A preliminary biological assessment of Kirwin National Wildlife Refuge

              A Preliminary Biological

Assessment of Kirwin

National Wildlife Refuge

Biological Technical Publication

BTP-R6004-2004

U.S. Fish & Wildlife Service

A Preliminary Biological Assestment of Kirwin

National Wildlife Refuge

Biological Technical Publication

BTP-R6004-2004

Murray K. Laubhan

U.S. Geological Survey, Northern Prairie Wildlife Research Center

8711 37th Street SE, Jamestown, North Dakota, USA 58401

U.S. Fish & Wildlife Service

Author Contact information:

Murray K. Laubhan, U.S. Geological Survey,

Northern Prairie Wildlife Research Center, 8711 37th

St. SE, Jamestown, ND 58401.

Phone: 701/ 253 5534,

Fax: 701/ 253 5553

Email: murray_laubhan@usgs.gov.

For additional copies or information, contact:

Wayne J. King, USFWS Region 6, Regional Refuge

Biologist. P.O. Box 25486 Denver Federal Center

Denver, Colorado 80225-0486

Phone: 303/236 8102

Fax: 303/236 4792

Email: wayne_j_king@fws.gov

Recommended citation:

Laubhan, M. K. 2004. A preliminary biological

assessment of Kirwin National Wildlife Refuge,

Phillipsburg, Kansas. U.S. Department of Interior,

Fish and Wildlife Service, Biological Technical

Publication, BTP-R6004-2004, Washington, D.C.

Series Senior Technical Editor:

Stephanie L. Jones, USFWS, Region 6 , Nongame

Migratory Bird Coordinator, P.O. Box 25486 Denver

Federal Center Denver, Colorado 80225-0486

Associate Editor:

Wayne J. King, Regional Refuge Biologist USFWS,

Region 6, P.O. Box 25486, Denver Federal Center,

Denver, Colorado 80225-0486

Table of Contents

Summary .......................................................................................................................................................... 1

Acknowledgments ........................................................................................................................................... 2

Introduction ..................................................................................................................................................... 3

Description ....................................................................................................................................................... 4

Refuge Establishment and Authorities .......................................................................................... 4

Location and Topography ................................................................................................................. 4

Climate .............................................................................................................................................. 5

Geology .............................................................................................................................................. 5

Groundwater ..................................................................................................................................... 6

Soils ................................................................................................................................................... 6

Vegetation ......................................................................................................................................... 6

Wildlife Conservation ...................................................................................................................................... 8

Kirwin National Wildlife Refuge ...................................................................................................... 8

State of Kansas ................................................................................................................................. 8

Bird Conservation Region ................................................................................................................. 9

Birds of Conservation Concern ........................................................................................................ 9

North American Waterfowl Management Plan ............................................................................. 9

Partner’s In Flight North American Landbird Conservation Plan............................................10

Shorebird Conservation Region ......................................................................................................10

Waterbird Conservation Region .....................................................................................................10

Playa Lakes Joint Venture ..............................................................................................................10

Community Types ..........................................................................................................................................11

Reservoir (Deepwater) ....................................................................................................................11

Managment Potential ....................................................................................................................13

Shoreline ..........................................................................................................................................13

Managment Potential .....................................................................................................................15

Riparian ............................................................................................................................................15

Managment Potential .....................................................................................................................18

Upland ..............................................................................................................................................19

Managment Potential .....................................................................................................................21

Public Use .......................................................................................................................................................22

Literature Cited ..............................................................................................................................................23

Appendix A. Potential information needs that may be required to address the recommendations

provided in the evaluation section ..................................................................................................................27

Appendix B. Scientific and common names of animals and plants known to occur on Kirwin NWR.

Information on fish, amphibians, and reptiles obtained from Kansas Department of Wildlife and

Parks (refuge files dated 01/30/2003) ..............................................................................................................29

Appendix C. Total annual use days, average annual populations, and peak populations, respectively

for the following waterfowl groups using Kirwin National Wildlife Refuge between 1983 and 2001:

American Coot and dabbling ducks excluding Mallard (a – c), diving ducks (d – f), Canada Goose

and Mallard (g – i), and White-fronted Goose and Snow Goose (j – k). .......................................................33

Appendix D. Conservation status for avian species based on regional and national plans ...................... 39

1

Summary

This report represents an initial biological

assessment of Kirwin National Wildlife Refuge

(NWR) conducted as part of the pre-planning phase

for development of a Comprehensive Conservation

Plan (CCP). Stimulation for the report was based on

the concept that future decisions related to the

biological portion of the CCP will be based on the best

available science. Therefore, an attempt is made to

integrate information from many different scientific

disciplines (e.g., geology, hydrology, biology) to help

identify ecological constraints and opportunities

imposed by the land base being considered.

Consequently, there is a greater likelihood of

identifying potential shortcomings of various

management actions during the decision making

process.

Information in this report is based on a relatively

limited number of published articles, past notes, and

observations. An attempt was made to locate

sufficient relevant information necessary to formulate

more definitive ideas and provide additional context.

Thus, the information provided below is known to be

incomplete and a more thorough synthesis will be

required. Further, interpretation of published

information can vary among individuals and the refuge

staff is encouraged to review the documents cited in

this report. Their years of observation and experience

managing the refuge are invaluable to ensuring that

information used to make decisions is applicable.

Consequently, some sections contain information that

was not fully explored in the evaluation section;

however, it was retained because it may be useful as

the refuge staff and core CCP team examines

different management options. Finally, decisions

regarding management of each individual community

should be combined and evaluated collectively to

identify potential conflicts. Although this may seem

simple and straightforward, this task often is difficult

because it frequently requires an iterative approach to

ensure that important issues have not been omitted.

This report does not contain conclusions, nor does it

advocate any opinions (favorable or unfavorable)

regarding the biological program. Further, concepts

such as alternatives, goals, and objectives, are not

discussed. The core CCP team will address these

topics. Rather, it represents a summary that

hopefully will be used to focus future discussion

regarding biological data needs and approaches for

using this information to make decisions. Ultimately,

however, scientific information alone will not lead to a

definitive decision regarding future direction, because

biology is only one of many components that must be

considered in the evaluation. Therefore, it is

recommended that U. S. Fish and Wildlife Service

(FWS) personnel responsible for determining the

future direction of the refuge be consulted to establish

guidelines and agree on the approach that will be used

in evaluating the biological program prior to

proceeding.

2

Acknowledgments

Prior to writing the report, the author was invited to

the refuge for a two-day meeting with Erich Gilbert

(refuge manager), Craig Mowry (assistant refuge

manager), and Wayne King (regional refuge biologist).

The purpose was to become familiar with the site,

discuss management opportunities and constraints,

and identify potential types of information that would

assist the staff in developing a credible biological plan

to guide future management. These individuals

contributed significant time and insight regarding

management of the refuge. The assistance of Rachel

Lambhan and Mark Fly in locating and processing

information used in this report is greatly appreciated.

Thanks also to the following individuals for providing

reviews of an earlier draft: E. Gilbert, R. A. Gleason,

S. L. Jones, W. J. King, R.A. Laubhan, C. Mowry, D.

M. Mushet, and J. D. Petty.

3

Introduction

Stimulation for this report was generated by passage

of the 1997 National Wildlife Refuge System

Improvement Act (NWRSIA) that requires each

refuge in the National Wildlife Refuge System to

develop a CCP that includes goals and objectives that

are based on the best available science. To accomplish

this mandate, Region 6 of FWS contracted with the

Biological Resources Division of the U. S. Geological

Survey (USGS) to inspect refuge habitats and

synthesize available information pertinent to the

management of Kirwin NWR as part of a pre-planning

phase to guide development of a CCP. This report

represents such a synthesis.

The brevity of the site visit did not allow for detailed

discussions between USGS and FWS personnel, but it

did provide the opportunity to exchange thoughts

regarding the information needed to evaluate the

biological program. Thus, the ideas contained within

this report are of a general nature and should be

viewed as a collaborative effort that involved the

refuge staff. Additional work will be required to

objectively evaluate the biological program, and this

report should be viewed as an initial effort to start this

process. In addition, there are alternative ways of

approaching an evaluation that would require different

levels and types of information. Therefore, the

responsibility of FWS is to review the report and

other relevant materials, discuss available options

with appropriate personnel, and determine if the

identified information needs and recommendations

outlined in this report are acceptable and represent

the preferred manner of proceeding.

General descriptive information on refuge

establishment, topography, climate, geology, soils,

vegetation, and wildlife is intended to provide a brief

background of the refuge with regard to functions,

processes, and values. This information is important

as a baseline for understanding the impact of past

land alterations and for developing management

guidelines for the future. In contrast, the section on

conservation is intended to provide perspective

regarding potential refuge contributions to natural

resources based on conservation plans that have been

developed for application at larger geographic scales

that encompass the refuge. The section on evaluation

of community types discusses in more detail the

attributes of various communities that occur within

the refuge boundary. For ease of discussion, four

broad communities were delineated as follows: (1)

Reservoir Pool, (2) Shoreline, (3) Riparian, and (4)

Upland. For each community, a brief synopsis of

historic and current conditions is provided. Also,

potential management options are discussed along

with some associated benefits and detriments.

Appendix A summarizes the information needs that

may be required to address the recommendations

provided in the evaluation section. These

recommendations largely are those of the author and

are based on thoughts that resulted from discussions

with FWS personnel during the site visit trip in

March. Therefore, the list is incomplete from a

biological perspective and largely ignores recreational

and other considerations. Additional effort will be

required by FWS personnel to identify and integrate

issues, concerns, and recommendations through

internal discussions and public scoping. Although

some scoping has already occurred, hopefully this

report will assist future efforts by providing some

background biological information. Obviously, this

represents only the first step in a long process and

additional work is necessary.

4

Description

Refuge Establishment and Authorities

Kirwin NWR is an overlay project on a U. S. Bureau of

Reclamation (BOR) irrigation and flood control

reservoir known as Kirwin (U. S. Fish and Wildlife

Service, URL http://kirwin.fws.gov/refugemap.htm).

The reservoir was constructed before irrigation

became widely practiced (Christensen 1999).

Dedicated in 1955, the reservoir has a contributing

drainage area of 3,540 km2 (1,367 mi2) and a total water

storage capacity of 38,797 ha-m (314,550 ac-ft). The

storage capacity is allocated among flood control

(36.532 ha-m [215,115 ac-ft]), conservation storage

(11,057 ha-m [89,650 ac-ft]), and inactive or dead

storage (1,207 ha-m [9,785 ac-ft]) (Christensen 1999).

Kirwin NWR was established in 1954 pursuant to the

Fish and Wildlife Coordination Act (16 U.S.C. ¤ 664)

for the “…conservation, maintenance, and

management of wildlife, resources thereof, and its

habitat thereon,…” (U. S. Fish and Wildlife Service,

URL http://refuges.fws.gov/policymakers/). The

BOR owns the land and controls reservoir water

levels, whereas the refuge staff manages all other

activities on the land and water (U. S. Bureau of

Reclamation, URL http://www.usbr.gov/gp).

Location and Topography

The 4,366-ha (10,788 ac) refuge includes Kirwin

Reservoir and bordering areas in southeast Phillips

County, Kansas. Topography of the area is

characterized by rolling hills, the gently sloping

Kirwin terrace, and a narrow river valley formed by

the North Fork of the Solomon River (Leonard 1952,

Christensen 1999). The refuge encompasses portions

of the North Fork Solomon River and Bow Creek.

These rivers drain an area of 359,874 ha (889,248 ac)

above the reservoir (U.S. Bureau of Reclamation 2002).

The altitude of the terrace ranges from an elevation of

602 m (1,975 ft) in Phillips County to 445 m (1,460 ft) in

Mitchell County. The gradient of the North Fork

Solomon River channel is about 1.3 m/km (7.1 ft/mi) in

Phillips County.

The refuge is located in the North Fork Solomon

River Sub-Basin. The North Fork Solomon originates

in western Thomas County, Kansas, approximately

193 km (120 mi) west of Kirwin Dam and drains an

area of 3,556 km2 (1,373 mi2) (U.S. Bureau of

Reclamation 2002). Elevation of the area ranges from

about 430 m (1,410 ft) at the mouth of North Fork

Solomon River in Mitchell County, to about 640 m

(2,100 ft) in the western part of Phillips County. Like

other valleys in north-central Kansas, the North Fork

Solomon valley and its tributaries are asymmetrical

and typically have precipitous south walls and gently

sloping north walls. According to Leonard (1952) the

valley consists of broad undissected terraces and a

floodplain width that varies from 201-805 m (660-2,640

ft). The relief from the stream channel to the top of

the Kirwin terrace, which lies above flood level, is as

much as 12 m (40 ft). The Kirwin terrace slopes

gently (gradient = 1.4 m/km [7.3 ft/mile]), is

moderately well drained, and represents the primary

area of cultivated farmland. Historically, the elevation

of the floodplain varied between 4.6-7.6 m (15-25 ft)

below the Kirwin terrace and 3.7-4.9 m (12-16 ft) above

the water level of the river. Many of the ephemeral

streams that drain the uplands disappear on the

surface of the terrace, which suggests these streams

contribute a large part of their water to the

groundwater reservoir in the area.

