The effects of changes in flow on the ecological conditionof two Arizona streams: analysis of trends in waterchemistry and structure of biological assemblages

Item Type Thesis-Reproduction (electronic); text

Authors Bymers, Leah.

Publisher The University of Arizona.

Rights Copyright © is held by the author. Digital access to this materialis made possible by the University Libraries, University of Arizona.Further transmission, reproduction or presentation (such aspublic display or performance) of protected items is prohibitedexcept with permission of the author.

Download date 05/05/2018 18:13:48

Link to Item http://hdl.handle.net/10150/191336

THE EFFECTS OF CHANGES IN FLOW ON THE

ECOLOGICAL CONDITION OF TWO ARIZONA STREAMS: ANALYSIS OF

TRENDS IN WATER CHEMISTRY AND STRUCTURE OF

BIOLOGICAL ASSEMBLAGES

by

Leah Bymers

A Thesis Submitted to the Faculty of the

SCHOOL OF NATURAL RESOURCES

In Partial Fulfillment of the Requirements for the Degree of

MASTER OF SCIENCE

WITH A MAJOR IN FISHERIES AND WILDLIFE

In the Graduate College of

THE UNIVERSITY OF ARIZONA

2005

STATEMENT BY AUTHOR

This thesis has been submitted in partial fulfillment of requirements for anadvanced degree at The University of Arizona and is deposited in the University Libraryto be made available to borrowers under rules of the Library.

Brief quotations from this thesis are allowable without special permission,provided that accurate acknowledgment of source is made. Requests for permission forextended quotation from or reproduction of this manuscript in whole or in part may begranted by the head of the major department or the Dean of the Graduate College when inhis or her judgment the proposed use of the material is in the interests of scholarship. Inall other instances, however, permission must be obtained from the author.

2

SIGNED:

APPROVAL BY THESIS DIRECTOR

This thesis has been approved on the date shown below:

Matter DateProfessor of the School of Natural Resources

Dr. 'William/(t, 6" 0 6

2

ACKNOWLEDGEMENTS

I would like to thank the official and unofficial members of my committee, Dr. David

Walker, Dr. Bill Matter, Dr. Kevin Fitzsimmons and Gail Cordy for their feedback and

guidance. Also, I would like to acknowledge David Aiming and Nick Paretti of the USGS

for their help in obtaining and understanding the hydrological data and biological

collection procedures. Chris Goforth was also integral in analysis of macroinvertebrate

data.

3

TABLE OF CONTENTS

ABSTRACT 8

INTRODUCTION 9

Seasonal precipitation patterns 11

Streamflow and base flow 12

Effect of precipitation on flow 13

Study sites 13

Ecological assessment 16

METHODS 19

Data 19

Periphyton collection 20

Macroinvertebrate collection 21

Fish collection 22

Metrics 22

RESULTS 23

Historical precipitation and flow 23

Water chemistry 26

Biological assemblages 29

DISCUSSION 31

Fish 31

Macroinvertebrates 33

Periphyton 36

4

TABLE OF CONTENTS - Continued

CONCLUSION 39

APPENDIX A 40

REFERENCES 42

5

LIST OF FIGURES

FIGURE 1, Historical precipitation patterns 10

FIGURE 2, Study sites on topographical map of Arizona 14

FIGURE 3.1-3.6 Historical precipitation and flow at study sites 23

FIGURE 4, Regression of temperature and DO with flow at the San Pedro River at

Charleston 27

FIGURE 5, Regression of temperature and DO with flow at West Clear Creek near

Camp Verde 28

FIGURE 6, Regression of native:non-native fish with streamflow at the San Pedro

at Charleston 32

FIGURE 7, Regression of native:non-native fish with streamflow at West Clear

Creek near Camp Verde 32

FIGURE 8, Regression of H/QDEQ Biotic Index with flow at the San Pedro River

at Charleston 34

FIGURE 9, Regression of H/ADEQ Biotic Index with flow at West Clear Creek

near Camp Verde 34

FIGURE 10, Regression of FFG with flow at West Clear Creek near Camp Verde.35

FIGURE 11, Regression of periphyton total abundance with flow at West Clear

Creek near Camp Verde 37

6

LIST OF TABLES

TABLE 1, Regression table for water chemistry parameters and flow at the San

Pedro River at Charleston 28

TABLE 2, Regression table for water chemistry parameters and flow at West Clear

Creek near Camp Verde 28

TABLE 3, Regression table for all taxa with flow at the San Pedro River at

Charleston 29

TABLE 4, Regression table for FFG with flow at the San Pedro River at

Charleston 30

TABLE 5, Regression table for all taxa with flow at West Clear Creek near Camp

Verde 30

TABLE 6, Regression table for FFG with flow at West Clear Creek near Camp

Verde 30

7

ABSTRACT

Analysis of climate observations have led researchers to predict that precipitation

in the Southwest, USA will become more extreme in its variability. This prediction

combined with increasing demand for water for human use support the claim that streams

in the Southwest will experience flows lower than historic minima in the future. In order

to protect stream communities, the effects of decreasing flow need to be studied. I

investigated how changes in streamflow and base flow of the San Pedro River and West

Clear Creek, AZ, affect water chemistry, fish, macroinvertebrates, and periphyton.

Regression analysis showed the macroinvertebrate pollution-tolerance index and

percent Collectors were significantly negatively related to flow. This indicates a

depression of pollution-intolerant species with lower flows. Several other strong

relationships were noted, however, long-term data are needed in order to make

conclusions about the future effects of decreased flow on the biological assemblages of

arid perennial streams.

8

INTRODUCTION

The climate of southern and central Arizona is characterized as warm and arid

(Liverman and Merideth 2002, Sheppard et al. 2002 ). Average annual rainfall in Arizona

is only 32.3 cm (Sheppard et al. 2002) and temperatures are consistently among the

highest in the country. Climate data, either actual measurements of temperature and

precipitation, or those derived from other sources (e.g., tree rings and models) show

marked fluctuations through time (Hereford and Webb 1992, Fenbaio et al. 2002,

Sheppard et al. 2002). Both summer and winter precipitation are highly variable across

years. Periods of low or high precipitation have occurred on an annual to multi-decadal

time scale. When examining precipitation patterns back to 1700, it is apparent that in the

recent past (1900 to the present), years of extreme high and low temperature and

precipitation have become even more extreme (Sheppard et al. 2002). In other words, wet

periods are becoming wetter and dry periods are becoming drier. This phenomenon has

been observed in several arid ecosystems around the world (Humphries et al. 2003).

