Printer Friendly

Assessing small streams in the upper Ocmulgee watershed using the Georgia adopt-a-stream macroinvertebrate monitoring protocols.


Using Georgia Adopt-A-Stream's (AAS) volunteer macroinvertebrate monitoring protocol, we examined how several streams' macroinvertebrate communities differed with the land usage surrounding each stream reach. Our study sites included various headwater streams and larger tributaries of the South River within the upper Ocmulgee watershed. We sampled at different locations from January 2007 through June 2007 in a parking lot, in a wetland, several forests, and suburban parks within Clayton, Henry, and Rockdale counties, including Panther Creek, Big Cotton Indian Creek, Bush Creek, Martin Creek, and an unnamed tributary of Alexander's Lake at Panola Mountain State Park. Sites in parking lots and suburban parks had macroinvertebrate communities which scored in the poor range on the AAS scale, whereas sites in forests and wetlands scored in the range considered fair or good.

Key Words: Stream Ecology, Macroinvertebrate, Impervious Surface, Georgia Adopt-A-Stream, Volunteer Monitoring, Ocmulgee, land use, disturbance


Both professional stream ecologists and volunteer stream monitors are interested in knowing what makes a stream suitable for an aquatic macroinvertebrate community. Many published studies suggest the watershed surrounding a stream is a key impacting factor, and in fact, deforestation and impervious surface can degrade a stream. Removing forest vegetation surrounding a stream decreases woody debris input, and thus reduces habitat and food available for macroinvertebrates (1). Any form of land development that involves the removal of forested areas increases the amount of impervious surface which increases runoff, peak discharge, and pollutants into streams (2). Beyond these well-established principles, there is some debate as to how much an entire watershed affects a particular stream reach, and recent studies are focusing and whether local land use makes any difference at all. Burcher and Benfield (3) sampled 3rd and 4th order streams from agricultural and recently suburbanizing watersheds, but found only very subtle differences between them for macroinvertebrate assemblages. Roy et al. (4) reported that the presence or absence of forest canopy cover at the stream reach scale had no effect upon habitat quality or macroinvertebrate richness in study sites within urban catchments. On the other hand, Schiff and Benoit found that amount of impervious surface within the 100 m buffer area within the 5 [km.sup.2] surrounding a stream, negatively influenced the stream's water quality and macroinvertebrate indices more than entire upstream watershed (5). Because some studies show that the immediate land use adjacent to a stream reach is important, whereas others suggest that that the land use over an entire watershed is more important, the importance of local land usage is under debate. Thus, in our investigation, we asked the question, "does the type of local land usage immediately adjacent to the stream affect the macroinvertebrate community?"

We examined several different streams in sites differing in levels of suburban development ranging from nearly pristine forest preserves and wetlands to suburban parks, and a site surrounded by 100% impervious surface (a parking lot). Small streams and their headwaters were chosen for this project because 80% of the stream network in North America consists of this stream order, and are generally overlooked for protection, despite their importance (6, 7). Usually, small headwater streams depend upon vegetative input such as leaves, branches, and logs that are deposited from the surrounding watershed, and these inputs are influenced by stream order (8).

Being quick and relatively inexpensive, Georgia Adopt-a-Stream's volunteer monitoring protocols were ideal for this two-semester research project (2.9, 10). The biological protocol for instance, requires volunteers to identify macroinvertebrates to taxonomic order, thus eliminating the daunting task of identification to species. Several published studies have investigated this simple, but reliable approach. Engel and Voshell (11) confirmed that volunteers in Virginias Save Our Streams program could correctly categorize an acceptable stream using coarser taxonomic levels such as order, but found that unacceptable streams are sometimes overrated, due to the way that the scores are calculated. Winn et al. (7) found that the Georgia Adopt-a-Stream (AAS) protocol was valid as an indicator of stream macroinvertebrate quality, and the use of coarse taxonomic levels involved less ecological noise. Muenz et al. (12) verified that the AAS water quality index corresponded well to professional metrics, and showed that the volunteer protocol correctly identified streams as either protected or impaired based upon the macroinvertebrates present.


