 1. Establish the survey schedule. Breeding bird surveys will focus on the time period that coincides with the peak breeding activity of most landbird species in the study area. The PC will establish the survey time period.

2. Sampling schedules often change due to weather events. It is wise to plan for weather days throughout the month and always have a backup plan. Under good conditions, and depending on logistical considerations, observers can complete 8 to 12 points per day.

Organizing Supplies and Equipment Compile an equipment inventory before the field season.

Organize and prepare all field equipment several weeks before beginning field work. Table 1 summarizes the list of field equipment. All equipment should be examined for functionality and completeness before field work begins.

It is useful to attach brightly-colored flagging to items like thermometers, binoculars, GPS units, and any other equipment that may accidentally be left behind or dropped between points.

Bird field guides. Though birds are most often heard during point counts, observers may find it helpful to have a field guide and/or a CD/mp3 player of bird songs with them for reference after the survey period.

Data Forms – see SOP #5.

Table 1. List of field equipment for point counts.

Binoculars Bird Field Guides Clipboards (recommended: aluminum with storage area) Compasses Digital camera (with extra batteries and storage card) Digital timer with clip (some GPS units have a built-in stopwatch) Extra batteries (for radios, GPS, camera) Field Data Sheets (printed on Rite-in-the-Rain© paper) First Aid Kit Flagging (for marking points, gear that might be lost, etc.) GPS unit for navigation Pencils/pens Radio and/or cell phone for field-to-base and field-to-field communications Thermometers (Celsius scale) Topographic or other maps

Standard Operating Procedure #3: Hiring and Training Observers

Please cite this document as:

Knutson, M., B. Route, and T. Gostomski. 2010. Standard operating procedure #3: Hiring and training observers. In Gostomski, T., M. Knutson, N. P. Danz, B. Route, and T. W. Sutherland. 2010. Landbird monitoring protocol, Great Lakes Inventory and Monitoring Network. Natural Resource Report NPS/GLKN/NRR—2010/225. National Park Service, Fort Collins, Colorado.

Contents Page

Introduction ...... 1

Recruiting Observers ...... 1 Hearing Acuity ...... 1 Certification of Bird Observer Skills ...... 1 Issue Field Gear Before Training Begins ...... 1

Field Training...... 3 Learning to Use the Equipment ...... 3 Review Field Sampling Methods ...... 3 Recording Time of Detection ...... 3 Distance Estimation ...... 4 Field Testing ...... 4 Training for Vegetation Monitoring ...... 4 Invasive Species and Decontamination Procedures ...... 4 Procedure at Remote Locations Known or Suspected to be Infested with Exotic Species ...... 5 Procedure at Non-remote Locations, When Tap Water is Available ...... 5 Regardless of Remoteness ...... 5

Literature Cited ...... 7

iii

Introduction

This Standard Operating Procedure (SOP) explains recruitment of and training requirements for landbird monitoring participants. This SOP is adapted from a protocol developed for the National Park Service, Northeast Temperate Network (Faccio and Mitchell 2007).

Recruiting Observers

The park’s monitoring coordinator (Park Coordinator, or PC) is responsible for recruiting observers either by hiring qualified individuals or attracting experienced volunteers. Observers can be recruited from a variety of local sources, including bird clubs, Audubon Chapters, birding listserves, and state Breeding Bird Survey (BBS) coordinators. Depending upon the location of the survey area, the University of Wisconsin-Green Bay Birder Certification Program may have a list of certified bird observers (see below).

Various studies have shown that observer bias is one of the most noteworthy factors affecting the accuracy of trend estimates for songbird populations (Kepler and Scott 1981, Baker and Sauer 1995). Therefore, the most essential component for the collection of credible, high-quality bird data is well-trained and experienced observers. Proficient bird observers obtain species estimates within 90% of total species known to be present and estimate abundance with 80% accuracy (Ralph et al. 1993).

Hearing Acuity Hearing is fundamental to bird identification because a large proportion of birds detected during surveys are heard and not seen. Observers must possess excellent visual and auditory bird identification skills and should be capable of identifying by song 90% of the bird species likely to be encountered. Several months prior to the field season, observers should review and practice bird identification skills. Though not required, it is reasonable to ask potential observers to have a hearing test; hearing loss that exceeds 20 dB will compromise the ability to effectively survey landbirds (Kepler and Scott 1981).

Certification of Bird Observer Skills Various software programs are available to help observers learn the birds by sight and sound. Observers who conduct point counts in Great Lakes Network parks will be required to complete an online certification program developed by the University of Wisconsin-Green Bay (http://www.uwgb.edu/birds/certification/index-1.htm ) for the western Great Lakes region (Bird Conservation Regions [BCR] 12 and 23). The website allows observers to certify their skills in identifying birds by sight and sound, and allows parks to verify that certification. Bird observers should evaluate and improve their birding skills well before the field season begins. The PC should keep records of each observer’s Birder Certification Number, which is assigned by the certification website.

Issue Field Gear Before Training Begins See SOP #2 for a list of field gear. PCs may wish to provide all trainees with a copy of the protocol narrative and SOP #5, along with field gear and other training materials, before training begins. Ensure that crew members examine the list of necessary personal gear and that they have

all the appropriate equipment by the time field work starts. Ensure that trainees are informed and prepared for the training and survey events; everyone should bring binoculars, raingear, appropriate footgear and clothing, field guides, and other equipment appropriate for each day of training.

Field Training

If a park requires observers to have safety, first aid, computer security, or other training, the PC should plan on a minimum of two weeks of training prior to beginning surveys. Review safety procedures, including first aid, what to do in case of an accident or injury, and appropriate use of park radios.

The following suggestions are for parks wishing to conduct a brief “on-the-job training” in survey procedures. Start training days early and conduct the majority of the training in habitats and topography similar to those encountered during surveys. Problems enumerating individual birds during the dawn chorus can lead to erroneous survey results. Thus, adjust the training time to best prepare observers, incorporating the peak singing time for most species. While emphasizing distance estimation and how to use the laser rangefinder, point out as many birds as possible with the initial objective of maximizing the trainee’s exposure to the species most likely to be encountered during the surveys. Discuss personal techniques for accurate data recording, dealing with busy points, flyovers, moving birds, and potentially confusing scenarios.

Observers from previous seasons normally do not need as much annual training in bird identification skills as new observers. However, it is recommended that, when possible, all observers participate in annual training or refresher training. Experienced observers can assist as additional teachers with less experienced trainees, giving them an opportunity to continue practicing distance estimation, working on identification of birds by call notes and partial songs, and generally improving their identification skills.

Learning to Use the Equipment The PC will ensure that observers understand how to use all necessary equipment while they are in the field, including GPS navigation, laser rangefinder, and emergency communication equipment (e.g., park radios or cell phones). Observers are encouraged to practice using navigation equipment and will also be encouraged to locate their point count sites the day before the official counts are conducted. This will allow them to become familiar with survey locations and save time when they are conducting their counts.

Review Field Sampling Methods The PC should provide each observer with a list of bird species likely to be encountered at the park. Appendix A in SOP #5 can be used to create a park-specific list. In addition, accurate data recording will require practice during the training period so PCs should train observers in data recording methods while in the field.

Recording Time of Detection We will be primarily using the time removal method (Farnsworth et al. 2005) to estimate detection probabilities. This method assumes that a bird has an equal probability of first being detected within any 1-minute time interval. To avoid biases, observers must be trained to record individual birds the first time they are detected and attribute them to the minute indicated by the digital timer. In practice, it may be difficult to record all the birds detected within the first minute before the timer rolls over to the second minute, but every effort should be made to do so.

Distance Estimation Distance estimation is a secondary method of estimating detection probability in this protocol. Therefore, we need to record birds as accurately as possible within our distance bands. The majority of birds are usually heard but not seen, and estimating distances to birds that are only heard is often the greatest source of error in point counts (Simons et al. 2007). Practicing distance estimation before the field season may be helpful in making observers comfortable with the measure. Two ways of practicing distance estimation are provided below. For a detailed discussion, refer to Kepler and Scott (1981).

1. Use a range-finder to practice estimating distances in each general habitat type that will be included in the survey. Placing flagging at four or five locations visible from a central point and ask observers to estimate the distance band (e.g., 0-50m, 51-100m, >100m) that each flag falls within. Then use a range-finder to evaluate the accuracy of each estimate. Repeat this exercise several times until observers can consistently estimate distances.

2. Standing at the central point, listen for vocalizing birds. Choose one consistently vocalizing individual and estimate the distance band in which it is singing. Try to visually identify the tree or branch where you think the bird is, and use a range-finder to estimate the distance to an object near where you think the bird is vocalizing. Mark your location. Walk toward the vocalizing bird until you can either see it or accurately estimate its location. Compare your initial estimate to the actual distance between your survey location and the bird. Repeat this exercise for several birds at various distances.

Field Testing New observers should be tested before conducting point counts. New observers should accompany a qualified observer into the field and simultaneously recording the birds detected. This does not have to be done at point count sites. In addition, an examination may be given using a recording of bird vocalizations or bird vocalization software (for current resources, see http://www.partnersinflight.org/education.cfm ). Even observers certified by the Birder Certification Program should have their skills evaluated in the field.

Training for Vegetation Monitoring Review the vegetation monitoring protocol (SOP #6) with the observers. Verify that they can correctly record vegetation monitoring information. If you are using a detailed vegetation monitoring protocol, reserve more time for training. PCs should: 1. Prepare field data sheets, 2. Review procedures for collecting digital photos (optional), and 3. Decide when vegetation monitoring will be done (before, during, or after the day when bird observations are made).

Invasive Species and Decontamination Procedures Though the likelihood of doing so is low, it is very important for observers to avoid transporting non-native invasive species (both terrestrial and aquatic) to uncontaminated areas. The PC should discuss with the observers any potential for transporting non-native invasive species (especially garlic mustard, Alliaria petiolata; spiny water flea, Bythotrephes cederstroemi; and zebra or quagga mussels, Dreissena spp.) and review procedures for decontaminating equipment in the event that an infested area is visited for purposes of conducting point counts. A suggested

decontamination procedure follows, but parks may use any existing procedures they currently have established.

In most cases, observers will only have to inspect their clothing, boots, and possibly vehicles to ensure no plant materials are being transported. However, in those parks where boats or canoes are used in the course of landbird monitoring, it may be necessary to wash the boat(s)/canoe(s) after completing a survey. If your boat/canoe could potentially transport aquatic invasive species (AIS) to uninfested lakes, please do the following (from Elias 2008):

Procedure at Remote Locations Known or Suspected to be Infested with Exotic Species Unless observers will be near an uninfested lake or river from which they can collect water to use in decontaminating a boat or canoe, they should take a large container of clean tap water with them into the field on the day of the survey.

1. Upon return to shore, inspect boat, trailer, and all equipment that has come into contact with the water for invasive species. 2. Move the boat/canoe to a vegetated area away from the lake or river and take out the container of clean tap water you brought with you, or use a clean and dry container to collect lake/river water from an uninfested source. 3. Remove and rinse mud, plant material, mussels, and other visible organic material from boat/canoe, oars/engine/paddles, boots, and from all sampling equipment. 4. Scrub all equipment with a (clean) scrub-brush, paying particular attention to ropes, cracks, and crevices in equipment. Rinse well. Avoid rinsing on impervious surfaces or slopes where rinse water might run directly into the lake. 5. Run hand up and down ropes attached to all equipment and anchor to dislodge mud, plant material, and organisms. 6. Visually inspect all equipment and gear to verify it has been cleaned and rinsed.

