TURNHOUT C VAN, DIJK A VAN, WEIDE M VAN DER & BEUSEKOM R VAN (2006) Booming Bitterns Botaurus stellaris in the Netherlands: detection, trends and numbers.. LIMOSA 79 (1): 1-12.

In this paper the results of the Dutch national Bittern census in 2003 are presented. Also trends in numbers and distribution, as derived from monitoring and atlas studies, are summarised, and detection probabilities and accuracy of Bittern surveys are quantified. During the 2003 census 279 booming males were recorded in 77 sites (Fig. 1). Core areas were Oostvaardersplassen (49 booming males) and De Wieden (33), together holding over 25% of the total Dutch breeding population. Since the mid 1970s the distribution of Bitterns has evidently declined in the Netherlands, especially in the provinces of Friesland, Groningen and Flevoland (Fig. 2). Also former breeding areas in the western part of the country and in the riverine district were abandoned. Only few areas were colonized, mainly in the Lauwersmeer and the Delta area, which were successively cut off from the sea by large water works, and around Lake IJsselmeer. Numbers probably peaked in the 1970s, when large areas of reed marsh were created in the reclaimed Flevopolders (Tab. 2), but have been decreasing ever since as a result of habitat loss and degradation, accelerated by severe winters (especially in 1978/79 and the mid 1980s) from which the population never fully recovered. In 1996-97 the Dutch Bittern population was at its lowest. Numbers have been increasing since, with an annual growth of 3% in 1990-2004 (Fig. 3). Mild winters, nature restoration projects and improved habitat management are held responsible for the increase. The positive trend was observed in all regions but one: in the river district numbers decreased by 10% per year in this period (Fig. 4).
      Detection probabilities based on weekly visits in 11 sites (38 booming males) vary from 39% (SE=7%) per visit in the period April 1st- May 15th to 20% (SE=6%) per visit in the period 15 May - 30 June. However, detection probabilities strongly differ between sites, partly as a result of differences in Bittern densities. Detection probability in the peak period is 47% (SE=8%) per visit in sites where other Bitterns are present within a range of 1 kilometer, and 29% (SE=9%) for Bitterns without neighbours within 1 kilometer.
      Timing and number of visits determine the accuracy of different types of Bittern surveys. Numbers are underestimated by on average 40% in case of low-intensity surveys (2 visits), and by 15% in high-intensity surveys (9-10 visits) (Tab. 1). Correcting for underestimation, depending on type of survey per site, results in a national population estimate of 335-375 booming males in 2003. Eurasian Bittern Botaurus stellaris

[free pdf] [dutch summary]