The influence of colour-rings on recovery rates of Herring and Lesser Black-backed Gulls

C. B. Shedden, P. Monaghan, K. Ensor and N. B. Metcalfe

Ringing and Migration. (1985) 6: 52-54.

Comparison of recovery rates of Herring and Lesser Black-backed Gulls with and without colour-rings in addition to a metal ring shows that colour- ringed birds are twice as likely to be recovered and reported to the British Trust for Ornithology. Implications for studies of mortality rates are discussed.

C. B. Shedden, P. Monaghan, K Ensor and N. B. Metcalfe, Zoology Department, Glasgow University, Clasgow G12 8QQ, UK

Received 18 October 1984; revised and accepted 6 December 1984.



The most commonly used methods of estimating wild bird mortality rates rely upon ringing recoveries, and are based on "ratio" or "maximum likelihood" analyses. An underlying assumption of such analyses is that the probability of a ringed individual of a particular species being reported when it dies is the same for all age classes, and constant between years (Anderson et al. 1981, Lakhani and Newton 1983). Differences between age classes in recovery rates are thus held to reflect differences in mortality rates. This assumption may not always be valid, since behavioural differences between age classes may affect the recovery rate. A further problem may be introduced if the method of marking influences the probability of recovery. GossCustard et al. (1982) suggested that the presence of colour-rings may increase the conspicuousness of corpses and thereby increase the recovery rate, and Harris (1984) reported this to be the case for adult Puffins Fratercula arctica. We report a similar enhancement of the recovery rate of colour-ringed Herring and Lesser Black-backed Gulls Larus argentatus and L. fuscus, and discuss how this affects the interpretation of ringing recoveries.


This study examines the recovery rates of Herring Gulls caught outside of the breeding season (between October and March) at refuse tips in the Strathclyde and Central Regions of Scotland, and Lesser Black-backed Gulls caught in the same area during the breeding season (between March and September). All birds caught were ringed on the tarsus with a British Trust for Ornithology numbered metal ring. A proportion of the birds were additionally marked with individual combinations of either three or four Darvic colour-rings, 15-20 mm. in height. All recoveries were received in the normal manner through the B.T.O. ringing scheme.

To avoid any biases resulting from differential mortality rates between age classes, only birds ringed as adults are considered here. Different proportions of the captured gulls were colour-ringed in different years. It was essential therefore to ensure no further biases were introduced due to birds ringed by the two methods being eligible for recovery over different periods of time. This was done by keeping the recovery period constant. Thus the recoveries of Herring Gulls ringed between October 1978 and March 1979 have been analysed over the period April 1979 to August 1983 and those for birds ringed between October 1979 and March 1980 have been analysed for the period April 1980 to August 1984. In each case the recovery period extended over 53 months. Lesser Black-backed Gulls ringed between March - September from 1979 to 1981 were considered for this analysis only if the recovery was made in the 36 month period after the 30 September in the year of ringing.

To check whether or not there were any differences in mortality rates of birds marked by the two methods, we examined the rate at which we recaptured metal and colour-ringed Herring Gulls during the periods specified above. To ensure an adequate sample size, birds caught throughout central Scotland were included; catches were made at least once a month during this time. Since all of the birds were caught at refuse tips, we have treated males and females separately, to avoid any bias due to differences in the extent to which they use refuse tips, and thus in the probability of their being recaptured (Monaghan 1980). Birds were sexed using the head and bill measurement as specified by Coulson et al. (1983).


Table 1 gives the recovery rates of metal ringed Herring Gulls with and without colour rings, during the 53 month period specified. The differences between the proportions recovered is significant (X2 = 9.1, 1 df, p < .01), with over twice as many of the colour-ringed birds being recovered as compared with the metal ring only birds. There was no significant difference in the recapture rates, and presumably therefore mortality rates, in relation to marking method for either males or females (Table 2).

TABLE l.The proportions of Herring Gulls ringed as adults with metal rings, and metal plus colour rings. which were recovered in a 53 month period .

No. recovered No. ringed

Colour-rings 55 (7.5%) 730

No colour-rings 20 (3.6%) 561

TABLE 2. The proportions of male and female Herring Gulls ringed as adults with metal rings,and metal plus colour-rings, which were recaptured in a 53 month period. There were no significant differences in the recapture rates for either males or females (males X2 = 0.16, 1 df, NS: females X2 = 0.02, 1 df, NS).


No. recaptured No. ringed No. recaptured No. ringed

Colour-rings 9 (2.3%) 398 14 (3.0%) 465

No colour-rings 4 (3.4%) 117 9 (3.5%) 258

Table 3 gives the recovery rates of metal ringed Lesser Black-backed Gulls with or without colour rings, during the 36 month period specified. Again, almost twice as many of the colour-ringed birds were recovered as compared with the metal only birds, though in this case, due to the smaller sample size, the difference is not statistically significant (X2 = 2.4, 1 d.f.).


These data clearly demonstrate that the presence of colour-rings greatly enhances the proportion of ringed, dead birds which are reported to the British ringing scheme. This is presumably because the presence of colour- rings attracts the attention of the casual observer to the dead bird, and thus increases the probability of the metal ring being noticed and reported. These findings have implications for the calculation of mortality rates based on ringing recoveries. For example, if differing proportions of birds have been colour-ringed during two time periods, a comparison between the two periods may lead to the erroneous conclusion that mortality rates have changed, when it is in fact the recovery rate which has altered.

TABLE 3. The proportions of Lesser Black. backed Gulls ringed as adults with metal rings - and metal plus colour rings, which were recovered n a 36 month period.

No. recovered No. ringed

Colour rings 22 (6.38%) 345

No colour rings 7 (3.35%) 209

Similarly, if the proportion of birds of different age classes which carry colour-rings is not the same, as could arise due to loss of colour-rings by older birds or differential colour-ringing at the outset of a study this could lead to additional errors in the calculation of mortality rates. Thus care should be taken by ringers to ensure that comparisons of recovery rates are made only between birds marked in the same manner.


We would like to thank the British Trust for Ornithology for use of recovery details, District Authorities in central Scotland for access to rubbish tips, all those who assisted us in catching gulls, and Drs. R. W. Furness and J. C. Coulson for comments on an earlier draft of the manuscript.


ANDERSON, D. R., WYWIALOWSKI, A. P. and BURNHAM, K. P. 1981. Tests of the

assumptions underlying life table methods for estimating parameters from

cohort data. Ecology 62: 1121-1124.


and SHEDDEN, C. B. 1983. The use of head and bill to sex live gulls Laridae.

Ibis l25: 549-557.


1982. Age structure and survival of a wintering population of Oystercatchers.

Bird Study 29: 83-98.

HARRIS, M. P. 1984. Movements and mortality patterns of North Atlantic Puffins

as shown by ringing. bird Study 31: 131-140.

LAKHANI, K. H. and NEWTON, 1. 1983. Estimating age-specific blrd survival

ratios from ring recoveries - can it be done? J. Anim. Ecol. 52: 83-92.

MONAGHAN, P. (1980). Dominance and dispersal between feeding sites in the

Herrng Gull (Larus argentatus). Anim. Behav. 28:521-527.