Ardea
Official journal of the Netherlands Ornithologists' Union

login


[close window] [previous abstract] [next abstract]

Rappoldt C., Kersten M. & Smit C. (1985) Errors in large-scale shorebird counts. ARDEA 73 (1): 13-24
Field experiments were carried out to examine the stochastic and systematic error in the counts of single flocks of birds. For flocks of sitting shorebirds the use of a relative stochastic error of 37% is recommended, independent of flock size. For flying flocks of shorebirds this is 17%. The systematic error in both cases could not be quantified accurately, but is smaller than some tens of per cents. For abundant shorebird species the stochastic error in the result of a large-scale count can be calculated from the errors in the counts of the largest occurring flocks. The addition of counts of single flocks leads to a decrease of the (relative) stochastic error. Calculations for the Dutch Wadden Sea show a stochastic error of 5 to 10 per cent in the counts of abundant species. For these species, the systematic error caused by missing of birds is unlikely to be larger than the stochastic error for single flocks (about 37%). For species which are common in one or a few parts of the Wadden Sea area only, the stochastic error may be somewhat larger than for the abundant species. The counts of relatively scarce species, of which the individuals are widely spread over an area, will generally be very inaccurate. Large errors resulting from missing of bird during the count will dominate. The count may just indicate an order of magnitude.


[close window] [previous abstract] [next abstract]