Procedure at Non-remote Locations, When Tap Water is Available 1. On a vegetated area away from the lake, use a hose with a spray nozzle and a scrub brush to rinse and scrub with tap water all sampling equipment, boat/canoe, oars/engine/ paddles, and boots. Pay particular attention to ropes, cracks, and crevices in equipment. Rinse well. 2. Discard water on land, away from lakeshore, avoiding impervious surfaces and slopes where rinse water might run directly into the lake. 3. Visually inspect all gear to verify that all equipment and gear has been cleaned and rinsed.

Regardless of Remoteness 1. When possible, avoid using boats and other equipment on both infested and uninfested waters. 2. Designate a set of equipment for both types of waters. 3. Always begin sampling with lakes/rivers not known to harbor exotic species; sample lakes/rivers known or suspected to contain exotic species last. 4. Thoroughly clean or disinfect all equipment between use in infested and uninfested waters.

5. When possible, allow boat/canoe and oars/engine/paddles to dry for five days and all other sampling equipment (including boots) to dry for 10 days after sampling a water body known to be infested with AIS.

Literature Cited

Baker, R. J., and J. R. Sauer. 1995. Statistical aspects of point count sampling. Pages 125-130 in C. J. Ralph, J. R. Sauer, and S. Droege, editors. Monitoring bird populations by point counts. General Technical Report PSW-GTR-149. USDA Forest Service, Pacific Southwest Research Station, Albany, California.

Faccio, S., and B. R. Mitchell. 2007. Developing vital signs of ecological integrity: A monitoring plan for the National Park Service Northeast Temperate Network. National Park Service and Vermont Institute of Natural Science, Quechee, VT.

Elias, J. E. 2008. Standard operating procedure #5, Decontamination of equipment to remove exotic species. In J. E. Elias, R. Axler, and E. Ruzycki. Water quality monitoring protocol for inland lakes, Version 1.0. Natural Resources Technical Report NPS/MWR/GLKN/ NRTR- 2008/109. National Park Service, Fort Collins, Colorado.

Farnsworth, G. L., J. D. Nichols, J. R. Sauer, S. G. Fancy, K. H. Pollock, S. A. Shriner, and T. R. Simons. 2005. Statistical approaches to the analysis of point count data: A little extra information can go a long way. Pages 736-743 in C. J. Ralph and T. D. Rich, editors. Bird conservation implementation and integration in the Americas: Proceedings of the Third International Partners in Flight Conference, Volume 2. General Technical Report PSW-GTR- 191. USDA Forest Service, Albany, California.

Kepler, C. B., and J. M. Scott. 1981. Reducing bird count variability by training observers. Studies in Avian Biology 6:366-371.

Ralph, C. J., G. R. Geupel, P. Pyle, T. E. Martin, and D. F. DeSante. 1993. Handbook of field methods for monitoring landbirds. General Technical Report PSW-GTR-144. USDA Forest Service, Albany, CA. Available online (www.fs.fed.us/psw/publications/ documents/gtr- 144/).

Simons, T. R., M. W. Alldredge, K. H. Pollock, and J. M.Wettroth. 2007. Experimental analysis of the auditory detection process on avian point counts. Auk 124:986-999.

Standard Operating Procedure #4: Using GPS to Locate and Mark Sampling Points

Please cite this document as:

Sutherland, T., B. Route, and T. Gostomski. 2010. Standard operating procedure #4: Using GPS to locate and mark sampling points. In Gostomski, T., M. Knutson, N. P. Danz, B. Route, and T. W. Sutherland. 2010. Landbird monitoring protocol, Great Lakes Inventory and Monitoring Network. Natural Resource Report NPS/GLKN/NRR—2010/225. National Park Service, Fort Collins, Colorado.

Contents Page

Introduction ...... 1

Procedures ...... 1 Generate Sampling Points ...... 1 Upload Sampling Points Into GPS Unit ...... 1 Navigate to the Sample Point...... 2 Mark Sample Point With a Permanent Marker ...... 3

Appendix A. Example Sampling Point Location Table...... 5

iii

Introduction

This SOP provides instructions for preparing maps and using a GPS to locate and mark sampling points associated with the bird point counts. Staff and volunteers conducting point counts should complete GPS training before field work begins. Upon completion of the training, all observers are expected to know how to mark a point, enter new waypoint coordinates, and navigate to a waypoint using a GPS.

We strongly encourage parks to establish and mark their sampling points before the bird surveys begin. Fall or early spring are ideal times for this work, as leaf-off conditions enhance the performance of GPS units.

Procedures

The following are responsibilities of the Park Monitoring Coordinator (PC), who should have some knowledge of ArcGIS. However, point count observers should notify the PC if they note any changes or information to be added to the navigation data while conducting point counts.

Generate Sampling Points 1. Generate a GIS layer of sampling points. Use one of the sampling designs discussed in SOP #1.

2. Create a map of sampling points, overlaid on land cover map or a rectified aerial photo. The map should show useful landmarks that can be located on the ground by the observer.

3. Create a list of your sampling point locations. Use the UTM or Geographic (Latitude/Longitude) coordinate system referenced to either the horizontal datum of NAD83 or WGS84 (see Appendix A). Be sure to specify the datum used and UTM zone if UTM coordinates are used. Each point must have a unique identification number.

4. Use a unique naming format for points. To distinguish point count locations from other data that may be in GPS units, the following naming format is recommended: “GLKNbird_park_transect name_01” where the two-digit number (01) indicates the point number. Using two-digit numbers keeps the points in numerical order in the GPS unit.

5. Pre-print field data sheets with the GPS locations (see SOP #5).

Upload Sampling Points Into GPS Unit 1. Use ArcGIS to identify the projection and datum being used by the GIS layer that contains your sampling points. If no metadata exists for this layer, contact the source who created the layer.

2. Change the projection and/or datum if necessary. If the GIS layer that contains your sampling points is not employing the UTM or Geographic (Latitude/Longitude)

1 coordinate system (referenced to either the horizontal datum of NAD83 or WGS84) use the “Project” tool in ArcGIS to change the projection and/or datum of the layer. Do not upload data into your GPS device from a GIS layer that is referenced to the horizontal datum of NAD27. Consult with the Great Lakes Network GIS Specialist or Data Manager if necessary.

3. Delete all existing waypoints in your GPS device.

4. Upload the sampling coordinates into your GPS device as waypoints. Do not manually enter the coordinate values by hand into the GPS device as many mistakes are likely. There is software available to automate the uploading process.

Navigate to the Sample Point The following steps are for navigating with a GPS unit. It is best to have navigation information supplemental to the data sheets (i.e., written format) with compass bearings, approximate distance between points, approximate travel time between points, and major landmarks to watch for. Remember, assume that the GPS unit will break down or not work under the forest canopy, and plan accordingly! 1. Ensure your GPS device is set to the same datum as the GIS layer used during the upload process described above.

2. If applicable to your device, ensure the Wide Area Augmentation System (WAAS) capabilities are enabled.

3. Ensure your distance units are set to METERS.

4. Ensure your GPS device is turned on and remains stationary in an open area for at least 10 minutes before you start to use the device for navigation.

5. Select the desired waypoint you wish to navigate to and begin navigating to the waypoint.

6. As you approach the waypoint, verify that at least four satellites are being used to determine your current position (3D). Once your distance to the waypoint is within 10 meters, you should begin looking for the permanent marker (if used) that identifies the location of the sampling point.

a. If this is the first visit to a new random sampling point, use the navigation mode on your GPS device and navigate to within 5 meters of the point (if navigation mode is not used, verify the estimated positional error value (EPE) is less than 5 meters). b. If you cannot get within 5 meters of the sampling point. Options are returning another day to establish this location as a permanent sampling point or moving the sampling point to avoid barriers (e.g., water, cliffs). If the point is to be moved, use your GPS device to record the coordinates of the new location. Ensure you collect the new location using a 3D fix with an EPE of 5 meters or less (PDOP of <6 using Trimble devices).

Mark Sample Point With a Permanent Marker Before the field season, and if allowed by the park’s general management plan and other guiding documents, establish a permanent marker at the sampling point using a metal tag or other permanent marker. Tags should be affixed so they are visible and permanent, given anticipated any disturbances to the site (e.g., flooding, burning, logging, wind-throw). Metal stakes with bright paint and a permanent label are ideal but may not be appropriate in all situations. Keep in mind that brightly colored flagging or paint will make the location easier to locate in the short term, but flagging rapidly degrades. The PC, in consultation with his/her supervisor and in line with park policies, will decide the most appropriate type of permanent markers to use.

Record a description of the permanent marker location using bearings and distances from natural features. For example, use a compass and tape measure or range finder to determine the distance and azimuth to the permanent marker from a nearby natural feature such as a distinctive tree. Record this information in the NOTES section of Appendix A.

Appendix A. Example Sampling Point Location Table

Identify Horizontal Datum Being Used: WGS84 or NAD83 UTM ZONE: 15

SAMPLE PT. # LATITUDE_DD LONGITUDE_DD UTM(EASTING) UTM(NORTHING) NOTES From Northern Red Oak tree, go 8.5 feet at a 1 43.82169 -91.26671 639382 4853527 bearing of 186 degrees 2 43.82255 -91.26782 639291 4853621 3 43.81996 -91.26706 639358 4853335 4 43.81691 -91.26629 639427 4852997 5 43.81424 -91.26377 639636 4852705 6 43.81885 -91.26515 639514 4853215 7 43.81785 -91.26418 639595 4853105 8 43.81787 -91.26270 639713 4853110 9 43.81581 -91.26146 639817 4852884 10 43.81408 -91.26068 639884 4852692 11 43.81216 -91.26290 639711 4852476

5 12 43.81366 -91.26346 639662 4852641 13 43.96902 -91.30264 636156 4869831 14 43.96905 -91.30319 636112 4869834 15 43.96886 -91.30259 636161 4869814 16 43.96941 -91.30225 636187 4869875 17 43.97779 -91.28964 637179 4870827 18 44.02302 -91.32248 634443 4875796 19 43.97774 -91.28917 637217 4870823 20 43.93258 -91.25810 639815 4865858 21 43.81991 -91.26394 639609 4853334 22 43.82169 -91.26671 639382 4853527

Standard Operating Procedure #5: Conducting the Bird Point Count

Please cite this document as:

Knutson, M., N. Danz, B. Route, and T. Gostomski. 2010. Standard operating procedure #5: Conducting the bird point count. In Gostomski, T., M. Knutson, N. P. Danz, B. Route, and T. W. Sutherland. 2010. Landbird monitoring protocol, Great Lakes Inventory and Monitoring Network. Natural Resource Report NPS/GLKN/NRR—2010/225. National Park Service, Fort Collins, Colorado.

Date Correction Made Correction Made By 25 Feb 2011 Weather codes on data sheet re-arranged to match Table 1. T. Gostomski

2 Feb 2012 Corrected four-letter codes for GRAJ and EWPW in T. Gostomski Appendix A. Added column for “# of individuals” to data sheet. Added category for non-vocal detection (N) to cover tapping woodpeckers, drumming grouse, and winnowing snipe). Make related changes to text and figures in body of SOP. Removed “end time” from data sheet.

