Steven G. Summersell, Department of Earth Sciences, University of South Alabama, Mobile, AL 36688. Email: 


A physical change has occurred within the Dog River watershed in Spring Creek, a tributary to Halls Mill Creek and finally to Dog River. A bank stabilization project has been completed on the lower portion of Spring Creek just above a pristine, bottomland forest at the confluence with Halls Mill Creek. This study will record the physical characteristics of the construction area before, during, and after the stabilization project. This study will also aid in identifying on going watershed degradation and provide base-line data for future studies. In-situ data was collected using a water quality data sonde, which records the following characteristics: salinity, conductivity, pH, dissolved oxygen (mg/L), temperature and water depth.  Turbidity samples were collected and later analyzed using a turbidity meter. Content analysis of the data shows the level of significance in changes to the construction area’s physical characteristics over the length of the construction project. Data presentation is in the form of graphs and tables. Turbidity levels were documented that, at times, exceeded state regulations. Water quality characteristics varied with seasonal change. Evaluation of the overall impact to Spring Creek and the Dog River watershed is difficult to define and is compounded by a separate construction project located downstream of the study area.


Keyword: Channelization, Spring Creek, Dog River Watershed.




Spring Creek is a tributary to Halls Mill Creek within the Dog River watershed, located in Mobile, Alabama.  In mid February 2002 the lower portion of Spring Creek underwent physical modifications. Thirty thousand cubic yards of sediment were removed from Spring Creek and gabions were installed for bank stabilization. This project extended from US Highway 90 to south of Halls Mill Road and upstream of the confluence of Spring and Halls Mill Creeks. The confluence area is within the largest remaining bottomland forest in the Dog River watershed.  A study of the modifications, before, during and after is necessary to assess the overall health of the sub-watershed and any effect modification might have on Spring Creek, Halls Mill Creek, and Dog River.

            In 1994 and 1995 the Alabama Department of Environmental Management (ADEM) conducted two studies on the Dog River Watershed.  These studies identified problems and documented the water quality of the Dog River Watershed (DRW), yet Spring Creek had only a vague reference in the second study (Alabama Department of Environmental Management 1995). With the lack of prior documentation and an impending construction project, a study concerning the Spring Creek sub-watershed was of the utmost importance.

By assessing Spring Creek’s physical condition, I have acquired baseline data that will be useful for future studies and have documented the effect of physical changes to Spring Creek and their relation to the Dog River watershed.


Research Question


Did the modification project to the lower portion of Spring Creek have a significant impact on water quality of Spring Creek itself and subsequently Halls Mill Creek and Dog River? I expect that clear cutting of the riparian zone will cause thermal heating resulting in an increase of water temperature and lower dissolved oxygen. I also expect that significant amounts of sediment were deposited downstream of the construction site.




A qualitative and quantitative assessment of Spring Creek was necessary to determine any impacts from the construction project. The qualitative portion consisted of documenting land use within the sub-watershed from its origin, just north of Cottage Hill Road, to its confluence with Halls Mill Creek. Field determinations will quantify Spring Creek’s physical characteristics.

            The qualitative assessment included a watershed reconnaissance, using topographic and street maps.  Reconnaissance began with identifying Spring Creeks sub-watershed  by determining high elevations that separate Spring Creek from other creeks using a topographic map.  Street maps were used to map out a reconnaissance route.  Photographs and detailed field notes recorded initial observations. 

Next, a habitat assessment and physical characterization was conducted in accordance with the Alabama Department of Environmental Management’s Field Office Annual Stream Sampling Quality Assurance Quality Control Training Workshop Manual on the proposed construction site and the area upstream of it. Observations made include the following: water quality indicators, flow conditions, riparian land use, and vegetation. It appeared that both areas were previously channelized and allowed to stabilized as evidenced by the presence of fish, an indication of water quality (Cole 1983). It was also observed that there was abundant canopy cover in the riparian zone for the entire length of the creek.

