3.3 Presence/Not Detected and Relative Abundance

3.3.1 Call Playback

A call playback/call response survey is the most economical monitoring survey for bitterns and rails. Call playback surveys improve detectability of secretive-species compared to passive (look-and-listen) observations, or detection through responses to disturbances such as clapped hands or rocks thrown into the water (Glahn 1974; Gibbs and Melvin 1993). As noted in the species life history descriptions, call playback has been used successfully in monitoring populations of most marsh-inhabiting birds. Visual surveys have a high degree of detection error due to the densely vegetated habitat and secretive nature of the birds.

Advantages

• Call playback surveys will increase the chances of locating birds that may otherwise remain quiet and undetected.
• Call playback stations can be established at the edges of marsh habitats, thus reducing the risk of damaging habitat by trampling the vegetation.

Disadvantages

• Call playback does not sufficiently sample females or young in a population, therefore it only provides an index of a breeding population.
• The relationship between the number of calling or responding birds and the actual breeding population is unknown.
• The use of tapes and broadcast equipment introduces an additional source of variation (i.e., in addition to observers, weather etc.).

Office Procedures

• Review the section, Conducting a Wildlife Inventory, in the introductory manual Species Inventory Fundamentals No. 1. For further discussion on statistical power in waterbird surveys refer to Gibbs and Melvin (1997).
• Obtain relevant maps of the project area (e.g., 1:50 000 air photo maps, 1:20 000 forest cover maps, 1:20 000 TRIM maps, 1:50 000 NTS topographic maps).
• Outline the Project Area on a map and determine Biogeoclimatic zones and subzones, Ecoregion, Ecosection, and Broad Ecosystem Units for the Project Area from maps.
• Delineate one to many Study Areas within this Project Area. Study Areas should be representative of the Project Area if conclusions are to be made about the Project Area. For example, this means if a system of stratification is used in the Sampling Design then strata within the Study Areas should represent relevant strata in the larger Project Area.
• Generally, establish Study Areas by stratifying the Project Area by size of wetland (see Habitat Description for definition). One Study Area may consist of one or more marsh habitats (see Glossary for definition). Recent colour aerial photographs of the habitat that will be surveyed will be especially useful (see Sampling Standards above).
  For each Study Area:
• Determine Biogeoclimatic zones and subzones, Ecoregion, Ecosection, and Broad Ecosystem Units for areas surrounding the wetland. Determine wetland type using the Riparian Management Area Guidebook (B.C. Min. Environ. and For. Serv. 1995) as a general reference.
• Determine the size of each marsh habitat by estimating the area covered by non-woody emergent vegetation.
• Identify access points and potential call playback stations for surveys from large scale, colour aerial photographs.
• Review the life histories of the marsh birds that may potentially be in your Study Areas. Since marsh birds have generally not been studied in B.C., use the known range only as a general guideline, and rely more on general habitat descriptions for each species as an indication of potential presence.
• Prepare a survey tape of the target species' calls.
• Tapes should include, for each species, three series of 20 seconds of calls (c) followed by 30 seconds of silence (s), (except after the last call in a sequence, which should have one minute of silence before the next species), [sequence 20(c1)-30(s1)-20(c1)-30(s1)-20(c1)-60(s1)-20(c2)-30(s2)-etc.]. Each species can be recorded consecutively on one tape.
• If you are recording from a commercial CD or tape, do not include the announcement of the species!
• For comparable results, broadcast tapes should be identical in terms of duration, sequence and quality of calls. Some birds (e.g. B-VIRA) will reply to calls of other species, as well as conspecifics, and so whether they are first or last on the tape will influence the duration of time available for a response. Although it has not been well explored, it is also possible that loud calls (e.g. B-AMCO) may suppress calling of quieter species (e.g. B-AMBI and B-LEBI). It is probably best to be conservative and play calls from quietest species to loudest.
• Tapes degrade with exposure and should be replaced at least once per field season.
• Test the call playback equipment to ensure that it works (see Call Playback Equipment section below).
• Complete the following Data Forms that are included in the Species Inventory Fundamentals No. 1 [Forms] (previously referred to as the Dataform Appendix):
• Wildlife Inventory Project Description Form.
• Wildlife Inventory Survey Description Form - General.
• Locate capable Field Personnel.