Contents Page

Figures...... v

Tables ...... v

Appendices ...... vii

Introduction ...... 1

Procedures ...... 1 Before Surveys Begin ...... 1 Weather and Time Considerations ...... 1 Conducting the Point Count and Recording Data ...... 2 After the Count ...... 5 Field Descriptions for Point Count Field Data Form ...... 5

Literature Cited ...... 9

iii

Figures Page

Figure 1. Example of how data are entered on the point count data form, both in text form and mapping...... 4

Figure 2. Example of how to note the number of individuals observed in a single-species flock on the map portion of the Point Count Data Form...... 4

Tables Page

Table 1. Codes used to record sky conditions during bird counts1...... 6

Table 2. Codes used to record wind strength during bird counts (Beaufort Scale)1...... 6

Table 3. Codes used to record levels of background noise during bird surveys...... 6

Table 4. Codes used to record detection type during bird surveys ...... 7

Table 5. Codes used to record detection distance during bird surveys...... 7

Appendices Page

Appendix A. Four-letter AOU codes1 for birds in the Great Lakes Network National Parks2. .... 11

Appendix B. Point count data form ...... 15

vii

Introduction

This SOP provides step by step instructions for conducting landbird point counts, including how to complete the Point Count Data Form (Appendix B). The protocol is designed for either one person who does both the observing and the recording, or a two-person team consisting of an observer (who conducts the survey) and the recorder (who records the detections made by the observer). Details regarding sampling designs are found in Standard Operating Procedure (SOP) #1.

Procedures

Before Surveys Begin 1. If the route has been sampled in the past, examine the point sequence and times of surveys. The sequence and times should remain consistent with the original sampling sequence.

2. Discuss the sequence and route with field staff. Determine the survey plan for field staff. Always plan to survey the maximum number of points possible each day and take advantage of good survey conditions.

3. Ensure the GPS unit is loaded with the appropriate coordinates. This should be done before entering the field. See SOP #4.

4. Practice estimating distances using the laser rangefinder. Do this at the points if possible, but regardless of where the practice is done, it should be in an area where near and distant targets can be measured. To use distance sampling theory in analyzing bird data, it is critical that all distance estimations are the horizontal distance to each bird. The closer the bird, the more accurate the distance estimation should be (Buckland et al. 1993).

5. Review safety information relevant to your field situation. Review safety considerations with field staff, including acceptable conditions for conducting the point counts and how to contact help in case of an emergency. Field staff should carry food, water, rain gear, a map and compass and/or GPS unit, and a first aid kit, and they should wear appropriate footwear. Supplying volunteers or contractors with a park radio and providing them with appropriate training in radio procedures and etiquette is recommended.

6. Organize equipment. See Table 1 in SOP #2.

Weather and Time Considerations 1. Consider weather conditions before leaving. When survey conditions are questionable, the primary consideration is the observer(s)’s safety; the ability to hear birds is always secondary. High winds (>10 mph), heavy rain, or fog may prevent or delay surveys for several hours or even days. If the weather is unacceptable for conducting surveys, postpone them. It may be necessary to assess survey conditions from the actual survey points, not from the departure point.

2. Discuss options for continuing surveys when weather is questionable. It may be necessary to go to the first survey point to determine if survey conditions are acceptable. The crew leader is responsible for deciding if conditions are unacceptable for surveys. Some form of communication among crews and with the office (cell/satellite phones or hand-held radios) is recommended.

3. Record your float plan or field itinerary according to park protocols. At a minimum, record who is in the field crew, where you will be sampling that day, time of departure, when you expect to return, and who is responsible for checking that you have returned. Record your field and home contact information (phone number or pager) for use if you do not return.

4. Whenever possible, complete the entire route before moving on to the next route.

5. Complete all surveys between 0.5 hr before sunrise and 4 hrs after sunrise. Survey as many points as the time and weather allow each day. Arriving at a survey point on time often requires leaving your base 1.5 hrs before sunrise or earlier, depending on the time it takes to reach the first survey point. Be prepared with several extra data sheets.

Conducting the Point Count and Recording Data 1. Navigate to the survey point using map, compass, and a handheld GPS unit. See SOP #4.

2. After arriving at a survey point, rest for at least one minute. Resting for a minute allows the heartbeat to slow down after hiking, the observer to orient themselves, and birds to acclimate to your presence. Make a note of any birds observed flushing when the crew approaches the survey point.

3. Record location and weather data. During the one-minute pre-count period, place a thermometer on a branch or other elevated area (i.e., off the ground) and out of direct sunlight. Record temperature at the end of the count period. Fill in the location and weather data on the Point Count Data Form (Appendix B; see “Field Descriptions for Point Count Field Data Form” below), and record any bird species detected between points in the Comments section. If mapping bird locations, orient the Point Count Data Form to north. Prepare the digital timer, binoculars, and compass. If you are using a 2- person crew, the recorder and observer should divide up these “on-scene” tasks ahead of time.

4. Identify landmarks to use in estimating horizontal distance from the observer to the bird. Do this before beginning the survey. Select landmarks for the 50 m boundary and the 100 m boundary in each of the four cardinal directions if possible. To use distance sampling theory in analyzing bird data, it is critical that all distance estimations are the horizontal distance to each bird. The closer the bird, the more accurate the distance estimation should be (Buckland et al. 1993). Thus, pay particular attention to accurately recording the distance band (Table 5) of birds closest to the point. All individual birds are recorded

2 the first time they are observed and assigned a distance code (1, 2, or 3) that corresponds to a distance band (0-50 meters from the point, 51-100 m, and >100 m).

5. Conduct the point count as a “snapshot in time.” The survey results should represent the actual distribution of the birds relative to the point when they are first detected. The underlying theory of distance sampling requires that each point be recorded as close to a “snapshot in time” as possible. Some movement is acceptable, so long as a bird is only counted once, and the observer does not cause movement. Any birds that flush upon approaching the point, or birds that seem to be attracted by the presence of the surveyors, should be noted under Comments.

6. Use a digital timer to record the minute associated with each bird observation. Small digital timers can be attached to the clipboard. Alternatively, some GPS units have a stopwatch function. The observer records the minute (0, 1, 2, 3,….9) during which each bird is first observed on the data sheet next to the species code. The first time period is coded ‘0’, since that is what the timer will indicate. This is relatively easy to do and allows the data analyst to group times together in an optimal way.

7. If using a 2-person crew, announce all detections to the recorder in a clear yet quiet voice, including species, detection type, distance and (if mapping) direction. The observer stands at the survey point and quietly announces each detection to the recorder (for example “Red-eyed Vireo, singing, 50-to-100 m”). The recorder writes this information on the Point Count Data Form along with the minute interval that the bird was first detected (Figure 1). If you are mapping, the minute is noted as a superscript after the 4-letter species code (Figure 1).

On the data sheet, the species is identified by its 4-letter AOU code (see Appendix A). For example, “NAWA” is written for Nashville Warbler. All unknown birds, including unknown calls, should be recorded as “UNBI,” or as specifically as possible (e.g., UNSP for “unknown sparrow,” UNWO for “unknown woodpecker”). There is a “non-vocal” category (N) to use for detections such as tapping woodpeckers, drumming grouse, and winnowing snipe. A copy of Appendix A should be placed in the clipboard with the data sheets. The recorder is responsible for informing the observer when the count period has ended.

8. For most species, record each individual bird as a separate observation. For species that occur in clusters or flocks, the appropriate unit to record is the cluster or flock size, and not the individual birds. When mapping species locations, you may find it helpful to parenthetically note the number of individuals in the flock after the four-letter species code on the map portion of the Point Count Data Form (Figure 2). This will assist you in accurately tallying the total number of individuals after the survey is completed.

Figure 1. Example of how data are entered on the point count data form, both in text form and mapping. Note that the map indicator includes the minute of detection as a superscript after the species code.

CEDW(6)

Figure 2. Example of how to note the number of individuals observed in a single-species flock on the map portion of the Point Count Data Form.

9. Complete all fields on the datasheet before departing for the next point. After a point count is completed, record the temperature reading from the thermometer (indicate if reading is in Celsius or Fahrenheit), and check that all fields have been completed on the datasheet. Any additional comments, particularly regarding factors that might affect the quality of the data or detections made at the point after the point count period, should be recorded in the Comments section. Teams can also record observations of other fauna in the Comments section. Identify detections to species if possible.

10. Complete a datasheet for each point, even if no birds are detected. If no birds were detected at a point, complete as many data fields as possible and write “no birds were detected” in the Comments to document that the point was surveyed.

11. Ensure that no equipment is left behind. It is useful to attach brightly-colored flagging to items like thermometers, binoculars, GPS units, and any other equipment that may accidentally be left behind or dropped between points.

4 12. Navigate to the next point. Use the GPS to attain a bearing to the next point and hike to the next point at a reasonable pace. The pace between points should be fast enough to get a maximum number of points in days of good weather, but repeatable by other field crews in future surveys. Do not race to points. A good survey day may range from completing as few as seven points to as many as 14 points, depending on a variety of other factors, including weather, topography, and vegetation.

After the Count 1. Proofread all data forms and update daily field notes. For those parks who map bird locations during the point counts, we suggest transcribing count data from the map to the data form list after the day’s surveys are completed. It is also suggested that the observer do this (especially for two-person crews) as a check on the recorder or, for single-person surveyors, to prevent mistakes made when a second person enters the data into the database. Write bird names out completely on the lists if a second person (e.g., SCA, intern, or anyone who is otherwise uninvolved with the breeding bird surveys) is entering data into the database. Include any birds seen later in the day, but not during the survey period, in the field notes.

2. Enter data into the park database.

3. Once all bird surveys are completed, convert data to an Excel file and send to the Great Lakes Network landbird monitoring coordinator.

Field Descriptions for Point Count Field Data Form All field notes should be recorded legibly to prevent mistakes due to unreadable entries.

1. Date (mm/dd/yyyy): Record the month (2 digits), day (2 digits), and year (4 digits) in the format shown.

2. Park: The name of the park is used in the database to label all the point counts associated with this effort.

3. Location: Record the unique transect or route name.

4. Point: Record the point number.

5. Observer/Recorder (Obs/Rec): Record the initials of the observer and recorder. Names and contact information for all observers and recorders are stored in the database.

6. Start Time (hhmm): Record the time the point count begins: use military time and fill in all four digits. For instance, 0630 (6:30 am).

7. Sky: Record the sky condition as it applies during the point count period from Table 1.

8. Wind Speed: Record the wind code (Table 2) as it applies to the average wind speed during the point count period.

Table 1. Codes used to record sky conditions during bird counts1.

Sky Code Description 0 Clear or a few clouds 1 Partly cloudy (scattered) 2 Cloudy (broken) or overcast 4 Fog or Smoke 5 Drizzle 7 Snow 8 Showers 1 These are the same codes used in the North American Breeding Bird Survey (i.e., no #3 or #6). Acceptable conditions for counting birds include a sky code of 0-2.

Table 2. Codes used to record wind strength during bird counts (Beaufort Scale)1.

Wind Speed Wind Code (miles/hr) Description 0 < 1 Calm; smoke rises vertically 1 1-3 Light air; wind direction shown by smoke drift 2 4-7 Light breeze; wind felt on face 3 8-12 Gentle breeze; leaves in constant motion, light flag extended 4 13-18 Moderate breeze; raises dust; small branches move 5 19-24 Fresh breeze; small trees sway, crested wavelets on inland waters 6 25 or more Strong wind; large branches in motion 1 Acceptable conditions for counting birds include a wind code of 0-3.

9. Background Noise: Record the background noise code (Table 3) that describes the background noise conditions relative to the observer’s ability to hear birds.

Table 3. Codes used to record levels of background noise during bird surveys.

Background Noise Code Description 0 No background noise 1 Barely reduces hearing 2 Noticeable reduction of hearing 3 Prohibitive (greatly reduced hearing)

10. Temperature: Record the ambient air temperature at the end of the count, rounded to the nearest degree. Indicate whether reading is in Celsius or Fahrenheit. The thermometer should be placed above the ground, but not in direct sunlight.

6 11. Detection Type: Describes the type of detection associated with each bird. This is recorded on the data form as shown in column 1 of Table 4. Column 2 shows symbols to use if mapping bird distribution around a point.

