Introduction
Prespawning mortality events in Pacific (genus Oncorhynchus ) salmon are well-documented throughout the ranges of species, but the mechanisms and consequences of these events for populations and ecosystems, including people, are not well known (Bowerman et al. 2016). Prespawning mortality events can occur at various stages of the upstream homeward migration. Upon reaching the spawning grounds, prespawning mortality (PSM) appears linked to spawner density and streamflow, which in turn influences the availability of life-supporting dissolved oxygen (Sergeant et al. 2017). Our understanding of PSM on the spawning grounds is largely limited to years with anomalously large returns (Quinn et al. 2007; Tillotson and Quinn 2017) or locations where spawner densities are influenced by returns of hatchery-produced salmon (McConnell et al. 2018). While our understanding of PSM on the spawning grounds is in its early phases, most of what we know has been garnered by studies tracking fish en route upstream during the freshwater phase of the homeward migration (Hinch et al. 2012).
Substantial work on large watersheds such as the Fraser River (British Columbia, Canada), Klamath River (USA), and the Columbia River (USA) has revealed that en route mortality events are the result of complex interactions between the phenotype of migrants (e.g. run timing and physiology) and their environment, with water temperature and flow being master variables (Keefer et al. 2008). As a poignant example, early migrating sockeye salmon (O. nerka ) in the Fraser River tend to die at much higher rates than later returning individuals, which appears to be associated with a mismatch between current warm water conditions and thermal adaptation in terms of aerobic scope to historical migration conditions (Eliason et al. 2011). In contrast, later-timed migrants of Sockeye Salmon died at higher rates in the Columbia River during 2000, likely due to delayed migration and increased susceptibility to parasite and pathogens (Keefer et al. 2007). Perhaps the best known example of a temperature-related disease outbreak is from the Klamath River in September of 2002, where an estimated 34,056 migrating Chinook salmon (O. tshawytscha ) died as a result of heavy infections ofIchthyopthirius multifiliis (Ich) and/or Flavobacterium columnare (columnaris) (Guillen 2003). Of course, a perennial challenge is disentangling whether pathogen presence and disease is the proximate agent of mortality or, more commonly in wild fish, a response to other unquantified stressors. Taken as a whole, strong evidence indicates that the combination of high water temperatures, accelerated maturation and senescence, elevated stress, ionoregulatory dysfunction, and disease are the causal agents behind en route mortality in migrating salmon (Hinch et al. 2012). Moreover, patterns observed thus far indicate that populations at the southern margins of the ranges may be most susceptible to en route mortality given regional warming that increasingly push water temperatures beyond tolerance thresholds. In this paper I document, a large-scale en route mortality event by summer Chum Salmon (O. keta ) returning to spawning grounds on the edge of the Arctic in the Yukon River watershed, Alaska (Fig.1).
The summer of 2019 will forever be burned in the consciousness of Alaskans. Most apparent was the extreme air temperature anomalies with many regions, including the Yukon, experiencing many days with maximum temperatures in excess of 32°C. Temperatures in July and August were far above average compared to long-term records (Fig. 2). In addition, ocean temperatures in Alaska were 5 degrees warmer than average, wildfires burned throughout the state choking the air with smoke, droughts plagued rain forests, a lack of sea ice in the Bering Sea, and anomalous biological patterns were detected across the region (Thoman and Walsh 2019). Beginning on July 20th, the residents living along the Koyukuk River that have long been the stewards and first responders on their lands began reporting a large scale die-off of Chum Salmon on social media outlets. Hundreds of comments and similar observations streamed in, and on July 26th a dispatch of western-trained scientists, including the author, traveled to the Koyukuk River drainage with the goals of i ) documenting the extent of the die-off,ii ) inferring any apparent abiotic or biotic causes of the event, and iii ) quantifying whether mortality was associated with body size or sex by comparing to observations of fish that survived to the spawning grounds.