|
Résumé :
|
Strodam Reserve was established in 1925 in the most southwesterly part of Grib Forest in NE Zealand. The reserve consists predominantly of deciduous forest that has remained largely untouched for the last 60-70 years. In the southernmost part of the reserve, a 26 ha bird census area was established in 1986 and monitored yearly for breeding birds ever since. The main part of the census area consists of almost 200 year old beech Fagus sylvatica forest (Fig. 1).This area was open parkland until the middle of the 20th century, but has since developed into genuine old growth forest with glades, undergrowth of different ages and scattered oak Quercus robur and wild cherry Prunus avium etc. Other parts are dominated by European ash Fraxinus excelsior, birch Betula sp. and willow Salix sp. thicket. During the first decade of the study period, a number of relatively small spruce stands were removed, after which birch and other species established on these areas.The census method was the territory mapping method with 10 censuses performed each year (Lundstrom 1978). However, since this method clearly underestimates the populations of many species, the census taker (BGH) estimated the 'real' populations more subjectively based on the same territory mapping each year. These estimates varied considerably from +2% to +16% in 'consistently' singing species like Willow Warbler Phylloscopus trochilus, Eurasian Blackcap Sylvia atricapilla, Eurasian Wren Troglodytes troglodytes, Eurasian Blackbird Turdus merula and Common Chaffinch Fringilla coelebs to several hundred percent in much less 'self-promoting' species like Marsh Tit Poecile palustris, Eurasian Treecreeper Certhia familiaris and Spotted Flycatcher Muscicapa striata (Tab. 1). It should be noted, however, that in about half the more 'regularly' breeding species, the underestimates resulting from a strict adherence to the territory mapping method are around or less than 20% (Tab. 1). Throughout this paper it is these 'upgraded' estimated census results that are used (Fig. 2, Tab. 2, Tab. 3 and Appendix 1).In total, 64 bird species bred or most likely bred in the census area during the study period. The average number of breeding species per year was 38.8, 34.5 and 38.0 in 1986-1995, 1996-2005 and 2006-1014, respectively. The 'dip' in the middle of the study period was caused by a statistically significant reduction in the numbers of long-distance migrants (Fig. 3; see further below).Simpson's diversity index shows slightly increasing values of 0.936, 0.938 and 0.940 for the years 1986-1995, 1996-2005 and 2006-2014, respectively (Tab. 2). These values lie at the absolutely higher end for Danish forests (Appendix 2), and the increase during the study years is statistically significant (R-2 = 0.153, p = 0.036, N = 29).Of 25 species for which we have sufficient data for statistical analysis, eight species were stable/slightly fluctuating, five species increased more or less markedly, while three species declined (Fig. 2, Appendix 1). The remaining nine species fluctuated more widely.When all species are considered together, the breeding bird population was relatively stable, fluctuating between about 320 pairs and 420 pairs throughout the study period (Fig. 3). However, when sedentary/short/medium distance migrants were compared with long-distance sub-Saharan migrants, the populations of the latter decreased from around 11% of the total population during the first years of the survey to about 6% in the middle of the period (Fig. 3). After this decline, the sub-Saharan migrants have recovered by a few percentage points, primarily as a result of recovery of Wood Warbler Phylloscopus sibilatrix and Common Redstart Phoenicurus phoenicurus (Fig. 2).The proportion of hole-nesters was about 40% (Tab. 2; no nest boxes were found in the study area), which is at the high end of the range for Danish forests.A pairwise comparison of the population estimates of the 25 species across the 29 years showed significant positive correlation between the population sizes of 33 pairs of species (Fig. 4). Similarly, there was a significant negative correlation between population sizes of 25 species-pairs. Here it is perhaps noteworthy that the negative correlation between Common Starling Stumus vulgaris on the one hand and Eurasian Nuthatch Sitta europaea and Common Redstart on the other could be the result of competition for nesting holes in that decrease in Common Starling may have left room for more Nuthatches and Redstarts.Of the 25 species analyzed for population trends, there was a significant positive correlation between the census results from Strodam and the national bird population indexes from the Danish Ornithological Society's point counts in 13 species (Tab. 1).For Eurasian Wren the classical effect of severe winters (1986, 1987, 1996, 1997, 2010 and 2011) is seen (Heldbjerg
|