Sampling Design

• Superficially evaluate the wetlands within the Project Area. Gibbs and Melvin (1993) suggest stratifying effort among large, typically species-rich, wetlands and wetlands that are smaller, of marginal habitat quality, or were historically occupied.
• Randomly choose Study Areas to be surveyed from each of the different strata. The objective here is to strike a balance between maximizing species detections and surveying a sample of habitats which are representative of the Project Area (Gibbs and Melvin 1993).
• For each marsh habitat within a Study Area, establish Point Count Stations systematically, either throughout or along the edges of the marsh habitat(s) of interest. To avoid replication of counts, stations are placed a minimum of 250 m apart. This also means that stations will be placed at approximately one station for every 5 ha of marsh habitat, a sampling intensity suggested by Gibbs and Melvin (1993).

Sampling Effort

• Each Point Count Station must be visited at least three times during the peak of the breeding season. To ensure sufficient coverage throughout the breeding season, survey visits should be separated by a minimum of 10 days.
• Samples are based on a 100 m radius plot. Depending on whether the observer is at the edge of a wetland or within it, s/he may sample either a complete circular plot (3.14 ha) or a semi-circular plot (1.57 ha), either of which must be predominately marsh habitat.
• Gibbs and Melvin (1993) surveyed approximately 10-15 Point Count Stations of 10 minutes in 4.5 hours. Many of these were done from a canoe while others were completed on foot.
• The number of Point Count Stations sampled in a marsh habitat is based on a sampling intensity of one station for every 5 ha (see above), to a maximum of 10 stations.

Personnel

• All field personnel should be familiar with identifying features and life histories of the species described in this manual. To avoid mis-identification, personnel should also be familiar with other marsh birds. The project biologist should train all observers with appropriate books, tapes, study skins, etc.
• At least one person should be familiar with the collection of habitat data. This will require someone who is familiar with wetland plants and wetland classification.

Equipment

• Several copies of Point Count Data Form - Marsh Birds on waterproof paper attached to a wooden clipboard (metal ones are too noisy when handled and may scare the birds)
• Several copies of appropriate forms for habitat description (see Species Inventory Fundamentals No. 1)
• Nest Site Description Forms (you may find nests during your surveys; follow the RIC guidelines for completing these forms that are found in the Species Inventory Fundamentals No. 1 [Forms])
• Small kayak, canoe, or inflatable if traveling through wetland to do surveys
• Puncture-resistant hip or chest waders
• 8-10x wide angle binoculars
• Compass
• Bright flash light (headlamp preferable) for nocturnal surveys
• Black permanent ink marker to mark flagging tape and posts
• 3 m sections of ½" electrical conduit piping to mark Point Count Stations (one for each station)
• Blaze orange flagging tape, or reflective tape to mark stations so they can be relocated at night
• 35 mm camera for photodocumentation of the habitat
• Call playback equipment (waterproofing the equipment may be a good idea) -see below

Call Playback Equipment

The broadcast of each species' calls should be heard by birds 100 m from the point of broadcast. Gibbs and Melvin (1993) utilized a standard maximum sound pressure of 80 dB at one metre from the source. A cruder means of assessing sound level is to subjectively evaluate the sound level at 100 m from the speaker, but there is no way of assuring that you will be hearing what the birds will be hearing. A small portable tape player with two speakers (i.e., a portable stereo) should be sufficient for the purpose of these surveys. An alternative is to use a walkman attached to an amplified speaker. Describe whatever equipment is used by brand name, model name, and power of sound output (these specifications can be found in the owner's manual or from the dealer). Play calls at a maximum volume where there is minimal tape-noise in the background. This usually seems to be a step below maximum output. To keep your hands free during the survey you may want to attach the tape player and speaker to a strap and hang them around your neck.

Field Procedures

• Visit your proposed field routes during daylight to familiarize yourself with the habitat features and to see if there will be any particular transportation problems both to and through the Study Area. Determine if the wetland can be sufficiently sampled along its edge, or if a boat will be required for full coverage. Visit the survey sites prior to the peak of the breeding season, but after plants have emerged so habitat data can be completed. Do not try to do both at the same time!
• Establish the Point Count Stations as described below.