9

Figure 1. Long-term precipitation fluctuations. PDSI = Palmer Drought SeverityIndex; higher number =more precipitation. From Sheppard et al. (2002).

The effects of these aspects of climate change on the biological communities of

streams and rivers are largely unknown (Lake 2003). Organisms living in perennial

streams are well-adapted to the characteristic variable stream flow of the Southwest;

however, a change in the extent of variability may alter the structure of the communities.

This is especially true in regions that already experience climate extremes such as the

desert Southwest (Brown et al. 1997). Intense droughts predicted for the future and the

likely increase in anthropogenic water demand could lead to streamflow dropping below

historic minima in coming years. Therefore, research to determine if stream assemblages

have potential to be negatively impacted (e.g., loss of diversity, invasion of non-native

species) by reduced flow, is a necessity. Accurate and in-depth information about how

climate variability affects Arizona's aquatic communities can help in planning for

allocation of water resources (Fenbaio et al. 2002).

10

11

I investigated how precipitation patterns affect annual streamflow and base flow

of two perennial streams of southern and central Arizona and in turn, how those changes

in flow are related to the water chemistry and biological assemblages of the two streams.

Seasonal precipitation patterns

The geographic location of the state of Arizona combined with its diverse

topography results in variable precipitation patterns across the state (Sheppard et al. 2002,

Skirvin et al. 2003). In southeastern and central Arizona, precipitation, although

regionally variable, follows a very pronounced seasonal pattern. Most of the annual

rainfall occurs in winter and summer. Spring and autumn are usually extremely dry.

Winter rains, which provide roughly 30% of the annual total (Sheppard et al.

2002), result from stotins that come from the northwest and pass through California

(Sheppard et al. 2002). Winter rains are greatly affected by El Nino-Southern Oscillation

(ENSO) processes. ENSO is the result of an increase in sea-surface temperature of the

eastern equatorial Pacific Ocean concurrent with movement of the active center of

atmospheric convection from western to central equatorial Pacific. La Nina years occur

when there is a decrease in the sea-surface temperature and no shift in the convection

center in the equatorial Pacific Ocean. El Nino years result in wet winters, whereas La

Nina years are typically more dry (Brown et al. 1997, Sheppard et al. 2002). Another

factor determining the amount of precipitation that falls in the winter is the Pacific

Decadal Oscillation (PDO). The PDO is an index that represents long-term fluctuations in

sea-surface temperature of the northern Pacific Ocean. It can work in conjunction with

ENSO to increase winter precipitation variability throughout the Southwest (Sheppard et

12

al. 2002). Winter rain in Arizona is generally more long-lasting and gentle than rain that

falls in summer.

Summer rain in Arizona occurs during what is known as the monsoon season.

Monsoonal storms are a result of a change in wind direction (a subtropical high-pressure

ridge replacing the westerlies) that begins in Mexico and extends northeast through

Arizona and New Mexico (Hereford and Webb 1992, Sheppard et al. 2002). Precipitation

that occurs during the monsoon season is characterized by short, heavy bursts of rain and

provides up to 50% of Arizona's annual precipitation (Sheppard et al. 2002). These

storms usually occur between early July and early September.

Streamflow and base flow

Both summer and winter rains have a large effect on streamflow. Streamflow is

the total volume of water flowing in a stream cross section at any particular time. This

includes groundwater input, rainfall into the stream and runoff from precipitation or

snowmelt. An important parameter to monitor is base flow. Base flow of the stream is the

component of the total flow that comes solely from groundwater. Base flow cannot be

directly measured, but rather is calculated from streamflow records. I chose to use the

BFI model (Base Flow Index) created by the British Institute of Hydrology. This program

uses local minimum streamflow data and a recession slope test to estimate base flow.

Base flow is recognized as a good comparative measure when examining long-term

hydrological trends because it is less affected by rain events than streamflow (Wahl and

Wahl 1988). Extreme events must be taken into consideration when investigating short-

13

term implications of changes in flow on the biological assemblages of streams, therefore,

mean streamflow is also used in analyses.

Effect of precipitation on flow

Monsoon rains that occur in Arizona in the summer are heavy and ephemeral and

greatly affect streamflow. A monsoonal rain event often elevates discharge in a stream by

100 or even 1000 fold, but because this water runs off quickly, and because evaporation

rates are high, summer rains do not contribute as much to base flow, as do winter rains

(Bryson and Hare 1974). The more gentle and longer-lasting rains in winter, and snowfall

in higher elevations are more likely to seep into the ground and recharge groundwater

reserves, thus elevating base flow. This makes it imperative to not only monitor annual

precipitation, but also season precipitation when tracking potential drought trends in the

context of its effect on streams.

Study sites

Study sites were the San Pedro River at Charleston and West Clear Creek near

Camp Verde, Arizona. These streams had been selected by the National Water Quality

Assessment program as study sites because they are perennial and considered minimally

affected by anthropogenic disturbance. They also are representative of two different

ecoregions of Arizona; basin and range lowlands and central highlands (Cordy et al.

1998, Gebler 2004). Long-term data from US Geological Survey (USGS) stream gages

are available for both sites.

Figure 2. Southern "X" is the San Pedro River at Charleston and central Arizona

"X" is West Clear Creek near Camp Verde.

The San Pedro River is a tributary of the Gila River and its watershed is roughly

10,000 km2 . The river begins in Sonora, Mexico and flows north into Arizona. It flows

near the towns of and Cananea, Mexico, and Sierra Vista and Fort Huachuca, AZ in its

upper reaches and then progresses northward, to the east of the Huachuca and Santa

Catalina Mountains. The Charleston site is roughly 10 km east of Sierra Vista. The

14

15

substrate of the San Pedro River is characterized as sand and coarse-grained sediment or

basin fill (Cordy et al. 1998). Riparian vegetation in the upper reaches is dominated by

Populus fronontii and Salix goodingii (Gebler 2004). Tamarix spp and Prosopis spp. can

also be found. Chihuahuan desertscrub and semi-arid grassland occur outside of the

riparian corridor (Gebler 2004). Much of the area outside of the riparian corridor is

rangeland used for livestock production. The San Pedro watershed receives an average of

38.2 - 51.0 cm of rain/year with an average of 152.9 - 165.6 cm of free water-surface

evaporation (Cordy et al. 1998, Skirvin et al. 2003). The upper reaches of the river have

been designated a National Riparian Conservation Area since 1988.

