J. Brouwer et Wc. Mullie, A method for making whole country waterbird population estimates, applied to annual waterbird census data from Niger, OSTRICH, 2001, pp. 73-82
Regular dry season counts of waterbirds have taken place in Niger since 199
2, as a part of the African Waterbird Census. These counts included large s
ections of the river Niger and its floodplains, as well as up to fifty isol
ated wetlands. We were involved in these counts in a personal capacity from
1992-1997. In addition to identifying wetlands of international and nation
al importance and Important Bird Areas,the results of the counts can be use
d to estimate total numbers of birds in Niger for many waterbird species. I
n this paper we present a method for doing so, which includes (a) separatin
g isolated wetlands from irrigated floodplains and from sections of the riv
er Niger; (b) separating open areas from vegetated areas at each site; (c)
estimating the percentage cover that was achieved for each type of habitat
at each site; (d) estimating the proportion surveyed of each of these three
types of wetland, as part of the total areas present in the country; (e) t
aking into account relative visibility of each species (e.g. very high for
cranes, very low for crakes); (f) using all this information to extrapolate
the actual numbers counted to an estimate of total numbers present in Nige
r at the time of counting, for each species. We also present the resulting
information for 97 species of waterbird in Niger, plus three raptors and th
ree kingfisher species largely dependent on wetlands. This information incl
udes where possible a population estimate for the January-February period,
comments on population trend, and comments on subspecies concerned. For man
y species the results are in agreement with the draft Third Waterbird Popul
ation Estimates (Delany in lit.), but for others there is a considerable de
viation. Species that are possibly considerably more numerous than official
estimates indicate include Pholacrocorax africanus, Dendrocygna bicolor, D
.viduata, Sarkidiornis melanotos and Himantopus himantopus. Species that ar
e possibly considerably less numerous include several Charadriidae. Further
validation of the extrapolation method is desirable and ways of improving
it are suggested.