Preston Ellison Department of Earth Sciences, University of South Alabama, Mobile AL.  Email: ape301@jaguar1.usouthal.edu

The foci of this study area, is the tributaries of Moore Creek within the Dog River Watershed (DRW).  The DRW drains most of the city of Mobile; in fact, 60% of the watershed lies within city limits.  Consequently, Moore Creek’s tributaries have undergone many transformations including channelization and streambed engineering.  Another tremendous influence is the increase of impermeable surfaces, such as road and parking lots.  The growing area of impermeability has led to increased runoff rates and high turbidity levels, including the siltation of many of Dog River’s Tributaries.  This research studies the problem of turbidity in conjunction with significant rainfall events.  Turbidity is a measurement of the cloudiness or murkiness of a stream due to suspended and colloidal materials.  Specifically, the central question determined how long it took certain tributaries to regain normal or “background” levels of turbidity after various intensities of rainfall.  This question is also be correlated to certain types of tributaries (natural, gabion lined, and concrete armored) and how they influence turbidity fluctuations over time.  Results show that among the three main types of drainages; turbidity reached higher levels but also regained background levels much quicker in concrete armored channels.

Keyword:  turbidity, stream engineering, Dog River



            The focus of this study area is the tributaries of Moore Creek within the Dog River Watershed (DRW). The map in Figure 1 displays the study area as well as the sampling sites. The DRW drains most of the city of Mobile; in fact, 60% of the watershed lies within city limits (ADEM 1994). Moore Creek and its tributaries, which include Montlimar Canal, is one of the larger contributing sources to Dog River. Moore Creek’s watershed, excluding the southernmost extent, drains primarily the developed areas of Mobile. Consequently, Moore Creek’s tributaries have undergone many transformations such as streambed engineering and channelization. Another tremendous influence is the increase of impermeable surfaces including roads and parking lots, which have led to increasingly high runoff rates. The changes brought about by urbanization have undoubtedly led to the overall degradation of the Moore Creek Tributary as well as many areas of the DRW.
This research is intended for the community of Mobile and anybody concerned with the health of the DRW. Its purpose serves the community in testing selected areas of Moore Creek’s watershed for turbidity after significant rainfall events. Turbidity is a measurement of the cloudiness or murkiness of a stream due to suspended and colloidal particles composed of algae, detritus (decaying organics), and sediment (Caplinger 2002).
            Many of Dog River’s tributaries are sources of sediment and remain somewhat turbid, which alone can greatly degrade the overall water quality. Turbid waters tend to be warmer as suspended particles absorb heat from sunlight, causing oxygen levels to fall. Less light penetration also results in less photosynthesis, which further decreases oxygen levels. Turbidity can not only affect plants but also can clog fish gills and prevent egg and larval development (fivecreeks.org 2004). In addition to the reduction in a stream’s health, turbid waters detract from the aesthetic qualities of our streams.
The fact that a large majority of Moore Creek and Dog River is an urban watershed has led to such problems as high turbidity that has contributed to the overall decline in water quality. It is evident that channelization and high runoff rates have increased siltation in the watershed, from the costly dredging projects in the past.  The dredging project of 2001 cost taxpayers $1,900,000 (Ford 2004).  Dredging exemplifies the consequences facing an urban watershed, although the project may have reflected a political agenda rather than a genuine effort to improve the quality of Dog River.
The Alabama Department of Environmental Management (ADEM) has established standards for overall water quality, which includes turbidity. ADEM states, “there shall be no turbidity other than natural origin” furthermore, “in no case shall turbidity exceed 50 Nephelometric Units (NTU’s) above background”. ADEM defines “background” as the natural condition of the stream without the influence of man-induced causes (McIndoe 2002). This research will therefore be useful in determining whether Moore Creek’s tributaries comply with ADEM standards.

