The decline of the grey partridge in Europe: comparing demographies in traditional and modern agricultural landscapes

Count data were collected from available studies on the European grey partridge (Perdix perdix) populations with at least 3 years of consecutive data. According to habitat characteristics, the period of study and the management practice the data were combined in two main subsets: the British populations from 1930 to the end of the 1960's and the declining continental populations from 1970 to the present. The latter populations are located in areas where the traditional landscape has been severely modified by important changes in agricultural practices. The goal of this work is: (a) to compare demographies of the two sets of populations; (b) to design and calibrate a stochastic demographic model on the basis of available data; (c) to use it to assess the risk of extinction under different management alternatives; (d) to test some of the most credited hypotheses on the grey partridge decline. Population dynamics are investigated via regression analyses of different factors on demographic rates. Spring-to-summer rate of increase and autumn to winter survival are shown to be density dependent in both UK and continental populations. Over-winter losses in UK prior to 1970 are positively correlated with the young-to-adult ratio, suggesting an important role of dispersal. Reproduction and survival rates were systematically larger in the traditional UK populations. Based on the regressions, stochastic demographic models were then built to assess the probability of extinction under various harvesting and management conditions, and the possibility of developing conservative hunting strategies. The stochastic modelling confirms that the UK populations prior to 1970 were intrinsically much more resilient than the continental populations in recent decades. Even very low rate of harvesting cannot be tolerated by the present continental populations, suggesting that the persistence of hunting activity, although with a limited effort, has probably contributed to the extinction of many sub-populations and is critically threatening the remaining ones. We show that the introduction of stochasticity into the model is fundamental for assessing the real extinction risk for partridge populations under different management scenarios. The comparison of UK and continental demographies does not provide evidence of a single main bottleneck for the continental populations, but the viability of continental populations is endangered by the concurrent deterioration of reproduction and survival rates. (C) 2004 Elsevier B.V. All rights reserved.