Table 4. Codes used to record detection type during bird surveys. Four-letter code is used to indicate species (e.g., CEDW=Cedar Waxwing, RUGR=Ruffed Grouse).

Detection Type1 Mapping Symbol2

Singing CEDW

Calling CEDW

Visual CEDW

Flyover (indicate direction of travel) CEDW

Non-vocal (drumming grouse, tapping RUGR woodpeckers, winnowing snipe) 1 First letter of Detection Type is used on the data sheet species list. 2 If a single-species flock is observed, indicate number of individuals in parentheses after the species code but use the appropriate detection type indicator.

12. Distance (Category): Describes the horizontal distance in meters (Table 5) of each bird detection. This is recorded on the species list portion of the data sheet.

Table 5. Codes used to record detection distance during bird surveys.

Category Distance 1 0-50 m 2 51-100 m 3 >100 m

Literature Cited

Buckland S. T., K. P. Burnham, D. R. Anderson, and J. L. Laake. 1993. Density Estimation Using Distance Sampling. Chapman and Hall, London.

Pyle, P., and D.F. DeSante. 2010. English and scientific alpha codes for North American birds through the 51st AOU supplement (2010). Online at www.birdpop.org/AlphaCodes.htm (accessed 26 January 2011).

Appendix A. Four-letter AOU codes1 for birds in the Great Lakes Network National Parks2.

Common Name Species Code Common Name Species Code Acadian Flycatcher ACFL Chestnut-collared Longspur CCLO Alder Flycatcher ALFL Chestnut-sided Warbler CSWA American Crow AMCR Chimney Swift CHSW American Goldfinch AMGO Chipping Sparrow CHSP American Kestrel AMKE Chuck-will's-widow CWWI American Pipit AMPI Clay-colored Sparrow CCSP American Redstart AMRE Cliff Swallow CLSW American Robin AMRO Common Grackle COGR American Three-toed Woodpecker ATTW Common Nighthawk CONI American Tree Sparrow ATSP Common Raven CORA American Woodcock AMWO Common Redpoll CORE Baird's Sparrow BAIS Common Yellowthroat COYE Bald Eagle BAEA Connecticut Warbler CONW Baltimore Oriole (Northern Oriole) BAOR Cooper's Hawk COHA Bank Swallow BANS Dark-eyed Junco DEJU Barn Swallow BARS Dickcissel DICK Barred Owl BADO Downy Woodpecker DOWO Bay-breasted Warbler BBWA Eastern Bluebird EABL Bell's Vireo BEVI Eastern Kingbird EAKI Belted Kingfisher BEKI Eastern Meadowlark EAME Bewick's Wren BEWR Eastern Phoebe EAPH Black-and-white Warbler BAWW Eastern Screech-Owl EASO Black-backed Woodpecker BBWO Eastern Towhee (Rufous-sided Towhee) EATO Black-billed Cuckoo BBCU Eastern Wood-Pewee EAWP Black-billed Magpie BBMA European Starling EUST Blackburnian Warbler BLBW Evening Grosbeak EVGR Black-capped Chickadee BCCH Ferruginous Hawk FEHA Black-headed Grosbeak BHGR Field Sparrow FISP Blackpoll Warbler BLPW Fox Sparrow FOSP Black-throated Blue Warbler BTBW Golden Eagle GOEA Black-throated Green Warbler BTNW Golden-crowned Kinglet GCKI Blue Grosbeak BLGR Golden-winged Warbler GWWA Blue Jay BLJA Grasshopper Sparrow GRSP Blue-gray Gnatcatcher BGGN Gray Catbird GRCA Blue-headed Vireo (Solitary Vireo) BHVI Gray Jay GRAJ Blue-winged Warbler BWWA Gray Partridge GRAP Bobolink BOBO Gray-cheeked Thrush GCTH Bohemian Waxwing BOWA Great Crested Flycatcher GCFL Boreal Chickadee BOCH Great Gray Owl GGOW Boreal Owl BOOW Great Horned Owl GHOW Brewer's Blackbird BRBL Greater Prairie-Chicken GRPC Brewster's Warbler BRWA Gyrfalcon GYRF Broad-winged Hawk BWHA Hairy Woodpecker HAWO Brown Creeper BRCR Harris's Sparrow HASP Brown Thrasher BRTH Henslow's Sparrow HESP Brown-headed Cowbird BHCO Hermit Thrush HETH Canada Warbler CAWA Hoary Redpoll HORE Cape May Warbler CMWA Hooded Warbler HOWA Carolina Wren CAWR Horned Lark HOLA Cedar Waxwing CEDW House Finch HOFI Cerulean Warbler CERW House Sparrow HOSP

Appendix A. Four-letter AOU codes for birds in Great Lakes Network National Parks (continued).

Common Name Species Code Common Name Species Code House Wren HOWR Red-breasted Nuthatch RBNU Indigo Bunting INBU Red-eyed Vireo REVI Kentucky Warbler KEWA Red-headed Woodpecker RHWO Kirtland's Warbler KIWA Red-shouldered Hawk RSHA Lapland Longspur LALO Red-tailed Hawk RTHA Lark Sparrow LASP Red-winged Blackbird RWBL Le Conte's Sparrow LCSP Ring-necked Pheasant RNPH Least Flycatcher LEFL Rock Dove See Rock Pigeon Lincoln's Sparrow LISP Rock Pigeon (Rock Dove) ROPI Loggerhead Shrike LOSH Rock Wren ROWR Long-eared Owl LEOW Rose-breasted Grosbeak RBGR Louisiana Waterthrush LOWA Rough-legged Hawk RLHA Magnolia Warbler MAWA Ruby-crowned Kinglet RCKI Marsh Wren MAWR Ruby-throated Hummingbird RTHU Merlin MERL Ruffed Grouse RUGR Mississippi Kite MIKI Rufous-sided Towhee See Eastern Towhee Mourning Dove MODO Rusty Blackbird RUBL Mourning Warbler MOWA Sandhill Crane SACR Nashville Warbler NAWA Savannah Sparrow SASP Nelson's Sharp-tailed Sparrow NSTS Scarlet Tanager SCTA Northern Bobwhite NOBO Scissor-tailed Flycatcher STFL Northern Cardinal NOCA Sedge Wren SEWR Northern Flicker NOFL Sharp-shinned Hawk SSHA Northern Goshawk NOGO Sharp-tailed Grouse STGR Northern Harrier NOHA Short-eared Owl SEOW Northern Hawk Owl NHOW Smith's Longspur SMLO Northern Mockingbird NOMO Snow Bunting SNBU Northern Oriole See Baltimore Oriole Snowy Owl SNOW Northern Parula NOPA Solitary Vireo See Blue-headed Vireo Northern Rough-winged Swallow NRWS Song Sparrow SOSP Northern Saw-whet Owl NSWO Sora SORA Northern Shrike NSHR Spruce Grouse SPGR Northern Waterthrush NOWA Summer Tanager SUTA Olive-sided Flycatcher OSFL Swainson's Hawk SWHA Orange-crowned Warbler OCWA Swainson's Thrush SWTH Orchard Oriole OROR Swallow-tailed Kite STKI Osprey OSPR Swamp Sparrow SWSP Ovenbird OVEN Tennessee Warbler TEWA Palm Warbler PAWA Townsend's Solitaire TOSO Peregrine Falcon PEFA Tree Swallow TRES Philadelphia Vireo PHVI Tufted Titmouse TUTI Pileated Woodpecker PIWO Turkey Vulture TUVU Pine Grosbeak PIGR Upland Sandpiper UPSA Pine Siskin PISI Varied Thrush VATH Pine Warbler PIWA Veery VEER Prairie Warbler PRAW Vesper Sparrow VESP Prothonotary Warbler PROW Virginia Rail VIRA Purple Finch PUFI Warbling Vireo WAVI Purple Martin PUMA Western Kingbird WEKI Red Crossbill RECR Western Meadowlark WEME Red-bellied Woodpecker RBWO Western Tanager WETA

Appendix A. Four-letter AOU codes for birds in Great Lakes Network National Parks (continued).

Common Name Species Code Common Name Species Code Wester Wood-Pewee WEWP Wood Thrush WOTH Whip-poor-will EWPW Worm-eating Warbler WEWA White-breasted Nuthatch WBNU Yellow Warbler YWAR White-crowned Sparrow WCSP Yellow-bellied Flycatcher YBFL White-eyed Vireo WEVI Yellow-bellied Sapsucker YBSA White-throated Sparrow WTSP Yellow-billed Cuckoo YBCU White-winged Crossbill WWCR Yellow-breasted Chat YBCH Wild Turkey WITU Yellow-headed Blackbird YHBL Willow Flycatcher WIFL Yellow-rumped Warbler YRWA Wilson's Snipe WISN Yellow-throated Vireo YTVI Wilson's Warbler WIWA Yellow-throated Warbler YTWA Winter Wren WIWR 1 Species codes from Pyle and DeSante (2010). 2Species lists derived from each park’s inventory in the NPSpecies database (https://science1.nature.nps.gov/npspecies/web/main/start).

Appendix B. Point count data form

We recommend printing the data form as a double-sided page. Parks that do not map their point count data do not have to print the map side.

Date: Park: Location: Point: No. of Detection Observer: Species Indiv. Minute Type Distance

Recorder:

Start Time:

Sky:

Wind Speed:

Background Noise:

Temperature (indicate F or C):

Detection Type

Singing

Calling

Observed

Flyover

Non-vocal e.g., tapping, drumming, winnowing

Detection Distance

1 = 0 - 50 m

2 = 51 - 100 m

3 = >100 m

SKY CONDITION CODES WIND SPEED CODES BACKGROUND NOISE CODES 0 Clear or a few clouds 0 < 1 mph 0 No background noise 1 Partly cloudy (scattered) 1 1-3 1 Barely reduces hearing

2 Cloudy (broken) or overcast 2 4-7 2 Noticeable reduction of hearing

4 Fog or Smoke 3 8-12 3 Prohibitive (greatly reduced hearing) 5 Drizzle 4 13-18 7 Snow 5 19-24

8 Showers 6 >25 mph

Comments: (any major changes to habitat around the point? What noise contributed to reduced hearing?)

100 m

Detection Symbols NAWA Observed

NAWA Singing

NAWA Calling

NAWA Flyover (arrow indicates direction of travel)

RUGR Non-vocal sound (drumming grouse, tapping woodpeckers, winnowing snipe)

Indicate time (minute) of observation with a superscript number, 0-9. e.g., NAWA5 Indicate number of individuals in a single-species flock in parentheses after the four-letter species code

Standard Operating Procedure #6: Vegetation Monitoring

Please cite this document as:

Knutson, M., N. Danz, T. Sutherland, B. Route, and T. Gostomski. 2010. Standard operating procedure #6: Vegetation monitoring. In Gostomski, T., M. Knutson, N. P. Danz, B. Route, and T. W. Sutherland. 2010. Landbird monitoring protocol, Great Lakes Inventory and Monitoring Network. Natural Resource Report NPS/GLKN/NRR—2010/225. National Park Service, Fort Collins, Colorado.

Contents Page

Appendices ...... v

Introduction ...... 1

Procedures ...... 3

Train Field Staff ...... 3 Record Land Cover Class (Required) ...... 3 Take Photographs at Each Sampling Point (Optional) ...... 4 Record Additional Vegetation Metrics (Optional) ...... 4

Literature Cited ...... 5

iii

Appendices

Page

Appendix A. NLCD Classifications...... 7

Appendix B. Rapid Vegetation Measurements (Forests)...... 11

Introduction

This SOP provides instructions for conducting vegetation monitoring associated with the bird point counts. The Network requires one vegetative characteristic to be recorded for each point. That is the third order land cover class, using the National Landcover Database (NLCD), in which the bird monitoring point is located (see Appendix A). This information will be permanently linked to the bird data at each point in the Network database.