West Clear Creek is in a deep valley that cuts through the Mogollon Rim at the

southwest edge of the Colorado Plateau and is a tributary of the Verde River. It remains

minimally impacted primarily because it is remote and difficult to access. A very small

portion of the adjacent area is used as rangeland. The narrow band of riparian vegetation

includes various species of Pinus and Juniperus in the upper elevations and Populus

fremontii, Fraxinus velutina, Juglans major, Platanus wrightii, Acer negundo, Alnus

oblongifolia, and Salix spp. in mid-lower elevations (Galuszka and Kolb 2002). The

substrate of West Clear Creek is a combination of basin fill of gravel, cobbles, and

boulders of sedimentary bedrock, and bedrock of the mountains. The West Clear Creek

region receives an average of 38.2 cm/year of precipitation and has 127.4— 152.9 cm of

evaporation (Cordy et al. 1998).

Ecological assessment

There are many ways to quantitatively estimate the health of a freshwater

biological community (Boulton 1999). There are, in fact, many ways to define the

"health" of a freshwater body, though most authors agree that a healthy system will lack

significant distress, but would be resilient when faced with stress (Karr 1999, Norris and

Thoms1999). In recent years, the health of a river has been determined by assessing the

integrity of its biological assemblages. Karr (1999) defines biological integrity as the

state in which a community is the product of evolutionary and biogeographic processes

with minimal influence from modern human society. Many researchers agree that

sampling the biological assemblages is an essential part of assessing the integrity and

thus the health of a freshwater body (Karr 1999, Norris and Thorns 1999).

Some researchers have simply used abundance and community distribution data

to asses biological integrity (Wood et al. 2000, Alimov 2001, Nelson and Lieberman

2002, Daufresne et al. 2003, Chen et al. 2004, Clavero et al. 2004, Paperno and Brodie

2004, Pegg and McClelland 2004, Wood and Armitage 2004), where as others have

calculated diversity or similarity indices ( Pegg and Pierce 2002, Paperno and Brodie

2004, Pegg and McClelland 2004, Steffy et al. 2004, Wood and Armitage 2004). Species

richness has been recognized as a particularly valuable measure by the UN Convention

on Biological Diversity (Foggo et al. 2003). More recently, the ratio of native fish species

to non-native fish species has become an indication of the integrity of the aquatic

environment (Bernardo et al. 2003, Eby et al. 2003, Feyrer and Healey 2003, Kennard et

16

17

al. 2005). Researchers have also considered species turnover rates in fish assemblage

samples when assessing an aquatic environment (Eby et al. 2003, Adams et al. 2004).

Macroinvertebrates can be used as biotic indicators of the integrity of a water

body. Many researchers use the tolerance levels of macroinvertebrates to various

pollutants to formulate a biotic index (Davis et al. 2003). An analysis of the proportion of

the macroinvertebrate community that is represented by each of the several functional

feeding groups (FFG) has also been used to assess the health of an aquatic environment

(Merritt et al. 2002, Stepnuck et al. 2002, Habidija et al. 2003). Macroinvertebrates are

classified as 1.Collectors (those that gather or filter fine particulate matter out of the

water, such as tubificid worms and Ephemeropterans in the Baetidae family); 2.Shredders

(those that break plant material and some animal material into smaller particles through

their feeding and digestive process, such as members of the Plecoptera or stonefly order);

3.Scrapers (those that feed on algae and plant material that is on rocks or plants, such as

Coleopterans in the Elmidae family); 4.0mnivores (those that eat plant and animal

material, such as the Trichopterans in the genus Nectopsyche); and 5.Predators (those that

feed on other macroinvertebrates such as Hemipterans and Odonotans).

When deciding which metrics to use as representatives of the aquatic assemblages

in this study, I considered biological and ecological significance, ability to detect

temporal changes, and frequency of use among recently published studies. Data for fish

were analyzed for abundance, species richness, Simpson's Diversity Index, proportion of

native to non-native species, and species turnover rate. For macroinvertebrates,

abundance and species richness, as well as a modified Hilsenhoff Biotic Index (regional-

18

specific input from Arizona Dept. of Environmental Quality), composition of functional

feeding groups, and Simpson's Diversity Index were calculated. Abundance and species

richness, as well as Simpson's Diversity Index were calculated for periphyton data. These

metrics are considered good representations of diversity, and are ecologically relevant

(Washington 1984).

METHODS

Data

I used hydrological and biological data collected by the U.S. Geological Survey

(USGS). Daily flow data were obtained from long-established stream gages. The daily

flow data set for the San Pedro River at Charleston and West Clear Creek near Camp

Verde, AZ is quite complete over the long-term with only a few small gaps. The base

flow data were calculated by entering the daily stream flow into the Baseflow Index

model (BFI). BFI can be downloaded from:

http://www.usbr.gov/pmts/hydraulics lab/twahl/bfilindex.html.

Precipitation data were collected by the Western Regional Climate Center. They

maintain precipitation-monitoring stations throughout Arizona. I chose the two stations

most representative of the stream sites (Tombstone for the San Pedro River, and Happy

Jack for West Clear Creek). Trends in precipitation and stream and base flow were

analyzed using JMP version 4 statistical software.

Biological and water chemistry data were collected as part of the National Water

Quality Assessment (NAWQA) program facilitated by the USGS. This program was

designed to comprehensively analyze the condition of various flowing waters throughout

North America. NAWQA began in 1991 and was primarily concerned with collecting

data from many different locations to assess the occurrence and distribution of various

biological, physical, and chemical features of flowing waters. Recently, the focus of

19

20

NAWQA shifted to collecting the same comprehensive data at a few fixed sites in order

to monitor long-term trends.

Biological and water chemistry data collection at the San Pedro River at

Charleston and West Clear Creek near Camp Verde began in 1995 and continued

somewhat inconsistently until the present. Periphyton and macroinvertebrate data are

available from 1995-1997, and 2001 at the San Pedro and West Clear Creek sites. Data

on fish are available from 1995-1997, 2002, and 2003 from San Pedro River and West

Clear Creek. Fish data were also collected in 2001 from the San Pedro River. Because

water chemistry data were not collected on the same day as the biological samples, the

data used in analysis were taken from the sampling date closest to the biological sampling

day. If a water chemistry sample was not taken within five days of the biological sample,

an average of the measurements from the closest days pre- and post-sample was

averaged.

The collection of fish, macroinvertebrates, and periphyton at West Clear Creek

and the San Pedro River followed the NAWQA protocol. Each community collection was

performed within the same stream reach for each sample date. See Moulton et al. (2002)

for a more detailed description of current sampling protocol. The first two years of

NAWQA (1995/1996) protocol can be found in Cuffney et al. (1993), Meador et al.