Locations of study sites

All of our study sites were within 45 minutes driving distance of Clayton State University in the city of Morrow, a suburb to the south of Atlanta, Georgia (see inset of Figure 1). We specifically selected nine study sites that had safe and easy stream access (2), and were located in the upper Ocmulgee watershed. Hydrologic Unit Code 03070103, in Clayton. Henry, and Rock-dale counties. We registered all of these sites with Georgia Adopt-A-Stream, so that their locations and all data we collected could be available in the future for public use on the Internet. We classified study sites by land use type; one study site was adjacent to a parking lot, two were in suburban parks, one was downstream from a wetland, and the remaining five were in forest. As shown in Figure 1, our parking lot site (CSU) was an unnamed tributary that we considered to be the headwaters of Panther Creek, between a parking lot and a busy road on Clayton State University's campus in Morrow. Sites in suburban parks included a small unnamed tributary of Panther Creek at Indian Springs/Duffey Park in Morrow (IS) and part of Bush Creek in Gardner Park (GP), a suburban day-use park in Stockbridge. Our wetland study site (MR) was just downstream of a wetland near Maddox Road on Panther Creek in Rex. Our forest study sites included Panther Creek behind Liberty Baptist Tabernacle in Stockbridge (LB), two reaches on Big Cotton Indian Creek located on a floodplain on the grounds of Stockbridge High School (SHS) and farther downstream in at the northern most edge of J.P. Mosely Park (JP), Martin Creek located in Hidden Valley Park (HV), and a portion of Panola Mountain State Park (PM) on an unnamed tributary of Alexander's Lake, which was the only rocky-bottom stream in the study.


Stream monitoring

Macroinvertebrates include the larval stages of aquatic insects, crustaceans, mollusks. and aquatic worms, and are important to study because they serve as good indicators of the long-term physical, chemical, and biological conditions within a stream whereas the physical and chemical characteristics of a stream reach can change daily (9). Studying stream physics and chemistry properly would take many more site visits over several years, so in this paper we chose to focus mainly on our macroinvertebrate data, but we included stream flow and chemistry data in our results because the local macroinvertebrate community depends upon the abiotic conditions. As shown in Table I, we were able to visit each sire from one to three times during the Spring and Summer semesters of 2007.
Table I. Study Site Visit Schedule

Site Name                   Site Type      Dates Visited in 2007

Clayton State University  Parking Lot    Jan. 16, Mar. 24, Mar. 26

Gardner Park              Suburban Park  Jun. 14

Hidden Valley Park        Forest         Jun. 12

Indian Springs            Suburban Park  Jun. 8

JP Mosely Park            Forest         Feb. 20, Mar. 27

Liberty Baptist Church    Forest         Jan. 30, Mar. 13

Maddox Road               Wetlands       Jan. 23, Mar. 6

Panola Mountain Park      Forest         Jun. 28

Stockbridge High School   Forest         Feb. 6, Mar. 20

Under the Georgia Adopt-A-Stream protocol, which is based on sensitivities to dissolved oxygen, sensitive macroinvertebrates receive 3 points, moderately-tolerant taxa receive 2 points, and tolerant taxa receive 1 point. Thus, streams with a diverse group of sensitive macroinvertebrates score higher than streams having only a few tolerant macroinvertebrates. Equipment used for biological assessment included collapsible D-frame nets, sorting pans, forceps, and pipettes (9). All of the study sites except Panola Mountain State Park were muddy-bottom streams, therefore sampling primarily involved sampling near vegetated margins, woody debris with organic matter, and the middle of the streambed where sand, rock, and gravel accumulate. Because it was categorized as a rocky-bottom stream, sampling at Panola Mountain State Park required the use of a kick seine, which samples a 2x2 foot area. Following the AAS methodolgies, each D-frame sample covered one foot of area, sampling only those habifats that were submerged. We sorted macroinvertebrates in the field to the AAS taxonomic order and preserved them in 70% ethyl alcohol as voucher specimens. We consulted Vosheil's "A Guide to Common Freshwater Invertebrates of North America" where necessary (13). A water quality rating for each of the nine sites was calculated using GA Adopt-A-Stream's Macroinvertebrate Count Form (9).

Statistical Analysis

We used chi-square to test (a = 0.05) for difference in distributions was calculated, to test for a significant difference between the number of individual macroinvertebrates found across each of the four study site land use types. All assumptions and conditions were met in order to do this type of statistical test (i.e. data are independent of each other, data are nominal and discrete, no more than 20% of the expected values are less than five, and no expected value is less than 1 etc.).


Our data showed no clear relationship between average water quality score and average water temperature at each site (Figure 2). Sites with low average flow had poor, fair, and good water quality scores, whereas the two sites with the highest flows both scored in the good range. On the other hand, average score seemed to correlate positively with both pH and dissolved oxygen.


Interestingly, the site at Clayton State University stood out from the rest, as it had the lowest average pH, dissolved oxygen, and water quality score in the entire study (Table II).
Table II. Average water quality score, dissolved oxygen, pH,
flow, and water temperature.