In addition, we suggest that observers take digital photographs of the vegetation in the four cardinal directions at each point each year. This can be done at any time after the landbird surveys are completed and before 1 September (i.e., during the nesting season). Establishing a photographic record of vegetation is quicker and requires less training than recording more in- depth vegetation metrics. The low price and light weight of a simple digital camera makes it an inexpensive data logger. The strengths of repeated photographs in monitoring vegetation changes are that vegetation change can be assessed qualitatively from an annual series of photos, and the field portion of the process can be executed rapidly and easily. A weakness of repeated photographs in monitoring vegetation is that observable changes in species composition are limited to obvious species such as woody plant invasion and large herbaceous species replacing small species. As well, species identification is difficult at best, except for those nearest the point and/or the camera. Even those species that can be identified are limited to trees and shrubs with obvious characteristics (e.g., paper birch bark).

All other vegetation documentation or monitoring related to point count locations is optional, depending on the park’s purpose and objectives for their landbird monitoring program. For those who are interested in recording vegetation data, we provide two protocols for a rapid vegetation/habitat assessment (Appendix B).

Procedures

Train Field Staff The park monitoring coordinator (PC) will train field staff to recognize land cover classes at each point using the National Landcover Database (NLCD). Other habitat measurements may be designed for specific purposes; a companion (MS Access) database will hold these additional habitat variables.

We are using the NLCD because there will soon be a consistent national map of these land cover classes. The classes encompass all the major habitats of interest with regard to landbirds and include some agricultural and developed classes. At a minimum, each park will be able to overlay their property boundaries on this map, no matter where they are located. Additionally, unlike the NLCD, the National Vegetation Classification System (NVCS) is not as exhaustive a classification, and The Nature Conservancy’s Ecological Systems is a very detailed classification system that require intensive training for field staff to accurately record vegetation using this system. Plus, agricultural cover and cultural land cover types are not defined (Comer et al. 2003; http://www.natureserve.org/explorer/index.htm).

Record Land Cover Class (Required) Record the name of land cover class and the associated number (11 – 99) at each bird monitoring point immediately after conducting the point count. Recording at least one dominant land cover class at each point is required and will be permanently linked to the bird data at each point at each visit in the National Point Count Database. Record the data in the Comments section of the point count data form (SOP #5, Appendix B). 1. Field 1: Primary land cover class (required) 2. Field 2: Secondary land cover class (record if the area within 100 m of the point has more than one land cover class) 3. Field 3: Comments

Take Photographs at Each Sampling Point (Optional) Set the digital camera to record *.jpg images at minimum image resolution of 1260 X 990 pixels (approximately 14 X 11 inches; file size ≥ 500 KB) and to display date and time information on the photos. Set the camera to use the ‘automatic’ focusing and exposure settings. 1. Stand at the center of the bird sampling point, close to the permanent marker (stake, tree). 2. While standing at the center of the bird sampling point, take one photo with the camera pointed in each cardinal direction (N, E, S, W, in that order) along a trajectory parallel with the ground, at a height that is eye level for the observer. Note the photo numbers on the data sheet. 3. In the office, edit the digital images so that the transect, point, direction, date, and NLCD 2001 land cover classification information are affixed permanently to the photos associated with that point (bottom left corner of the photo).

Record Additional Vegetation Metrics (Optional) You may consider collecting additional vegetation/plant community/habitat structure variables in association with your bird point counts. We recommend thinking carefully about your management objectives and how you will use the vegetation data to meet your objectives. See Appendix B for an example of a vegetation monitoring protocol used in association with bird point count data.

Appendix B includes rapid vegetation measurements used by the University of Minnesota Natural Resources Research Institute (NRRI) with bird point counts in forested landscapes. NRRI has an MS Access database for these data and will make it available to parks.

Literature Cited

Anderson, J. R., E. E. Hardy, J. T. Roach, and R. E. Witmer. A land use and land cover classification system for use with remote sensor data. USGS Professional Paper 964.

Comer, P., D. Faber-Langendoen, R. Evans, S. Gawler, C. Josse, G. Kottel, S. Menard, M. Pyne, M. Reid, K. Shulz, and others. 2003. Ecological systems of the United States: A working classification on U.S. terrestrial systems. NatureServe, Arlington, VA, USA.

Homer, C., C. Huang, L. Yang, B. Wylie, and M. Coan. 2004. Development of a 2001 national landcover database for the United States. Photogrammetric Engineering and Remote Sensing 70:829-840.

New Jersey Department of Environmental Protection (NJDEP). 1998. Land use land cover classification system. (Online). www.state.nj.us/dep/gis/digidownload/metadata/ lulc95/anderson.html. Accessed in 2004 and October 2006.

Appendix A. NLCD Classifications. Available online at www.mrlc.gov/nlcd_definitions.php

11. Open Water - All areas of open water, generally with less than 25% cover of vegetation or soil.

12. Perennial Ice/Snow - All areas characterized by a perennial cover of ice and/or snow, generally greater than 25% of total cover.

21. Developed, Open Space - Includes areas with a mixture of some constructed materials, but mostly vegetation in the form of lawn grasses. Impervious surfaces account for less than 20 percent of total cover. These areas most commonly include large-lot single-family housing units, parks, golf courses, and vegetation planted in developed settings for recreation, erosion control, or aesthetic purposes

22. Developed, Low Intensity - Includes areas with a mixture of constructed materials and vegetation. Impervious surfaces account for 20-49 percent of total cover. These areas most commonly include single-family housing units.

23. Developed, Medium Intensity - Includes areas with a mixture of constructed materials and vegetation. Impervious surfaces account for 50-79 percent of the total cover. These areas most commonly include single-family housing units.

24. Developed, High Intensity - Includes highly developed areas where people reside or work in high numbers. Examples include apartment complexes, row houses and commercial/industrial. Impervious surfaces account for 80 to100 percent of the total cover.

31. Barren Land (Rock/Sand/Clay) - Barren areas of bedrock, desert pavement, scarps, talus, slides, volcanic material, glacial debris, sand dunes, strip mines, gravel pits and other accumulations of earthen material. Generally, vegetation accounts for less than 15% of total cover.

32. Unconsolidated Shore* - Unconsolidated material such as silt, sand, or gravel that is subject to inundation and redistribution due to the action of water. Characterized by substrates lacking vegetation except for pioneering plants that become established during brief periods when growing conditions are favorable. Erosion and deposition by waves and currents produce a number of landforms representing this class.

41. Deciduous Forest - Areas dominated by trees generally greater than 5 meters tall, and greater than 20% of total vegetation cover. More than 75 percent of the tree species shed foliage simultaneously in response to seasonal change.

42. Evergreen Forest - Areas dominated by trees generally greater than 5 meters tall, and greater than 20% of total vegetation cover. More than 75 percent of the tree species maintain their leaves all year. Canopy is never without green foliage.

43. Mixed Forest - Areas dominated by trees generally greater than 5 meters tall, and greater than 20% of total vegetation cover. Neither deciduous nor evergreen species are greater than 75 percent of total tree cover.

51. Dwarf Scrub - Alaska only areas dominated by shrubs less than 20 centimeters tall with shrub canopy typically greater than 20% of total vegetation. This type is often co-associated with grasses, sedges, herbs, and non-vascular vegetation.

52. Shrub/Scrub - Areas dominated by shrubs; less than 5 meters tall with shrub canopy typically greater than 20% of total vegetation. This class includes true shrubs, young trees in an early successional stage or trees stunted from environmental conditions.

71. Grassland/Herbaceous - Areas dominated by grammanoid or herbaceous vegetation, generally greater than 80% of total vegetation. These areas are not subject to intensive management such as tilling, but can be utilized for grazing.

72. Sedge/Herbaceous - Alaska only areas dominated by sedges and forbs, generally greater than 80% of total vegetation. This type can occur with significant other grasses or other grass like plants, and includes sedge tundra, and sedge tussock tundra.

73. Lichens - Alaska only areas dominated by fruticose or foliose lichens generally greater than 80% of total vegetation.

74. Moss - Alaska only areas dominated by mosses, generally greater than 80% of total vegetation.

81. Pasture/Hay - Areas of grasses, legumes, or grass-legume mixtures planted for livestock grazing or the production of seed or hay crops, typically on a perennial cycle. Pasture/hay vegetation accounts for greater than 20 percent of total vegetation.

82. Cultivated Crops - Areas used for the production of annual crops, such as corn, soybeans, vegetables, tobacco, and cotton, and also perennial woody crops such as orchards and vineyards. Crop vegetation accounts for greater than 20 percent of total vegetation. This class also includes all land being actively tilled.

90. Woody Wetlands - Areas where forest or shrubland vegetation accounts for greater than 20 percent of vegetative cover and the soil or substrate is periodically saturated with or covered with water.

91. Palustrine Forested Wetland* -Includes all tidal and non-tidal wetlands dominated by woody vegetation greater than or equal to 5 meters in height and all such wetlands that occur in tidal areas in which salinity due to ocean-derived salts is below 0.5 percent. Total vegetation coverage is greater than 20 percent.

92. Palustrine Scrub/Shrub Wetland* - Includes all tidal and non-tidal wetlands dominated by woody vegetation less than 5 meters in height, and all such wetlands that

occur in tidal areas in which salinity due to ocean-derived salts is below 0.5 percent. Total vegetation coverage is greater than 20 percent. The species present could be true shrubs, young trees and shrubs or trees that are small or stunted due to environmental conditions.

93. Estuarine Forested Wetland* - Includes all tidal wetlands dominated by woody vegetation greater than or equal to 5 meters in height, and all such wetlands that occur in tidal areas in which salinity due to ocean-derived salts is equal to or greater than 0.5 percent. Total vegetation coverage is greater than 20 percent.

94. Estuarine Scrub/Shrub Wetland* - Includes all tidal wetlands dominated by woody vegetation less than 5 meters in height, and all such wetlands that occur in tidal areas in which salinity due to ocean-derived salts is equal to or greater than 0.5 percent. Total vegetation coverage is greater than 20 percent.

95. Emergent Herbaceous Wetlands - Areas where perennial herbaceous vegetation accounts for greater than 80 percent of vegetative cover and the soil or substrate is periodically saturated with or covered with water.

96. Palustrine Emergent Wetland (Persistent)* - Includes all tidal and non-tidal wetlands dominated by persistent emergent vascular plants, emergent mosses or lichens, and all such wetlands that occur in tidal areas in which salinity due to ocean-derived salts is below 0.5 percent. Plants generally remain standing until the next growing season.

97. Estuarine Emergent Wetland* - Includes all tidal wetlands dominated by erect, rooted, herbaceous hydrophytes (excluding mosses and lichens) and all such wetlands that occur in tidal areas in which salinity due to ocean-derived salts is equal to or greater than 0.5 percent and that are present for most of the growing season in most years. Perennial plants usually dominate these wetlands.

98. Palustrine Aquatic Bed* - The Palustrine Aquatic Bed class includes tidal and nontidal wetlands and deepwater habitats in which salinity due to ocean-derived salts is below 0.5 percent and which are dominated by plants that grow and form a continuous cover principally on or at the surface of the water. These include algal mats, detached floating mats, and rooted vascular plant assemblages.

99. Estuarine Aquatic Bed* - Includes tidal wetlands and deepwater habitats in which salinity due to ocean-derived salts is equal to or greater than 0.5 percent and which are dominated by plants that grow and form a continuous cover principally on or at the surface of the water. These include algal mats, kelp beds, and rooted vascular plant assemblages.

