9371

• UCREFRP Catalog Number: 9371
• Author: Levee Removal Evaluation Group.
• Title: Green River levee Removal and Floodplain Connectivity Evaluation.
• USFW Year: 1998.
• USFW - Doc Type: CAP-6 LR,
• Copyright Material: NO

1

I

1
1
1
1
1
1

1

1
APPROACH
The purpose of this study is to determine whether the removal of floodplain levees will
benefit endangered fishes in the middle Green River. A design to test whether nutrients,
vegetation, invertebrate biomass, and/or fish respond to levee removal and floodplain inundation
was developed for selected sites and reaches in the Green River. The study design was stratified
to test for differences in nutrients, vegetation, invertebrate biomass, and fish abundance in natural
floodplains versus breached areas among sites that hold water following recession of flood
waters and those that do not, within areas that were determined to have high and low nonnative
fish densities (according to unpublished data by Todd Crowl, Utah State University). Study sites
were assigned using the following hierarchical categories.
1) Low nonnative fish densities versus high nonnative fish reaches
2) Sites that hold water following flood flow recession (depressions) versus sites that do
not (terraces)
3) Natural versus breached (application) sites
During the first year of the levee removal study, six sites were evaluated including three
terrace and three depression sites (Table 1.1). In 1997, two sites were added (two depressions
were added, one in the low nonnative density reach and one in the high nonnative density reach),
and in 1998 one depression (low nonnative density) and one terrace (low nonnative density) were
added. As will be discussed in later chapters, sites did not always function as terraces or
depressions.
The analytical approach described in the initial scope of work involved sampling within
individual floodplain sites and riverine environments adjacent to the floodplain both during
inundation and after river flows receded to baseflow conditions. Although sampling within the
floodplains was effective in determining the response of inundation inside study floodplains
(hypotheses 1-3), riverine collections did not represent a response to the adjacent floodplain
(hypothesis 4). Following the first year of the study it was evident that fish abundance and
distribution in the river adjacent to study floodplains were influenced more by the surrounding
environment than the relatively small floodplain connected at the site. Irving and Burdick (1995)
identified approximately 5,050 acres of middle Green River floodplain inundated in May 1993,
whereas between 100 and 200 floodplain acres were added in 1997 (depending on flows) and
between 950 and 1,600 acres were added in 1998. In an effort to adjust our sampling design
accordingly, the design of the riverine collections in 1997 and 1998 was modified to randomly
sample a given distance of river within six longitudinal reaches in the river. In this manner, the
riverine samples represented a longitudinal gradient of the river rather than an immediate
response to the adjacent floodplain study site. Thus, analysis of fish abundance and distribution
in the river reflects responses to the river reaches in general and not solely to the influence of tl
study floodplains. B
Following the analysis of fish abundance and distribution data, it will be critical to
determine the criteria in which floodplain enhancement is considered beneficial or deient
a?
4