            The study’s quantitative portion took explicit measurements in a controlled setting (the construction zone).  During this phase, data was collected to address the research question in the form of a time series experiment, “data collected on the same element for the same variable at different points in time” (Mann 2001).  Two sampling points were established and are discussed in the results.

            In-situ data have been collected using a Hydrolab Surveyor 2/H2O water quality data sonde, which records the following parameters: salinity, conductivity, pH, dissolved oxygen (mg/L), temperature and water depth (meters). The data sonde was calibrated prior to data collection and post-calibrated after data collection to assess the meter’s performance and accuracy. Air temperature was collected using a Fisher Celsius thermometer, placed in the shade before reading. The performance of both the data sonde and the air thermometer were determined to be within acceptable limits throughout the study. Turbidity samples were collected in an individual ¼ gallon plastic containers, preserved on ice at 4°C, and analyzed in the ADEM Field Operations office Laboratory on a Hach Ratio Turbidmeter (Model 18900-00), reported in Nephelometric (NTU) units. The performance of the turbidity meter was determined to be within acceptable limits by the ADEM Laboratory’s internal quality control evaluation. 

            The parameters that were evaluated are key to evaluating water quality and were collected by the data sonde in the following ways: Salinity was determined by the meter after calibration against a standards solution of standard seawater. This parameter is a measure of the mass of dissolved salt content of water. Conductivity (specific conductance) was also determined by the meter using the same standards solution to calibrate for salinity. Conductivity is a measure of the ability to carry an electric current which is greatest in the presence of inorganic compounds and less with organic compounds (Clesceri, Greenberg, and Eaton  1998).

            The data sonde was also used to determine pH. The meter was calibrated against buffer solutions with a pH of 7.00 and 10.00 (pH 4.00 was read at both pre and post calibrations but not calibrated). The best pH range for aquatic life is 6.5  to 8.5 and less than 4.0 or more than 10.0 are considered lethal (Alabama Water Watch 2003). 

            Dissolved Oxygen (DO) was collected by the data sonde using the electronic method. Membrane electrodes were calibrated using water temperature and barometric pressure to determine a DO saturation percentage (DO %), then calibrated to 100% and reported in mg/L. Aquatic animals and plants require oxygen which enters water from the atmosphere. Streams with adequate shade and that are free of pollution in the form of organic matter have high DO levels and support aquatic life (Alabama Water Watch 2003).

            Turbidity is a measure of water clarity, which is impacted by suspended and colloidal matter. Clay, silt, and sand were the greatest cause of high turbidity levels in Spring Creek due to the removal of sediment and channelization. Turbidity was collected as a grab sample that was preserved on ice and later read with a Hach Ratio Turbidmeter. The meter utilized the Nephelometric Method in which the intensity of light scattered, as compared to a standard, will be increased as suspended materials reflecting or absorbing light are also increased. This intensity is recorded in NTU’s or Nephelometric Turbidity Units (Clesceri, Greenberg, and Eaton  1998).

            A content analysis of the data collected looks for parameter levels that vary from the norm (parameters collected before the construction project), and how parameters after the project compare to the norm to identify any degradation.  Data are presented in the form of graphs to assist in interpreting changes in physical characteristics (Ott 1995).



            During the reconnaissance of Spring Creek, I determined that it has been channelized for some time from its origin to just south of Halls Mill Road (See Figures 1, 2, and 3).  The land use is mostly residential housing and storm water drainage. Though stabilized, concrete lining is failing in some areas and re-stabilization is in Spring Creek’s future.

The construction project was titled “Spring Creek Phase 1B” and encompassed the removal of 30,000 cubic yards of sediment from Spring Creek and the construction of stream bank stabilizers (gabions). The total disturbed area was 4.9 acres and best management practices (BMPs) included a large sediment trap, silt fences, hay bales, grass matting, hydroseeding of slopes, and a total bypass (pumping around the site) of the creek during sediment removal and gabion placement. The sediment trap is a permanent fixture and will allow for the removal of future construction sediment. Routine maintenance will also allow for the removal of garbage from the creek.