Site          Water   Dissolved  pH        Flow          Water
             Quality    Oxygen            [cm.sup.3]/sec  Temperature
              Score                                     [degrees]C

Panola Mtn      22.0        6.6  6.5             0.3         21.5
State Park

JP Mosely       19.5        8.5  6.8            29.1         13.1

Stockbridge     19.5        9.1  6.8            26.7          9.8

Hidden          19.0        6.1  6.8             0.6         20.0
Valley Park

Maddox Road     17.5        8.9  6.6             6.1          8.3

Liberty         12.0        8.9  6.8             6.5          8.6

Indian           9.0        6.6  6.3             0.1         22.0

Gardner          8.0        6.9  6.5             0.5         20.5

Clayton          3.0        4.1  5.9             0.0         15.0

Average         13.6        7.5  6.5             7.0         13.8

Figure 3 shows the average Georgia Adopt-A-Stream (AAS) water quality index scores from our study sites. Streams with scores less than 11 indicate poor water quality; scores ranging from 11-16 are rated fair; and scores ranging from 1 7-22 are rated as good, while scores greater than 22 indicate excellent water quality (9).
Average GA Adopt-A-Stream Water Qualily Rating

Panola Mtn (Forest)          22
Stockbrulge HS (Forest)    19.5
JP Mosely (Forest)         19.5
Hidden Valley (Forest)       19
Maddox (Wetland)           17.5
Liberty Baptist (Forest)     12
Indian Springs (Sub Park)     9
Gardner Park (Sub Park)       8
CSU (Parking Lot)             3

Figure 3. Water Quality scores for all study sites based on
surrounding land use type (forest, wetland, suburban park,
parking lot).

Note: Table made from bar graph.

The parking lot and suburban parks have poor water quality scores, but the wetlands and all of the forested study sites range from fair to good water quality.

Tables III and IV show the observed and expected frequencies from the chi-square test of difference between distributions.
Table III. Observed chi-square values of macroinvertebrates in sites
with different land usage.

Observed   Parking  Wetlands  Suburban  Forest    Row
Values       lot                park            Totals

Sensitive        0        14         3     308     325

Moderate         7        70        14      80     171

Tolerant        15        19        22     183     239

Column          22       103        39     571     735

Table IV. Expected chi-square values of macroinvertebrates
in sites with different land usage.

Hxpected values  Parking lot  Wetlands  Suburban park   Forest

Sensitive                9.7      45.5           17.2    252.5

Moderate                 5.1      23.9           9.07    132.9

Tolerant                 7.2      33.5           12.7    185.7

                                                       Total =  735

Since there are six degrees of freedom, the critical [X.sup.2] value is 12.6 and the calculated [X.sup.2] value is 190.71, yielding a probability < 0.001, so based on the data from our samples there is a significant difference between the amounts of macroinvertebrates found across all four types of land use categories.


We conclude that there is indeed a significant difference in the groups of macroinvertebrates found at sites having different types of land usage. The site near the parking lot had the lowest score. On the day that we monitored it, there was no observed flow, yet the banks at this particular site were steep and had a lot of exposed roots, suggesting erosion. During rainfall events subsequent to our monitoring, we noticed that rain greatly swelled this small creek, and its flow rapidly decreased when the rain stopped. The sudden flow changes, combined with a ready supply of silt would explain the poor macroinvertebrate community at this study site. Unfortunately, any vehicles leaking oil, antifreeze, and other automotive chemicals could also have negatively affected the health of the stream (2).

Interestingly, the streams that were located in suburban parks also possessed depauperate macroinvertebrate communities, as evidenced by low scores. Streams in these areas lacked natural vegetation, and the adjacent areas were constantly mowed. These areas often had patchy vegetation cover (grasses) and steep banks.

The site near the wetlands and sites within forested areas had water quality scores that ranged from good to excellent. Wetlands are known for filtering pollutants so streams that flow through these areas tend to have good water quality (6). Streams that are located in forested areas with little anthropogenic disturbance also have good water quality, because forested banks with plenty of shade cover add to the amount of diverse habitats for the macroinvertebrates to live within. We observed woody debris and leaf packs in the forest streams which would presumably serve as a food source and good refugia for macroinvertebrates.

Even though the results of our study are clear, further investigation is needed to confirm our conclusions. Although there were plenty of forested sites that had good water quality, more sites are needed in disturbed areas that have poorer water quality in order to accurately assess the affects of impervious land cover on smaller streams. A larger sample size could also lead to correlation studies to find out which variables are more likely to affect stream health.