Agriculture and ranching have been the primary

economic forces in the area since the early 1800’s (U.S.

Bureau of Reclamation 2002). In Phillips County, the

land in farms (220,398 ha [554,603 ac]) accounted for

97.8% of the county land base (229,478 ha [567,040 ac])

in 1997. Further, of the land in farms, cropland

(130,329 ha [322,043 ac]) accounted for 56.8% of the

county land base (U. S. Bureau of Reclamation 2002).

Representative crops produced in Phillips County

during 1997 included corn, sorghum, wheat, oats,

soybeans, and hay (U. S. Department of Agriculture

National Agricultural Statistics Services, URL http://

www.nass.usda.gov/census/census97/highlights/ks/

ksc074.txt).

The drainage area below the reservoir is 348,004 ha

(859,918 ac), of which 6,325 ha (1.8%; 15,630 ac) is

irrigated (Christensen 1999). The BOR operates

Kirwin Irrigation District Number 1 downstream

from the reservoir and provides surface water outflow

to irrigate as much as 4,650 ha (11,490 ac) of farmland

(U.S. Bureau of Reclamation 2002) that generally is

within 8 km (5 mi) of the North Fork Solomon River

(Christensen 1999). As mentioned previously, the

drainage area above the reservoir is 359,874 ha (889,248

ac), of which 24,525 ha (6.8%; 60,600 ac) is irrigated.

The use of surface water for irrigation first occurred

in 1928 with the establishment of two pumping plants,

but by 1946 a total 13 plants had been established

(Leonard 1952). Currently, the primary source of

irrigation water is from alluvial wells, which have

increased from <10 in 1949 to>150 in 1993 (U.S.

Bureau of Reclamation 2002).

The North Fork Solomon Sub-Basin is part of the

Solomon River Basin, which extends across parts of 17

Kansas counties and includes the Solomon River and

its major tributaries, the North Fork and South Fork

Solomon Rivers. The rivers and associated tributaries

drain approximately 17,716 km2 (6,840 mi2) of mainly

agricultural land (Christensen 1999). In addition to

Kirwin Reservoir, the BOR also operates two

additional reservoirs in the basin; Webster Reservoir

on the South Fork Solomon River and Waconda Lake

at the confluence of the North and South Fork

Solomon Rivers. The three reservoirs provide water

for irrigation, municipal, industrial, and domestic use;

flood control; recreation; and fish and wildlife habitat

(Christensen 1999).

The elevation of the basin varies from approximately

1,006 m (3,300 ft) in the west to 351 m (1,150 ft) in the

east near the town of Solomon, Kansas. The average

gradient of the region is approximately 2.7-2.8 m/km

(14-15 ft/mi). The average gradient for the main stem

123°

5

Solomon River is 0.9 m/km (5 ft/mi) (U.S. Bureau of

Reclamation 2002). Both the North Fork and South

Fork Solomon Rivers derive their flows from

precipitation runoff and groundwater discharge from

underlying aquifers (U.S. Bureau of Reclamation

2002).

Finally, the Solomon River Basin is located in both the

Smoky Hills and High Plains Physiographic Regions

of the Great Plains Physiographic Province

(Launchbaugh and Owensby 1978). The High Plains

region of Kansas extends from the western state line

eastward into Graham County. The topography is

characterized by flat to gently rolling hills with

narrow, shallow valleys of low relief. Sand, gravel, and

porous rock cover most of the region. The Smoky

Hills Region, which encompasses Kirwin NWR, is

composed of three distinct hill ranges. Steep chalk

bluffs characterize the western range of hills, whereas

Greenhorn Limestone and Dakota Sandstone cap the

middle and eastern range of hills, respectively. The

eastern boundary of the Smoky Hills Region is in Clay

County and is adjacent to the Flint Hills

Physiographic Region (Kansas Geological Survey,

URL http://www.kgs.ukans.edu/Physio/physio.html).

Vegetation communities within this region are

classified as mixed-grass prairie with forested river

bottoms (Kuchler 1974).

Climate

The climate of the Solomon Basin is classified as

subhumid. Summers are characterized by hot days

and cool evenings, whereas winters are normally

moderate with light snowfall and only occasional short

periods of severe cold. The average length of the

growing season is about 167 days (Leonard 1952) and

the frost-free period extends from 29 April to 13

October (Albertson 1937). Long-term climate data (5/

1/1952 – 12/31/2002) was obtained from the National

Climate Data Center cooperative station located at

Kirwin Dam (http://lwf.ncdc.noaa.gov/oa/climate/,

station # 144357). The mean monthly maximum

temperature ranged from 3.1 C(37.5°F) in January to

33.4°C (92.2°F) in July, whereas mean monthly

minimum temperature ranged from -11.3°C (11.6°F) in

January to 17.8 °C (64.0°F) in July.

Average annual precipitation during this period was

58.5 cm (23.0 in), with 44.2% of total annual

precipitation occurring in May (mean = 10.2 cm [4.0

in]), June (mean = 7.8 cm [3.1 in]), and July (mean =

7.7 cm [3.0 in]). However, not all of this moisture is

necessarily available for plant growth because

evaporation is also occurring during these months.

Graphs obtained from the National Oceanic and

Atmospheric Administration (URL http://

www.noaa.gov), which depict ranges of evaporation for

the entire U.S, were used to obtain a coarse estimate

of 28.2 cm (11.1 in) of annual evaporation for north-central

Kansas. Months with highest evaporative

losses were June (5.6 cm [2.2 in]), July (6.1 cm [2.4 in]),

and August (5.6 cm [2.2 in]).

The Palmer Drought Severity Index (PDSI) was

developed to represent the severity of dry and wet

periods based on monthly temperature and

precipitation data as well as the water holding

capacity of soils at a location (Palmer 1965). Thus, this

measure provides a method integrating the above

information. For north-central Kansas, the long-term

PDSI (1895-2002) indicates cyclic patterns of drought

and wetness. During 1989 and 1990, portions of 1991,

and for a brief period in 2001, this region experienced

severe drought, but from late 1993 through mid-1994

the area was extremely wet (National Climate Data

Center, URL http://lwf.ncdc.noaa.gov/oa/climate/).

Geology

The surface geology of the Solomon Basin consists of

unconsolidated and consolidated rocks. The

unconsolidated surface deposits consist of Quaternary

alluvium, loess, and the Tertiary Ogallala Formation,

whereas Cretaceous and Permian rocks form the

bedrock. In general, the basin is underlain by strata

of marine origin (Christensen 1999). The dendritic and

asymmetrical drainage pattern of the Solomon River

suggests the lack of faults and folds and the presence

of flat underlying rock units (U.S. Bureau of

Reclamation 1984).

The Greenhorn Limestone, Graneros Shale, and

Dakota Sandstone outcrop as far east as western Clay

County, Kansas. Permian beds outcrop in counties

farther east. The Greenhorn Limestone consists of

alternating beds of calcareous shale and chalky

limestone. The Graneros Shale is noncalcareous,

fissile shale with sandstone lenses. The Dakota

Formation consists of lenticular sandstone bodies that

are embedded in mudstone. Generally, the sandstones

are fine to medium grained, well sorted, and exhibit

cross-bedding (Kansas Department of Agriculture

Division of Water Resources), URL http://

www.accesskansas.org/kda/dwr/).

The North Fork Solomon River is underlain by, or

incised into, Cretaceous beds that generally dip to the

west, whereas the erosional surface generally slopes

to the east. The oldest subsurface rocks at the

eastern end of the basin are of the Sumner Group.

Above the Sumner Group is Cretaceous marine

sediment beginning with the Dakota Formation, which

is overlain by the Cheyenne Sandstone, Kiowa Shale,

Graneros Shale, Greenhorn Limestone, and Carlile

Shale. The Carlile Shale is exposed in stream valleys

in Phillips County. Above the Carlile Shale is the

Niobrara Formation, which is exposed in much of the

North Fork Solomon River Basin (Leonard 1952), and

the Pierre shale, of which there is only one known

small outcrop in the basin upstream from Webster

Reservoir (Moore and Landes 1937, Ross 1991). The

Pierre Shale lies conformably on the Niobrara Chalk,

which is a gray, shaly, fossiliferous chalk with

weathered surfaces. The chalk contains bentonite

beds and limonite concretions (Kansas Department of

Agriculture Division of Water Resources).

The north and south divides of the Solomon River are

capped by remnants of the Ogallala formation in the

western part of the Solomon Basin, whereas the

uplands and valley walls over much of north-central

6

Kansas are composed of loess of the Sanborn

formation that was deposited during glacial retreat

(Leonard 1952). The Ogallala formation was formed

during the Pliocene by eastward flowing streams that

filled pre-existing valleys with alluvial sediments.

Continued deposition of alluvial sediments formed a

broad alluvial plain. The Ogallala consists mainly of

silt, sand, gravel, and “mortar beds” formed by

cementation of sediments with calcium carbonate.

However, lenticular beds of well-sorted sand, gravel,

bentonite, and volcanic ash also exist. The Ogallala

lies unconformably on the Pierre Shale in the western

part of the basin and on the Niobrara Formation in the

eastern part of the basin. The surface of the Ogallala

dips to the east-northeast and the average gradient is

2.1 m/km (11 ft/mi) (Kansas Department of

Agriculture Division of Water Resources).

Narrow belts of alluvium, most of Recent age, are

adjacent to the Solomon River channel and its

tributaries and occupy the floodplain (Leonard 1952).

The alluvium consists mainly of gravel, sand, silt, and

some clay. However, loess also may occur along

major streams. The loess is underlain by stream-deposited

sands that are in a high terrace position

with respect to the valleys (Leonard 1952). At several

places in the floodplain, wind has deposited sand from

the alluvium into dunes or in thin layers that cover the

terrace surfaces (Leonard 1952). These areas of sand

deposition occur in Phillips County, but thickness of

the fluvial and loess deposits is < 3.0 m (10 ft) (Kansas

Department of Agriculture Division of Water

Resources). A map illustrating the locations of these

geologic features on the refuge was developed by

Johnson and Arbogast (1993) and can be obtained from

the Kansas Geological Survey (URL http://

www.kgs.ukans.edu/General/Geology/County/nop/

phillips.html).

Groundwater

The Sanborn formation, which consists of a thin layer

of loess that overlies Cretaceous rocks, is a locally

important source of groundwater (Leonard 1952). The

most important aquifer in the area, however, occurs in

the deposits underlying the Kirwin terrace surface.

In general, this terrace is underlain by 9.1-27.4 m (30-

90 ft) of unconsolidated deposits (e.g., coarse textured

sand and gravel) that is quite permeable and lies

below the water table (Leonard 1952). The broad,

nearly flat terrace surface constitutes a large

recharge area and streams that originate in nearby

hills contribute additional recharge. Groundwater

moves laterally through the terrace deposits and into

the alluvium or into the channel of North Fork

Solomon River. Thus, the water table in the Recent

alluvium is continuous with the water table in the

terrace deposits and with the water level in the

flowing streams. The coarse nature of the alluvium

makes it an important potential source of

groundwater (Leonard 1952). Hydraulic conductivity

has been estimated at 51.8 m/day (170 ft/day) with an

average transmissivity of 241.5 m2/day (2,600 ft2/day)

(Phillips 1980). Well yields vary from 38-1893 l/min

(10-500 gal/min).

The water table in the valley slopes from east to west,

and from the sides of the valley toward the center.

The downstream slope of the water table varies from

about 2.2 m/km (11.5 ft/mi) in western Phillips County

to about 1.2 m/km (6.4 ft/mi) near the town of Kirwin

(Leonard 1952). Most ephemeral streams in the area

are above the water table and, when flowing, probably

contribute some water to the groundwater. In

contrast, the Solomon River and Bow Creek are

gaining streams (e.g., flow in these streams is

partially maintained by groundwater that seeps into

the channel) (Leonard 1952).

Soils

In north-central Kansas, soils are composed primarily

of Mollisols in the suborder Ustolls. A dark surface

horizon rich in bases are primary characteristics of

Mollisols. Nearly all have a mollic epipedon, but

many also have an argillic, nitric, or calcic horizon.