(1993), and Porter et al. (1993).

Periphyton collection

Periphyton was sampled from rock surfaces and was collected using the SG-92, a

modified syringe sampling device. Five cobbles from five locations (total of 25 cobbles)

21

representative of each stream reach were collected. The SG-92 device was secured to a

smooth surface of each of the cobbles and a combination of suction, brushing, and

flushing with water, was used to dislodge the periphyton within the barrel of the SG-92.

The periphyton and water solution was then transferred to a sample bottle and preserved

with buffered formalin. The area sampled on each of the cobbles was measured and

recorded. Samples were sent to the Philadelphia Academy of Sciences Laboratory for

processing.

Macroinvertebrate collection

Riffles, from which five distinct sample sites were determined, were selected for

collection. Substrate type, water velocity, depth, and debris accumulation were kept

constant among sample sites within a particular riffle.

To collect macroinvertebrates, a disturbance-removal technique was used. A

sample area was defined and the substrate within that area was disturbed. A 1.25 m2 area,

directly upstream from the slack sampler, was disturbed by digging to a depth of 10 cm

with a hand rake and gently scrubbing the rocks and other debris. The macroinvertebrates

were dislodged washed downstream into a net. The net used, as part of NAWQA

protocol, is called a slack-sampler. It is a rectangular net with a handle and a detachable

collection receptacle.

Everything collected in the net was transferred to a bucket where large debris was

removed and the water volume was reduced. Large invertebrates such as crayfish,

hellgrammites, and mussels were removed and processed separately. The remaining

organisms were transferred to sample bottles and preserved with buffered formalin.

22

Samples were sent to the USGS National Water Quality Laboratory where

macroinvertebrates were counted and identified.

Fish collection

Fish were collected as per NAWQA protocol for "riffle-pool streams." A

combination of electro-fishing and seining were used. In the early years of the NAWQA

program, some fish were sacrificed after collection. In recent years, fish were stunned,

caught, identified and counted, and then released.

Metrics

Before metric calculation, each macroinvertebrate species was assigned a

tolerance value (Hilsenhoff 1988, Arizona Dept. of Enironmental Quality) and functional

feeding group (ADEQ). Abundance, species richness, Simpson's Diversity Indices,

Hilsenhoff Biotic Indices, native to non-native species ratios, and species turnover rates

were calculated using Excel. All index formulas are found in the Appendix.

Regression analyses to detect relationships between flow and each of the biological

metrics were performed using JMP version 4 statistical software. Although the time it

takes for an environmental stressor to impact a biological assemblage may vary among

taxa, for the sake of consistency and because samples were collected roughly yearly, the

average flow of the year prior to each biological sample was used to test for relationships

between biological metrics and flow.

RESULTS

Historical precipitation and flow

Before biological data were analyzed, flow and precipitation were graphed to

detect trends.

---- Base Flow; ---- Strea m ---- Precipitation

CD CO CO 01 r Chi (T) c,c) cor— co min C'") 1,11 C001 CO 1.-(N (7, 511,0 (D CD COCO -

(D CD CD CD CD 1^--CO CO 00 CO CO CO CO CO COCO CO 0101 CO 0101 CO COO) 01=1=CO CO CD CD CD 01 a) CO 07 CO 0101 CO MO, CO CO CO CO (N

year

Figure 3.1. Mean annual precipitation (cm) at Tombstone and mean annualstreamflow and base flow in cubic meters per second in the San Pedro River atCharleston.

23

7.0 7

6.0

5.0

4.0

3.0 -

2.0

1.0

0.0

40 CO N- N CO In CO CO DI (N CD CO al N 14)0) CO 0) 0) N(DID (ON h-- N- COO) CO CO 0=1 OD GO CO CO CO CD 0) a) a) a) a) ar) w a)

0) Cf) 0) a) 0) CO 07 0) 0) 0) 01 07 a) co a) a) a-Jt a) al al a) a) co cr) a) 01 01 a) Cr) 0) 0) a) a)c-w N

year

Figure 3.2. Mean winter precipitation (cm) at Tombstone and mean winterstreamflow and base flow in cubic meters per second in the San Pedro River atCharleston.

U) CD N- OD 0)0) N 1,11 CD DI VI V' 411 CD CO CO C) N V- 1.10 (D 0) 0) (NCO CO CO ID ICD N- C--ID CD (00) OD (0 0) CO CO CO a) 01 a) 07 0) 01 a) 07 01 a) CI iMa) a) a) a) a) ar) a) CI) CI CI) 07 01 C!) 0) a) a) 00 0) CO 01 CO 01 07 01 al 0) a) 0) 0) 0) CI) 01 CD 0 (1)

N (N N

year

Figure 3.3. Mean summer precipitation (cm) at Tombstone and mean summerstreamflow and base flow in cubic meters per second in the San Pedro River atCharleston.

24

C) CN CO •ql- CCI 00 0) D N "4" In CD I,- CO a) SD1"-- I,- OD CO CO 00 CO CO CO CO CO CO 0)

CO 07 al 0) 0) a) al cn 0) 0) a) 0) 01 0) 07 0) 0) 07 a) a) al

N CO ,± CON-COO,C),— N0) CT) 0) Or) Cr) 07 CI Cr) 0, D D0) a) a) 0) 0) a) CD 07 0) D D .. CN CN N

yearFigure 3.4. Mean annual precipitation (cm) at Happy Jack and mean annualstreamflow and base flow in cubic meters per second in West Clear Creek nearCamp Verde.

I's A- F.-- P., F..- I--- CO CO 00 CO CD CO OD CO 00 CO CD 0) 07 0) 0) 01 07 CD 0) CD DCO 0) CO 07 CD 0) DC) DC) CD 01 0) 01 0) CS) 07 0) DC) 07 07 07 07 07 CO 07 07 07 07 Di Di

CN

yearFigure 3.5. Mean winter precipitation (cm) at Happy Jack and mean winterstreamflow and base flow in cubic meters per second in West Clear Creek nearCamp Verde.

25

7.0 —

6.0

5.0

4.0 =

3.0 —-

2.0 --

1.0 —

0.0

li1CONCOM Dv-NM"..1- C) (CINNNNN MMMMM COCOMMCOMMCOM(J) C) MMMMMMMMMMMMMMMMMMM

In CO0) 01 0) 0) 0)01 0) CD 0) 01....

CO y- rgO) CD CD CDCO CD CD CDy- CA C4 CA

year

Figure 3.6. Mean summer precipitation (cm) from Happy Jack and mean summerstreamflow and base flow in cubic meters per second in West Clear Creek nearCamp Verde.

