ELSEVIER Biological Conservation 84 (1998) 157-165 0 1998 Elsevier Science Ltd All rights reserved. Printed in Great Britain PII: SOOOS-3207(97)00097-9 0006-3207/98 $19.00+0.00 POPULATION DEVELOPMENT AND CHANGES IN WINTER SITE USE BY THE SVALBARD LIGHT-BELLIED BRENT GOOSE, Branta bernicla hrota 1980-1994 Preben Clauseqa* Jesper Madsen,= Steve M. Percival,b David O’Connorc & Guy Q. A. Anderso& aNational Environmental Research Institute, Department of Coastal Zone Ecology, Kalti, DK-8410 R&t&, Denmark bNorthumbrian Water Ecology Centre, University of Sunderland, Science Complex, Sunderland SRI 3SD, UK cEnglish Nature, Northminster House, Peterborough PEl IUA, UK (Received 25 February 1994; accepted 9 June 1997) Abstract The Svalbard breeding population of light-bellied brent goose Branta bernicla hrota, wintering in 4-5 sites in Denmark and at Lindisfarne in north-east England, increased from c. 3SO&4000 in the early 1980s to 4000- 5800 in the early 1990s. The increase was mainly due to years with good breeding success, but probably also to improved survival in mild winters. Thus, in cold winters it was observed that mortality, of at least first-winter birds, increased. Major changes in winter site use were observed during the 1980s; the use of the Danish W ad&n Sea decreased, and in consequence, the brent geese arrived earlier at Lindisfarne and at the main winter site, Maria- ger- Randers Fjords. In the course of winter the population now disperses to more sites and new wintering sites have been established; in spring the majority of the population is concentrated in two sites in north-west Jutland in Den- mark. The changed pattern of movement reflects decrea- ses in distribution of Zostera food resources and depletion of the remaining resources due to waterfowl grazing, which may negatively affect condition of the geese in winter and ultimately survival. On the other hand, the population is now more dispersed during winter and hence less vul- nerable to local environmental pertubations. The w ider dispersal and changedphenology have several implications for management of the population which are discussed. 0 1998 Elsevier Science Ltd. All rights reserved. Keywords: brent goose, population, survival, phenology, distribution. INTRODUCTION The Svalbard population of light-bellied brent goose Branta bernicla hrota, wintering in Denmark and north- east England, is one of the smallest and more vulnerable *To whom correspondence should be addressed. 157 goose populations in the world, numbering 3500-4000 individuals in the first half of the 1980s (Madsen, 1984 1987). Traditionally, the population made successive use of sites within the winter range. On arrival from the breeding grounds in September the majority of geese first stayed in the Danish part of the Wadden Sea. From there they moved to Mariager-Randers Fjords in the Kattegat where they remained throughout mild winters. In cold winters, however, the majority of the geese moved on to Lindisfarne in north-east England. Small tlocks also moved to the Netherlands in cold winters (Lambeck, 1981; van den Berg, 1984). From February until departure to the breeding grounds in late May the population concentrated in fjords in north-west Jutland in Denmark (Madsen, 1984, 1987). Since 1987, studies have been intensified in Denmark, England and Svalbard in the Barents Sea to investigate factors affecting the population size and distribution of the Svalbard light-bellied brent goose. In this paper we examine and discuss the population development during 1980-1994 and changes in winter site use by the popu- lation which have been observed in the 1980s and early 1990s. METHODS Counts and population parameters Since 1980, mid-monthly simultaneous ground counts have been made by teams of voluntary and professional observers in the Danish and English wintering sites of the population (Fig. 1); in the Danish Wadden Sea the coverage was incomplete until 1985. During 1980/198 l- 1982/1983 and 1988/1989-1991/1992 l-3 monthly sup- plementary counts were made on the main wintering grounds, whereas during 1983/1984-l 987/l 988 and 1992/1993-l 994/ 1995 counts in Denmark were irregular at some sites while intensive at others (Mariager Fjbrd, Agers, Nibe Bredning). At Lindisfarne, 3-monthly