Specifically, soils of the North Solomon Valley are

primarily fertile, silty clay loams derived from

reworked loess (Leonard 1952), some of which are rich

in selenium (U. S. Geological Survey, URL http://

ks.water.usgs.gov/Kansas/studies/ressed/). The soils

in valleys are slightly sloping, friable, and generally

have high agricultural productivity. In the western

and central parts of the basin, soils are generally

friable and relatively impermeable, with some silt

loam and loess. The more level soils in the western

and central parts of the basin are used for grain

cultivation and are moderately productive. The soils

in the eastern part of the basin range from shallow

sands to thick clays and generally have low

agricultural productivity (U.S. Bureau of Reclamation

1984).

Vegetation

Historically, the floodplains of both rivers supported

woody vegetation, tall grasses, and forbs, whereas the

uplands largely were mixed-grass prairie (Kuchler

1974). However, human settlement and associated

land use activities altered historic processes and,

ultimately, the plant and wildlife communities.

Although construction of the reservoir in 1952

represents an obvious perturbation that altered

various ecological aspects of the rivers and associated

floodplains, alteration of the uplands also occurred

prior to refuge establishment. The area surrounding

Kirwin NWR was cultivated extensively and also was

used to pasture cattle prior to 1945 (Leonard 1952).

The general location of these activities was related to

topography and soils. Much of the cultivation

occurred on the terrace due to the presence of fertile,

moderately well drained soils. In contrast, cattle

production typically occurred on areas bordering the

valley where cultivation was prevented because of

steep gradients or presence of rocky soil formed on

chalk and limestone rocks (Leonard 1952). Although a

detailed history of past land use was not developed for

the refuge, a study conducted by Rezsutek (1990)

supports this scenario. He reports that lands

composing the refuge were formerly family farms and

much of the native grassland in the area had been

turned under for the production of crops prior to

establishment. Further, although J. Launchbaugh (in

7

K. Launchbaugh’s field notes) described some areas

as “native prairie”, Rezsutek (1990) suggests they

may have been “restored prairie”.

Grassland, cropland, deepwater and shoreline habitats

of the reservoir, and riparian zones bordering the

tributary rivers are dominant communities on the

refuge. In addition, shelterbelts, palustrine wetlands,

and chalk bluffs also occur within the refuge

boundary. When the reservoir is at conservation pool

(527.1 m [1729.25 ft]) there are 2,050 ha (5,065 ac) of

surface water and about 60 km (37 mi) of shoreline.

According to refuge staff, croplands occupy about 486

ha (1,200 ac) at conservation pool. Remaining

community types encompass 1,826 ha (4,513 ac) at

conservation pool and include grasslands,

shelterbelts, and riparian habitat. However, the area

of each community was not available.

Of the community types on the refuge, the reservoir

(e.g., deepwater habitat) and associated shoreline,

cropland, and shelterbelts were not present prior to

human settlement. Consequently, the area of native

communities on the refuge has been reduced.

Further, the grassland and riparian communities,

although historically present, have been severely

altered floristically and structurally. For example,

two plant species that may occur on the refuge

(western prairie-fringed orchid and Meads milkweed;

scientific names of all species mentioned in this report

are given in Appendix B) are listed as threatened or

endangered under the Endangered Species Act (ESA)

of 1973 (16 U.S.C. 1531-1544, 87 Stat. 884; URL: http://

laws.fws.gov/lawsdigest/esact.html). More detailed

information is presented in the section entitled

Evaluation of Community Types.

8

The 1997 NWRSIA mandates that each refuge

develop a CCP that is consistent with the principles of

sound fish and wildlife management and available

science. Further, this act also specifies that each CCP

shall identify and describe the purposes of each

refuge; the distribution, migration patterns, and

abundance of fish, wildlife, and plant populations and

related habitats; significant problems that may

adversely affect the populations and habitats of fish,

wildlife, and plants and the actions necessary to

correct or mitigate such problems; and, to the

maximum extent practicable and consistent with this

Act, be consistent with fish and wildlife conservation

plans of the State in which the refuge is located.

The purpose of this report was not to fully develop

information on all species potentially occurring on the

refuge. However, some general future direction must

be specified with regard to wildlife given the purpose

for refuge establishment. Therefore, this report

concentrates on the importance of the refuge to

migratory birds because these species represent a

primary FWS responsibility according to the

requirements of the Migratory Bird Treaty Act of

1918 (16 USC 703-711; 40 Stat. 755; URL: http://

laws.fws.gov/lawsdigest/migtrea.html). This should

not be interpreted as meaning other vertebrates,

invertebrates (e.g., butterflies), and plants can be

ignored since these organisms are important as

individual entities and also because they are critical to

proper system function. Rather, information

regarding the habitat requirements of these species

also should be used in evaluating the direction of

future refuge management to ensure that valuable

opportunities are not overlooked.

Baseline information on the avian community of

Kirwin NWR was developed using a checklist of bird

species sighted on the refuge (Igl 1996). Scientific

names for all species mentioned are provided in

Appendix B; birds follow the American Ornithologists’

Union Committee on Classification and Nomenclature

(American Ornithologists’ Union 1998, 2000, 2002,

2003). In addition, refuge files of duck, goose, and

swan counts were used to generate graphs of total

annual use days, average annual populations, and

average peak populations spanning a 20-year period

(Appendix C). There are several qualifying factors

that must be considered when considering this

information. First, the list from the website

represents a composite of all birds that have been

sighted over a long time; thus, the list may not

accurately represent the current avian community.

Second, the list only designates occurrence; thus, the

contribution (e.g., source/sink) of the refuge to the

species population is not known. Although refuge

waterfowl counts contribute information on

abundance, it is difficult to compare data among years

with any certainty because routes and areas surveyed

are not available and likely have varied among years.

In addition, count information was not collected based

on community type or bird activity. Thus, it was not

possible to use this information to determine local use

patterns or assess relative importance of different

communities. Similarly, the avian list (Appendix B)

was not developed using standardized protocols and

methods, but based on the long-term refuge bird list,

with an additional 29 species added from a research

study conducted in the riparian zone on the refuge in

the mid-1990s (Sevigny 1998). Regardless of these

constraints, this list is valuable because it can help

focus discussion among individuals (e.g., FWS

personnel, core CCP team) responsible for

determining the future management direction of the

refuge.

Finally, depending on the source and type of

information sought, Kirwin NWR is located in many

different regions. In all cases, there is considerable

overlap among boundaries although the area

encompassed tends to vary greatly. To provide an

overall perspective, relevant information regarding

species of concern and population targets contained in

these plans has been summarized, but no attempt has

been made to prioritize or make decisions regarding

species or guilds that should receive attention. In

some cases, species considered to be of conservation

concern at a regional level may not be of concern at a

national level, or vice versa. Such differences do not

indicate discrepancies; rather, it suggests differences

in distribution and population status at different

geographical scales. However, the small size of the

refuge precludes providing quality habitat for all

species and decisions likely will be required to

evaluate tradeoffs in management approaches and for

development of detailed habitat objectives.

Kirwin National Wildlife Refuge

The refuge bird list includes 233 species, of which 45

are recorded as nesting and four (Piping Plover, Bald

Eagle, Whooping Crane, and Least Tern) are listed as

threatened or endangered under the ESA. Kirwin

NWR also is recognized as a globally Important Bird

Area (IBA) by the American Bird Conservancy (URL

http://www.abcbirds.org/iba/kansas.htm). The IBA

program, initiated by BirdLife International in

Europe during the mid-1980’s, was developed to

recognize and support sites of importance to birds.

Based on the criteria developed by BirdLife

International, an IBA must maintain and support one

or more of the following: (1) species of concern (e.g.,

threatened and endangered), (2) species with

restricted ranges, (3) species vulnerable because of

population concentration, and (4) species vulnerable

because they occur at high densities due to their

congregative behavior (Kushlan et al. 2002).

State of Kansas

Wildlife resources in the state of Kansas were

historically rich and varied. Circa 1865, it is estimated

that the Great Plains portion of Kansas supported as

many as 407 bird species, including about 178 breeding

species (Fleharty 1995). Fish and herpetofauna were

not as rich due to the paucity of water, but included 30

fish species, 13 amphibian species, and 46 species of

reptiles (Fleharty 1995). Although Kirwin NWR is

small relative to the geographic area of Kansas, the

historic communities composing the refuge likely

supported many species mentioned in earlier

Wildlife Conservation

9

accounts. Appendix B contains a list of animals and

plants known to occur on the refuge.

Bird Conservation Region

Kirwin NWR is located within the Central Mixed-grass

Prairie Bird Conservation Region (BCR 19), an

ecologically distinct region of 535,734 km2 (206,848 mi2)

with similar bird communities, habitats, and resource

management issues. The area encompassed by BCR

19 extends from the edge of shortgrass prairie on the

west to the beginning of tallgrass prairie to the east

and includes portions of Texas, Oklahoma, Kansas,

and Nebraska (North American Bird Conservation

Initiative, URL http://www.nabci-us.org/map.html).

Although large areas in this region have been

converted to agriculture, areas of high-quality

grassland (e.g., Nebraska Sandhills) still remain,

including some of the best habitat for the Greater

Prairie-Chicken and Henslow’s Sparrow, and

sandbars along the larger rivers host a large

percentage of the continent’s breeding Least Tern

population. The region also is a spring migration area

for the American Avocet, Semipalmated Sandpiper,

and Buff-breasted Sandpiper.

Birds of Conservation Concern

The Birds of Conservation Concern (BCC) is the most

recent effort to satisfy the 1988 amendment to the

Fish and Wildlife Conservation Act, which mandates

FWS to “identify species, subspecies, and populations

of all migratory nongame birds that, without

additional conservation actions, are likely to become

candidates for listing under the Endangered Species

Act of 1973” (U. S. Fish and Wildlife Service 2002).

The document provides species lists at three

geographic scales: national, FWS regions, and BCRs.

Species considered for inclusion include nongame

birds, game birds without hunting seasons, and

numerous categories (candidate, proposed endangered

or threatened, and recently delisted) used in the ESA.

Parameters considered in determining if species

within these categories are of concern include

population size, extent of range, threats to habitat,

and other factors. The BCC should be consulted for

details regarding the assessment process.

There are 28 species known to occur on Kirwin NWR

that are considered to be of national conservation

concern in the BCC (U. S. Fish and Wildlife Service

2002). Among these are 8 shorebirds, 5 hawks and

falcons, 2 owls, and 2 sparrows. Twenty-one of these

28 species also are considered to be of conservation

concern at either FWS Region 6 or BCR 19 scale

(Appendix D). Special note should be made that some

species of conservation concern listed for BCR 19 are

not on the regional list and, likewise, several species of

regional conservation concern were not included in the

BCC list for BCR 19 (Appendix D).

North American Waterfowl Management Plan

The national goals set forth in the 1998 update of

North American Waterfowl Management Plan

(NAWMP) includes (1) maintaining the diversity of

duck species throughout North America and achieving

a mid-continent breeding population of 39 million

ducks during years with average environmental

conditions, and (2) reaching or exceeding mid-continent

populations for ten individual species,

including Gadwall, American Wigeon, Mallard, Blue-winged

and Cinnamon Teal, Northern Shoveler,

Northern Pintail, Green-winged Teal, Canvasback,

Redhead, and Greater and Lesser scaup. These

target populations are presented in Appendix D. The

plan also establishes population objectives for 30

populations of six goose species, three populations of

Trumpeter Swans, and two populations of Tundra

Swans. Of these, relevant objectives applicable to

Kirwin NWR include reducing all five populations of

Canada Geese that migrate through the central

flyway and also reducing mid-continent populations of

the Snow and Greater White-fronted geese to 1,000,000

and 600,000, respectively. The plan also sets forth

objectives to increase the interior population of

Trumpeter Swans to 2,500 and reduce the eastern

population of Tundra Swans to 80,000 (Appendix D).

Partner’s In Flight North American Landbird

Conservation Plan

The North American Landbird Conservation Plan

(NALCP) is a synthesis of priorities to guide national

and international conservation actions targeting 448

native landbirds from 45 families that breed in the

United States and Canada (Rich et al. 2004). Each

species is assigned scores ranging from 1 (low

vulnerability) to 5 (high vulnerability) for 6 factors

(population size, breeding distribution, nonbreeding

distribution, threats to breeding, threats to

nonbreeding, and population trend). These scores

subsequently are used to calculate a combined score

that represents relative conservation importance

(range = 4 [low concern] to 20 [high concern]).

Species with a combined score > 14, or combined

score = 13 and population trend score = 5, are

assigned to the Watch List that includes species of

highest conservation concern. In addition, a

Stewardship List was developed based on avifaunal

biomes in North America. These biomes were

delineated using cluster analyses to identify groups of

BCRs that share similar avifaunas. For each biome,

Stewardship Species are those species that have a

proportionately high percentage of their world

population within a single region during either the

breeding or wintering season. Kirwin NWR occurs in

the Prairie Avifaunal Biome, which is composed of

BCRs 11, 17-19, and 21-23 (Rich et al. 2004). The

NALCP should be consulted for details on the scoring

and assignment process, national and regional

population objectives, and other pertinent information

(Rich et al. 2004, URL http://

www.partnersinflight.org, URL http://www.rmbo.org/

pif/pifdb.html).

