The population decline of Gyps vultures in India and Nepal has slowed since veterinary use of diclofenac was banned

Abstract

Populations of oriental white-backed vulture (Gyps bengalensis), long-billed vulture (Gyps indicus) and slender-billed vulture (Gyps tenuirostris) crashed during the mid-1990s throughout the Indian subcontinent. Surveys in India, initially conducted in 1991-1993 and repeated in 2000, 2002, 2003 and 2007, revealed that the population of Gyps bengalensis had fallen by 2007 to 0.1% of its numbers in the early 1990s, with the population of Gyps indicus and G. tenuirostris combined having fallen to 3.2% of its earlier level. A survey of G. bengalensis in western Nepal indicated that the size of the population in 2009 was 25% of that in 2002. In this paper, repeat surveys conducted in 2011 were analysed to estimate recent population trends. Populations of all three species of vulture remained at a low level, but the decline had slowed and may even have reversed for G. bengalensis, both in India and Nepal. However, estimates of the most recent population trends are imprecise, so it is possible that declines may be continuing, though at a significantly slower rate. The degree to which the decline of G. bengalensis in India has slowed is consistent with the expected effects on population trend of a measured change in the level of contamination of ungulate carcasses with the drug diclofenac, which is toxic to vultures, following a ban on its veterinary use in 2006. The most recent available information indicates that the elimination of diclofenac from the vultures' food supply is incomplete, so further efforts are required to fully implement the ban.

Figure 1. Population indices and trend of Gyps bengalensis.

Circles show indices of population density, relative to that in 1992, estimated by log-linear Poisson regression preformed on data from six road transect surveys in India. Vertical lines show 95% bootstrap confidence limits. The curve shows the cubic log-linear population trend fitted to data for the period 2000–2011.

Figure 2. Population indices and trend of Gyps indicus and G. tenuirostris combined.

Circles show indices of population density, relative to that in 1992, estimated by log-linear Poisson regression preformed on data from six road transect surveys in India. Vertical lines show 95% bootstrap confidence limits. The curve shows the quadratic log-linear population trend fitted to data for the period 2000–2011.

Figure 3. Counts of Gyps bengalensis on a road transect through western Nepal.

Figure 4. The rate of population decline (% per year) of Gyps bengalensis in India.

Circles show average annual rates between each pair of consecutive surveys with their 95% bootstrap confidence limits (vertical lines) plotted midway between the dates of the two surveys. The curve shows the first derivative of the cubic log-linear population trend fitted to the survey data for the period 2000–2011.

Figure 5. The rate of population decline (% per year) of Gyps indicus and G. tenuirostris combined in India.

Circles show average annual rates between each pair of consecutive surveys with their 95% bootstrap confidence limits (vertical lines) plotted midway between the dates of the two surveys. The line shows the first derivative of the quadratic log-linear population trend fitted to the survey data for the period 2000–2011.

Figure 6. Average numbers of Gyps bengalensis (filled circles) and Gyps indicus (open circles) seen per kilometre of road transect at selected sites.

Results are for transects in and near National Parks in India which contained the majority of records of these species in 2011.

Figure 7. Simulated changes in the annual survival rate of adults (solid lines), the annual survival of pre-adults (dashed lines) and the mean number of fledglings per adult pair per year (dotted lines) required to account for the observed change in the population index derived from road transects between 2007 and 2011.

Results are shown separately for Gyps bengalensis (blue) and Gyps indicus (red). In these simulations the annual survival of adults in the absence of additional mortality was assumed to be 0.90.

Figure 8. Simulated changes in the annual survival rate of adults (solid lines), the annual survival of pre-adults (dashed lines) and the mean number of fledglings per adult pair per year (dotted lines) required to account for the observed change in the population index derived from road transects between 2007 and 2011.

Results are shown separately for Gyps bengalensis (blue) and Gyps indicus (red). In these simulations the annual survival of adults in the absence of additional mortality was assumed to be 0.97.

Figure 9. Change in the rate of decline of the Gyps bengalensis population in India observed and expected from the ban on diclofenac.

The annual decline rate in two time periods after the ban on veterinary diclofenac, is shown relative to the decline rate before the ban. The horizontal axis shows the expected rate of decline calculated from the prevalence and concentration of diclofenac in liver samples from carcasses of domesticated ungulates sampled in April – December 2006 (T2) and in January 2007– December 2008 (T3), relative to the expected rate from samples collected before the ban (T1: May 2004– July 2005). The vertical axis shows the ratio of the vulture population decline rates at the mean sample collection dates of surveys T2 and T3 to that at T1, estimated from the cubic regression model fitted to the road transect data. The vertical and horizontal lines associated with each point show 95% bootstrap confidence limits. The diagonal line shows the result if the observed and expected ratios of decline rates were equal.