Sampling site selection was decided on by allowing for an upstream (of construction site) station designated SC-1 and a station down stream of the sediment trap, SC-2b. Figure 4 depicts Spring Creek’s location and sampling points. SC-2a is a downstream sampling station at the Halls Mill Road bridge were Alabama Water Watch has an existing sampling point. This was a very useful location because SC-2b was at times an inaccessible construction area.

For safety reasons, I established an alternate upstream sampling point at Maudelayne Drive (SC-1alt). This was due to interruptions from persons in the area as well as dangerous surface conditions at SC-1. SC-1 is located in the back parking lot of the shopping center on Demetropolis Rd. and Girby, a parking lot drainage ditch leads directly to the creek. Moving upstream of the storage facility’s parking lot drainage outfall pipe, a large pool is the sampling point. During high rain events, this access becomes impassable and SC-1alt was used.  Table 1 contains all characteristics evaluated for the length of the project.

It should be noted that after the establishment of these upstream sampling points, I discovered an unnamed tributary that enters the creek between SC-1 and SC-1alt. On April 14th, 2003, I investigated the unnamed tributary and determined that its source is natural springs that originate just south of Gordon Oaks Retirement Home on Knollwood Rd (Figure 5). Along its course, the tributary is stable and has no significant impact on SC-1.

SC-1 was first evaluated on January 27, 2002. Field parameters and a turbidity sample were collected along with an abbreviated flow measurement using a Pygmy flow meter and staff gauge, I found no detectable flow (Alabama Department of Environmental Management 2001). The lack of flow was due to a beaver dam located south of the Halls Mill Creek Bridge as observed during evaluation of SC-2.

Both sites were revisited on February 6, 2002 during a rain event. Turbidity at SC-1 was 13.0 NTU and 7.2 NTU at SC-2. Construction began on the down stream portion of Spring Creek (SC-2) on February 18, 2002 and a rain event on February 20 yielded turbidities of 29 NTU at SC-1 and 44 NTU at SC-2. Turbidity limits as set forth by the Alabama Department of Environmental Management are 50 NTUs above background for an Alabama water body with a Fish and Wildlife classification (Alabama Department of Environmental Management 2001). One sample is not enough to establish a proper background, however, it was collected on a rain free day and prior to construction. Therefore, for the purpose of this study, the turbidity limits should be set at 63.0. These limits were exceeded 10 times during the study.

            Other physical characteristics limits as set forth by ADEM for water bodies with Fish and Wildlife designations are: pH-range of 6.0 to 8.5 or one unit from the normal or natural pH, water temperature- 90°F maximum (there are no industrial thermal discharges to Spring Creek), and Dissolved oxygen (DO)-5 mg/l minimum (Alabama Department of Environmental Management 2001). At no time were any of these criteria exceeded during the study.

            The importance of turbidity levels in Spring Creek relates construction work performed and the transportation of material as suspended solids.  Turbidity measures the “cloudiness caused by suspended matter” and is “caused by soil erosion and runoff” (Alabama Water Watch 2003). As construction activities clear-cut riparian zone vegetation, the exposed stream banks were allowed to erode into the stream (Figures 6a before and 6b during construction). Also, as equipment entered the creek, the bed load was affected and caused further suspension of sediment. Given these causes, turbidity levels during the construction phase did exceed their limits. To gauge the effectiveness of the construction company’s BMPs to keep erosion on site, portions of the creek must be investigated for deposition of sediment.

On March 23, 2002, I investigated a portion of Spring Creek located between the construction site and the confluence area. Entering the woods just south of Negus Marine on Demotropolis Road, I observed an old garbage dump. I then traveled east along an old survey line. The terrain was of a marshy, delta like area with multiple, small channels.  After 40 minutes time, the terrain and thick vegetation had become impassible. The channels had no sign of deposition and were observed to have detritus (organic matter). This area is dangerous. At times, I sank in boggy areas approximately 3 feet deep. However, deposition will reach this area before the confluence and this is the best approach via land, as other areas would require a boat. Future studies should use this site as a benchmark to judge deposition.