The Watch List and Stewardship List of continentally

important species in the United States and Canada

currently include 100 and 158 species (66 of these

species also occur on the Watch List), respectively

(Rich et al. 2004). Of these species, two (Greater

Prairie-Chicken, Dickcissel) of the Watch List species

and five (Mississippi Kite, American Tree Sparrow,

Lark Bunting, Grasshopper Sparrow, and Chestnut-

10

the Prairie Avifaunal Biome, respectively. Of these,

two species (Greater Prairie-Chicken and Dickcissel)

on the Watch List and six species (Mississippi Kite,

American Tree Sparrow, Lark Bunting, Grasshopper

Sparrow, Chestnut-collared Longspur, and Lapland

Longspur) on the Stewardship List are known to

occur on Kirwin NWR.

Shorebird Conservation Region

Kirwin NWR is located in the Central Plains/Playa

Lakes Region (CP/PLR) (United States Shorebird

Conservation Plan, URL http://

shorebirdplan.fws.gov/RegionalShorebird/

RegionsMap.asp). This region is larger than the area

encompassed by BCR 19 and includes portions of

Texas, eastern New Mexico and Colorado, western

Oklahoma, Kansas, Nebraska, and southeast

Wyoming. Thirty-eight shorebird species use habitat

in the CP/PLR during migration and 13 of these

species also breed in the region (Fellows et al. 2001).

Of these, 16 species have been identified as species of

primary concern. Based on the Kirwin NWR bird list,

13 of these priority species have been documented on

the refuge, including Snowy Plover, Long-billed

Curlew, Upland Sandpiper, White-rumped Sandpiper,

Baird’s Sandpiper, and Piping Plover (Appendix D).

The national and regional plans should be consulted

for national population objectives, justifications for

species designations, and other pertinent information

(U. S. Shorebird Conservation Plan, URL http://

shorebirdplan.fws.gov/RegionalShorebird/downloads/

CPPLR.doc).

Waterbird Conservation Region

Kirwin NWR is within the Central Prairies Region of

The North American Waterbird Conservation Plan

(NAWCP). In North America, separate initiatives

exist for waterbirds, shorebirds, and waterfowl.

Thus, the NAWCP focuses on seabirds, coastal

waterbirds, wading birds, and marsh birds (Waterbird

Conservation for the Americas, URL http://

www.waterbirdconservation.org/waterbirds/). There

are 23 species listed in the Central Prairies that have

been documented on Kirwin NWR. Of these, the

conservation status of six species are designated as

“not currently at risk”, five are considered of “low

conservation concern”, nine are of “moderate

conservation concern”, and three are of “high

conservation concern” (Kushlan et al. 2002; Appendix

D). The entire plan should be consulted to fully

understand how scores were determined.

Playa Lakes Joint Venture

At a smaller geographic scale, Kirwin NWR is part of

the Playa Lakes Joint Venture (PLJV; URL http://

www.pljv.org/about01.html), which encompasses most

of BCR 18 (Shortgrass Prairie) and BCR 19 (Central

Mixed-Grass Prairie). Joint ventures originally were

conceived by FWS in 1986 to implement the NAWMP.

URL http://northamerican.fws.gov/NAWMP/

nawmphp.htm). However, in addition to waterfowl,

many joint ventures (including the PLJV) are now

incorporating an “all bird” approach. National and

international bird plans provide the foundation for the

PLJV master plan, which sets direction for

conservation activities within the region. Established

in 1989, the mission of the PLJV is to create

sustainable landscapes for the benefit of birds, other

wildlife, and humans. More specifically, the PLJV

directs effort to restore, preserve and protect myriad

habitats, including playas, saline lakes, riparian areas,

and grasslands for resident, wintering, migrating, and

breeding birds. More than 400 species of birds use the

PLJV, including continentally important populations

of waterfowl, shorebirds, other waterbirds, and

grassland birds. The objectives established for the

PLJV in the 1998 update to the NAWMP are to

restore and enhance 4,047 ha (10,000 ac) and 10,117 ha

(25,000 ac) of habitat, respectively, and protect 20,639

ha (51,000 ac) of existing habitat.

collared Longspur) of the Stewardship list species are

known to occur on Kirwin NWR (Appendix D). In

contrast, there are 21 and seven species of continental

importance on the Watch List and Stewardship List in

11

This section has been divided based on general

community designations. This was done to improve

clarity, but it should be recognized that such

distinctions are arbitrary. Communities do not occur

as distinct entities but grade together as evidenced by

the movement of nutrients, energy, and wildlife within

and among communities. Therefore, information for

one community may be relevant for other

communities. Attempts were made to identify these

relationships, but some may have been overlooked.

For each community, a brief history is provided on

historic and current conditions and potential concerns

and opportunities are identified. The inclusion of

historic information is not intended as an effort to

direct refuge management toward restoring

presettlement conditions. Rather, historic

descriptions have been incorporated because they can

provide valuable insight regarding the original

location, extent, and vegetation composition of

communities and changes in abiotic factors (e.g.,

hydrology) that have occurred through time.

Together with information on current conditions,

historic descriptions can be used to (1) identify

important changes in processes that may require

further evaluation and (2) better understand the

potential of refuge lands.

Reservoir (Deepwater)

Historically, there was no deepwater habitat on the

area that now constitutes the refuge. Flows from the

North Fork Solomon River and Bow Creek flowed

unimpeded through refuge lands and occasionally

inundated the floodplain during wet periods.

Construction of Kirwin Reservoir has changed these

conditions. Obviously, damming the flows of the

Solomon River and Bow Creek and impounding water

in the historic floodplain of the rivers created

deepwater habitat. The surface acreage of the

reservoir varies dramatically from about 2,024 ha

(5,000 ac) at conservation pool (527.7 m [1731.25 ft]

elevation) to 356 ha (879 ac) during drought periods

(refuge staff). During the site visit the reservoir

elevation was 522.9 m (1715.6 ft), but water is expected

to drop an additional 3.4 m (11 ft) in the summer of

2003 (refuge staff). These fluctuations likely are due

to a combination of frequent drought periods coupled

with upstream pumping from the aquifer (U.S. Bureau

of Reclamation 2002). Since the mid-1960’s, inflows to

Kirwin Reservoir have declined significantly as

evidenced by a long-term reduction in average annual

inflow of 3,577 ha-m (29,000 ac-ft) between 1920-1964

(6,562 ha-m [53,200 ac-ft]) and 1965-1999 (4,070 ha-m

[33,000 ac-ft]; U.S. Bureau of Reclamation 2002). In

addition, >150 alluvial wells occur above the refuge

(U.S. Bureau of Reclamation 2002). The majority of

these wells are for agricultural uses, but municipal

wells also exist.

There also have been less obvious influences. Prior to

settlement, some amount of sediment was

transported from the uplands to the channel during

storm events. The amount of sediment varied, but

intact upland and floodplain vegetation probably

reduced the amount of sediment that entered the

channel. Following settlement, increased agricultural

activity likely altered the amount and pattern of

sediment transport and deposition in the valley. More

than 50% of the basin is currently cropland (U.S.

Bureau of Reclamation 1984) and cultivation and

intensive grazing in some areas (e.g., areas with large

topographic relief) have likely increased the amount of

erosion, and therefore sediment, entering the

floodplain. Prior to constructing reservoirs in the

Solomon Basin, the distribution of this sediment

varied depending on antecedent conditions and

magnitude of the current event. Following

construction, however, the dams functioned as a

barrier to sediment transport downstream. Thus,

although sediment deposition can occur at various

locations upstream of Kirwin Dam, the dam itself now

represents a terminal location that likely traps the

majority of sediment that enters the reservoir.

The potential impacts of increased sedimentation at

one location are numerous. In terms of quantity,

sediment is the major pollutant of wetlands, lakes,

estuaries, and reservoirs in the United States (Baker

1992). Sediment quality also is an environmental

concern because sediment may act as both a sink and

source for water-quality constituents (U. S. Geological

Survey, URL http://ks.water.usgs.gov/Kansas/

studies/ressed/). Once in the food chain, sediment-derived

constituents may bioaccumulate and pose an

even greater concern to fish, wildlife, and humans. In

addition, sediment loads may never consolidate with

bottom materials. The surface waters in the basin of

the North Fork Solomon River are reported as turbid

with moderate to high concentrations of dissolved

solids (U.S. Bureau of Reclamation 2002). Thus,

increased sedimentation may increase turbidity even

more due to wind and wave action that periodically

suspends sediment throughout the water column.

This could lead to other impacts, including reduced

dissolved oxygen concentrations, altered nutrient

availability, and reduced sunlight penetration. If

sufficient, these changes can eliminate or reduce

growth of submerged aquatic vegetation (SAV) (Robel

1961, Kullberg 1974, Dieter 1990).

The extent that sediment impacts have occurred, or

potentially could occur, in Kirwin Reservoir has

received attention recently. In 1998, the BOR initiated

a sampling program to assess the presence or absence

of organic and inorganic compounds in reservoir

waters. Part of this study involved collecting two

groups of four sediment cores near the dam

(Christensen 1999). Sediment thickness estimated

from these cores ranged from 2.9-3.4 m (9.5-11.3 ft) in

the first group of 4 cores to 2.1-2.3 m (6.9-7.4 ft) in the

second group. Unfortunately, it was not possible to

accurately determine sedimentation rates due to core

shortening (i.e., essentially a partial collapse or

compression of the core that prevented accurate

dating). However, a visual examination of the cores

revealed thick layers of sediment in the deepest

Community Types

12

intervals, which indicates sediment deposition was

greater in the early years of the reservoir. This is

consistent with other reservoir studies (Ritchie et al.

1986; Callendar and Robbins 1993) that have

demonstrated decreased sedimentation rates with

reservoir age. Historical stream flow data indicate

that much of the early sediment deposition in the

reservoir may have been caused by floods during the

1957 and 1960 water years (Christensen 1999).

Another objective of the BOR sampling program was

to determine potential environmental effects due to

elevated levels of total organic carbon (TOC), trace

metals, and major nutrients. The Environmental

Protection Agency has established two threshold

concentrations for many of these elements. The

threshold effect level (TEL) is assumed to represent

the concentration below which toxic effects rarely

occur, whereas the probable effect level (PEL)

indicates the concentration that usually or frequently

results in toxicity. Adverse effects occasionally occur

at concentrations between the TEL and PEL. Both

the TEL and PEL are guidelines used to screen for

possible hazardous chemical levels, but are not

regulatory criteria.

Total organic carbon was measured because various

organic solutes can form complexes that affect metal

solubility (Hem 1992). The median TOC concentration

in the reservoir was 11,600 mg/kg and the trend was

not increasing. There are no published TEL and PEL

limits for TOC; thus, the impact of existing levels is

not readily apparent. However, further investigation

should be conducted to obtain information from other

reservoir studies to ascertain potential impacts of

current TOC levels.

Selenium (Se) is a naturally occurring trace element

common in the marine shales underlying the Solomon

River Basin (see section on Geology). This metal is of

concern because irrigation in other areas underlain by

marine shales has resulted in elevated Se

concentrations that have caused birth defects,

reproductive failure, and death in fish and wildlife (U.S.

Bureau of Reclamation 2002). No TEL or PEL has

been established for Se, but concentrations >4.0 mg/

kg in sediment can result in bioaccumulation in fish

and wildlife (Lemly and Smith 1987). Concentrations

of Se in Kirwin Reservoir bottom-sediment ranged

from <0.3 to 2.2 mg/kg, indicating low potential for

bioaccumulation (Christensen 1999). However, Se did

exhibit a significant increasing trend (P = 0.006) in one

of the two cores, suggesting that concentrations may

be of concern in the future.

Reports by Christensen (1999) and Christensen and

Juracek (2001) also indicate median arsenic (As)

concentrations (range = 4.6-10.0 mg/kg) exceeded the

TEL (7.24 mg/kg) but not the PEL (41.6 mg/kg)

established for this element. However, a significant

increasing As trend was not evident. Arsenic could

originate from many potential sources, including

underlying geologic features (Christensen and Juracek

2001) or pesticides and industrial activities (Pais and

Jones 1997). However, industrial sources are not likely

given the predominance of agriculture in the basin

(Christensen and Juracek 2001). The median

concentration of copper also exceeded the TEL (18.7

mg/kg) as did cadmium in four samples. In contrast,

chromium, lead, nickel, silver, zinc, and mercury

either were not detected or did not exceed TEL limits.