Most likely due to the relatively short time scale of available data, no obvious

precipitation drought trends were detected. It is interesting to note, however, the similar

patterns of precipitation and flow, and the variability in affect of seasonal precipitation on

base flow and streamflow.

Water Chemistry

No statistically significant relationships were found between flow and various

water chemistry parameters, though some trends were detected. At both West Clear

Creek and the San Pedro River at Charleston, dissolved oxygen was strongly positively

associated with flow. Temperature was also somewhat dependent upon flow. At both

sites, temperature increased as flow decreased. The other water chemistry parameters

analyzed showed inconsistent relationships to flow. At the San Pedro River at Charleston,

26

9.25

6 0

Streamflow Base flow

C 9 -CD0)›.5 75-Xo> 5.25-

o

7.25

10

Base flow

R2= - 915

pH was strongly negatively related to flow and at West Clear Creek, total nitrogen and

suspended sediment decreased with flow.

Figure 4. Regression of temperature and DO with flow at the San Pedro River atCharleston.

27

R2= - 547

le 5 17 5

Base flow

ca)r _

-0_0 10_.) > . _.

.

16 5 17 5

Base flow18 5

12 2 = 635

12 2 = - 77R

12

,Streamflow6 15

28

Fig. 5. Regression of temperature and DO with flow at West Clear Creek nearCamp Verde.

Table 1. Regression table for water chemistry parameters and flow (First number =R2 ; second number = p value) at San Pedro River at Charleston.Water quality parameter Stream flow Base flowDO +.657; .190 +.729; .146Temperature -.128; .642 -.235; .515PH +.628; .208 +.726; .148Total Nitrogen +.036; .810 +.002; .958Total Phosphorous +.031; .823 +.001; .968Suspended sediment +.033; .819 +.001; .968

Table 2. Regression table for water chemistry parameters and flow (First number =R2; second number = p value) at West Clear Creek near Camp Verde.Water quality parameter Stream flow Base flowDO +.403; .365 +.635; .203Temperature -.228; .522 -.542; .263PH -.002; .951 -.086; .707Total Nitrogen -.566; .248 -.873; .066Total Phosphorous -.119; .656 -.000; .988Suspended sediment -.629; .207 -.439; .338

29

Biological assemblages

Two statistically significant relationships were found between flow and a

biological metric. At the San Pedro River, the Hilsenhoff Biotic Index and streamflow

were positively related (R 2 = .916 with a p value < .05). Pollution-tolerance level of the

macroinvertebrate community significantly decreased with streamflow, i.e more

macroinvertebrates that cannot tolerate poor water quality were present in the San Pedro

River when streamflow was higher. The percentage of the macroinvertebrates classified

as Collectors was significantly positively related to streamflow (R2 = .912; p < .05) at

West Clear Creek. The proportion of the macroinvertebrate community made up of

Collectors significantly decreased as streamflow increased. Though not statistically

significant, many other strong and therefore potentially ecologically significant,

associations were found.

Table 3. Regression table for all taxa with flow (first number = R2 ; Second number= n value) at San Pedro River at Charleston._Taxa Metric Stream flow Base flowFish Simpson's D +.123; .563 +.323; .318

Number of species -.077; .651 -.133; .547Total abundance +.031; .777 -.001; .965N:N-n ratio +.607; .120 +.009; .881Turnover rate +.275; .476 +.654; .191

Macroinvertebrates Hilsenhoff BioticIndex -.916; .043 -.065; .744Simpson's D -.103; .679 -.036; .810Number of species +.149; .614 -.277; .474Total abundance +.803; .104 +.419; .352

Periphyton Simpson's D -.087; .705 +.646; .196Number of species +.237; .513 -.375; .388Total abundance +.542; .264 +.776; .119

30

Table 4. Regression table for functional feeding group and flow (First number = R2 ;second number = p value) at San Pedro River at Charleston.Group proportion Stream flow Base flow% Collectors -.441; .336 +.159; .601% Shredders +.362; .398 -.307; .446% Scrapers +.281; .469 -.123; .649

Table 5. Regression table for all taxa with flow (first number = R2 ; second number =value) at West Clear Creek near Camo Verde.

Taxa Metric Stream flow Base flowFish Simpson's D -.347; .411 -.338; .419

Number of species -.178; .578 -.184; .571Total abundance +.062; .752 +.034; .851N:N-n ratio +.637; .202 +.864; .070Turnover rate +.913; .191 +.572; .454

Macroinvertebrates Hilsenhoff BioticIndex -.511; .285 -.813; .098Simpson's D +.137; .629 +.265; .486Number of species -.267; .484 -.447; .331Total abundance -.036; .817 +.067; .740

Periphyton Simpson's D +.465; .318 +.231; .591Number of species -.251; .499 -.058; .759Total abundance -.485; .303 -.371; .391

Table 6. Regression table for functional feeding group and flow (First number = R2 ;second number = p value at West Clear Creek near Camp Verde.Group proportion Stream flow Base flow% Collectors -.912; .045 -.544; .263% Shredders -.101; .682 .000; .998% Scrapers +.814; .098 +.394; .373

DISCUSSION

Fish

I found no statistically significant relationships between fish metrics and

streamflow or base flow, though several regressions revealed fairly strong associations.

The ratio of native to non-native fish increased with streamflow at the San Pedro River,

as well as with streamflow and base flow at West Clear Creek. Native fish make up a

larger percentage of the total fish assemblage when average yearly flow is greater.

Though the San Pedro River and West Clear Creek are perennially-flowing streams, they

do experience bursts of increased flow due to rain events. Native fish species such as

roundtail chub (Gila robusta), dace (Rhinichthys osculus, Agosia chtysogaster), and

Sonoran and desert sucker (Catostomus insignis, Catostomus clarki) are morphologically

and behaviorally adapted to survive these events. Perhaps, when insufficient water is

retained in the stream, the floods do not increase streamflow to the levels that usually

wash away non-natives fish. Non-native fish populations that are adapted to more lentic

conditions such as the green sunfish (Lepomis cyanellus), yellow and black bullhead

(Ameiurus natalis, Ameiurus melas), and smallmouth bass (Micropterus dolomieu) are

able to persist and grow when flows are not high enough to remove them.

31

15-

a)> -co 10-

R2= 607

o

a; 5-..>

coZ

610 20 25

Stre am flow

Figure 6. Regression ofNative:non-native fish with streamflow at the San Pedro at Charleston.

