Off-campus South Dakota State University users: To download campus access theses, please use the following link to log into our proxy server with your South Dakota State University ID and password.

Non-South Dakota State University users: Please talk to your librarian about requesting this thesis through interlibrary loan.

Document Type

Thesis - University Access Only

Award Date

2008

Degree Name

Master of Science (MS)

Department

Wildlife and Fisheries Science

First Advisor

Steven R. Chipps

Keywords

geographical distribution, sturgeons, missouri river, south dakota, nebraska, habitat

Abstract

Monitoring and assessment of the endangered pallid sturgeon Scaphirhynchus albus population is essential to the recovery of the species. Movement and distribution of pallid sturgeon is generally documented using radio telemetry methods. However, because of the time and cost involved in tracking individual fish (i.e., small sample size), it is often difficult to evaluate spatial distribution of groups of fish over long time periods (>3 years). Standardized random sampling, which relies on a variety of gear types including trammel nets, otter trawls, gillnets and setlines, has been conducted in the reach of the Missouri River downstream of Fort Randall Dam, South Dakota and Nebraska annually since 2003 to evaluate the pallid sturgeon population. This river reach is unique in that it has riverine, depositional delta, and reservoir habitats. Although data on catch rates and capture locations of pallid sturgeon have been routinely collected since 2003 in the Fort Randall Reach of the Missouri River, there have been no attempts to use this information to evaluate spatial distribution patterns and habitat use in the riverine and depositional habitats that exist in the reach. The objectives of this study were to: 1) quantify spatial distribution patterns of juvenile pallid sturgeon using non-telemetry methods, 2) develop a habitat assessment tool that is useful for predicting potential capture locations for juvenile pallid sturgeon, and 3) apply the model to quantify the relative distribution of potential capture locations in the Fort Randall reach, Missouri River. Using presence/absence data, I evaluated the spatial distribution of juvenile pallid sturgeon. Spatial scan statistics were used to analyze pallid sturgeon distribution patterns based on 1) singular gear collections per season, 2) pooled gears per season 3) singular gears pooled across seasons, and 4) the overall pooled dataset. Three significant areas, characterized by ‘clusters’ of pallid sturgeon captures, were identified from analysis of the overall pooled dataset. These same areas were also identified using summer trammel net sampling implying that this gear/season combination could be used to reliably identify areas with a high probability of pallid sturgeon presence. This methodology could be used to effectively identify areas with a high probability of pallid sturgeon presence elsewhere in the Missouri River basin using existing data from the ongoing Pallid Sturgeon Monitoring and Assessment Program. Areas of pallid sturgeon habitat use could then be evaluated to better understand why pallid sturgeon use these habitats. Understanding the habitat requirements of juvenile pallid sturgeon represents an important part of the monitoring and assessment program. In this study, I developed a spatially explicit habitat assessment tool that integrated information on physical habitat, aquatic invertebrate abundance, and pallid sturgeon occurrence to predict pallid sturgeon capture. The model that I developed is the first attempt at integrating both habitat and prey parameters in predicting sturgeon capture (i.e., occurrence). Maps for water depth, bottom water velocity, directional variation in bottom flow, percent sand substrate composition, bottom slope, and ephemeropteran and larval dipteran abundance were created for eight 3.2-km river segments using 20 m grid cell resolution. Pallid sturgeon capture and non-capture areas were then overlain onto physical habitat and prey availability maps to derive means or proportional availability for these variables. I used discriminant function analysis (DFA) to distinguish between capture (n=25) and non-capture (n=49) locations from the Missouri River downstream of Fort Randall Dam in South Dakota and Nebraska. Four variables successfully discriminated capture from non-capture locations and included: 1) water depth > 2 m, 2) percent sand substrate, 3) dipteran abundance and 4) ephemeropteran abundance. Classification functions were then used to predict pallid sturgeon occurrence in the eight 3.2-km river segments. Correct classification rates were 88% for pallid sturgeon capture areas and 100% for non-capture areas for an overall classification rate of 95%. The percentage of area in each of the eight Fort Randall reaches (n=8) that were classified as pallid sturgeon capture areas ranged from 22 to 64%; the percent area classified as potential capture locations was not significantly different between riverine and depositional habitats. Pallid sturgeon captures often occurred within close proximity to each other in this study suggesting that the combination of physical habitat conditions (i.e., deeper water, sandy substrates) and prey availability (i.e., dipteran and ephemeropteran abundance) influenced their spatial distribution. The model developed here could be used to identify areas where there is high potential for pallid sturgeon captures that would enhance future monitoring and restoration efforts.

Library of Congress Subject Headings

Sturgeons -- Missouri River -- Geographical distribution
Sturgeons -- Habitat -- Missouri River
Sturgeons -- South Dakota -- Geographical distribution
Sturgeons -- Habitat -- South Dakota
Sturgeons -- Nebraska -- Geographical distribution
Sturgeons -- Habitat -- Nebraska

Format

application/pdf

Number of Pages

110

Publisher

South Dakota State University

Rights

Copyright © 2008 Bryan D. Spindler. All rights reserved.

Share

COinS