* Coastal NLCD class only

Appendix B. Rapid Vegetation Measurements (Forests).

Used by the University of Minnesota Natural Resources Research Institute for bird point counts in forested landscapes. The objective of this vegetation sampling method is to get information on habitat structure and plant composition as quickly and as accurately as possible while conducting a bird survey. With some experience and familiarity with the trees and shrubs of the region, vegetation data can be gathered in less than 3 minutes either before or after the bird survey. A Microsoft Access database is available for entering the data. The following information is recorded:

1. Canopy Height – The average canopy height (in meters) within the 100 m radius should be estimated. This does not include shrub or subcanopy layers. Note: If the point is in a new clearcut with residual trees, estimate canopy height for the residual trees. If the point is in a regenerating aspen stand with DBH greater than 2.5 cm, estimate the canopy height of the regenerating aspen. If the regenerating aspen is less than 2.5 cm DBH, estimate the canopy height of the residual trees. Be sure to include code 9, 10, or 11 under Special Features. 2. Tree Density – Estimate tree density by counting all the trees (> 2.5 cm DBH) within a 10 m radius and assigning the corresponding density code (see code sheet) by abundance. 3. Shrub Density – Estimate shrub density in the same manner as above except count those woody plants with a DBH of 2.5 cm or less within a 10 m radius. 4. High Canopy Cover – Estimate the percent coverage of the high canopy layer within the 100 m radius, using percent estimated in increments of 10. 5. High Canopy % Deciduous – Estimate the percent of deciduous species found within the high canopy, using percent estimated in increments of 10. 6. Subcanopy Cover – Estimate the percent coverage of the subcanopy layer within the 100 m radius, using percent estimated in increments of 10. 7. Subcanopy % Deciduous – Estimate the percent of deciduous species found within the subcanopy, using percent estimated in increments of 10. 8. Understory Cover – Estimate the percent coverage of the understory layer within the 100 m radius, using percent estimated in increments of 10. 9. Understory % Deciduous – Estimate the percent of deciduous species found within the understory, using percent estimated in increments of 10. 10. Ground Cover - Estimate the percent coverage of the ground layer within the 100 m radius, using percent estimated in increments of 10. 11. Tree Species – List up to five tree species (> 2.5 cm DBH), beginning with the most abundant species. 12. Shrub Species – List up to five shrub species (< 2.5 cm DBH; this can include tree species), beginning with the most abundant species. 13. Special Features – List up to three special features (see code sheet). Note: Special feature 9, 10, or 11 should be used when residual trees are found within the 100 m radius.

Rapid Vegetation Measurements (Forests) Field Data Sheet

Tree Density Shrub Density Special Features (> 2.5 cm DBH) (< 2.5 cm DBH) <10 in 10m 1 None 1 1 Beaver flooding 7 Natural opening in site radius <5 in 10m Large downed 2 2 11 to 100 2 8 Rock outcrop radius logs 3 6 to 20 3 101 to 500 3 Small openings 9 Residual hardwood trees 4 21 to 40 4 501 to 1000 4 Snags 10 Residual conifer trees Wetland pocket in 5 >40 5 >1000 5 11 Residual patches site Woodland pond in 6 12 Roads, buildings site

COMMENTS Vegetation Structure

Canopy height (m)

Tree Density (1 = few to 5 = dense) - 10 m radius

Shrub Density (1 = few to 5 = dense) - 10 m radius

High canopy cover (%)

High canopy % deciduous (0% = all conifer)

Subcanopy cover (lower layer of trees: %)

Subcanopy % deciduous (0% = all conifer)

Understory cover (3 ft to 12 ft: %)

Understory % deciduous (0% = all conifer)

Ground cover (3 ft to ground: %)

Tree species (up to 5)

Shrub species (up to 5)

Special features (codes)

Standard Operating Procedure #7: After the Field Season

Please cite this document as:

Knutson, M., T. Sutherland, B. Route, and T. Gostomski. 2010. Standard operating procedure #7: After the field season. In Gostomski, T., M. Knutson, N. P. Danz, B. Route, and T. W. Sutherland. 2010. Landbird monitoring protocol, Great Lakes Inventory and Monitoring Network. Natural Resource Report NPS/GLKN/NRR—2010/225. National Park Service, Fort Collins, Colorado.

Contents Page

Introduction ...... 1

Procedures ...... 1

Returning Field Equipment ...... 1 Datasheets and Data Entry ...... 1 Field Notes and Reports ...... 1

iii

Introduction

This Standard Operating Procedure (SOP) describes procedures for returning field equipment, entering and proofing data, and writing reports. This SOP also describes the procedures to prepare data sheets for data entry and data archival, including compilation of field notes.

Procedures

Returning Field Equipment 1. Clean all equipment before storage. Clean all field gear, including binoculars. Re-shelve reference materials. Wash any government vehicles and boats that were used to access field sites and address any needed repairs or maintenance.

2. Sign in and return field equipment. Sign in all gear that was issued to at the beginning of the season, assuring that issued items are returned and/or accounted for. Organize any damaged or incomplete equipment, repair equipment if possible, and make notes about those items that cannot be repaired. Hand notes in to park monitoring coordinator (PC), along with a list of any items that need to be purchased.

Datasheets and Data Entry 1. Photocopy the original data sheets. Upon returning from the field, make a copy of each original field data sheet using the double-sided copy feature of copying machines. Review each copied data sheet for clarity. The copied data sheets are used for data entry, so it is important that the copied data sheets are readable.

2. Archive the original data sheets. Organize the original field data sheets by grid and point sequence and store in a designated fireproof safe or cabinet.

3. Proofread the copied datasheets, making sure that they are filled out completely. All data sheets should have been reviewed for completeness while in the field. However, some deficiencies in data may not be identified until all data sheets have been reviewed as a group.

4. Mark corrections on copied data sheets with red pen. Any corrected errors, or changes made by the data “proofer” (that are entered differently into the database than they appear on the data sheet) should be circled using a red, fine-point, permanent Sharpie™ marker. Write notes/corrections, in red ink, in the margins or in the comments section to document the corrections, and the reason for correcting, if necessary.

Field Notes and Reports 1. Photocopy any field notes. Store the copies of the field notes with the original data sheets.

2. Compile a complete list of bird species detected between points or otherwise outside of a designated point count period. Also compile a list of any other fauna observed, other than birds. Use the backsides of the field datasheets, along with all field notes to compile the list.

3. At the end of the season, the PC will compile a field season summary report and file it with the original data sheets and field notes. Field season summaries should include the following:

a. Dates of the sampling events. b. The sequence and times of each point count. c. Map of sampling locations (GPS coordinates). d. Crew members, full names and initials, and their responsibilities. e. A list of bird and other faunal species encountered; include useful standard reports produced by the database (SOP #9). f. Nests found (including egg/chick counts, fledging dates, if known). g. General observations of weather, bird behavior, and other animals. h. Any discrepancies that might affect data integrity/consistency. i. Potential hazards. j. Unique or noteworthy events. k. Advice for future survey crews.

4. Write an Annual Report. Prepare a report summarizing the bird and vegetation data from the survey (refer to SOP #9 and SOP #10).

5. Archive data, field notes, and photographs. The PC will archive raw survey data, field notes, and photographs in compliance with relevant NPS data standards.

6. Archive the protocol and data together. At the termination of a survey, or every 3-5 years, archive the protocol and any amendments, associated metadata, maps, photographs, field notes, data sheets, and electronic data files together with interim and final reports, as a package. This package is also stored in two places, one at the park and one at the Great Lakes Network office.

7. Archive any vegetation photographs taken at each sampling point. Edit the digital images so that the transect, point, date, and NLCD 2001 land cover classification information are affixed permanently to the photos associated with that point (bottom left corner of the photo).

8. Archive any additional photographs. Archive additional photos taken during the field season, linking information about park, transect, point, date, and subject with each photograph.

Standard Operating Procedure #8: Data Management

Please cite this document as:

Sutherland, T., M. Knutson, B. Route, and T. Gostomski. 2010. Standard operating procedure #8: Data management. In Gostomski, T., M. Knutson, N. P. Danz, B. Route, and T. W. Sutherland. 2010. Landbird monitoring protocol, Great Lakes Inventory and Monitoring Network. Natural Resource Report NPS/GLKN/NRR—2010/225. National Park Service, Fort Collins, Colorado.

Contents Page

Introduction ...... 2 Database Description ...... 2 Who can use it? ...... 2 Who manages it? ...... 2 Data Sources ...... 2 Users ...... 2 Getting Started ...... 2 Data Entry and Quality Assurance/Quality Control ...... 2 Data Maintenance, Archiving, and Storage ...... 3

iii

Introduction

This SOP describes the landbird monitoring database and provides instructions for data entry, data validation, and database administration. The bird point count data will reside in a Great Lakes Network database. We may, in the future, submit it to the USGS Patuxent Bird Point Count Database (http://www.pwrc.usgs.gov/point/index.cfm).

Database Description

Who can use it?

Who manages it?

Data Sources

Users

Getting Started

Data Entry and Quality Assurance/Quality Control

Data entry and validation is also discussed in SOP #7.

1. Transfer map data to list format. Upon returning from the field, for those parks that map bird locations on the data sheet, surveyors should transfer their map data to the list format, noting species, number of individuals recorded, and distance from point count center in the appropriate columns. Be sure that count totals in the list format match your notations on the map. In other words, if you recorded 10 birds on your map, be sure 10 birds are on your list. It is also a good idea to be sure counts of individuals by species also match.

2. Photocopy data sheets. Organize the data sheets by transect and point sequence, make photocopies, and give the originals to the park monitoring coordinator (PC).

3. Archive the original data sheets. The PC organizes the original field data sheets by transect and point sequence and stores them in a designated cabinet.

4. Proofread the copied data sheets. Proofread the copied datasheets, making sure that they have been filled out completely. All data sheets should have been reviewed for completeness while in the field. However, some deficiencies in data recording may not be identified until all data sheets have been reviewed as a group and some errors are inevitable.

5. Mark corrections on copied data sheets with red pen. Any corrected errors, or changes made by the data “proofer” (that are entered differently into the database than they appear on the data sheet) should be circled and corrected using a red pen. Notes, in red ink, should be

written on the margins or in the comments section whenever necessary to document the reason for the corrections.

Data Maintenance, Archiving, and Storage

Each year, after all data for the survey year has been verified, the park’s Responsible Party or Data Manager will download all of their data residing in the master database and store as an archived copy. This archived copy must be stored at the park and a copy sent to the Great Lakes Network office.

Standard Operating Procedure #9: Data Analysis

Please cite this document as:

Danz, N., M. Knutson, and B. Route. 2010. Standard operating procedure #9: Data analysis. In Gostomski, T., M. Knutson, N. P. Danz, B. Route, and T. W. Sutherland. 2010. Landbird monitoring protocol, Great Lakes Inventory and Monitoring Network. Natural Resource Report NPS/GLKN/NRR—2010/225. National Park Service, Fort Collins, Colorado.

Contents Page

Introduction ...... 1

Overview ...... 1

Sources of Variability ...... 1 The Index Assumption ...... 1 Sampling Design ...... 2

Analyses ...... 3

Breeding Bird Composition ...... 3 Vegetation Data ...... 3 Population Trend and Habitat Analyses ...... 3 Birds as Indicators of Ecosystem Integrity ...... 4

Literature Cited ...... 5

iii

Introduction

This SOP provides instructions and guidelines for analyzing bird data collected using point counts. Different parks will have different and often multiple objectives, including describing the composition and distribution of bird communities, calculating population trends, and developing bird-habitat relationships. A wide variety of appropriate statistical methods can be used to meet these objectives. Here, we discuss some of the alternatives and provide some guidance, although the analytical method ultimately chosen to meet a specific objective will depend upon the particular details of each monitoring effort and the expertise of the analyst.

