Definitions Related to Hunter-shot Banding Data

F = Recovery Rate

It is probability that a banded bird is encountered dead and the band turned in for a calendar year given that it was alive at the start of the year. It is the percent of banded birds that are shot and the bands turned in for processing. The base for computation is the population of banded birds.

Example:

10,000 banded mallards alive at start of hunting season .

1,000 bands turned in by hunters who said they took the band off a bird they shot.

F= (1000 / 10,000) x 100 = 10%

This statistic is used as an index to the harvest rate (see below).

r = Band Reporting Rate

It is the percent of the bands encountered by hunters that are actually turned in. Not all bands are turned in! If the band reporting rate were assumed to be 100%, then the recovery rate (F) would equal the harvest rate (see below). The base for computation is the population of bands encountered.

The only way to establish a band reporting rate is through a reward band study. That is, two similar samples of birds are banded (e.g. 2-10,000 bird samples). One sampled is banded with regular bands while the other sample says, for example, REWARD $10.00 for turning it into the banding laboratory. If we assume that all the reward bands are turned in, then a correction factor can be established for those bands encountered by hunters but not turned in:

r = (F_nonreward / F_reward ) x 100

Reporting rates are generally not established. Consequently, reporting rates are assumed to be constant even though they probably vary by geographic area and distance the bird is recovered from the banding site. Special “campaigns” to turn in bands seriously affect band reporting rates.

Example:

10,000 mallards were banded with reward bands.
10,000 mallards were banded with regular bands (nonreward bands)

3,000 of the reward bands were turned in within the next calendar year
1,000 of the regular bands were turned in within the next calendar year

• F_nonreward = (1,000/10,000) x 100 = 10%

• F_reward = (3,000 / 10,000) x 100 = 30%

• Band reporting rate (r) = (10% / 30%) x 100 = 33%

  or

r = (1,000 / 3,000) x 100 = 33%

This says that hunters are actually recovering three times more birds than the recovery rate indicates!

H_r = Harvest Rate

It is the percent of the banded population killed and retrieved by hunters. It includes both reported and unreported bands.

H_r = Recovery Rate / Band Reporting Rate

if F_nonreward = 10% and r = 33%, then H_r = (10 / 33) 100 = 30%

Since band reporting rates are generally not available, annual recovery rates are usually compared to assess the harvest of a particular species, i.e., recovery rate is used as an index to harvest rate. When recovery rates go up, the data imply a bigger harvest. Of course, this is based on the assumptiong that reporting rate is constant.

K_r = Kill Rate

This is the percent of the population that die either directly or indirectly from hunters. It includes birds that are recovered or lost. Lost birds can be either dead on impact or crippled. Some crippled birds may eventually survive and should not be included in the Kill Rate but those that do not survive are lost to: (1) mortal wounds, (2) starvation, or (3) predation.

Example:

10,000 banded mallards alive at start of season.

5,000 banded birds “knocked down” by hunters or mortally wounded.

K_r = (5,000 / 10,000) = 50% kill rate

If we go back to the Reporting Rate example in which 3,000 of the 5,000 knocked down were recovered, then we can calculate a crippling loss for this example:

Crippling loss for population = (5,000 - 3,000) / 10,000 = 20%

Crippling loss for birds "knocked down" = 2,000 / 5,000 = 40%

In many marshes, a crippling loss of 25% is not uncommon (i.e., for every 3 ducks a hunter retrieves, they lose 1 in the marsh).

M_r = Annual Mortality Rate

This is the percent of the population dying during the year from all causes (including hunting, predators, starvation, etc.).

Conclusion

Although the recovery rate (F) suggests how a population is behaving from one year to the next, it is probably a very low estimator of total mortality. By the time you consider natural mortality (which does not include hunter loss), the crippling rate, the reporting rate, and the harvest rate, it adds up to a sizable proportion of the population being removed each year. The question the manager must be constantly concerned about is whether or not the number being removed exceeds the number being added.

Return to Lab Notes : Band Recovery Analysis

Revised August 25, 2011