Bird Count Estimator

Corrected density divides the same average bird count by a smaller area than raw density does. Raw density uses the full circle defined by your entered detection radius, while corrected density uses the effective detection radius — that same radius scaled down by the square root of your detection probability (which is always 1 or less). Dividing the same bird count by a smaller area always produces a bigger density figure, which is the calculator's way of accounting for birds present but missed during the count.

Effective detection radius equals your entered detection radius multiplied by the square root of your detection probability (entered as a percentage, converted to a 0–1 fraction). So a 100m radius with a 50% detection probability gives an effective radius of about 71m, and a 100m radius with 100% detection probability gives back the full 100m radius unchanged. This is a simplified stand-in for the half-normal detection function formal distance-sampling software fits from your actual observed detection distances.

It assumes your counts follow a Poisson distribution, meaning the variance of the count equals its mean — a standard assumption for count data of rare or randomly distributed events. Standard error is calculated from that variance divided by your number of stations, so adding more point-count stations shrinks the interval width (and the coefficient of variation) even if the average birds-per-station count stays the same.

Sampling adequacy is judged purely by two thresholds — number of stations and number of replicate visits — regardless of how many birds you actually observed or how tight your confidence interval came out. You need at least 40 stations and 3 replicates for "Excellent," at least 20 stations and 2 replicates for "Good," and at least 10 stations for "Adequate"; anything below 10 stations is always "Insufficient" no matter what your bird counts looked like.