Summary: In 2019 the welsh red kite population is thought to be about 2500 breeding pairs. The post below shows how this estimate has been made.
Previous population estimates
It was not that long ago that the breeding population of red kites in Wales was known to within one or two pairs; the small range (never published in detail: see BTO atlas 1968-72 right) could be covered by a team of dedicated observers and confident population figures could be produced annually. But as the kite population expanded its range and increased in number during the 1990’s the task of locating all pairs became too onerous, and since the year 2000, the population figures have been derived from estimates.
Initially, during the early 2000’s the estimates could be confidently made to within + or – 25 pairs based on a collection of field observations collated by WKT covering most of the recorded breeding distribution, but it wasn’t long before the number of breeding pairs climbed to more than 1000, and it was no longer possible to make any sensible estimate by these methods. It is currently only possible to make realistic estimates by extrapolating from the results of monitored samples, and/or modelling.
This estimate
The population changes of any species depend on the breeding productivity and the survival of full-grown birds. The WKT has produced productivity figures annually throughout the period 1994 to 2019. These figures can be considered reliable as they are based on a large sample of monitored nests (6000+ in total) taken from both core areas and the periphery of the breeding range. Survival is harder to monitor, but a study of tagged kites from the Welsh population (Newton, Davis & Davis 1989) concluded that during their first year, about 66% of birds survive once they have fledged. Survival increases to 80% for the second year and is typically around 95% for territorial adults. These figures are similar to those measured in other kite populations.
Knowing the productivity and subsequent survival means that the population can be modelled, and the size can be estimated based on an extrapolation. In the graph below the model estimate is shown in orange. It was indexed to the field data in 1994 (111 pairs) and then used the annual measured productivity figures collected by WKT (adjusted between 2007 and 2011 to allow for birds removed from welsh nests for release schemes), and the above survival figures to estimate the population up to 2019.
As can be seen, the population estimates made by the model from 1995 to 2006 are an almost exact match to the figures derived from dedicated fieldwork (WKT). This gives confidence to the survival estimates made by Newton et al.
Then from 2006 they are broadly similar up to 2012 when WKT stopped producing estimates based on fieldwork. The projected trend then shows the population continuing to increase exponentially to around 2770 pairs in 2019.
How reliable is it?
At first, the figure of 2770 pairs may seem surprisingly high, and it should be born in mind that the survival rate of individuals post-fledging is likely vary from year to year depending on weather, and is also expected to decrease as the population approaches the carrying capacity due to competition for food. The model does not take this into account, so the figure should perhaps be treated as an upper limit rather than mid-range. Another factor that is likely to be density dependent is productivity, which is also predicted to decrease as the population increases due to increased competition for food and nest sites. However, productivity has been annually monitored with a large sample size and is taken into account by the model. To date there has been no long-term change in productivity and throughout the period it has remained at around 0.9 chicks per breeding attempt, implying that density dependent factors are yet to kick in, at least during the breeding season.
The BTO’s Breeding Bird Survey (BBS) project monitors randomly selected 1km squares which are covered by two transects each 1km in length. The number of red kites recorded on monitored 1km squares in Wales has been increasing and in the graph above the grey line shows the trend from 2003 when the number of kites recorded on BBS squares was indexed to the mid-range of the WKT population estimate of 375 pairs (+ or – 25). The trend is more variable than the model, but it broadly agrees and will become more reliable in the future as the species is recorded more widely and more frequently by the BBS.
All models extrapolating from a single point over a long period (in this case 26 seasons) are going to be prone to error, but the almost identical match between the model predictions and the figures derived from detailed fieldwork up to 2006 suggest that the model is at least plausible. The figure of 2770 pairs is likely to be a maximum, because survival may have been lower in cold winters, and may have decreased since those published in 1989 as competition for food increases. Also, some of the birds included in the estimate are breeding outside of Wales, as the breeding range of the welsh population has expanded east wards into Shropshire and Herefordshire.