The Performance of Quasi-Experimental BACI Studies of Fish Populations in Coupled Human-Environmental Systems

Ecological research extensively uses a quasi-experimental framework, as perfect replication of and complete control over vast natural systems is unfeasible.  Marine reserve evaluation is dependent on quasi-experiments because reserve sites are never assigned randomly.  Sites are selected based on biological and economic goals as well as unique features of conservation interest.  Marine reserve monitoring programs are tasked with assigning a control site or sites that can serve as a comparison to detect a treatment effect due to the creation of a reserve.  The reserve (impact) and control sites are commonly compared with a Before-After Control-Impact (BACI) analysis that tracks changes in stock over time before and after the establishment of a reserve.  These assessments overwhelmingly focus on the biological outcomes of reserves, but omit human behavior, a critical driver in these biological systems.  Marine reserves only affect stocks by decreasing human presence and fishing pressure.  In addition, fishers respond dynamically to the creation of reserves and changing stock abundances, influencing stock levels throughout their entire spatial range – including within marine reserves.  I will use a model of a simulated marine system to illustrate several ways human behavior dynamically affects fish populations.  Omitting human behavior when modeling biological changes in coupled systems can confound the ability to estimate unbiased parameters and detect treatment effects.  Further, BACI analyses follow several assumptions that may be challenging to meet in coupled systems.  Through modeling, I will discuss how quasi-experimental BACI analyses of changes in fish populations perform in coupled systems.