Annual survival and recruitment in black brant (Branta bernicla nigricans) have declined since the 1990s, yet aerial surveys of the global population have been stable or even increasing over the past decade. We used a combination of a Lincoln estimator based on harvest information and band recoveries, and marked‐unmarked ratios in bag checks in 1 harvest area in Mexico to estimate the number of adults in the population during 1992–2015. We produced weighted means from the 2 kinds of estimates for years in which we had data for both, with weights equal to the inverse of the variance of the individual estimates. We treated the black brant population as consisting of 2 subpopulations. One population consisted of breeding black brant on the Yukon‐Kuskokwim Delta (YKD), Alaska, USA, and the other consisted of Arctic (northern Alaska, western Canada, and eastern Russia) breeders, and nonbreeders and failed breeders from the YKD that underwent molt migration to the Arctic. For the global population estimates, we assessed potential bias due to differential marking and harvest of the 2 subpopulations, which was approximately 1%, probably because band recovery rates were similar for the 2 subpopulations. Population estimates declined from 229,980 (average for 1999–2002) to 161,504 (average for 2012–2015). Population estimates based on estimated harvest were variable but more stable in the later years of the study, when larger numbers of brant hunters were included in the sample. We suggest that the combination of Lincoln estimates and bag check data provides a reasonable and cost effective approach to monitoring the population.