Misapplied survey data and model uncertainty result in incorrect conclusions about the role of predation on alewife population dynamics in Lake Huron: a comment on He et al. (2015)
Drastic recent and ongoing changes to fish populations and food webs in the Great Lakes have been well-described ( Riley et al. 2008 ; Barbiero et al. 2009 ; Nalepa et al. 2009 ; Fahnenstiel et al. 2010 ; Evans et al. 2011 ; Gobin et al. 2015 ), and uncertainty regarding their potential effects on fisheries has caused concern among scientists and fishery managers (e.g., Dettmers et al. 2012 ). In particular, the relative importance of “bottom-up” (e.g., lower trophic level changes) versus “top-down” (e.g., predation) factors to fish community changes in the Great Lakes have been widely debated (e.g., Barbiero et al. 2011 ; Eshenroder and Lantry 2012 ; Bunnell et al. 2014 ). In Lake Huron, recent ecosystem changes have been particularly profound, and populations of alewife ( Alosa pseudoharengus ), an offshore pelagic prey fish, collapsed in 2003 and have yet to recover ( Riley et al. 2008 , 2014 ). He et al. (2015) recently used a series of linked ecological models to assess the role of predation in the dynamics of the offshore prey fish community in Lake Huron. While we believe that they provide a novel method for combining bioenergetics and stock assessment modeling, we question the validity of their conclusions because of the misapplication of survey data and the lack of critical interpretation of their modeling efforts. Here we describe how He et al. (2015) have misapplied bottom trawl data from Lake Huron, and we provide examples of how this has resulted in erroneous conclusions regarding the importance of predation to the population dynamics and collapse of alewife in Lake Huron.