These results clearly indicate that subsequent

monitoring of heavy metals and other water quality

parameters are warranted. Although no significant

effects have been documented, potential for future

issues may arise given evidence of increasing trends

for some metals.

Phosphorous (P) and nitrogen (N) are nutrients

required for plant growth, but excessive amounts can

enter reservoirs from fertilizer runoff or other non-point

pollution sources and create problems. The

median P and N concentration in core samples from

Kirwin was 616 mg/kg and 1,700 mg/kg, respectively.

However, only P exhibited a significant increasing

trend. Although no TEL and PEL limits have been

established for P and N, additional information could

be obtained to better understand potential impacts.

For example, excessive P has been shown to cause

algal blooms that can reduce dissolved oxygen

concentrations and cause fish mortality, or reduce

light penetration to levels that prevent growth of

some aquatic plant species. Information regarding

nutrient levels that result in algal blooms, or cause

other changes in aquatic biota (e.g., plants, animal)

could be used to develop desired thresholds.

Although reservoir water levels are managed by the

BOR, it may be possible for FWS to recommend or

help establish water quality criteria that would

ensure the needs of fish and wildlife are met.

Published information on the type and amount of SAV

in the reservoir was not located during the initial

literature search and the refuge staff was unable to

provide any qualitative observations. This is

unfortunate because plant composition and biomass

occurring in the deepwater community greatly

influences potential wildlife values. Plants capable of

growing in deep water provide substrate for

invertebrates (Krull 1970, Voigts 1976) that, in

combination with plant parts, provide foods for many

different vertebrates (e.g., fish, waterbirds). In

contrast, if SAV is not present, the deepwater

community may only provide roosting and loafing

habitat for birds. To better evaluate this community,

additional searches for information on plant resources

should be pursued, including contact with personnel

from Kansas Department of Wildlife and Parks and

BOR to obtain any reports or documents that may

exist.

Waterfowl counts conducted between 1983 and 2001

document ducks, geese, and swans occurring on the

refuge in varying numbers (Appendix C). On an

annual basis, the primary periods of use occur during

spring and fall migration; however, some species,

primarily Canada Geese and Mallards, remain on the

refuge during some winters (Appendix C, U.S. Bureau

of Reclamation 2002). Both diving ducks and geese use

the deepwater portion of the reservoir. Plant

13

composition and biomass information is lacking; thus,

it is not possible to determine if foraging habitat is

available. Waterfowl surveys only provide weekly

estimates on the entire refuge. Information on

numbers and activities (e.g., foraging, resting) of each

species in individual habitat types (e.g., deepwater

versus shoreline) are lacking. Therefore, this data

cannot be used to speculate on the type and

availability of resources and it is impossible to arrive

at any definitive conclusions. However, at a minimum

it is likely that the deepwater community provides

roosting and loafing habitat for waterfowl (ducks,

geese, swans), as well as sanctuary from shooting

during hunting season (U.S. Bureau of Reclamation

2002). This zone also could provide additional benefits

in the form of foraging habitat if SAV beds or

invertebrates are present. The types of foraging

habitat available would largely depend on the types

and locations of food items. For example, the

presence of pondweed drupelets or invertebrates

within about 46 cm (18 in) of the water surface is

available for dabbling ducks, whereas swans can

access pondweed foliage at greater depths

(Fredrickson and Reid 1986, Fredrickson and Laubhan

1994).

Based on conversations with refuge staff, another

interesting issue that would merit further

investigation is the value of the deepwater zone as

refuge during the hunting season. Currently, this

area is closed to hunting and the staff thinks this

increases the number and duration of time that geese

are in the area. A competing idea is that goose

numbers near the refuge are positively correlated

with the amount of row crops. Sufficient data may be

available to conduct a correlation analysis between the

size and availability of the closed zone, cropland acres,

and goose numbers. Although correlation analysis is

only a “measure of association” and does not prove

cause-and-effect, this analysis may provide some

insight or help identify variables to monitor in the

future.

Management Potential -The ability of FWS to

manage the deepwater habitat is minimal. Reservoir

elevations are determined by other federal entities

that must consider several factors (e.g., irrigation,

flood control) other than wildlife. Hydrology,

including the direction, magnitude, and time of water

level fluctuations, is the primary factor influencing

resource production and availability (Mitsch and

Gosselink 1993, Fredrickson and Laubhan 1994). The

inability of FWS to influence these hydrologic

parameters prevents the ability to reliably stimulate

or maintain desired plant communities and associated

food resources, or influence resource availability (e.g.,

water depth between food resources and water

surface). Even if FWS could establish guidelines

specifically for wildlife, the dramatic fluctuations in

surface area caused by uncontrollable factors (e.g.,

precipitation, upstream groundwater pumping) would

make it difficult to reliably and consistently achieve

desired outcomes.

Although direct management is minimal, the

deepwater community still provides resources that

contribute to the overall value of the refuge.

Therefore, FWS should consider options that can be

used to indirectly influence the values that are

provided as a result of annual reservoir operation.

Potential options to consider include working

cooperatively with BOR to establish water quality

criteria and continuing to maintain existing

agreements that designate a portion of the deepwater

community as a closed area during hunting.

Shoreline

The shoreline and deepwater communities are both

part of the reservoir and, therefore, are functionally

connected both spatially and temporally. They also

share some of the same ecological attributes,

including source and quality of water. However, these

communities have been separated because they are

very different with respect to some hydrologic

parameters (e.g., water depth), plant communities,

and wildlife resources.

Definitions vary, but the shoreline community is

defined in this report as the portion of the reservoir

(excluding the riparian zone) with water depths that

range from saturated soils to <61 cm (24 in). The

general shape of the shoreline is linear, but the width,

and spatial position of this area change both annually

and seasonally depending on reservoir water levels

and bathymetry (i.e., topography of reservoir bottom

sediments). For example, the bathymetry of the

shoreline has been differentially influenced by erosion

following reservoir impoundment. The presence of

deep and shallow cutbacks caused by wave action can

significantly influence habitat suitability for some

species that tolerate a narrow range of water depths

(e.g., shorebirds). The paucity of palustrine wetlands

on the refuge means that the shoreline is the only

community that potentially can provide substantial

foraging habitat for dabbling ducks (unless SAV

occurs in the deepwater community), shorebirds, and

wading birds. However, a sufficient area of suitable

(e.g., proper substrate, water depth) shoreline must

be available if the community is to provide limiting

resources (aquatic invertebrates, seeds, tubers) in

quantities that benefit different waterbird guilds.

Although area estimates of the shoreline community

were not available, the bathymetry and water level

data needed to develop estimates is likely available

from the BOR. The use of this data to develop curves

for estimating shoreline area is highly recommended

because it would likely prove useful in future

discussions regarding management potential and

values of this community. For example, the range of

water depths along the shore at different elevations

could be compared to water depths used by foraging

waterbirds to determine area of suitable habitat

available. In addition, the refuge staff reported that

an agreement exists between certain entities to

maintain conservation pool at 527.7 m (1731.30 ft),

rather than the legal elevation of 527.1 m (1729.25 ft).

The potential impacts (positive or negative) of this

increase on available foraging habitat for various avian

guilds could be estimated using the above mentioned

curves, but would be difficult otherwise.

14

Because area is important, a coarse estimate of 91-271

ha (224-670 ac) for the shoreline community at

conservation pool was derived to provide some

perspective. The accuracy of this estimate should be

viewed with extreme skepticism for several reasons.

First, the estimate uses an assumed average width of

15-46 m (50-150 ft). Exact widths are unknown and

likely vary extensively around the perimeter of the

reservoir. Second, the estimate is based on a shoreline

length of 60 km (37 mi) at conservation pool which

likely includes portions of the shoreline that would be

more appropriately considered the riparian

community. Also, and perhaps most important,

fluctuations in the reservoir surface are known to be

dramatic. Consequently, a single estimate at

conservation pool does not capture the full range of

shoreline area that occurs within and among years.

However, this estimate does suggest that the area of

shoreline is sufficient to warrant further

consideration.

The shoreline potentially can provide unique

resources for a diverse array of avifauna. According

to refuge files, Double-crested Cormorants and Great

Blue Herons have nested on the refuge since 1952 and

1963, respectively. Reproductive effort varies

annually, but between 1960 and 1995 the number of

Great Blue Heron nests has ranged from 1-20 with

production of 2-90 young. During the same period,

Double-crested Cormorant nests and young have

ranged from 3-37 and 40-60, respectively. The current

location of rookeries occurs within, or adjacent to, the

shoreline community near the main reservoir body in

the eastern portion of the refuge. Trees currently

used for nesting appear to be adjacent to stream

channels that were inundated when water was

impounded by the reservoir. Many of these trees were

killed as a result of high water in the 1990’s, but many

remain standing and still provide suitable nesting

habitat.

In addition, Least Terns occasionally nest within the

shoreline community and protection of ground nests is

required. Exposed sandbars constitute the preferred

nesting substrate of Least Terns. However,

substrates similar to sandbars are exposed along the

shoreline when reservoir elevation recedes and some

Least Terns occasionally nest in these areas.

The primary value of the shoreline community, based

on the geographic location of the refuge, likely is

foraging habitat for a variety of waterbirds. This area

constitutes a zone of high biological productivity. The

growth of plants during drawdown results in the

production of food resources (e.g., seeds, tubers) and

the release of nutrients when vegetation decomposes

upon reflooding can be assimilated by small aquatic

organisms (e.g., microinvertebrates) (Fredrickson and

Laubhan 1994). These organisms constitute the

forage base for macroinvertebrates, fish, and

amphibians, which are the primary foods of many

waterbirds. In addition, the hydrologic fluctuations

that occur within this area create numerous

microhabitats that can be used by numerous species.

Herons and other wading birds forage primarily on

aquatic animals, including fish, amphibians, and

macroinvertebrates. Although some species are

capable of capturing prey in water >60 cm (24 in), the

majority of foraging typically occurs within the

shoreline community (e.g., shallow water or along the

water-mud interface) (Fredrickson and Reid 1986).

Except for extreme fluctuations, changes in reservoir

levels likely do not alter production or availability of

fish, the primary food item of these species (but see

Gawlik 2002 for impacts that can occur given the right

circumstances). During drought years, it is

conceivable that fish kills may occur. This may or

may not impact foraging efficiency and nest success of

waders depending on the biomass of foods that remain

during these periods. Although pertinent data was

not located during this initial investigation, Kansas

Department of Wildlife and Parks may have relevant

information.

Ducks (diving and dabbling) and shorebirds also

forage within the shoreline community (Fredrickson

and Reid 1986, Skagen and Knopf 1994). In fact, the

paucity of palustrine wetlands suggests that these

species rely almost exclusively on the shoreline for

foraging when using the refuge. However, refuge

survey data does not provide information to confirm

this assumption. Further, the production and

availability of resources for these species is difficult to

predict because of the dynamic water fluctuations that

occur in this edge community. For example, during

March 2003 vegetation along the northern shoreline

included reed canary grass, saltcedar, and Canada

thistle. There also were extensive areas of bare

ground. Although the production of foods (browse,

seeds, tuber, etc.) for ducks and geese appears

minimal, the growing season had not yet started and

additional plants may germinate. If additional

germination does not occur, the extensive areas of

unvegetated shoreline likely will provide foraging

habitat for spring migrant shorebirds. However,

optimum foraging depths vary among shorebirds

depending on size (i.e., tarsus length); thus, not all

shorebirds would benefit equally. In contrast, a very

different plant community was evident during a field

visit to the refuge in 1999. This visit was conducted

during the growing season and water levels were

much higher. My field notes indicate minimal bare

ground and the presence of smartweed, millet,

bulrushes, cattail, beggarticks, ricecut grass,

spikerush, cocklebur, sedges, and panic grass among

other species. Depending on fall water conditions, the

shoreline would have provided excellent foraging

habitat for dabbling ducks and geese due to the large

biomass of seeds and browse produced. In contrast,

shorebird foraging habitat would likely have been

minimal due to excessive vegetation cover.

The above observations fromt two different years

provide evidence that the seed bank within the

shoreline community is diverse and includes both

desirable (e.g., browse, seed-bearing) and undesirable

(e.g., invasive, exotic) plant species. The species that

germinate from the seed bank, and the ultimate

densities of species that survive, are determined by a

multitude of factors. Most species that germinate in

15

the shoreline area require substrates that are moist

to wet, but not flooded (van der Valk and Davis 1978).

