Document Type

Thesis - Open Access

Award Date


Degree Name

Master of Science (MS)


Natural Resource Management

First Advisor

Nels H. Troelstrup, Jr.


benthic macroinvertebrates, index of biotic integrity, random forests


Biological assessments are used to evaluate the biotic integrity of a system. Great Plains prairie streams are harsh environments due to variable climate, topography, vegetation, and human development. This presents a challenge to water resource managers who must detect impairment and determine causation. The objectives of this study were: 1) to develop a macroinvertebrate Index of Biotic Integrity (IBI) for the Northwestern Great Plains (NWGP) and compare it to an existing Northern Glaciated Plains (NGP) IBI, and 2) to identify and evaluate statewide abiotic drivers of biotic integrity. The IBIs were hypothesized to differ regionally, as were abiotic drivers explaining their variation. Physical, chemical, and biological data were collected at 65 perennial, wadeable stream sites in the NWGP. Biological metrics were calculated and then optimized using six statistical screening tests. Random forest modeling, cluster analysis, and nonmetric multidimensional scaling (NMDS) were employed to evaluate abiotic drivers. Final IBI metrics in the NWGP were family richness, richness of non- Insecta, richness of collector-filterers, and richness of Plecoptera, Odonata, Ephemeroptera, and Trichoptera. IBI scores ranged between 3 and 93 (x̅ = 37) and successfully differentiated between least impaired and most impaired sites (Kruskal- Wallis ANOVA p < 0.01). All metrics comprising the NWGP and NGP IBIs differed. The NWGP IBI yielded a similar range of scores to the NGP IBI (range: 24 to 100, x̅ = 58). No significant differences were found in IBI scores among level IV ecoregions of either the NWGP or NGP. Random forests identified local physical and chemical drivers of biotic integrity and watershed-scale drivers of in-stream abiotic conditions. Major drivers of IBI scores included salinity, specific conductance, fine and gravel substrates, total riffle length, and stream discharge (46% of variation explained). Each component IBI metric was related to a unique combination of drivers (17 to 56% of variation explained). Ecoregion (level III and level IV) and river basin variables generally improved model performance for local abiotic drivers (x̅ = 9% increase in variation explained). Cluster analysis groupings largely separated NGP and NWGP streams, while NMDS ordinated groups along a land use gradient differentiating level III ecoregions. Human use, pasture and hay, and nitrogen loading were more positively associated with NGP groups, while total and riparian herbaceous cover and turbidity were more positively correlated with NWGP groups. This study developed the first South Dakota ecoregion-wide IBI west of the Missouri River, and greatly expanded geographic coverage and use of the IBI in making impairment decisions. These results can also be incorporated in causal analysis of impairment.

Library of Congress Subject Headings

Water quality biological assessment -- South Dakota.
Ecological integrity -- South Dakota.
Stream ecology -- South Dakota.
Freshwater invertebrates -- South Dakota.
Biotic communities -- South Dakota.


Includes bibliographical references (pages 91-116)



Number of Pages



South Dakota State University


In Copyright - Educational Use Permitted