Establishing a Route and Marking the Stations

Take the time before conducting the surveys to establish the call playback stations and travel routes. Each Point Count Station (Call Playback Station) has to be clearly identified to aid relocation both within and between seasons. Remember that you may not be the one to relocate stations, so the location of each must be clearly described to anyone who is unfamiliar with the area. Recording only UTM coordinates (using NAD83) of each plot is useful, but generally insufficient alone because this will mean that future field personnel will need a differential GPS to relocate plots within 10 m. A few well-worded comments can be very helpful.

1. Establish a fixed marker at the initial point of access. Indicate this point on TRIM mapsheets, and aerial photographs if it improves relocation by other field personnel.
2. Once you reach marsh habitat, select a representative Point Count Station that has a 100 m radius:

• Position the station so that the maximum amount of marsh habitat is being sampled. Based on this, you will need to decide whether to count the station as a full plot (3.14 ha) or a half plot (1.57 ha). Since many wetlands can be surveyed from the roadside, it is convenient to count these as half plots. In these situations, only half of a circular plot may be marsh habitat, while the other half of the plot behind you may be road or forest. Situations like this can be accommodated since the measure of abundance will be standardized to the number of responding birds per full plot. It is important to realize however, that sampling along the edge of a wetland does have some disadvantages, including a lower ratio of effort:coverage and a bias toward species at the periphery of wetlands as opposed to the interior.
• Situate the Point Count Stations so that as much as possible of the 100 m radius sample area can be seen and to minimize obstructions to hearing. Slightly elevated points with an unobstructed view are good locations.

3. Permanently mark the station:

• Since each Point Count Station will be re-visited, the centre must be marked permanently with a metal or wooden stake so the stations can be relocated. Wooden 2x2" stakes can be used, but a better alternative is ½" metal electrical conduit piping. Three metre long stakes can be firmly pushed into the marsh bottom, and should be able to withstand wind, wave and ice activity (Marsh Monitoring Program 1997). A less permanent marker may be used if the station will only be visited for one season.
• Point Count Station identification numbers can be inscribed on aluminum tags that are used commonly in forestry cruise plots. Identification numbers should include a code for the Study Area, plus consecutively marked stations beginning with A (e.g., the first station in Whitty's Marsh Study Area would be WHIT-A). Attach the metal tags to the pole with binding wire, and attach some reflective flagging tape so the plot can be relocated for nocturnal surveys. With a black felt pen (e.g., Sharpie), label each stake with the same individual plot identification number that you put on the aluminum tag. This marking will only last one season, but will be a "back-up" to the aluminum tag and can often be read at a distance (Marsh Monitoring Program 1997).

4. To help avoid replication of counts, additional Point Count Stations must be separated by a minimum of 250 m. This can also be minimized by paying attention to where individual birds call from and how they may move in response to a broadcast.

Describing the Habitat

Habitat must be described according to guidelines specified in Species Inventory Fundamentals No. 1.

• Prepare photodocumentation of each site following the guidelines described in Photodocumentation for Aquatic Inventory (Resources Inventory Committee 1996). The intense level of photodocumentation described in that manual will not be required for habitat description, but it does provide some good guidelines to follow for description of marshes.
• Ensure that your plot is predominantly in marsh habitat (see Glossary for definition). To meet this requirement you may choose to use a half plot instead of a full plot or re-position your Point Count Station.
• Include the following in the description of the habitat around each Point Count Station:
• Percent cover of the dominant non-woody emergent vegetation found in the plot.
• Mean water depth within the 100 m radius plot.
• It is useful to draw a map of the habitat within the plot on the Point Count Form - Marsh Birds.