            The main BMP for the construction project was the sediment trap at the downstream construction point (Figure 7). This trap is a large basin that “slows water velocities, thereby allowing soil particles to settle out” (Fifield and Harding 1992). Though other BMPs were in place during the project, the trap collects the larger particles of sand, however, silt and clays are allowed to travel downstream into Halls Mill Creek and ultimately Dog River.  Over time and as future upstream bank stabilization projects are carried out, this trap will collect sediment and is designed to be a removal point with routine maintenance. Future studies may consider gauging the trap depth to test for sediment accumulations. The trap is unwadable and a small boat would have to be employed.

To further investigate deposition, the confluence of Spring Creek and Halls Mill Creek was reached via a small boat on April 5, 2003. The GPS coordinates were recorded as 30.60605 degrees North, 88.15131 degrees West. The confluence is a fanned out delta like area with no definite channel. No deposition was observed. However, I noticed a sand slug on Halls Mill Creek with sand similar to that found near Demetropolis Road.

With removal of the riparian zone vegetation and loss of cover, water temperature should reach a level equivalent to that of the ambient air temperature resulting in lowered Dissolved Oxygen (DO) rate exclusive from seasonal change (Alabama Water Watch 2003). Table 2 shows the relation of air and water temperature along with dissolved oxygen observed for this study. The average air temperature, water temperature, and DO before the project was air: 13, water: 16, and DO: 7.9. After the project (for the same month last year) air: 20, water: 20, and DO: 9.5. Note that the average air and water temperature is the same, 20 degrees, after construction. While this cannot be the final comparison, it is intriguing that after such a construction project, when air and water temperatures begin to mirror one another and DO should be lower, DO increased.




            Judging by the characteristics evaluated during this study, it could be inferred that the significance of modifications made in Spring Creek is relatively low immediately after completion of the construction project. However, lack of significance does not equal lack of importance. These established sampling points would need to be continually evaluated to determine the overall effect on Spring Creek.  More work needs to be done to answer questions discovered in this study. For example, there needs to be data collected to explain occasions of lower turbidity and higher DO at the downstream locations than upstream. Future studies will provide data to help answer new questions raised and further catalog the effects of channelization in Spring Creek.




Alabama Department of Environmental Management (1994). A Survey of the Dog River Watershed: An overview of land use practices and an assessment of the effects of development on the natural resources of the basin. (ADEM Technical Report) Mobile Branch, Mobile, Alabama.


Alabama Department of Environmental Management (1995). A Survey of the Dog River Watershed, Second year’s findings: A review of ongoing development in the watershed and assessment of the effects of urban non-point sources on the aquatic resources of the basin. (ADEM Technical Report) Mobile Branch, Mobile, Alabama.


Alabama Department of Environmental Management (2001). Methodology for coastal watershed assessments. (ADEM Technical Report) Mobile Branch, Mobile, Alabama.


Alabama Water Watch (2003) Basic certification workbook water quality monitoring. Auburn, AL: Alabama Water Watch Program Office


Clesceri, L.S., Greenberg, A.E., and Eaton, A.D. (Eds.) . (1998) Standard Methods for the Examination of Water and Wastewater (20th ed.) Washington, DC: American Public Health Association


Cole, Gerald A. (1983) Textbook of Limnology (3rd ed.) St. Louis: C.V. Mosby Company


Fifield, Jerald S., Harding, Michael V. (1992) Effective erosion and sediment control. Alabama Technology Transfer Program, Auburn, Alabama


Mann, Prem S. (2001). Introductory Statistics (4th ed.) New Jersey: John Wiley and Sons, Inc.


Ott, Wayne R. (1995). Environmental Statistics and Data Analysis. Florida: Lewis Publishers.


Rosgen, Dave (1996). Applied River Morphology. Minnesota: Printed Media Companies.