25

R 2 = 6'17

40

Stream7low

Figure 7. Regression of native:non-native fish with streamflow at WestClear Creek near Camp Verde.

32

35

The species turnover rate was also strongly connected to flow at both sites. It was

positively related with base flow at the San Pedro River at Charleston, as well as with

stream and base flow at West Clear Creek. The relationship is basically another reflection

of the effect of flow on the native:non-native species ratio. The turnover rate increases

with increased flow as native species are replacing non-native species.

Simpson's Diversity index was moderately sensitive to changes in flow. It was

slightly positively associated with streamflow and base flow at the San Pedro River at

Charleston, and was slightly negatively associated with streamflow and base flow at West

Clear Creek. This may be due to the magnitude of the high flow events that affect the

native and non-native fish composition. At the San Pedro River at Charleston, as flows

increase, native fish replace non-native fish and for a time diversity actually increases as

native and non-native species occupy the same area during the transitional period. Flows

at the San Pedro River during the years I studied were not quite forceful enough to

33

completely change the fish assemblage to all native fish, and thus lower diversity, as was

seen at West Clear Creek near Camp Verde.

Total abundance and species richness did not show a strong relationship to

streamflow or base flow at either site. These results show that although flow does have an

affect on species composition, it does not strongly impact the total number of species or

individuals within the range of conditions I studied.

Macroinvertebrates

The Hilsenhoff Biotic Index (HBI) was negatively related to streamflow at the

San Pedro River indicating that as flow increases, more less-tolerant organisms can

survive. This suggests that water chemistry is better with increased flow. The Hilsenhoff

Biotic Index was also strongly negatively associated with streamflow and base flow at

West Clear Creek. As streamflow changes, vital water chemistry parameters, such as

temperature and dissolved oxygen, will change. My results show that as those

correspondent water chemistry changes occur because of decreased flow, fewer

individuals of less-tolerant species survive. In fact, the entire population of a pollution-

intolerant species may disappear. My results show inconsistent relationships between

flow and species richness.

6,25

R 2 =-016

1cTo

5.75-

OE 5 5-

' 5.25-

15 20 25 30

Streamflow

34

6.25

Figure 8. Regression analysis of Hilsenhoff Biotic Index and flow at the San PedroRiver at Charleston.

66

66 R 2 =-R11

6.2

20

Base flow5

Figure 9. Regression of Hilsenhoff Biotic Index and flow at West Clear Creek nearCamp Verde.

In the context of using analysis of functional feeding groups to assess an aquatic

environment, researchers have suggested that a larger percentage of collector-gatherers

indicates impaired water quality, where as a larger percentage of scrapers and shredders

indicates a healthy stream (Rawer-Jost et al., 2000; Stepnuck et al, 2002). This is in part

due to the generalist nature of the gatherers and the specific food needs of the scrapers

and shredders. At West Clear Creek, the proportion of Collecters decreased significantly

with increased streamflow. This result supports the conclusion drawn from the Hilsenhoff

Biotic Index; water quality deteriorates at lower flows. The proportion of the community

composed of Scrapers was strongly positively associated with flow, thus supporting the

I R 2 =-cm I

210 30 40 50 70

streamflow

12°= 1114

20 30 40 50

streamflow

35

same conclusion. The percent of the community made up of Shredders was not dependent

upon stream or base flow at West Clear Creek.

50

,:33 40

90

80

30

25

520 —

1 E)' , 5 -—

a)2 5 —a)

Figure 10. Regression of functional feeding group and flow at West Clear Creeknear Camp Verde.

At the San Pedro River at Charleston, proportion of Collectors was negatively related to

stream flow, but slightly positively related to base flow. Scraper and Shredder

percentages were minimally related to flow; both positively to stream flow and negatively

to base flow. These results are fairly inconclusive. It is interesting to note, though, that

when looking at stream flow, the changes in Collector, Shredder, and Scraper populations

all reflect the same result as the Hilsenhoff Biotic Index and functional feeding group

analysis at West Clear Creek; increased flow results in better water quality. When

looking at base flow, however, macroinvertebrate populations shifted minimally in the

opposite direction as they did with streamflow. This may be an indication that the

contribution of base flow to the food source of the Collector animals is more significant

than that of streamflow, but this is not true for Scrapers and Shredders.

Total macroinvertebrate abundance was strongly positively related to flow at the

San Pedro River at Charleston, though at West Clear Creek there was minimal

36

relationship. This may be due to the geomorphology of the stream bed. At the San Pedro

River at Charleston, there is a wider, flatter flood plain, so when flow increases there is a

greater, though still quite shallow, area of substrate under water compared to the

constrained channel of West Clear Creek. This would translate into a relatively large area

for macroinvertebrates to live, and thus a greater total abundance with increasing flow at

the San Pedro River at Charleston, but not West Clear Creek.

The Simpson's Diversity Index was not strongly associated with flow at either

site. This could be due to the fact that this diversity index gives a lot of weight to the few

dominant species (Washington 1984).

Periphyton

Analysis of the periphyton community showed the biological metric most strongly

connected to flow was total abundance. At the San Pedro River at Charleston, total

abundance increased with flow, while at West Clear Creek it decreased. As was

suggested with the macroinvertebrate analysis, higher flow leads to wider stream channel

and greater wetted width at San Pedro River at Charleston. This may be the primary

reason total abundance increases with streamflow and base flow. At West Clear Creek, it

appears the total abundance also increases with flow, up to a moderate mean flow. The

sample year when mean streamflow was nearly double the next highest annual mean

flow, corresponds to the period of lower total abundance. My results show that in a low to

moderate flow regime, periphyton total abundance will increase with flow, but when

extremely high flows occur, perhaps because of scouring, less abundance is found.

122=-4R5

10

20 40 50 60

Strearnflow

CD275-

Ci

C, 250-

1-0

76 2 2 5

17:5

200

80

R2=-171

20225

Base flow175

8 275-

CO

-C3 250C

_aCO -

CO_ 225

oH 200

15

Figure 11. Regression of periphyton total abundance and flow at West Clear Creeknear Camp Verde.

Simpson's Diversity index was moderately related to base flow, though not to

streamflow, at the San Pedro River at Charleston. At West Clear Creek, Simpson's D was

slightly positively related to both streamflow and base flow. This may be due to the

greater surface area underwater resulting in availability of more diverse resources and

substrate.

Species richness was slightly positively related to streamflow at San Pedro River

at Charleston. It was slightly negatively related to base flow at the San Pedro River at

Charleston and to streamflow and base flow at West Clear Creek. These results were not

consistent enough to draw conclusions.