While some analyses may be accomplished by park staff with little formal statistical training, other more complex analyses, such as habitat associations or population trend modeling, may require the assistance of a professional statistician. Thus, we provide some recommendations about what analyses can be undertaken annually and which analyses should be conducted periodically (i.e., every 3-5 years). It is beyond the scope of this protocol to provide instructions for each analysis method. Statistical analysis techniques and software evolve rapidly, and each data set will require individual thought and attention at the time of analysis.

Overview

Perhaps more has been written in the scientific literature about the collection and analysis of bird point count data for any other biological taxon (Sauer and Droege 1990, Ralph et al. 1995). Several issues related to point counts should be explicitly considered during the analysis and will be briefly reviewed here.

Sources of Variability In addition to true variability in bird abundance in space or time, there are many additional factors that can directly influence the number of birds observed during a point count survey. These factors include wind, rain, ambient noise levels, time of day, observer differences, and spatial heterogeneity. While the point count protocol has features built-in to minimize many of these sources of variability, their potential influence should be incorporated into the analyses, and periodically the influence should be investigated explicitly. Of particular importance may be variability associated with different observers (Ramsey and Scott 1981). There are well- developed statistical methods that can account for observer variability when individual observers complete surveys in multiple years (Sauer et al. 1994, MacKenzie et al. 2002), but dealing with observer variability is more difficult when observers participate in only one year (Thompson et al. 1998).

The Index Assumption The ultimate objective of a population monitoring program is to make conclusions about the magnitude and direction of change in true population size through time. The point count is a method for surveying rather than completely enumerating (i.e., censusing) all birds in an area, so the raw number of observations recorded during a point count is not a measure of density. For trend analysis, if the raw bird counts are a constant proportion of true population size, it may be possible to use unadjusted counts as an index of the population. However, if the chance of observing a bird is not constant through time or between habitats (that is, if there is heterogeneity

in the detection probability), then the raw counts will not be an unbiased index of population size.

There is currently a disagreement in the scientific literature about the degree to which detection heterogeneity influences conclusions about trends. Some authors are adamantly opposed to the use of unadjusted counts for any conclusions, while others are unconvinced that accounting for detection heterogeneity improves the estimate of trend (Johnson 2008, Nichols et al. unpublished data). While it may be some time before there is scientific consensus on this issue, there currently exist several alternative methods for evaluating and incorporating detection probabilities into the analysis. The point count protocol described in detail in SOP #5 will allow analysis of detection probabilities using time-removal methods (Farnsworth et al. 2002) and distance methods (Buckland et al. 1993).

Sampling Design The conclusions that can be made from a statistical analysis are ultimately influenced by the methods used to collect the data. A common objective of using multiple survey stations is to combine the information from the locations into a regional or pooled estimate of population size or trend. Whether the pooled trend is representative of the intended area of inference will depend on how the survey sites were selected. If the goal is to make an inference about bird populations for an entire park, then all areas within a park must have had an equal chance of being included in the sample (i.e., a probabilistic design). Alternatively, the sample might have been randomly drawn from a subset of areas within a park, such as all areas within a specified buffer distance from roads or trails. In this case, the area of inference would be the portion of the park near roads and not the entire park. In some existing monitoring programs, sites may not have been selected following strict adherence to a probabilistic sampling design and qualifying statements should accompany the statistical analysis. New programs should avoid non-probabilistic sampling. Regardless of the design used, it will be critical to understand how sites were selected and the target population in order to make the correct statistical inference (see SOP #1).

Statistical power – the ability of a design to correctly conclude that a trend is occurring – is also influenced by the sampling design. The number of points, the duration of point counts, the number of revisits, and the plot radius are all features of a sampling design that may influence the ability to detect trends even if an appropriate site-selection technique is used to obtain a representative sample of sites. Statistical power should be periodically evaluated through simulation and retrospective power analysis to evaluate the adequacy of the design and possibly suggest design modifications.

Analyses

Breeding Bird Composition At the end of each field season, bird community composition and distribution for each park should be summarized using a variety of tables and graphs, which may include: 1. A field season summary that describes the year’s accomplishments, such as number of points surveyed, number of species observed, and number of participants; 2. A table of all species observed with total number of observations; 3. A table of observed counts of individuals aggregated into life history guilds; 4. Table of species by vegetation class; 5. Graphs of annual total abundance (corrected for effort) and frequency of occurrence (proportion of sites where species was detected) for each species and life history guild; 6. Reports of rare species or unusual occurrences. These can be forwarded to state natural heritage programs.

Vegetation Data 1. Summarize vegetation data collected relative to number of points in each class. 2. Calculate means and variances, or frequency distributions, depending upon whether the data are continuous, ordinal, or categorical. The vegetation data can be summarized by management unit, species or groups of species, depending upon objectives. 3. More detailed habitat analyses can be carried out periodically (e.g., every 3-5 years) using appropriate statistical methods, if desired. See habitat analyses below.

Population Trend and Habitat Analyses More complex analyses of trend and habitat associations should be carried out periodically (e.g., every 3-5 years) using appropriate statistical methods. In most cases, consultation with a qualified statistician will be needed at the time of the analysis, as statistical methods and software evolve rapidly. Alternatively, a statistician can be contracted to conduct the analysis.

There is well-developed literature on trend analysis of bird data to draw upon when deciding on the analytical approach (Ralph et al. 1995, Nur et al. 1999). Thomas (1996) summarized the wide variety of statistical approaches for evaluating bird population trends, which generally boil down to regression approaches that model population size (or an index) over time. Differences arise in how the methods incorporate covariables, the assumed distribution of residuals, variance estimation technique, and weighting approaches. More recently, MacKenzie et al. (2002) described a technique for evaluating trends in site occupancy rather than population size. There is currently a major emphasis being placed on developing statistical tools for incorporating detection probabilities into trend analysis (Buckland et al. 1993, Farnsworth et al. 2002, Thompson 2002, Bart et al. 2004). Regardless of the trend analysis method that is chosen, there will be three critical issues to address: 1) how to integrate the information from each of the Great Lakes Network parks into a regional estimate of trend, 2) how to account for sources of variability, and 3) how to deal with heterogeneity in detection probability.

There is also a long history of analytical approaches that develop relationships between bird abundance or occupancy and habitat characteristics (Morrison et al. 1992, MacKenzie et al. 2002, Scott et al. 2002). Usually, habitat is characterized by measurements of vegetation or land

use. Regression techniques (e.g., linear, logistic, Poisson) are the most common traditional method for developing habitat relationships (Morrison et al. 1992), but other accepted methods include indicator value analysis (Dufrene and Legendre 1997), classification and regression trees (De'ath and Fabricius 2000), and ordination techniques (McCune et al. 2002). Habitat analyses may use predictors summarized on varying spatial scales, for example micro-scale vegetation measurements or land-use in larger spatial buffers. Periodic assessments of bird population change in relation to changes in vegetation structure and composition will provide insight into the relationships among ecosystem components and improve our understanding of breeding bird- habitat relationships and the effects of large scale perturbations.

Birds as Indicators of Ecosystem Integrity Birds are often used as indicators of the health or integrity of an ecosystem. Healthy ecosystems have a “signature” set of bird species, while degraded systems have higher numbers of generalist species and fewer specialist species (Browder et al. 2002). The role of birds as indicators of ecosystem health has been investigated in specific habitats and ecoregions (Brooks et al. 1998, O'Connell et al. 2000, Glennon and Porter 2005, Gardali et al. 2006, Hanowski et al. 2007). An analysis of the bird community from this perspective is useful if there is a known benchmark for comparison with the target data set. For example, plant ecologists have assigned “coefficients of conservatism” (C-values) to plant species, and they conduct floristic quality assessments to rank wetland restorations and compare them with natural wetlands (Lopez and Fennessy 2002, Mushet et al. 2002).

Ideally, a scoring system could be applied to bird species that would integrate their level of specialization; their density or occupancy at a site; and perhaps trends in abundance, density, and occupancy, resulting in an overall site-specific “bird quality assessment” score. Howe et al. (2007a, 2007b) developed a probabilistic indicator of ecosystem condition by integrating individual bird responses to landscape disturbance into an overall bird assemblage metric. While this approach requires calibration for a particular ecological system, it is transparent, fairly simple to implement, and could be applied to any region.

Literature Cited

Anderson, M. J., and A. A. Thompson. 1921. Multivariate control charts for ecological and environmental monitoring. Ecological Applications 14:1921-1935.

Bart, J., K. P. Burnham, E. H. Dunn, C. M. Francis, and C. J. Ralph. 2004. Goals and strategies for estimating trends in landbird abundance. Journal of Wildlife Management 68:611- 626.

Brooks, R. P., T. J. O'Connell, D. H. Wardrop, and L. E. Jackson. 1998. Towards a regional index of biological integrity: The example of forested riparian ecosystems. Environmental Monitoring and Assessment 51:131-143.

Browder, S. F., D. H. Johnson, and I. J. Ball. 2002. Assemblages of breeding birds as indicators of grassland condition. Ecological Indicators 2:257.

Buckland, S. T., K. P. Burnham, D. R. Anderson, and J. L. Laake. 1993. Density Estimation Using Distance Sampling. Chapman and Hall, London, UK.

De'ath, G., and K. E. Fabricius. 2000. Classification and regression trees: A powerful yet simple technique for ecological data analysis. Ecology 81:3178-3192.

Dufrene, M., and P. Legendre. 1997. Species assemblages and indicator species: The need for a flexible asymmetrical approach. Ecological Monographs 67:345-366.

Farnsworth, G. L., K. H. Pollock, J. D. Nichols, T. R. Simons, J. E. Hines, and J. R. Sauer. 2002. A removal model for estimating detection probabilities from point-count surveys. Auk 119:414-425.

Gardali, T., A. L. Holmes, S. L. Small, N. Nur, G. R. Geupel, and G. H. Golet. 2006. Abundance patterns of landbirds in restored and remnant riparian forests on the Sacramento River, California, USA. Restoration Ecology 14:391-403.

Glennon, M. J., and W. F. Porter. 2005. Effects of land use management on biotic integrity: An investigation of bird communities. Biological Conservation 126:499-511.

Guthrie, B., T. Love, T. Fahey, A. Morris, and F. Sullivan. 2005. Control, compare and communicate: Designing control charts to summarise efficiently data from multiple quality indicators. Quality and Safety in Health Care 14:450-454.

Hanowski, J., N. Danz, R. W. Howe, and G. J. Niemi. 2007. Consideration of geography and wetland geomorphic type in the development of Great Lakes coastal wetland bird indicators. Ecohealth 4:194-205.

Howe, R. W., R. R. Regal, J. Hanowski, G. J. Niemi, N. P. Danz, and C. R. Smith. 2007a. An index of ecological condition based on bird assemblages in Great Lakes coastal wetlands. Journal of Great Lakes Research 33:93-105.

Howe, R. W., R. R. Regal, G. J. Niemi, N. P. Danz, and J. Hanowski. 2007b. Probability-based indicator of ecological condition. Ecological Indicators 7:793-806.

Johnson, D. H. 2008. In defense of indices: The case of bird surveys. Journal of Wildlife Management 72:857–868.

Lopez, R. D., and M. S. Fennessy. 2002. Testing the floristic quality assessment index as an indicator of wetland condition. Ecological Applications 12:487-497.

