Factors Affecting Sediment Oxygen Demand Dynamics in Blackwater Streams of Georgia’s Coastal Plain

Sediment Oxygen Demand (SOD) is an important process affecting dissolved oxygen concentrations in blackwater streams of the southeastern coastal plain. Because very few data on SOD are available, it is common for modelers to take SOD values from the literature for use with dissolved oxygen (DO) models. In this study, SOD was measured in seven blackwater streams of the Suwannee River Basin within the Georgia Coastal Plain for between August 2004 and April 2005. SOD was measured using four in-situ chambers, and was found to vary on average between 0.1-2.3 g O2m-2day-1 across the seven study sites throughout the study period. SOD was found to vary significantly between the watersheds within the Suwannee River Basin. However, land use was not found to be the driving force behind SOD values. Statistical analyses did find significant interaction between land use and watersheds suggesting that an intrinsically different factor in each of the watersheds may be affecting SOD and the low DO concentrations. Further research is needed to identify the factors driving SOD dynamics in the blackwater streams of Georgia’s Coastal Plain.

Methods
Figure 1. Map showing the location of the study sites within the Suwannee River basin. Watershed 1 is an agricultural study site while watersheds 2 and 3 are forested study sites within the Alapaha River 8-digit HUC. Watershed 4 is the agricultural site and watershed 5 is the forested site within the Little River HUC. Watersheds 6 and 7 are the forested study sites within the Upper Suwannee River HUC.

Figure 1. Map showing the location of the study sites within the Suwannee River basin. Watershed 1 is an agricultural study site while watersheds 2 and 3 are forested study sites within the Alapaha River 8-digit HUC. Watershed 4 is the agricultural site and watershed 5 is the forested site within the Little River HUC. Watersheds 6 and 7 are the forested study sites within the Upper Suwannee River HUC.

Within Georgia, the Suwannee River Basin contains four 8-digit HUCs (United States Geological Survey Hydrologic Unit Code): the Withlocoochee, Little, Alapaha, and Upper Suwannee (Figure 1).

Twenty potential study sites were identified within these HUCs. Of these twenty, seven sites were selected for the study. Three sites were located within the Alapaha River HUC, two within the Little River HUC, and two within the Upper Suwannee River HUC (Figure 1). Two of the seven study sites were selected to be in watersheds where 50% or more of the land use is agriculture. The other five study sites were selected to be in watersheds that have greater than 50% forested land use.

The SOD chambers used in the study were designed by Murphy and Hicks (1986) and were on loan from Georgia EPD (Figures 2 and 3).

Figure 2. Photograph of an SOD chamber used in this study showing the main components of the system used to measure SOD rates.

Figure 2. Photograph of an SOD chamber used in this study showing the main components of the system used to measure SOD rates.

Three chambers were used at each site to measure oxygen depletion in the sediment matrix while a fourth chamber was used as a control and measured oxygen depletion in the water column. All chambers remained in the stream for 3 hours. Throughout the study the SOD chambers were deployed between 11:00 AM and 12:00 PM and were removed from the stream between 2:00 and 3:00 PM. The study sites were visited two to four times during the study period from July, 2004 to April, 2005. However, we were not able to visit the sites during August and September 2004 and again in March and early April 2005 due to high water levels. Between the months of October and March all sites were visited monthly, and measurements were attempted at all sites before the next rotation began. More information about the SOD chambers and how SOD is calculated is given in the section titled Instream Swamps and their Effect on Dissolved Oxygen Dynamics.

Figure 3. Photograph of the underside of an SOD chamber showing the two diffusers, the lower flange and cutting edge used to ensure proper installation, and the pipe welded to the inside of the lid which converts the inside of the chamber into an annulus.

Figure 3. Photograph of the underside of an SOD chamber showing the two diffusers, the lower flange and cutting edge used to ensure proper installation, and the pipe welded to the inside of the lid which converts the inside of the chamber into an annulus.