Analysis of the effects of streamflow and base flow on water chemistry and

biological assemblage structure has revealed some strong relationships. These results,

though not statistically significant for the most part, suggest that there are trends

occurring in the San Pedro River and West Clear Creek. The trends we have detected,

such as loss of native fish species with lower flows and worsening water quality and

concurrent loss of water quality-sensitive macroinvertebrates with lower flows, are

serious environmental concerns. Not only are they serious concerns, but also potentially

38

widespread, as these negative changes have occurred at sites that represent two different

ecoregions of Arizona. The biological integrity of arid southwest streams has been

affected by lessening of streamflow and base flow as shown by the results of this study.

In order to determine their magnitude of real-life significance, biological samples

need to continue to be collected to enlarge the database and therefore the robustness of

the analysis. Lack of statistically significant results (high R2 values, but also high p

values) supports the claim that this study needs to be long-term. With more data, analyses

such as detection for differences among years and among sensitivity of metrics and taxa

could be done. More data would also help in the fine-tuning of the process of choosing

which biological metrics to use. Not only would one be able to test for their relative

sensitivity to environmental changes, but the shortcomings in their representation of the

community (such as the Simpson's Diversity Index and its insensitivity to rare species)

could be investigated. Long-term data would also enable researchers to differentiate

between natural temporal fluctuations in the aquatic community and more serious

irreversible changes (Bunn and Davies 2000, Wood and Armitage 2004).

Other aspects of this study that could be modified to increase clarity of results

would be time (season) of sampling and number of samples throughout the year. This

would give a better understanding of the physiological and behavioral traits of aquatic

organism that may affect their population numbers at specific times.

CONCLUSION

If preserving native species and diversity of aquatic ecosystem assemblages is a

concern, humans need to be aware of how various environmental factors will impact

organisms that populate freshwater bodies. One of the most apparent environmental

variables in the flowing waters of the arid southwest is the amount of water in the stream

at any one time, i.e. flow. This environmental variable is an important one to research

because, if current predictions are correct, it has potential to change significantly in the

future. An understanding of how stream assemblages change with fluctuations in flow is

important in predicting the future of the arid Southwest's aquatic organisms.

Research to fully understand this situation needs to be long-term and on-going.

This study has shown that lower flows in West Clear Creek and the San Pedro River are

associated with negative changes in the biological assemblages. Most significantly, the

macroinvertebrate community in the San Pedro was composed of more pollution-tolerant

organisms when the year prior to sampling was of relatively lower flow. This paper

documents what would be the initial steps in an investigation into effects of increased

variability in flow on aquatic communities of arid perennial streams.

39

APPENDIX A. Metric formulas

Simpson's D (Washingtion 1986)

Where D = E 1'4E; - 1) N(N-1)

n = number of individuals in a species

N = number of individuals in a sample

Hilsenhoff Biotic Index (Washington 1986)

= E

N = number of individuals in a sample

Ili =number of individuals of species i in a sample

Q = tolerance value

Turnover rate calculated using two consecutive samples (Eby et al. 2003)

T = (C + E)/(S 1 + S2)

T = species turnover rate

C = number of species colonized (species seen in later sample that was not seen in earlier

sample)

E = number of species extirpated (species not seen in later sample that were seen in

earlier sample)

Si, S2 --- number of species present in each sample period

Native/Non-native ratio

40

D ---- Total number of native fish in sample

A ----- Total number of non-native fish in sample

41

REFERNCES

Adams, S. B., M. L. Warren Jr., and W. R. Haag. 2004. Spatial and temporal patterns infish assemblages of upper coastal plain streams, Mississippi, USA. Hydrobiologia528:45-61.

Alimov, A. F. 2001. Studies on biodiversity in the plankton, benthos, and fishcommunities, and the ecosystems of fresh water bodies differing in productivity.Biology Bulletin 28:75-83.

Bernardo, J. A., M. Ilheu, P. Matono, and A. M. Costa. 2003. Interannual variation of fishassemblage structure in a Mediterranean river: Implications of streamflow on thedominance of native or exotic species. River Research and Applications 19:521-532.

Boulton, A. J. 1999. An overview of river health assessment: philosophies, practice,problems and prognosis. Freshwater Biology 41:469-479.

Brown, J. H., T. J. Valone, and C. G. Curtin. 1997. Reorganization of an arid ecosystemin response to recent climate change. Proceedings of the National Academy ofScience USA 94:9729-9733.

Bryson, R. A. and F. K. Hare. 1974. The climates of North America. In: Bryson R.A., F.K. Hare (eds). Climates of North America. Elsevier, New York. 1-47.

Bunn, S. E. and P. M. Davies. 2000. Biological processes in running waters and theirimplications for the assessment of ecological integrity. Hydrobiologia422/423:61-70.

Chen, L.-H., K. C.-M. Chu, and Y.-W. Chiu. 2004. Impacts of natural disturbance on fishcommunities in the Tachia River, Taiwan. Hydrobiologia 522:149-164.

Clavero, M., F. Blanco-Garrido, and J. Prenda. 2004. Fish fauna in Iberian Mediterraneanriver basins: biodiversity, introduced species and damming impact. AquaticConservation-Marine and Freshwater Ecosystems 14:575-585.

Cordy, G. E., J. A. Rees, R. J. Edmonds, J. B. Gebler, L. Wirt, D. J. Gellenbeck, and D.W. Anning. 1998. Water Quality Assessment of the Central Arizona Basins,Arizona and Northern Mexico-Environmental Setting and Overview of WaterQuality. National Water Quality Assessment Program: United States GeologicalSurvey. Water Resources Investigations Report 98-4097.

42

43

Cuffney, T. F., M. E. Gurtz, and M. R. Meador. 1993. Methods for collecting benthicinvertebrate samples as part of the National Water-Quality Assessment Program:U.S. Geological Survey. Open-File Report 93-406.

Daufresne, M., M. C. Roger, H. Capra, and N. Lamouroux. 2003. Long-term changeswithin the invertebrate and fish communities of the upper Rhone River: effects ofclimatic factors. Global Change Biology 10:124-140.

Davis, N. M., V. Weaver, K. Parks, and M. J. Lydy. 2003. An assessment of waterquality, physical habitat and biological integrity of an urban stream in Wichita,Kansas, prior to restoration and improvements (Phase I). Archives ofEnvironmental Contamination and Toxicology 44:351-359.

Eby, L. A., W. F. Fagan, and W. L. Minckley. 2003. Variability and dynamics of a desertstream community. Ecological Applications 13:1566-1579.

