Data Analysis

For each Study Area or wetland:

1. If total counts are used:

• Trends in abundance over time. Calculate the number of breeding pairs or broods or non-breeding birds (depending on survey type) of each species per wetland at a given time.
• Each bird count survey of a wetland on a given day is one replicate (all birds counted at all observation stations are summed).
• Sum the number of birds counted at each replicate survey at a wetland by species, then divide by the number of replicate surveys (birds by species/wetland).
• Precision is calculated as the square root of the variance around the mean of replicate counts.
• Comparison of abundance estimates between areas. Calculate abundance estimates for each wetland by dividing the number of birds by species per wetland by the area of the wetland that was sampled (birds/km²).

2. If sample counts are used:

• Trends in abundance over time. Calculate the mean number of birds/station at a given time. Sum the number of each species seen or heard within the set radius from the observation station for all of the stations, then divide by the number of stations sampled.
• Comparison of abundance estimates between areas. Calculate birds/km²: divide the mean number of birds/station by the area of the observation sample station.

3.4.3 Helicopter Surveys

Relative Abundance in remote or inaccessible areas for any species, but especially for breeding pair counts, brood counts, and incubating adults for the most conspicuous species (e.g., geese and swans), or breeding pair and brood counts for loons and dabbling ducks.

Helicopters are recommended to survey breeding pairs and broods in locations inaccessible by road (e.g., Boreal Forest, northwest BC). Generally, the lower flight speed of and better visibility from helicopters than fixed-wing aircraft improve the ability of observers to detect birds (Ross 1985).

Counts by helicopters or from the ground over smaller portions of the study area can be used to derive correction factors for aerial transect surveys.

Estimates are very sensitive to viewing angle, altitude, glare and flight speed therefore it is important to conduct surveys consistently (Caughley 1974; Briggs et al. 1985).

Example: Nixon and Majiski (1991) conducted breeding pair surveys in boreal forest wetlands with low waterfowl densities in the Liard Plain and Teslin River Basin. Wetlands were stratified on 1:50,000 topographical maps into three size classes (<1.5 hectares, 1.5-15 hectares and 15-300 hectares). Larger lakes were excluded because it was not feasible to obtain an accurate count. A sample of wetlands within each size class was chosen randomly but most distant wetlands were omitted due to limited time available for the survey. Observers sampled each wetland by circling it in a helicopter. The helicopter would hover while they recorded the number of birds and wetland characteristics. Counts at each wetland were repeated four times at six day intervals starting in mid- to late- May to ensure an accurate estimate of the number of breeding pairs (McKelvey 1989). The mean number of breeding pairs for each species was calculated using the four repeated surveys for each wetland. The relative abundance of breeding pairs was calculated from stratified samples. The results were used to estimate the density of breeding pairs within the entire Liard Plain and Teslin River Basin region. Stratifying by wetland size did not increase the precision of the population estimates because the sample sizes were too small (9 to 30 in each strata). It may have been better to survey more wetlands in each strata and do fewer replicate counts at each wetland.

Office Procedures

• Review the introductory manual No. 1 Species Inventory Fundamentals.
• 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). Any map which is used to record data should be referenced to NAD83.
• 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 on topographic maps or air photos 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. Generally, a Study Area will be equivalent to a single wetland.
• Determine a suitable sampling design (see next section).
• Determine the number of the Study Areas (wetlands) to be surveyed (sample size).

Sampling Design

• A Project Area may contain one or many wetlands. Each wetland area is considered a Study Area. Each Study Area should be able to be surveyed in one day.
• Depending on the size of the Project Area and number of wetlands (Study Areas) it contains, you may:
• Survey all Study Areas (wetlands) in the Project Area, e.g. for a defined wetland (Study Area) such as an inlet or bay.
• Survey only a selected number of the Study Areas (wetlands) in the Project Area. Note that the size of the wetland in each Study area that is surveyed must be known, or
• If appropriate, assign each Study Area into a stratum based on habitat type or expected bird densities, then select Study Areas (wetlands) from each stratum to be sampled. Note that the size of the wetland in each Study Area that is surveyed must be known.
• For each Study Area (wetland) conduct a total count of birds by species.

Sampling Effort

• Breeding Pair Counts: Each wetland should be surveyed repeatedly during the appropriate daily survey window until required precision is obtained.
• Brood Counts: Each wetland should be surveyed every three days during the appropriate daily survey window until required precision is obtained.
• Non-breeding Bird Counts: Each wetland/field should be surveyed repeatedly until required precision is obtained.

Care must be taken during the counts to avoid biases as much as possible (e.g. consistent over or under counting). When biases are inherent in the counting method, no amount of additional sampling will remove them.

The use of alternating observers may help to average some observer biases. Ideally, if several observers are participating in each survey, they should rotate areas between surveys; otherwise, all sites should be surveyed by one observer during a given survey. The order in which basins are surveyed should be rotated between surveys.

Time of Survey

• Breeding Pair Counts: Crews should survey for breeding pairs after migrants have passed through the area and before males desert incubating females (these times will vary between species).
• Brood Counts: Crews should survey for broods after hatch (these times will vary between species).
• Non-breeding Bird Counts: Surveys of moulting, migrating and wintering birds should be timed according to the chronology of these events for each species (see Table 1).

Personnel

• A pilot with previous aerial survey experience, preferably in the project area. Pilots with no previous survey experience should receive a training flight on a portion of the project area.
• Two highly trained experienced observers with recent practice at estimating numbers.

Equipment

• Helicopter (refer to Simpson et al. 1993 for logistical and financial considerations in choosing appropriate aircraft)
• Topographic maps (1:20,000 or 1:50,000) for the study area
• 1:5000 air photo maps
• Navigational system (GPS NAD83)
• Tape recorders for each observer, extra tape cassettes and extra batteries
• Intercom and headsets
• Binoculars
• Personal gear including watch, ear plugs, anti-air-sickness devices, warm clothing, sunglasses
• Survival gear
• Window-cleaning rag