Date of Award

Fall 12-1-2016

Document Type

Thesis

Degree Name

Master of Science in Fisheries and Wildlife Science

Department

Biological Sciences

College

College of Natural & Health Sciences

Committee Chair

Dr. John Jackson

Second Committee Member

Dr. Chris Kellner

Third Committee Member

Dr. Joe Stoeckel

Dean of Graduate College

Dr. Mary B. Gunter

Abstract

Channel Catfish Ictalurus punctatus have an important role in Arkansas’s commercial and recreational fisheries. This importance promoted sampling strategy investigations for obtaining information that assist with future management decisions. My study focused on determining effective sampling effort and sample size requirements needed for detecting changes in relative abundance within Channel Catfish populations, and explored relationships between catch per unit effort (CPUE) and reservoir habitat characteristics. My study was conducted during the spring of 2014 in six reservoirs (Lake Nimrod, Lake Overcup, Lake Catherine, De Queen Lake, Lake Erling, and Lake Columbia). General stock assessments consisting of length frequency distributions, growth rates, and mortality rates were also conducted for each reservoir. Sampling involved setting 16 tandem hoop net sets during April, May, and June in each reservoir. Mean CPUE (fish/tandem hoop net set) ranged from 1.9 (SD, 2.1) to 355.9 (SD, 137.2). Lake Erling’s CPUE was the greatest, while Lake Catherine’s CPUE was the lowest. In addition, catch rates were influenced primarily by water temperature. Shallow reservoirs had higher CPUE than deep reservoirs for each month sampled. In addition, the CPUE of each reservoir’s lower section (closer to the levee) was less than the upper section during April. However, both sections were equal during May and June. Due to April’s low catch rates, only May and June’s CPUE were used to determine effective sample sizes. The greatest mean back-calculated length-at-age for Channel Catfish sampled occurred in Lake Overcup, while the greatest mean length-at-capture occurred in Lake Columbia. Growth rates were also significantly different between reservoir types for back-calculated ages two, four, five, and seven. Finally, shallow reservoir types displayed higher mortality rates than deep reservoirs. Sample size simulations were conducted to determine the number of sets necessary to detect 10% to 50% changes in relative abundance at an alpha of 0.05 and 0.10 and power of 0.95 and 0.80. Simulations revealed CPUE assessments ineffective at describing populations without also assessing age and growth data and length frequency distributions. Furthermore, angler surveys and harvest estimates could be a potential alternative to CPUE for evaluating a population’s relative abundance.

Download
COinS