An essential management objective of the Yukon Delta and Koyukuk National Wildlife Refuges in Alaska is to conserve fish and wildlife populations and habitats in their natural diversity. In keeping with this objective, the U.S. Fish and Wildlife Service installed weirs in two tributaries of the Yukon River, the East Fork Andreafsky and Gisasa rivers, in 1994 to collect information on salmon populations that used them. The weirs have been in operation for >23 y. Chinook Oncorhynchus tshawytscha and summer Chum Salmon O. keta were counted and sampled for various demographic data each year as they migrated through the weirs to upstream spawning areas. Here we examine this record of population data to describe and compare long-term variation in run abundance, run timing, length and age structure, sex composition, and production for these salmon populations. Fishery managers often look to multiple monitoring projects in-season seeking corroboration of observed run qualities; therefore, we also considered whether Yukon River main-stem indicators of abundance were correlated with these tributary escapements. Our analyses suggest long-term stability of these populations despite large annual variations in most metrics we examined. Annual escapements have varied by factors of 3–5 for Chinook Salmon and >23 for summer Chum Salmon, yet only the Chinook Salmon population in the Gisasa River appears to be declining. Main-stem abundance indicators were not correlated with Chinook Salmon escapements but were strongly correlated with summer Chum Salmon escapements. Run timing has varied annually by as much as a week earlier or later than average for all four populations with no trend over time. Mean age of the Chinook Salmon populations declined over time but remained stable for the summer Chum Salmon populations. Chinook Salmon populations in the East Fork Andreafsky and Gisasa rivers averaged 35% and 28% female, respectively. Both summer Chum Salmon populations averaged close to 50% female. Length at age has been stable or slightly declining for all four populations. Production over time was strongly correlated within species for populations in the two rivers, and averaged >1 recruit/spawner for all populations except Chinook Salmon from the Gisasa River. We discuss these findings in the context of major changes in the fishery and the environments these populations experience.