Conducting the Survey

• Marsh bird surveys will be conducted between 20 May and 5 July throughout B.C. Repeat surveys should be separated by at least 10 days (Marsh Monitoring Program 1997).
• Begin your survey half an hour before dawn and continue until no later than four hours after sunrise. Evening surveys should start at twilight and finish shortly after dark. Nocturnal surveys should begin within one hour after dark and finish before sunrise. Travel to your first Point Count Stations quietly.
• Each count lasts 10 minutes.
• Do nothing for the first two minutes at the station. Give yourself and your surroundings time to settle down now that you have stopped moving. Use this time to listen for any unsolicited calls and to orient the Point Count Data Form-Marsh Birds - this orientation will be used in subsequent surveys. Note all locations of calling birds, or birds that you may see within the plot. List the birds outside of the study plot as incidental sightings. The symbology used on your field data cards should be the same as that used for songbird point count surveys (Resources Inventory Committee 1997b). This standard mapping symbology can also be found in several other sources including Ralph et al. (1993) or other bird survey manuals. Also take this time to record the start time, temperature, cloud cover, wind, etc.
• Hold the speaker at chest height. Broadcast a species' call for 20 seconds, then listen and watch for 30 seconds. Play a series of three calls for each call type. For a full circular (3.14 ha), broadcast the three calls at 60⁰, 180⁰ and at 300⁰ from your line of travel (this means rotating the speaker 120º after each 20 seconds of call of each species). For a half plot (1.57 ha), broadcast the calls at 45⁰, 90⁰ and 135⁰ from the straight edge of the half plot. Map the response locations and sightings as they occur on the Point Count Data Form-Marsh Birds (refer to prior step).
• Wait one minute before playing the series of calls for the next species. Repeat the procedure as above. Ideally, the surveyor should complete call playbacks for all of the species within the 10 minute visit to the plot, leaving two minutes before the end of the survey to listen for calls. This gives enough time to play calls for up to three different species (not including the initial two minute wait before starting broadcasts).
• Be conservative when counting responses from bitterns and rails. Unless they are simultaneous calls, two responses from approximately the same location should be counted as only one individual.
• Be aware that certain birds (especially B-VIRA and B-SORA) will move inward towards a call broadcast. This should make observers cautious about counting a bird at 100 m and then later another at 50 m when both have similar angles to the observer.

Also,

• Due to potential observer bias and variability, rotate observers among stations to ensure that species are not overlooked or misidentified.
• The field crew leader should ensure that field personnel are sufficiently confident in their field identification and data sheet completion. This may require some mentoring on the part of senior field crew.
• Standardize handling of difficult problems such as counts for flocks of birds (if this ever occurs), and potential double counts (e.g., potentially eliciting responses from the same bird at more than one point count station). Stress that observers should pay attention to where individual birds call from and how they may be moving in response to the tape broadcast. This will help to distinguish individuals. Standardize your procedures among all field personnel, and document these procedures in the final report.

Data Analysis and Report Preparation

• Within a 24 h period of completing the survey, transcribe the sightings from the Point Count Data Form - Marsh Birds to the Animal Observation Form - Marsh Bird Call Playback. The Animal Observation Form - Marsh Bird Call Playback is a record of the total number of individuals, by sex and age, heard or observed within 100 m of the Point Count Stations. It will be helpful to cross-off a counted individual from the Point Count Data Form - Marsh Birds with a highlighter. Crossing them out with a pen may make your raw data difficult to read if it is required at a later date.

The standard measure of abundance will be the number of each species seen or heard by sex and age (although admittedly unlikely) within a 100 m radius from the centre (station) of the sample plot. Birds outside of the 100 m radius will be counted on a present/not detected basis. This measure of abundance is similar to the index of abundance used in the Marsh Monitoring Program developed in eastern Canada and the United States of America (Marsh Monitoring Program 1997).

Many of the following suggestions for data analysis and report preparation are adapted from Robbins and Stallcup (1981):

• Start an initial draft of maps, tables and text before the end of the field season so major problems will become apparent. You may find that sampling procedures, sampling intensity or field personnel will have to be changed.
• Compare your results to other studies to be sure that your methods and results are consistent with other procedures.
• Check carefully for errors in transcribing field notes - this is especially important for transcribing the five-letter species codes (e.g., B-YERA or B-YRAI could either be different species, or different codes for Yellow Rail). Ensure consistency of field notes and code usage among observers (see RIC manual No. 2, Vertebrates of B.C. for proper species codes);
• In the report, include a list of all species detected in the Study Area. Include incidental observations, and a measure of survey effort wherever possible.

When submitting your completed data forms, the completed Animal Observation Form - Marsh Bird Call Playback should be accompanied by the required project and survey description data sheets, habitat descriptions, and TRIM map sheets with Point Count Stations plotted. It is expected that photocopies, or print-outs of digitized aerial photographs of the sample sites, with Point Count Stations indicated on the photos will be useful as well.