Thus, the most important factor controlling

germination likely is the annual changes in reservoir

water levels, including the magnitude, timing, and

rate of water level fluctuations. These hydrologic

parameters greatly influence recruitment from the

seed bank by affecting time of soil exposure, soil

temperature and oxygen levels, and the rate of soil

moisture loss (Leck 1989, Fredrickson 1991). Water

quality also may be important because constituents

(e.g., salts, iron, copper) in the water are bound by soil

particles at the soil-water interface and can affect

plant germination and growth. This deserves further

investigation because of water quality issues in the

reservoir. However, the diversity of plants that have

already been documented along the shoreline

suggests that water quality currently is not severely

impacting the germination or survival capability of

many species.

Management Potential – Similar to the deepwater

portion of the reservoir, the ability of FWS to manage

the shoreline community is constrained by the lack of

hydrologic control. Consequently, the value of the

shoreline community to waterbirds likely will vary

among species and years. Trees adjacent to the

reservoir and the presence of fish near the shoreline

are probably consistently available. Thus, suitable

habitat for breeding Great Blue Herons and Double-crested

Cormorants, as well as migrating and

wintering Bald Eagles, is usually present on the

refuge in most years. In contrast, foraging habitat for

ducks and shorebirds will be more variable for two

primary reasons. First, it is not possible to

manipulate water levels to match the germination

requirements of plants that produce a large biomass

of foods (e.g., seeds, tubers, browse) and provide

substrate for invertebrates. Second, water levels

cannot be intentionally manipulated to coincide with

duck and shorebird migration periods. Therefore, in

the absence of hydrologic control, some exposed and

vegetated shoreline habitat will be available to

shorebirds and ducks every year, but water level

changes that expose abundant foods during migration

will occur only sporadically. Finally, the availability of

habitat for Least Terns varies, but likely is more

predictable than ducks and shorebirds. This

statement is based on the reported long-term

drought/wet cycle of 30 years with about 23 years of

drought and seven years of wet conditions (refuge

staff). According to the refuge staff, reservoir pool

elevations tend to consistently decrease during the

drought phase. When this occurs, the availability of

substrates suitable for Least Tern nesting tends to

become more reliable, and the probability of nest

destruction due to flooding less likely, during a period

of several years. During the start of the wet period,

water levels in the reservoir start to increase,

available nesting habitat decreases, and, if nesting is

attempted, the likelihood of nests being destroyed by

flooding increases.

Another reality is the potential for an unfavorable

plant community to develop in the shoreline

community. The land-water interface in this zone is a

prime area for the establishment and proliferation of

many invasive species due to the frequent presence of

exposed soil, variable soil moisture, and high nutrient

availability. For example, along the north shoreline

numerous saltcedar seedlings and stems of Canada

thistle and reed canary grass were evident. Although

currently present only in small numbers, the

potential exists for expansion of these invasive species

(or others) along the shoreline, which could result in

the loss of current shoreline values (Sudbrock 1993,

Bailey et al. 2001). Evidence of this potential exists in

the floodplain of the lower riparian zone where reed

canary grass and Canada thistle currently dominate

the herbaceous vegetation (see below).

FWS cannot alter the hydrology of the reservoir to

minimize the potential for invasions of non-natives to

occur. Similarly, FWS cannot intentionally raise pool

elevation to eliminate invasions that do occur.

Nevertheless, the refuge staff is responsible for

addressing invasive species that do occur. Potential

control options (herbicides, fire, mechanical

equipment) exist, but implementation may not be

possible in some years (e.g., too wet). Also, many

techniques often are costly and require repeated

application to be successful. Thus, it is recommended

that decisions be made regarding the ability and/or

desire to conduct such operations under different

conditions. Information available in the literature and

refuge files should be adequate to develop working

hypotheses of potential tradeoffs (e.g., costs, benefits,

probability of success) related to active management

in this community. This information should be

consolidated prior to CCP development.

In summary, the shoreline community has the

potential to provide many values to waterbirds that

other communities on the refuge do not provide.

There also is potential for extensive, rapid colonization

of invasive species. These detrimental impacts are

common on many reservoirs, and approaches to

minimize impacts are often difficult to develop due to

constraints imposed by the reservoir operation plan.

In the case of Kirwin Reservoir, the only

recommendation is to consult with the BOR to

determine if annual operations can be altered slightly

to take advantage of existing conditions that occur in

some years. For example, it is likely that the release

of a relatively small volume of water in spring would

expose a large amount of shoreline habitat around the

reservoir for spring migrant shorebirds. Such

releases would not have to occur annually; rather, an

agreement could be developed that would result in

release only when pool elevations are above a certain

level. In many cases, such alterations may have only

negligible impacts on other reservoir uses but result

in significant wildlife benefits.

Riparian

The riparian community, which includes the floodplain

and channel of the Solomon River and Bow Creek, was

dynamic historically. Although both streams were

considered perennial (Leonard 1952), flows were

highly variable depending on precipitation cycles.

16

Stream hydrology was characterized by flood flows in

the spring and low flows or ponding during the

summer and fall (U.S. Bureau of Reclamation 2002).

These extremes in hydrology influenced the types of

flora that developed and the fauna that inhabited the

riparian system. During major floods the channel was

reworked, vegetation was uprooted, and sediment was

transported downstream and deposited at various

locations in the channel and floodplain. These actions

resulted in the creation of various channel habitats

(e.g., pools, riffles), marsh areas adjacent to the rivers,

and sites for regeneration and growth of various plant

types in the floodplain.

The historic floristic composition of the floodplain

included grasses, forbs, and woody vegetation.

Kuchler (1974) described this community as

“floodplain forest and savanna”, with scattered trees

and shrubs and a dominant ground cover of bluestem

prairie. However, he also states that “the prairie was

suppressed in areas of dense woody growth”,

suggesting that certain areas of the floodplain were

extensively forested. The wooded component

apparently was continuous but narrow based on

accounts of early settlers and one aerial photograph

(Plate 5, page 18 in Leonard 1952) of the Solomon

River near Glade, Kansas. Dominant woody species

included cottonwood, American elm, hackberry, and

peachleaved willow, whereas the dominant herbaceous

vegetation consisted of big bluestem, little bluestem,

switchgrass, and Indian grass. In contrast, marshes

were dominated by prairie cordgrass and lesser

numbers of myriad species, including bulrushes,

cattail, and rice cutgrass (Kuchler 1974).

The historic wildlife community inhabiting the

riparian community was diverse and unique. Forests

were rare in the Great Plains and the woody

vegetation provided cover, forage, and nesting

substrates for neotropical migrants that were not

available in other communities. The tall grasses

provided important resources for both migratory and

resident wildlife, and marshes provided resources for

a host of waterfowl. The stream fishery was not rich

and included only species (e.g., plains killifish, red

shiner, creek chub) that could tolerate extremes in

hydroperiod, temperature, current velocity, and

dissolved oxygen concentrations (U.S. Bureau of

Reclamation 2002).

However, as with other communities on the refuge,

human settlement and the accompanying changes

have greatly altered processes and influenced

vegetation in the riparian community. Among the

most important changes include reservoir

construction, increased groundwater pumping,

diversion dam construction, and irrigation canal

development (Christensen and Juracek 2001). The

potential range of impacts caused by these changes

varies from subtle to obvious depending on the year

and antecedent environmental conditions. The upper

portion of the riparian community differs greatly from

the lower portion due to the impacts of the above-mentioned

changes.

woody riparian vegetation on the Solomon River (3,112

ha [7,689 ac]) ranked second only to the Lower

Arkansas (5,083 ha [12,560 ac]) (Eddy 1994). The

composition of trees in the mid-1990’s was dominated

by eastern cottonwood (58%) and willow (25%) with

lesser amounts of American elm (4%) and green ash

(3%), hackberry, boxelder, and mulberry (Sevigny

1998, Eddy 1994). The shrub and vine component (5%)

also was evident, but some non-native trees have

invaded the system, including saltcedar (Eddy 1994),

Siberian elm, and honey locust (Sevigny 1998, refuge

staff). Perhaps the greatest change from historic

structure and composition has occurred in the ground

vegetation. The once dominant tall, warm season

grasses described by Kuchler (1974) have been

replaced by shorter cool season grasses (e.g., smooth

brome), which has altered structural and floristic

diversity (refuge staff, personal observation).

The avian community also remains diverse, which is

not surprising. The ability of riparian systems to

support a diverse assemblage of vertebrates is well

documented (Pashley et al. 2000). However, the

composition and relative abundance of species have

likely changed due to landscape level changes in land

use (e.g., agriculture). In 1997, a study of the riparian

bird community on the refuge during spring migration

resulted in the identification 87 species from 19

families (Sevigny 1998). A detailed inspection of this

list identified some intriguing (although not

substantiated) aspects that may be related to changes

in ground flora. The nine most abundant species

(>100 recorded) were the House Wren, Blue Jay,

Black-capped Chickadee, Mourning Dove, Northern

Cardinal, Common Yellowthroat, Red-winged

Blackbird, and Brown-headed Cowbird. Based on

Breeding Bird Survey (BBS) data for Region 6 of

FWS, the Black-capped Chickadee (n = 257 routes,

trend = 0.9, P = 0.19, 95% confidence interval [CI] = -

0.4 – 2.1), Mourning Dove (n = 568 routes, trend = 0.0,

P = 0.82, 95% CI = -0.4 – 0.5), Northern Cardinal (n =

73 routes, trend = 1.1, P = 0.06, 95% CI = 0.0 – 2.2),

Common Yellowthroat (n = 322 routes, trend = 0.1, P

= 0.89, 95% CI = -0.8 – 0.9), Red-winged Blackbird (n

= 525 routes, trend = 0.1, P = 0.84, 95% CI = -0.5 –

0.6), and Brown-headed Cowbird (n = 533 routes,

trend = 0.1, P = 0.71, 95% CI = -0.4 – 0.6) exhibited

stable populations trends, whereas the House Wren (n

= 410 routes, trend = 2.4, P = 0.00, 95% CI = 1.8 –

3.1) and Blue Jay (n = 179 routes, trend = 0.8, P =

0.02, 95% CI = 0.1 – 1.6), exhibited increasing

population trends between 1966 and 2003 (Sauer et al.

2004, URL http://www.mbr-pwrc.usgs.gov/bbs/

bbs.html). Further, most of these species are capable

of adapting to changes occurring in the riparian

communities throughout the western United States

(Saab 1999). In contrast, however, the list also

included 19 species whose status is of some concern

according to current regional and national plans

(Appendix D). The presence of these species in low

In the mid-1990’s, the floodplains of both streams

supported trees on the refuge, but the width varied

from a few scattered trees to areas as wide as 180 m

(590 ft) (Sevigny 1998). Of the nine river systems in

the western two-thirds of Kansas, the amount of

17

abundance suggests the riparian plant community has

not been completely altered, but subtle, significant

changes have occurred that has reduced habitat

suitability for some species. Additional investigation

to identify these changes and their causes would be

valuable for determining appropriate future

management actions. However, given the impacts of

high water during the mid-1990’s (see below) the

preceding statement only applies to the upper portion

of the riparian community.

Comparing shifts in avian communities between

historic and current periods often is used as a

technique to describe community changes. This

approach has much value, but it also has several

weaknesses. First, this technique cannot identify all

changes that potentially have occurred. Second,

changes in processes often can only be detected by

certain avian parameters (e.g., abundance, nest

density); thus, multiple parameters often must be

collected and this data is rarely available in historic

accounts. Finally, these comparisons only illustrate

past trends and do not identify cause-effect

relationships. Consequently, they provide little

information on future expectations, particularly in

environments that are subject to rapid and extreme

changes due to human events (see below).

To remedy this shortcoming, information regarding

change in processes that influence community

structure and function also must be developed.

Traditionally, ecologists assumed that the most

important processes affecting populations operated at

local spatial scales (Carothers et al. 1974, Urban and

Smith 1989). Recent research, however, has indicated

that larger scale assessment also should be

considered (Wiens 1989, Forman 1995). These large

scale assessments can help (1) identify changes that

are preventing desired conditions from being obtained,

(2) identify future management actions that are likely

to be most effective, and (3) determine if management

is feasible or warranted. Often, as is the case with the

riparian community, the evaluation largely is

subjective because all the required information is not

available. However, a combination of available

information integrated with logic and general

principles of how processes affect community

structure and function can still provide valuable

insight for making management decisions. The

following information has been developed based on this

perspective.

Increased groundwater pumping, canals, diversion

dams, and reservoir construction have all contributed

to altered stream flow in both streams (Christensen

and Juracek 2001). However, the impacts to the

riparian community caused by pumping, diversion

dams and canals differ in some ways from those

caused by the reservoir and will be discussed

separately. The first three activities occur above the

refuge, are associated largely with agriculture, and

have changed the annual hydrograph by reducing the

volume of water in the channel and changing the

timing of peak and low periods (Wis. Bureau of

Reclamation 2002).

