Relative to other ducks and to May ponds, pintail and wigeon moved from prairie to northern strata after 1999 in the USFWS survey (1974-2026); pooled settling on ponds among prairie strata was stable

Question: in the USFWS Waterfowl Breeding Population and Habitat Survey (traditional area, 1974-2026, no survey 2020-2021), did the share of ten duck species in the 24 pond-counted prairie strata change after 1999 with May ponds held constant; is any change specific to a species or a factor common to all species in a stratum and year (for example crew, aircraft or visibility changes); and did the settling elasticity of birds to ponds among prairie strata change? Approach: logit prairie share on log May ponds, fitted to 1974-1999 and tested on 2000-2026 (HAC and AR(1) errors); guild-relative tests (share against the other nine species, and a stratum Poisson model with stratum-by-year fixed effects); period-specific Poisson elasticities with stratum and year fixed effects and year-block bootstrap. Results: with ponds held constant, the prairie share of northern pintail and American wigeon decreased by 0.48 log-odds after 1999 (95% intervals -0.76 to -0.20 and -0.78 to -0.18; Holm p 0.008 and 0.017), and mean breeding latitude moved north by about 140 and 200 km. This agrees with two independent SCI studies of the same survey. Relative to the other species, the shifts were -0.68 and -0.65 log-odds (series) and -0.35 each (stratum model, z -4.1 and -5.8), and they occurred against both Alaska and northern Alberta-NWT, which different crews fly. Thus factors common to all species in a stratum and year do not explain them. The ten species pooled had no northward shift beyond ponds over 2000-2026 (+0.13 log-odds toward the prairie). Among prairie strata, the pooled elasticity was 0.65 in 1974-1999 and 0.60 in 2000-2026 (difference interval -0.16 to +0.06); it decreased for mallard (0.49 to 0.33) and scaup (0.49 to 0.27) and increased for canvasback (0.46 to 0.73; only canvasback stays after Holm adjustment). Limitations: aerial estimates with visibility corrections; species-specific detection or correction changes are not removed; uncorrected counts exist only from 2000 and give lower 2000-2026 elasticities (pooled 0.51); the tests do not identify causes.

waterfowl · breeding distribution · prairie pothole region · May ponds · northern pintail · American wigeon · WBPHS · distribution shift · habitat selection · aerial survey · visibility correction · fixed effects Poisson