Recommended method(s):
Call playback/drumming surveys during the breeding season may also be used to collect relative abundance data, especially for those species that are known to respond to call playback, occupy relatively large home ranges and/or are otherwise difficult to detect (refer to Section 2), but its uses are limited to obtaining an index abundance (Thompson et al. 1998).
Wildlife tree surveys for direct and indirect woodpecker sign are recommended for relative abundance data during the non-breeding season when birds are less conspicuous.
Measuring relative abundance is usually based on the number of detections of individual birds per unit of sampling effort (e.g., sampling hours or kilometres walked while sampling). To achieve adequate accuracy and precision among samples, it is essential that: (i) sampling effort is equal among study areas; (ii) observers move at a constant rate; and (iii) that a sufficient number of surveys are conducted. Thus, assuming that no differences exist among observers in detection ability, the main factors that influence bird detectability include habitat structure, weather, season, and inherent behavioural differences among birds.
General considerations: The primary assumptions to consider when conducting relative abundance surveys include:
If these assumptions are addressed appropriately in survey design, then replicate surveys should show (on average) the same relative bias, thus allowing robust calculation of trends and comparison between areas.
Sampling design and sampling effort: To determine relative abundance, a systematic sampling design must be followed. The survey area should be stratified according to the objectives and hypotheses of the study. Refer to the RIC Species Inventory Fundamentals manual for sampling effort required to determine relative abundance.
When using call playback, the number of call stations will vary according to the home range size for the woodpecker species of interest, the density of birds within the habitat and the quality of the broadcasting equipment. Hartwig (1999), found that 15 - 20 call stations were required for Pileated Woodpeckers surveys to ensure an adequate sample size (i.e., standard error began to stabilize) on south-eastern Vancouver Island in the Coastal Western Hemlock biogeoclimatic zone (CWHxm). Appropriate distances for spacing call stations have not been determined for each woodpecker species. A distance of 300 metres between call stations and 800 metres between transects was used by Aubry and Raley (1994) while surveying for Pileated Woodpeckers in Oregon forests. As a general rule, surveys for larger species have greater distances between call stations than small species inventories.
Power Analysis procedures for calculation of sample sizes must be integrated into the study design of all techniques. As described in the RIC Species Inventory Fundamentals manual, Section 2.5 - Sampling Effort, programs such as MONITOR, POWER AND PRECISION, and NQUERY are user friendly, and can be easily used in an adaptive fashion to calculate sample sizes needed for the ultimate analysis questions. Many other software developers are developing Power Analysis packages.
For relative abundance data, the statistical problems will be those concerned with the analysis of count data. Consult the RIC Species Inventory Fundamentals manual, Section 5.3.1 - Distribution of count data, for a detailed discussion of count data analysis. The quantification of sampling intensity and effort is fundamental to the use of indices and relative abundance measures. This way the assumptions of equal bias of surveys between areas and over time can be met. In addition, the usefulness of indices depends on the precision of estimates and standard measures of variance. For further discussion, refer to the RIC Species Inventory Fundamentals manual.
Of particular concern with woodpecker surveys is obtaining adequate sample sizes to allow the monitoring of trends. If a species initially exists at a very low density then in general it requires a great degree of survey effort to ensure precision is high enough to detect trends. For this reason it is essential that biologists conduct a statistical power analysis to determine optimal sampling effort.