ALTENBURG JF & VAN BEMMELEN RSA (2021) Can progress in primary moult be used to estimate local turn-over of Eurasian Curlew Numenius arquata at a roost site?. LIMOSA 94 (1): 66-76.

Eurasian Curlews moult their primaries after the breeding season. Several hundreds of Eurasian Curlews overnight at the Everdingerwaard, an important roost site in the mainland of the Netherlands. To estimate the total number of birds using a roost over the course of a season, not only regular roost counts are required, but also estimates of the 'turn-over' of birds between counts. When a substantial part of a population is colour-ringed, turn-over can be calculated from the duration of stay of colour-ringed birds. However, this method cannot be used for Eurasian Curlew, as the proportion of colour-ringed birds is very low (c. 1: 2800 in the Wadden Sea). This study presents a first step in using primary moult score to estimate minimal turn-over between roost counts, by investigating the feasibility of using photographs of birds in flight. The basic idea behind this approach is that arrival or departure of individual birds - even when the number of roosting birds remains the same - should be reflected in the observed moult scores, assuming that arriving birds have different moult scores than birds that were already present. During late summer and autumn of 2017-19, a total of 97 counts were performed and attempts were made to photograph arriving groups. For moult analysis over time, the results of 40 count/photo sessions were available. Curlews moult their primaries from the inside (P1) to the outside (P10). An 'outermost dropped primary index' (GGH) was developed as a measure of the progress of moult, as this allowed to collect a much larger dataset compared to other moult score indexes, which require a detailed view of each primary. GGH basically identifies the last shed or re-growing primary of the bird. This resulted in GGH scores of 1276 individuals. Overall, our data shows a clear progress in primary moult, in which birds start moulting P4 in late June, at the start of the counts, and finish moult in October (Fig. 1). However, moult progressed significantly slower in 2019 compared to 2017 and 2018. When numbers of Curlews were similar between subsequent roost counts, the changes in mean GGH scores were slightly positive, presumably reflecting the progress of moult of a population with little e- and immigration (Fig. 2). However, when numbers of Curlews increased between subsequent counts, a higher mean GGH score was recorded. On the other hand, a drop in Curlew numbers resulted in a decrease in mean GGH scores. These results suggest that both arriving and departing Curlews have higher GGH scores than stationary birds, and support the idea that primary moult scores can be used to estimate the minimum turn-over rates between counts. The authors have the ambition to collect an optimised dataset in order to develop a method to quantify the minimum turn-over and estimate the number of individual Curlews that roost in the Everdingerwaard during the post-nuptial stage.

[pdf only for members] [dutch summary]