MacKenzie, D. I., J. D. Nichols, G. B. Lachman, S. Droege, J. A. Royle, and C. A. Langtimm. 2002. Estimating site occupancy rates when detection probabilities are less than one. Ecology 83:2248-2255.

McCune, B., J. B. Grace, and D. L. Urban. 2002. Analysis of Ecological Communities. MJM Software Design, Gleneden Beach, Oregon.

Mohammed, M. A., P. Worthington, and W. H. Woodall. 2008. Plotting basic control charts: Tutorial notes for healthcare practitioners. Quality and Safety in Health Care 17:137-45.

Morrison, L. W. 2008. The use of control charts to interpret environmental monitoring data. Natural Areas Journal 28:66-73.

Morrison, M. L., B. G. Marcot, and R. W. Mannan. 1992. Wildlife-Habitat Relationships: Concepts and Applications. University of Wisconsin Press, Madison.

Mushet, D. M., N. H. Euliss, and T. L. Shaffer. 2002. Floristic quality assessment of one natural and three restored wetland complexes in North Dakota, USA. Wetlands 22:126-138.

Nichols, J. D., L. Thomas, and P. B. Conn. Unpublished paper. Inferences about landbird abundance from count data: Recent advances and future directions. (Online). http://www.creem.st-and.ac.uk/len/papers/NicholsSub.pdf.

Nur, N., S. L. Jones, and G. R. Geupel. 1999. A statistical guide to data analysis of avian monitoring programs. U.S. Department of the Interior, Fish and Wildlife Service, Washington, D.C.

O'Connell, T. J., L. E. Jackson, and R. P. Brooks. 2000. Bird guilds as indicators of ecological condition in the central Appalachians. Ecological Applications 10:1706-1721.

Ralph, C. J., S. Droege, and J. R. Sauer. 1995. Monitoring bird populations by point counts. General Technical Report PSW-GTR-149. USDA Forest Service, Albany, California.

Ramsey, F. L., and J. M. Scott. 1981. Tests of hearing ability. Studies in Avian Biology 6:341- 345.

Sauer, J. R., and S. Droege, editors. 1990. Survey designs and statistical methods for the estimation of avian population trends. U.S. Fish and Wildlife Service, Biological Report 90(1).

Sauer, J. R., B. G. Peterjohn, and W. A. Link. 1994. Observer differences in the North American Breeding Bird Survey. Auk 111:50-62.

Scandol, J. P. 2003. Use of cumulative sum (CUSUM) control charts of landed catch in the management of fisheries. Fisheries Research 64:19-36.

Scott, J. M., P. J. Heglund, M. L. Morrison, J. B. Haufle, M. G. Raphael, W. A. Wall, and F. B. Samson. 2002. Predicting Species Occurrences: Issues of Accuracy and Scale. Island Press, Covelo, California.

Thomas, L. 1996. Monitoring long-term population change: Why are there so many analysis methods? Ecology 77:49-58.

Thompson, W. L. 2002. Towards reliable bird surveys: Accounting for individuals present but not detected. Auk 1991:18-25.

Thompson, W. L., G. C. White, and C. Gowan. 1998. Monitoring Vertebrate Populations. Academic Press, San Diego, California.

Standard Operating Procedure #10: Reporting

Please cite this document as:

Danz, N., M. Knutson, B. Route, and T. Gostomski. 2010. Standard operating procedure #10: Reporting. In Gostomski, T., M. Knutson, N. P. Danz, B. Route, and T. W. Sutherland. 2010. Landbird monitoring protocol, Great Lakes Inventory and Monitoring Network. Natural Resource Report NPS/GLKN/NRR—2010/225. National Park Service, Fort Collins, Colorado.

Contents Page

Introduction and Overview ...... 1

Report Types ...... 1

Annual Reports ...... 1 Analysis and Synthesis Reports – Trends and Habitat Relationships ...... 2 Protocol Review Reports ...... 3 Scientific Journal Articles and Book Chapters ...... 3 Presentations at Symposia, Meetings and Workshops...... 4

iii

Introduction and Overview

This SOP defines the different types of reports and reporting schedules for purposes of summarizing and interpreting bird point count data. Refer to SOP #9 for a description of how to summarize and analyze the data.

The following are the different types of reports that summarize and interpret bird survey information:

• Annual reports • Analysis and synthesis reports – trends and habitat relationships • Protocol review reports • Scientific journal articles and book chapters • Presentations at symposia, meetings, and workshops

The park’s monitoring coordinator (PC) is responsible for assuring compliance at the park level with reporting procedures, including statistical or peer review or review by interested partners. PCs will also distribute, post, and archive their individual park reports, and coordinate with their supervisors as appropriate to integrate results into resource management decisions.

The Network program manager will work with park PCs and a statistician to analyze the data on a regional scale at five-year intervals. The Network program manager will assume primary writing responsibilities for synthesis reports, but may also work with park staff to have them write sections relevant to their individual parks. The Network program manager, in this case, will also be the editor with responsibilities for ensuring the final report is consistent in both style and voice. A draft of the report will be distributed for peer review, and the final version will be published, at a minimum, in the National Park Service Natural Resource Technical Report series. Publication of the data in a peer-reviewed journal will be considered and a manuscript submitted if appropriate.

At a minimum, an interim report will be produced at the close of each field season at both the park and Network levels. Park reports will include results of the interim data analysis and interpretation relative to issues raised by the management question(s) at hand. If data analysis is deferred for one or more years, the interim report will include a field season summary and identify any issues to be addressed prior to the next season (see SOP #7).

Report Types

Annual Reports

Overview An annual report is required for all surveys, and data analysis and report writing are typically accomplished after the field season. Purposes of the Annual Reports are to: • Document monitoring activities and archive data for the year (see SOP #7),

• Describe the current condition of the resource and provide alerts if data are outside bounds of known variation (see SOP #9), • Provide information about bird populations and their habitats associated with management actions and decisions, and • Document changes in monitoring protocols.

Prerequisites SOP #7 describes how to summarize the field season activities. SOP #9 describes various tables and graphs that summarize patterns in bird community composition and distribution. The narrative, tables, and graphs will be imported into an annual report.

Instructions 1. Parks will enter their data into their individual databases and then submit an Excel file to the Network office for inclusion in the Network database.

2. PCs may prepare an annual report that summarizes the field season and describes patterns of bird composition, distribution, and change for their respective parks. A hard copy of the report will be archived at the park and an electronic copy sent to the Network office.

3. PCs will meet with their supervisors annually to discuss how the survey results may be communicated to superintendents and other administrators with an eye toward improving management practices. Any needed modifications to the survey design or protocol should also be discussed at this time.

Analysis and Synthesis Reports – Trends and Habitat Relationships

Overview Periodic analysis and synthesis reports will be prepared every 3-5 years. Purposes of the analysis and synthesis reports are to: • Evaluate patterns/trends in condition of resources being monitored, • Discover new characteristics of resources and correlations among resources being monitored, • Analyze data to determine amount of change that can be detected by this type and level of sampling, • Interpret data for the park within a regional or national context, and • Recommend changes to management of resources (feedback for adaptive management).

Instructions 1. PCs and their supervisors should plan and budget time for preparation of an analysis and synthesis report at least every five years.

2. Network program manager will work with a statistician to analyze the data and create tables and graphs. The analysis and synthesis report will include documentation of statistical methods, results, and discussion of population trend analysis and/or bird-habitat relationships (see SOP # 9).

3. Network program manager will take the lead in writing the report but will work with park staff to share these duties. Park staff will be responsible for writing sections of the report relative to their individual parks. Network program manager will serve as the editor, with responsibility for ensuring consistent style and voice.

4. Network program manager will arrange for appropriate peer review. A copy of the final report will be provided to each park.

5. PCs will meet with their supervisor to discuss how the survey results can be used to improve management practices or decisions. Any needed modifications to the survey design or protocol will be discussed at this time.

Protocol Review Reports

Overview A protocol review will be conducted every five years. Purposes of the protocol review reports are to: • Formally review monitoring operations and results, • Review protocol design and products to determine if changes are needed and to document those changes, and • Be part of quality assurance by incorporating a peer review process.

Instructions 1. For the Network protocol, the Network program manager will plan and budget for a protocol review approximately every five years, beginning five years after the initial protocol is approved.

2. The Network program manager will follow SOP #11 in incorporating any changes recommended by the protocol review process.

Scientific Journal Articles and Book Chapters

Overview The National Park Service encourages periodic publication of monitoring program results in scientific journal articles and book chapters. Purposes of these peer-reviewed publications are to: • Document and communicate advances in knowledge, and • Be part of quality assurance by incorporating a peer review process.

Instructions 1. PCs and their supervisors should plan and budget time for preparing manuscripts suitable for submission to peer-reviewed scientific publications, as findings warrant.

2. The Network program manager will coordinate the preparation and submission of peer- reviewed publications that include data from all the Network parks. PCs will participate in this process, either as writers or as reviewers of the draft manuscript.

3. PCs can prepare and submit manuscripts about their own park data without involvement of the Network program manager. In such cases, the PC will prepare manuscripts and submit them to his/her supervisor for review and submission to an appropriate publication outlet.

Presentations at Symposia, Meetings and Workshops

Overview The National Park Service encourages participation in biological symposia, meetings and workshops, at regular intervals. Purposes of participation in these activities are to: • Review and summarize information on a specific topic or subject area, • Communicate latest findings with peers, and • Help identify emerging issues and generate new ideas.

Instructions 1. PCs and their supervisors should plan and budget for staff to present park landbird monitoring data at symposia, meetings and workshops.

2. The Network program manager and his/her supervisor should plan and budget for presentations of Network landbird monitoring data at symposia, meetings and workshops.

Standard Operating Procedure #11: Revising the Protocol

Please cite this document as: Knutson, M, N. Danz, T. Sutherland, and T. Gostomski. 2010. Standard operating procedure #11: Revising the protocol. In Gostomski, T., M. Knutson, N. P. Danz, B. Route, and T. W. Sutherland. 2010. Landbird monitoring protocol, Great Lakes Inventory and Monitoring Network. Natural Resource Report NPS/GLKN/NRR—2010/225. National Park Service, Fort Collins, Colorado.

Introduction

It will probably be necessary to revise either or both the protocol narrative or/and specific SOPs over time. Careful documentation of changes to the protocol is needed to maintain consistency in data collection and for appropriate treatment of the data during summary and analysis. Additionally, all versions of the narrative and SOPs will be archived in a Protocol Library maintained by the Network office.

Revision Process

A review of the protocol will be scheduled at least every five years by the Network’s landbird monitoring program manager, unless changes are warranted before that time. The protocol will also be revised if new research describing improved methods for monitoring landbirds seems relevant to how the Network parks collect data.

1. Changes in the protocol will be documented in the Revision History Log. Any implications for comparing historic and future data will be considered at that time.

2. Parks using the protocol will annually evaluate the usefulness of the survey and the application to management decisions. It may be that the survey is no longer needed or should be modified or replaced with more useful monitoring. a. If it is determined that modifications to the protocol are needed to better meet management objectives, the park monitoring coordinator will consult with the Network landbird monitoring program manager. b. If the Network landbird monitoring program manager agrees that the protocol should be modified, s/he will consult with appropriate statistical or other experts to determine how to implement any changes to the protocol to avoid difficulties with future interpretation of the data. These procedures will be documented in the revised protocol. When the Network’s landbird monitoring database comes online, any changes to the protocol should also be reflected in the database and clearly documented (SOP #8, Data Management) c. The Network landbird monitoring project manager will explicitly document the changes in the revised protocol and the rationale for them in a Protocol Review Report (SOP #10), and file the report with the original protocol. The rationale should include all issues driving the revision(s), whether biological or logistical.