Before-after, control-impact analysis of evidence for the impacts of water level on Walleye, Northern Pike and Yellow Perch in lakes of the Rainy-Namakan complex (MN, USA and ON, CA)

Water level (WL) fluctuations in lakes influence many aspects of ecosystem processes.
Concern about the potential impact of WL fluctuations on fisheries was one of the factors
that motivated the decision in 2000 to alter the management of WL in the Rainy-Namakan
reservoir complex (on the border between the U.S. state of Minnesota and the Canadian
province of Ontario). We used a Before-After, Control-Impact (BACI) framework to identify
potential impacts of the change in WL management to Walleye, Northern Pike and Yellow
Perch catch per unit effort (CPUE). The CPUE of these species from 1990±1999 and from
2005±2014 were compared in four impact lakes (Lake Kabetogama, Namakan Lake, Rainy
Lake and Sand Point Lake) and two control lakes (Lake of the Woods and Lake Vermilion)
using a simple Bayesian model. Changes in fish CPUE in the impact lakes were often similar
to changes that occurred in at least one control lake. The only change that was not similar to
changes in control lakes was an increase of Yellow Perch in Lake Kabetogama. The two
control lakes often differed substantially from each other, such that if only one had been
available our conclusions about the role of WL management on fisheries would be very different.
In general, identifying cause-and-effect relationships in observational field data is
very difficult, and the BACI analysis used here does not specify a causative mechanism,
so co-occurring environmental and management changes may obscure the effect of WL
management.