Fenbiao, N. I., T. Cavazos, M. K. Hughes, A. C. Comrie, and G. Funkhouser. 2002. Cool-Season precipitation in the southwestern USA since AD 1000: Comparison oflinear and nonlinear techniques for reconstruction. International Journal ofClimatology 22:1645-1662.

Feyrer, F. and M. P. Healey. 2003. Fish community structure and environmentalcorrelates in the highly altered southern Sacramento-San Joaquin delta.Environmental Biology of Fishes 66:123-132.

Foggo, A., S. D. Rundle, D. T. Bilton. 2003. The net result: evaluating species richnessextrapolation techniques for littoral pond invertebrates. Freshwater Biology48:1756-1764.

Gebler, J. B. 2004. Mesoscale spatial variability of selected aquatic invertebratecommunity metrics from a minimally impaired stream segment. Journal of NorthAmerican Benthological Society 23:616-633.

Galuszka, D. A. and T. E. Kolb. 2002. Tree growth and regeneration response to climateand stream flow in a species-rich southwestern riparian forest. Western NorthAmerican Naturalist 62:266-279.

Habidija, I., I. Radanovic, B. Primc-Habidija, R. Matonickin, and M. Kucinic. 2003.River discharge regime as a factor affecting the changes in community andfunctional feeding group composition of macroinvertebrates on a cobble substratein the Sava River. Biologia 58:217-229

Hereford, R and R. Webb. 1992. Historic Variation of Warm-Season Rainfall, SouthernColorado Plateau, Southwestern USA. Climate Change 22:239-256.

Hilsenhoff, W. L. 1988. Rapid field assessment of organic pollution with a family-levelbiotic index. Journal of the North American Benthological Society 7:65-68.

Humphries, P. and D. S. Baldwin. 2003. Drought and aquatic ecosystems: anintroduction. Freshwater Biology 48:1141-1146

Karr, J. R. 1999. Defining and measuring river health. Freshwater Biology 41:221-234.

Karr, J. R. and D. R. Dudley. 1981. Ecological perspective on water quality goals.Environmental Management 5:55-68.

Kennard, M. J., A. H. Arthington, B. J. Pusey, and B. D. Harch. 2005. Are alien fish areliable indicator of river health? Freshwater Biology 50:174-193.

Lake, P. S. 2003. Ecological effects of perturbation by drought in flowing water.Freshwater Biology 48:1161-1172.

Liverman, D. M. and R. Merideth. 2002. Climate and society in the US Southwest: thecontext for a regional assessment. Climate Research 21:199-218.

Meador, M. R., T. F Cuffney, and M. E. Gurtz. 1993. Methods for sampling fishcommunities as part of the National Water-Quality Assessment Program: U.S.Geological Survey. Open-File Report 93-104.

Merritt, R. W., K. W. Cummins, M. B. Berg, J. A. Novak, M. J. Higgins, K. J. Wessell,and J. L. Lessard. 2002. Development and application of a macroinvertebratefunctional-group approach in the bioassessment of remnant river oxbows insouthwest Florida. Journal of the North American Benthological Society 21:290-310.

Moulton, S. R. II., J. G. Kennen, R. M. Goldstein, and J. A. Hambrook. 2002. RevisedProtocols for Sampling Algal, Invertebrate, and Fish Communities as Part of theNational Water Quality Assessment Program. Open-File Report 02-150. UnitedStates Geological Survey.

Nelson, S. M. and D. M. Lieberman. 2002. The influence of flow and otherenvironmental

factors on benthic invertebrates in the Sacramento River, USA. Hydrobiologia489:117-129.

Norris, R. H. and M. C. Thoms.1999. What is river health? Freshwater Biology 41:197-209.

44

45

Paperno, R. and R. B. Brodie. 2004. Effects of environmental variables upon the spatialand temporal, structure of a fish community in a small, freshwater tributary of theIndian River Lagoon, Florida. Estuarine, Coastal, and Shelf Science 61:229-241.

Pegg, M. A. and M. A. McClelland. 2004. Spatial and temporal patterns in fishcommunities along the Illinios River. Ecology of Freshwater Fish 13:125-135.

Pegg, M. A. and C. L. Pierce. 2002. Fish community structure in the Missouri and lowerYellowstone rivers in relation to flow characteristic. Hydrobiologia 479:155-167.

Porter, S. D., T. F. Cuffney, M. E. Gurtz, and M. R. Meador. 1993. Methods forcollecting algal samples as part National Water-Quality Assessment Program:U.S. Geological Survey. Open-File Report 93-409.

Rawer-Jost, C., J. Bohmer, J. Blank, and H. Rahmnann. 2000. Macroinvertebratefunctional feeding group methods in ecological assessment. Hydrobiologia422/423:225-232.

Sheppard, P. R., A. C. Comrie, G. D. Packin, K. Angersbach, and M. K.Hughes. 2002.The climate of the US Southwest. Climate Research 21:219-238.

Skirvin, S. M., S. E. Marsh, M. P. McClaran, and D. M. Meko. 2003. Climate SpatialVariability and Data Resolution in a Semi-Arid Watershed, South-EasternArizona. Journal of Arid Environments 54:667-686.

Steffy, L. Y., A. L. McGinty, C. Welty, and S. S. Kilham. 2004. Connecting groundwaterinfluxes with fish species diversity in an urbanized watershed. Journal of theAmerican Water Resources Association 1269-1275.

Stepnuck, K. F., R. L. Crunkilton, and L. Z. Wang. 2002. Impacts of urban landuse onmacroinvertebrate communities in southeastern Wisconsin streams. Journal of theAmerican Water Resources Association 38:1041-1051.

Washington, H. G. 1984. Diversity, Biotic, and Similarity Indices; A review with specialrelevance to aquatic ecosystems. Water Research 6:653-694.

Wahl, K. L. and T. L. Wahl. 1988. Effects of Regional Ground-Water Declines onStreamflows in the Oklahoma Panhandle. Symposium on Water-Use Data forWater Resources Management. American Water Resources Association, Tucson,Arizona 239-249.

Wood, P. J., M. D. Agnew, and G. E. Petts. 2000. Flow variations and macroinvertebratecommunity responses in a small groundwater-dominated stream in south-eastEngland. Hydrological Processes 14:3133-3147.

46

Wood, P. J. and P. D. Armitage. 2004. The response of the macroinvertebrate communityto low-flow variability and supra-seasonal drought within a groundwaterdominated stream. Archiv fur Hydrobiologie 161:1-20.