Exact shifts should be determined by analyzing long-term

hydrographs (if available). However, compared

to historic conditions, the general effect is that larger

storm events or longer wet periods are required to

cause the same amount of overbank flooding and

channel scouring. The periodic occurrence of these

actions is critical to maintaining channel diversity

(e.g., pools, riffles) and creating conditions suitable for

germination of new woody and herbaceous vegetation.

In addition, the long-term average depth to the water

table underlying the floodplain has likely increased.

This change has occurred because the groundwater in

the floodplain is in direct connection with water in the

channel (see sections on Geology and Groundwater).

In general, these two entities are in dynamic

quilibrium. Due to upstream influences, the volume,

and therefore depth, of water in the channel has

decreased during non-flood periods. As water depth

has decreased, free water (i.e., not bound by soil

particles) in the ground has likely seeped into the

channel until equilibrium is reached or no more

groundwater is available. The greater the drop in

channel water depth (e.g., during extreme drought),

the more likely groundwater will seep into the

channel. Ultimately, these changes can result in

decreased soil moisture in the rooting zone of the

floodplain. This can impact plant community

composition and structure because altered soil

moisture can influence germination potential of seeds

and affect the growth of existing plants. Site-specific

information necessary to confirm these changes and

determine if the magnitude of change has been

sufficient to alter plant composition is not available,

but it is recommended that data be collected. This

can be accomplished by refuge staff using relatively

inexpensive methods and would be valuable in

determining future management actions.

Construction of the reservoir occurred immediately

downstream of the riparian community managed by

the refuge. Similar to upstream hydrologic

alterations, the dam has reduced flow velocity in the

stream because water no longer can be transported

downstream unobstructed. Historically, these events

were important because floodplain vegetation was

disturbed and areas suitable for new germination were

created. The reduced frequency or absence of these

events likely lowers the potential of bare, moist

substrate necessary for regeneration of species such

as cottonwood and willow (Scott et al. 1993). In

addition, the reservoir functions to store water during

wet years. During prolonged wet periods, or during

extreme precipitation events, the impoundment of

floodwaters can result in inundation of the floodplain

to deeper depths and for longer periods than

historically occurred. If inundation lasts a sufficient

time it can lead to the mortality of vegetation (Teskey

and Hinckley 1977). Also, the release of water from

the reservoir is timed to coincide with irrigation needs,

usually summer and early fall (U.S. Bureau of

Reclamation 2002). This, in combination with

upstream activities, has changed the period of

maximum stream flow from spring to summer. This

shift has several impacts, but one of the most

important is the potential effect on germination of

18

riparian vegetation. Seeds of many species, including

cottonwood and willow, are dispersed in spring, are

short-lived, and require bare, moist substrate for

germination. Thus, the shift from spring to summer

flows can negatively impact germination of these

species.

During certain years, or combinations of years, the

effect of these impacts can result in severe and long-lasting

effects on riparian vegetation. This is obvious

based on the damage to riparian vegetation caused by

the most recent wet period (1993-2000). This damage

was still evident in March 2003 and illustrates

potential management issues that should be

addressed during CCP development. A general

chronological description of events follows:

1. A period of high water starts in 1993 with a 140 cm

(55 in) rain event (U.S. Bureau of Reclamation 2002,

refuge staff). The riparian study (Sevigny 1998) was

being conducted at about this time.

2. Water levels in the reservoir increase enough to

back water into the lower portion of the riparian zone

immediately upstream of the reservoir.

3. The water remains at depths long enough in the

lower portion to kill all but 142 ha (350 ac) of riparian

vegetation (U. S. Fish and Wildlife Service 1996).

Although mortality was not as evident in the upper

riparian zone, it is likely that this area was also

impacted to some extent. For example, water in the

upper channel could not be evacuated downstream

until water was released from the reservoir. Thus,

some of the same plant species (e.g., reed canary

grass, Canada thistle) that became established in the

lower riparian zone also became established in low

elevation portions of the upper zone.

4. The wet period ends in 2000 and water recedes

between 2001 and 2003. To facilitate crop irrigation

downstream, releases occur primarily during summer

months (U.S. Bureau of Reclamation 2002).

5. The release of water is slow and scouring does not

occur in either the channel or floodplain.

6. Germination of woody vegetation is minimal

because bare, moist substrate is not available during

spring when seeds of cottonwood and willow disperse

from surviving parent trees.

7. Germination of herbaceous plants occurs, but

species composition is dominated by reed canary

grass and Canada thistle. This is expected since

seeds and rhizomes of these species are adapted to

growth in moist, warm soils that frequently occur

with summer removal of water.

8. In the spring of 2003, the floodplain consists of

standing, dead timber with an understory of Canada

thistle and reed canary grass.

This scenario likely occurs infrequently based on the

fact that it has occurred only once since reservoir

construction in 1952. This is not surprising based on

information provided by the refuge staff that the long-term

drought/wet cycle spans a 30-year period with

about 23 years of drought and seven years of wet

conditions. Because the most recent wet period (1993-

2000) ended three years ago, reservoir water levels

should continue to decline over the next 20 years.

However, even if these long-term predictions are

correct, the impacts of recent high water have been

severe. Tree mortality has been significant,

regeneration of the woody component is sparse, and

non-native vegetation has replaced natives in the

understory. Undoubtedly, such changes have also

altered the avian community from what was reported

in the mid-1990’s. Perhaps more important,

restoration will be costly in terms of both money and

labor. Some of the dead timber may have to be

removed, the invasive species controlled, and suitable

conditions for regeneration of desirable trees and

grasses created. Therefore, it is recommended that

more information be obtained regarding the

relationship between reservoir pool elevations and the

duration and depth of floodplain inundation for

different reaches of the riparian community. This

would help in determining the potential costs,

benefits, and reliability of managing different portions

of the riparian community. Several techniques could

be used to develop information on flooding depth in the

floodplain. The most comprehensive (e.g., full range

of estimates) and accurate technique would involve the

use of bathymetry or topographic data that

encompasses both the reservoir and tributary

streams. Alternatively, current aerial photographs of

the reservoir at different pool elevations could be used

to estimate relationships. Data on flood frequency

and duration can be developed based on the number of

times and duration the reservoir pool exceeded certain

elevations, respectively.

Management Potential – Streams, and their

associated floodplains, are complex ecological systems

that provide many benefits to society. Throughout the

western United States, these areas are valued as a

source of water, recreational activities, and the unique

plant and wildlife resources they support. Due to

these myriad values, they also are among the most

highly modified systems. Usually a single stream is

owned and managed by multiple entities for purposes

that often conflict to some extent. Thus, the ability to

successfully manage a reach for a specific outcome is

often influenced by uses both upstream and

downstream of the site.

Although the above description is generic and applies

to many streams, it aptly describes the portions of the

North Fork Solomon River and Bow Creek managed

by FWS. The values of this community are still

apparent based on available data. However, past

alterations both upstream and downstream of the

refuge have caused significant changes that affect the

ability of FWS to maintain the functions and processes

that supported the historic riparian community. Of

primary concern are the hydrologic alterations that

result in extreme water level fluctuations in the

floodplain. High water similar to that experienced in

the mid-1990’s may occur infrequently, but the cost of

restoring the native community following such events

19

likely will be time-consuming and costly. Further, this

effort may be required every 20-30 years based on

long-term predictions. Potential solutions that

address the entire riparian community are not readily

apparent because release of water from the reservoir

during high flow periods would be required. This is

not likely since a primary reason for reservoir

construction was to store this water for irrigation

below Kirwin. If a viable solution is not found, it is

recommended that the benefits and costs associated

with managing different portions of the riparian

community be evaluated. This may result in the

identification of portions that are relatively free of

reservoir impacts. Part of this analysis would require

determining the likelihood that certain reservoir

elevations will occur during a given time period and

estimating the riparian area impacted at these

elevations. For example, a reservoir elevation of 528.8

m (1735.0 ft) would likely occur 2 times every 20 years.

At this elevation, 20% of the riparian community

would be flooded to depths that cause significant

damage. Based on these probabilities, the riparian

community could be divided into areas that are

frequently and infrequently prone to reservoir

impacts. Each of these areas could be addressed

differently. For example, the area not prone to

reservoir impacts could be managed to provide the full

range of floristic composition and structure desired

with a high level of probability that progress would

not be impacted by the reservoir. In contrast,

management effort in the area frequently

experiencing reservoir impacts would differ because

the benefits would be short-term and the cost

excessive. This does not mean that this portion of the

riparian community would be abandoned; rather a

different set of management goals would be developed

that take into account uncontrollable factors.

Upland

Kirwin NWR is within the central dissected, or mixed-grass,

prairie region that historically was dominated

by the bluestem-grama association (Launchbaugh and

Owensby 1978). This association was prevalent on

uplands in west-central Kansas, but also extended

west on breaks into the dissected parts of the High

Plains where the grama-buffalo grass prairie

dominated the landscape. According to Kuchler

(1974), the bluestem-grama association is

characterized by dense communities of grasses and

forbs that often are in two distinct layers: one of low-growing

grasses and one of medium tall grasses and

forbs that is usually more open. Dominant species are

big and little bluestem, sideoats grama, and blue

grama. Other characteristic species include western

wheatgrass, western ragweed, leadplant, purple

threeawn, hairy grama, buffalo grass, Freemont’s

clematis, purple coneflower, and Canada wildrye

among others.

Factors historically controlling the distribution and

physiognomy of the mixed-grass prairie included

precipitation, fire, and herbivory. The plant species

composing this prairie are sensitive to major

precipitation fluctuations; thus, their distribution

tends to move east and west in response to alternating

periods of intense drought or wetness (Kuchler 1967,

1972). Summer fires (Sauer 1950) and herbivory

(Dyksterhuis 1958) also helped maintain the prairie by

suppressing woody vegetation. However, certain

woody plants were always present as natural

components in some areas (Kuchler 1974).

Herbivores, including bison and smaller vertebrates

such as prairie dogs, altered soil characteristics and

other factors that influenced plant establishment and

growth (Kuchler 1974).

Following the onset of human settlement, however,

processes were modified that profoundly affected the

prairie (Knopf and Samson 1997). Fire suppression,

development and expansion of agricultural crops,

changes in herbivores and herbivory, and planting of

trees have significantly altered the prairie landscape.

In addition, technological advances brought about

other less obvious but equally important changes,

including the development and introduction of new

grasses and crops, groundwater pumping, herbicides,

and fertilization. These and other actions have

resulted in significant loss and fragmentation of the

prairie community. For example, currently about 60%

of Kansas lands are used for agriculture. Of this 60%,

about 48% is cropland and the other 12% is non-native

grasses (e.g., brome) or CRP (seeded natives).

The condition on Kirwin NWR is representative of the

conditions for Kansas as a whole. The refuge

encompasses about 2,833 ha (7,000 ac) of uplands at

conservation pool and 2,712 ha (6,700 ac) at a pool

elevation of 527.7 m (1,731.3 ft). Grasslands dominate

this acreage, but the refuge staff reports that only

about 81 ha (200 ac) of native grass occur on the

refuge. The remainder is either pasture or reseeded

grass. Further, much of the native grass is isolated

(i.e., fragmented) and occurs in small blocks. Other

habitats occurring in the uplands include shelterbelts,

croplands, chalk bluffs, and a few temporary

wetlands. Although the exact area of shelterbelts is

not known, many appear to be 15-31 m (50-100 ft) wide

and extend for various distances along roads and fence

lines. The tree composition includes a mix of both

hardwood and evergreen species. Wheat, sorghum,

corn, and alfalfa are the dominant crops on the refuge

and approximately 486 ha (1,200 ac) are planted

annually when the reservoir is at an elevation of 527.1

m (1729.25 ft). The cropping program is designed to

prepare agricultural land for conversion to grass and

provide foods for migratory birds and resident

wildlife. Farming is accomplished using cooperative

farmers and arrangements vary depending on crop

(refuge staff). For example, the refuge share of row

crops is 25%, whereas stubble constitutes the refuge

share of wheat.

Chalk outcroppings occur at higher elevations in the

uplands, whereas only a few palustrine wetlands occur

in depressional areas. Both of these communities

exist as small, disjunct areas that compose only a

small percentage of refuge lands. Although not

covered in this report due to lack of information, both

support some distinctive plant species and constitute

unique habitats on the refuge. Thus, additional

information should be obtained regarding their

locations, sizes, and unique resources.

20

Most of the remaining discussion uses the grassland

component of the uplands as the baseline condition.

This approach was used because comparing and

contrasting the values and management options of

multiple plant communities without a standard for

comparison is confusing. Grassland was selected as

the standard because it represents the historic plant

community and presently dominates the uplands.

Also, the values and methods of managing other

habitats (e.g., corn, shelterbelts) in the uplands are

largely known and require little discussion.