Five cm diameter sediment cores were collected from the top 13 cm of the sediment at each sampling site. A particle size distribution analysis was completed using the hydrometer method as described by Gee (1986). Organic matter in each sample was destroyed using hydrogen peroxide; the percent organic matter was calculated by the difference in the dried sample before and after the hydrogen peroxide treatment. All large debris (sticks) was removed from the sample before mass was recorded and the procedure begun.

Results

Temperature-corrected SOD rates varied on average between 0.1 and 2.3 g O2 m-2 day-1 for the seven study sites (Figure 4).

Figure 4. Average SOD values and error bars for each study site and land use. ARag is the Alapaha River agricultural watershed. LRag is the Little River agricultural watershed. ARfr1 and ARfr2 are the Alapaha River forested watersheds. LRfr is the Little River forested watershed. USfr1 and USfr2 are the Upper Suwannee forested watersheds. AGavg and FORavg are means of all the individual SOD rates measured for the respective land uses.

Figure 4. Average SOD values and error bars for each study site and land use. ARag is the Alapaha River agricultural watershed. LRag is the Little River agricultural watershed. ARfr1 and ARfr2 are the Alapaha River forested watersheds. LRfr is the Little River forested watershed. USfr1 and USfr2 are the Upper Suwannee forested watersheds. AGavg and FORavg are means of all the individual SOD rates measured for the respective land uses.

However, this study showed some unexpected trends. For example, it was expected that on average, agricultural watersheds would have lower SOD values than forested watersheds due to lower rates of allocthonous organic matter. However, we found that in the Alapaha River watershed, the agricultural sites had higher SOD rates than the forested sites. This may be a consequence of higher levels of nutrients, erosion from farming operations, and legacy effects after decades of anthropogenic interference. In contrast, forested sites produced higher rates in the Little River watershed. Surprisingly, we found that when SOD rates were different between watersheds (Figure 5), The Upper Suwannee, a watershed where the majority of the land is densely forested, had the highest average SOD values.

Organic matter was found to be less than 2% of benthic sediments for all experimental sites and negligible at many sites. However, it is possible that a series of hurricanes that passed through our study area during the autumn of 2004 may have flushed much of the benthic organic matter from the tributaries we were studying. In this study, coarse organic debris was removed from the sample before the peroxide treatment of the sediment sample. Therefore, debris such as small sticks or leaves was not included in the mass and % organic matter calculations and may have contributed to low calculated values. This coarse debris may have been contributing to SOD through the mineralization process; however it is more likely that the organic debris was of significance to SOD by supplying a food source to microbes in the sediment. Their respiration would be the largest addition to the cumulative SOD.

Figure 5. Statistical comparison of HUC SOD rates in which Alapaha SOD rates were significantly lower than Upper Suwannee SOD rates. Letters indicate statistically significant differences.

Figure 5. Statistical comparison of HUC SOD rates in which Alapaha SOD rates were significantly lower than Upper Suwannee SOD rates. Letters indicate statistically significant differences.

The highest SOD rates and organic matter concentrations were measured near the Okefenokee Swamp – an area dominated by dense forests, frequently flooded land, and swamps. These features could be the driving forces for higher SOD rates in the Upper Suwannee River HUC and provide further evidence that streams in forested watersheds with fewer anthropogenic impacts may in fact have higher SOD rates.


Utley, Barbra C., George Vellidis, Richard Lowrance, and Matt C. Smith, 2008. Factors Affecting Sediment Oxygen Demand Dynamics in Blackwater Streams of Georgia’s Coastal Plain. Journal of the American Water Resources Association (JAWRA) 44(3):742-753.  Download .pdf

Download Barbra Utley’s M.S. Thesis

Barbra Utley, Ph.D. student, Biological Systems Engineering Department, Virginia Tech
George Vellidis, Professor; Biological and Agricultural Engineering, University of Georgia
Richard Lowrance, Ecologist, Southeast Watershed Research Laboratory, USDA-ARS, Tifton, Georgia
Matt Smith, Research Leader, USDA-ARS Animal Manure and By-Products Laboratory, Beltsville, Maryland