Additionally, the occurrence of an elevated AAS score just downstream of our one wetland site raises the question of whether this was due to effects of the wetland, or simply an artifact from small sample size. Lastly, although it is clear that parking lot streams are poor habitats for macroinvertebrates, a larger dataset could even address the question of how much impervious land cover near headwaters or smaller streams that an area can undergo before the health of a stream declines. A single research team would have a hard time addressing such a large question-perhaps the answer lies in a

confluence of volunteer data.


We thank the following for their cooperation and support of this project: Clayton State University, Henry County Parks and Recreation, Liberty Baptist Church, Panola Mountain State Park, and Stockbridge High School. Special thanks go to Georgia Adopt-A-Stream and the reviewers who provided


(1.) Palone RS and Todd AH (Eds): "Chesapeake Bay Riparian Handbook: A Guide for Establishing and Maintaining Riparian Forest Buffers". NA-TP-02-97. Radnor, PA: USDA Forest Service, 1998.

(2.) Georgia Department of Natural Resources: "Georgia Adopt-A-Stream: Getting to Know Your Watershed". Atlanta, GA: Georgia Adopt-A-Stream, 100 pp, 2005.

(3.) Burcher CL and Benfield EF: Physical and biological responses of streams to suburbanization of historically agricultural watersheds. J North Am Benthological Soc 25(2):356-369, 2006.

(4.) Roy AH, Faust CL, Freeman MC, and Meyer JL: Reach-scale effects of riparian forest cover on urban stream ecosystems. Can J Fisheries Aquatic Sci 62:2312-2329, 2005.

(5.) Schiff R, Benoit G: Effect of impervious cover at multiple spatial scales on coastal watershed streams. J Am Water Resources Assoc 43(3):712-730, 2007.

(6.) Meyer JL, Kaplan LA, Newbold D, Strayer DL, Woltemade CJ, Zedler JB, Beilfuss R, Carpenter Q, Semlitsch R, Watzin MC, Zedler PH: "Where Rivers are Bom: The Scientific Imperative for Defending Small Streams and Wetlands". Sierra Club and American Rivers, 2003.

(7.) Winn RT, Crisman TL, Golladay SW: Comparison of bioassessment methods of ecological condition using aquatic macroinvertebrate assemblages in Southwest Georgia headwater streams. In Proceedings of the 2005 GA Water Resources Conference (Hatcher KJ ed), Athens, GA: Institute of Ecology, UGA, 2005.

(8.) Vannote RL. Minshall GW, Cumminqs KW, Sedell JR and Cushinq CE: The river continuum concept. Can J Fisheries Aquatic Sci 37: 130-137, 1980.

(9.) Georgia Department of Natural Resources: Georgia-Adopt-A-Stream: Biological and Chemical Stream Monitoring. Atlanta, GA: Georgia Adopt-A-Stream, 66 pp, 2005.

(10.) Georgia Department of Natural Resources: Georgia Adopt-A-Stream: Visual Stream Survey. Atlanta, GA: Georgia Adopt-A-Stream, 74 pp, 2003.

(11.) Engel, SR and Voshell JR: Volunteer biological monitoring: can it accurately assess the ecological condition of streams? Am Entomologist 48(3):164-177. 2002.

(12.) Muenz TK, Golladay SW, and Veliidis G. Using Adopt-A-Stream in the coastal plain: a look at the macroinvertebrate index. In Proceedings of the 2005 GA Water Resources Conference (Hatcher KJ ed), Athens, GA: Institute oi Ecology, UGA. 2005.

(13.) Voshell. JR: A Guide to Common Freshwater Invertebrates of North America. Blacksburg. VA: McDonald & Woodward Pub. Co. 442 pp, 2002.

Anne Stahley and Christopher H. Kodani *

Department of Natural Sciences Clayton State University

Morrow, GA 30260

* Corresponding author (
COPYRIGHT 2011 Georgia Academy of Science
No portion of this article can be reproduced without the express written permission from the copyright holder.
Copyright 2011 Gale, Cengage Learning. All rights reserved.

Article Details
Printer friendly Cite/link Email Feedback
Author:Stahley, Anne; Kodani, Christopher H.
Publication:Georgia Journal of Science
Article Type:Report
Geographic Code:1USA
Date:Dec 22, 2011
Previous Article:Gas presidents comments and report from the academy council.
Next Article:Whitetail deer carrying capacity at a Georgia barrier island.

Terms of use | Privacy policy | Copyright © 2019 Farlex, Inc. | Feedback | For webmasters