Although much of the historic prairie on the refuge

was converted or degraded prior to establishment,

this community (excluding areas adjacent to the

reservoir) appears to be the least effected by the

reservoir. Consequently, FWS has more direct

control and can likely influence future conditions

more reliably. In general, the current condition of

refuge grasslands varies greatly. There are small

areas, many on the south side of the reservoir, that

contain a high proportion of native grass and forb

species. In contrast, other areas are primarily

composed of non-native, cool season grasses. The

dominant non-native species is smooth brome, but

small areas of Kentucky bluegrass also are present

(refuge staff). Finally, areas in various stages of

restoration also occur on the refuge. Species

composition of these stands is mixed, with the

presence of both warm season natives and cool season

non-natives. Unfortunately, more detailed information

on the current condition of the grasslands is lacking.

Maps depicting the locations and sizes of different

grassland types (native, reseeded, tame) are not

available, and both qualitative and quantitative

information regarding the floristic composition

(species of grasses, forbs) and structure of each type

are unknown. Although all information is rarely

available, the lack of at least some of the above

information renders any comments incomplete and

speculative. Therefore, the following information is

provided as a recommendation regarding the types of

information that should be developed prior to CCP

development. In addition, examples are provided

regarding potential uses of this information for

developing future management direction.

First, the locations and floristic composition of the

different grassland types should be determined. The

most proficient method would be to develop a digital

vegetation map based on floristics rather than generic

terms such as pasture, reseeded grass, or native

grass. The latter terms can be misleading if labels

are not applied based on floristic attributes and often

do not convey sufficient information to determine

current value or management options. For example,

the designation of native grass on the refuge appears

to be based on the premise that the ground has never

been plowed (e.g., unbroken sod is another frequently

used term). Such a designation may be correct and

provide some information, but definitions based solely

on past management actions (or lack thereof) do not

guarantee that floristic composition and/or structure

are representative of the historic native community.

For example, invasive grasses (e.g., smooth brome)

can invade areas never plowed and alter plant and bird

community composition and structure (Wilson and

Belcher 1989). Therefore, there is also no guarantee

that the plant community will provide suitable habitat

for vertebrates (e.g., birds) and invertebrates (e.g.,

butterflies) that are an important part of the

grassland community.

Second, the current value of the different grassland

communities identified during vegetation mapping

must be determined. Many vertebrates and

invertebrates exhibit plasticity and can tolerate

certain changes in plant composition and/or structure.

Thus, certain non-native communities, or mixed

native/non-native communities, may provide sufficient

value to warrant maintaining on the refuge. For

example, row crops and wheat provide foods for select

species of waterfowl, shelterbelts provide suitable

habitat for certain songbirds, and CRP plantings that

often include some non-native species provide benefits

to some grassland birds. The benefits provided by

these habitats must be compared to the potential

benefits that would result from converting portions of

these areas to other types. For example, shelterbelts

provide habitat for some songbirds, but represent

“hostile environments” that can hinder breeding by

some grassland bird species. Thus, removing

shelterbelts to increase the contiguous area of

grasslands above a threshold size may benefit

grassland birds but negatively impact songbird

numbers. Such tradeoffs should be documented prior

to making management decisions. Part of this

analysis should also involve evaluating the spatial

attributes (e.g., size, location) of each community.

Appropriate floristic composition and structure is

important, but many species also require areas of

certain size, or multiple communities in close

juxtaposition, to adequately meet life history

requisites (Herkert 1994, Helzer and Jelinski 1999,

Walk and Warner 1999). For example, some species of

grassland birds known to occur on the refuge exhibit

minimum area requirements for nesting. Therefore,

the grassland areas that have been developed for

public use may not constitute suitable habitat because

the extensive road network may have fragmented

these areas sufficiently, regardless of human activities

associated with these areas.

Although a large amount of information exists on

species-habitat relationships, attempts often are made

to assess values based on opinions and past

experiences. This often results in confusion and

conflict. Therefore, it is recommended that objective

information be developed prior to discussing the

relative merits of each community. This may seem

daunting, but much of the information on bird-habitat

relationships has already been compiled for many

species known to occur on the refuge, and additional

research could be conducted. The digital vegetation

map also would prove invaluable because it could be

used to assess size of each community or distances

between communities.

Third, the benefits and detriments of altering the

type, size, or spatial location of various communities

should be evaluated. For example, the removal of

shelterbelts and/or cropland in certain areas may not

21

result in the creation of contiguous grassland that

exceeds a threshold required to meet the needs of

certain avian species. In such cases, the costs of

conversion would not be warranted because it would

not result in expected benefits. Similarly, the public

use facilities may fragment grasslands in the eastern

portion of the refuge sufficiently to preclude nesting

by some wildlife. If this were true, the benefits of

managing these areas to provide specific structural

conditions may not be worth the cost. If a digital

vegetation map were available, various scenarios could

be portrayed graphically to determine the best

configuration and size of different community types.

Finally, information should be developed to evaluate

the potential of accomplishing different activities. For

example, statements made by the refuge staff suggest

that the current conditions of most existing

grasslands are not desirable. If an evaluation

confirms this statement, then decisions must be made

regarding the potential for different grassland areas.

An initial step could involve classifying each grassland

tract as being in need of either restoration or

maintenance. This is an important distinction because

maintenance implies that the existing floristic

composition is acceptable and management is directed

toward altering structural components (e.g., litter

depth, height, density). Although not everything is

known, the use of tools (fire, herbivory, mowing) to

maintain grasslands has received considerable

attention by the scientific community and general

concepts exist regarding appropriate techniques. In

contrast, true “prairie” restoration is much more

difficult. A review of the literature suggests some

techniques have reliably proven successful in

establishing some components of prairie (e.g., warm

season grass), but much remains to be learned

regarding restoration of all components (e.g., forbs,

native cool season grasses). This is particularly true

of mixed-grass prairie. Little definitive information

exists that provides guidance in determining

appropriate site preparation, seed mixtures, or time

and rate of seeding for different components relative

to specific site conditions (e.g., soil type, groundwater

table).

The lack of information for the existing grassland

communities, in combination with the lack of proven

restoration techniques for mixed-grass prairie,

suggests that attempts to convert existing upland

communities to native prairie will be difficult.

However, this does not mean that an attempt should

not be made. Rather, it indicates that success will

require much work over numerous years (Kindscher

and Tieszen 1998, Fuhlendorf et al. 2002), development

of appropriate monitoring techniques, and frequent

experimentation. An extensive search of the

literature and consultations with local experts likely

will provide initial guidance useful for identifying

potential attributes that may influence restoration

potential. This information could then be used to

refine additional searches for refuge-specific

information on these attributes. For example, if

depths to groundwater or certain soil characteristics

are deemed important, this information could be

obtained from various sources and evaluated to

determine the most likely locations to attempt

restoration.

Management Potential – In many respects, FWS can

exert the greatest influence on the upland community

compared to other community types. However,

constraints still exist that will influence future

conditions. Uplands adjacent to the reservoir are

wetter during high water years and extensive

groundwater pumping upstream of the refuge likely

has altered the subsurface hydrology of some upland

habitats. The effects of these alterations are

unknown, but research indicates change in the water

table can effectively alter environmental conditions

and, therefore, plant species occurrence (Currier

1988). Thus, literature reviews should be conducted

to determine potential changes in abiotic conditions

that would preclude certain management options. For

example, restoration of native grasses and forbs

adjacent to the reservoir may not be feasible due to

changes in soil characteristics. If this is the case,

another grassland type or habitat should be

considered. In addition, invasive species have altered

floristic and structural attributes of many grassland

tracts. Although techniques have been developed for

controlling many of these species, desirable

vegetation must be established following control or

there are no long-term biological benefits. The ability

to accomplish this task largely will depend on what is

considered a desirable community. Depending on the

types of natural resource benefits the core team

decides the refuge should provide, grasslands ranging

from native mixed-grass prairie to CRP or tame

grasses may be appropriate. Some of these types can

be developed with a high degree of certainty, whereas

others are very difficult to establish and maintain.

Finally, some plant communities (e.g., shelterbelts,

croplands) and the public use facilities were not

present historically. Although such features may be

considered obstructions from the perspective of

historic conditions, they provide values that cannot,

and should not, be ignored. Therefore, decisions must

be made regarding the appropriate mix of historic and

current conditions. These decisions will likely

influence the types and configuration of the different

communities composing the uplands.

22

Public Use

In addition to wildlife benefits, the refuge provides

various forms of public recreation that should be

mentioned. At conservation pool there are 1,402 ha

(3,465 ac) of surface water (68.4% of reservoir) open to

activities such as boat fishing, water skiing and jet

skiing. In addition, about 41 ha (100 ac) of grassland

adjacent to the main body of the reservoir are

developed for public use, including shore fishing,

camping, and picnicking (U. S. Bureau of Reclamation,

URL http://www.usbr.gov/gp). There is also an

extensive road network, including a tour loop around

the perimeter of the refuge. Finally, hunting (deer

and waterfowl primarily) is allowed in both the

riparian and upland communities. Given the number

and types of recreational visits to the refuge (U.S.

Bureau of Reclamation 2002), it is possible that human

activities could represent a disturbance that impacts

some species that use the refuge. Peak waterfowl and

shorebird migration periods do not appear to coincide

with peak recreational periods (refuge staff).

However, ground nesting ducks (e.g., Gadwall,

Mallard), Least Terns, grassland birds, and species

that nest in trees adjacent to deepwater habitat (e.g.,

Double-crested Cormorants, Great Blue Herons, and

Bald Eagles) are present during the summer

recreational period. Various regional and national

plans consider populations of some of these species to

be of concern (Appendix D) and information is

available that indicates human activities can interfere

with the successful rearing of young of these species.

Despite these observations, which only allude to

possibilities, there is not sufficient evidence to either

confirm or discount conflicts. However, the fact that

human-wildlife conflicts may occur identifies the need

to develop specific information regarding potential

detrimental impacts to wildlife. Scientific information

addressing the effects of disturbance is available, but

conditions on Kirwin NWR may differ from those

reported elsewhere. Thus, definitive resolution of this

concern may require conducting an appropriately

designed monitoring program or research study on

the refuge. Regardless of the source, this information

could be used to develop solutions that eliminate or

minimize impacts without negatively impacting public

use opportunities.

Finally, the refuge staff has responsibility for

enforcing regulations and maintaining facilities

associated with recreational activities. Currently, the

staff estimate that possibly 50 to 60% of their time is

annually devoted to tasks associated with the public

use program (refuge staff). This amount of effort,

although often warranted, should be evaluated relative

to the time required to achieve the biological goals

established in the CCP. FWS should use this

evaluation to determine an appropriate balance

between natural resource and public use

management, and adjust budgetary and time

requirements to meet these needs.

23

Literature Cited

Albertson, F. W. 1937. Ecology of mixed prairie in west central Kansas. Ecological Monographs 7:481-547.

American Ornithologists’ Union. 1998. Check-list of North American Birds. 7th edition. American

Ornithologists’ Union, Washington, D.C.

American Ornithologists’ Union. 2000. Forty-second supplement to the American Ornithologists’ Union Check-list

of North American Birds. Auk 117:847-858.

American Ornithologists’ Union. 2002. Forty-third supplement to the American Ornithologists’ Union Check-list

of North American Birds. Auk 119:897-906.

American Ornithologists’ Union. 2003. Forty-fourth supplement to the American Ornithologists’ Union Check-list

of North American Birds. Auk 120:923-931.

Bailey, J. K., J. A. Schweitzer, and T. G. Whitman. 2001. Saltcedar negatively affects biodiversity of aquatic

macroinvertebrates. Wetlands 21:442-447.

Baker, L. A. 1992. Introduction to nonpoint source pollution in the United States and prospects for wetland use.

Ecological Engineering 1:1-26.

Callender, E., and J. A. Robbins. 1993. Transport and accumulation of radionuclides in stable elements in a

Missouri River system. Water Resources Research 29:1787-1804.

Carothers, S. W., R. R. Johnson, and S. W Aitchison. 1974. Population structure and social organization of

southwestern riparian birds. American Zoologist 14:97-108.

Christensen, V. G. 1999. Deposition of selenium and other constituents in reservoir bottom sediment of the

Solomon River Basin, north-central Kansas. U.S. Geological Survey, Water Resources Investigations Report 99-

4230.

Christensen, V. G., and K. E. Juracek. 2001. Variability of metals in reservoir sediment from two adjacent basins

in the central Great Plains. Environmental Geology 40:470-481.

Currier, P. J. 1988. Plant species composition and groundwater levels in a Platte River wet meadow. Pages 19-24

in T. B. Bragg and J. Stubbendieck, editors. Proceedings of the 11th North American Prairie Conference.

University of Nebraska, Lin