Research Question

            The goal of this study is to research the problem of turbidity in conjunction with significant rainfall events.  Specifically, the central question is to determine how long it takes certain tributaries to regain normal or “background” levels of turbidity after various intensities of rainfall? This will also be correlated to the types of tributaries (natural, gabion lined, and concrete armored) and how they compare in turbidity fluctuations over time. Presumably, concrete armored drainages will regain background levels much quicker and have a relatively shorter and greater spike in turbidity during a rain event.
This research will add to the growing body of knowledge on the overall health of Dog River. The research could even broadly reflect the type development within the Moore Creek Watershed, from the overall amount of turbidity in the streams. This could be compared to the turbidity in Halls Mill Creek, where expanding growth has created large amounts of sediment run-off.  A good knowledge of background levels would also aid in recognizing turbidity sources whether it be poor BMP’s, failing engineering or even excessive areas of algal growth.


            In order to complete the proposed research, two major components were necessary. Proper data called for significant rainfall events of approximately .5 inches or greater within a 24/hour period.  Secondly, periods of drier weather were required to establish backgrounds. Background turbidity levels were collected at various times but were allowed at least 3 days without rainfall. Accurate precipitation events were measured for the study area using the online NOAA regional radar at the Mobile Regional Airport.           
During the precipitation event, samples were taken before, during and after at roughly 2-3 hour intervals.  Longer intervals were planned for up to 72 hours after an intense amount of precipitation, however only one significant rain event of .34 in. occurred during the study and required no additional samples.  The following sites were used in this study: Montlimar Canal (site 1) and West Eslava Creek (site 2) confluence, West Bolton Branch (site 4) at Montlimar Rd., the Spencer Branch (site 5) and Montlimar Canal (site 6) confluence and West Bolton Branch (site 7) at Azalea Rd.
Samples taken from the field were brought back to the lab at the University of South Alabama. They were measured using a standard Turbidity monitor and are represented in NTU’s.  The raw data was then combined with the timing of sampling, as well as intensity of rainfall.  This resulted in a correlation of turbidity levels versus rainfall and duration for the selected tributaries.


            Background turbidity levels were established by sampling all six sites at least 4 times during an extremely dry period.  Total precipitation for each day during the study is located in Appendix A. The four samples for each site were then calculated to give the average background for turbidity levels.  The results are listed in Table 1. Values of over 10 NTU’s can begin negatively affecting the aquatic environment within several hours. As little as five NTU’s can increase coughing rates in fish if it persists for several days (www.Waterontheweb 2004). In comparison, all the tested sites were relatively low, although site 1 may have had high enough turbidity rates to cause minor stress to the surrounding environment.

Table 1. Average background turbidity for all sites

Once background levels were established I could then determine the time it took for the tributaries to regain background levels after a rain event.  This would also allow me to verify whether the selected tributaries comply with ADEM standards.
            On April 11, Mobile received .34 in. precipitation that lasted roughly two hours.  Sampling occurred 7-8 times throughout the following 36 hours.  This rain event although under .5 in. produced sufficient data for analysis. A picture of the site and a graph displaying time vs. turbidity are as follows: Montlimar Canal (site 1), West Eslava Creek (site 2), West Bolton Branch (site 4) at Montlimar Rd., the Spencer Branch (site 5), Montlimar Canal (site 6), and West Bolton Branch (site 7) at Azalea Rd.
            Another aspect of the study was to determine if the type of drainage had any influence on the amount of time it took for turbidity to regain background levels. It was presumed that West Bolton Branch at Azalea (site 7) and West Eslava (site 2) would reach relatively quicker background levels than other sites. The reason for this assumption was due to the type of channel, both site 7 and 2 are concrete armored, which allows for increased flow rates. 
            This presumption was tested by taking samples at site 7 and West Bolton Branch at Montlimar confluence (site 4).  Site 4 is downstream from site 7, but becomes more of a natural drainage that is gabion lined and overgrown with vegetation after Azalea Rd. This allowed sampling of two different types of drainages, one concrete armored and the other gabion lined that were roughly the same distance. The sites were sampled during the same rain event of April 11. Site 7 resulted in reaching background levels within 10 hours of initial precipitation, while site 4 did not regain background for another 26 hours.
            The same test was applied to W. Eslava (site 2) and Montlimar Canal (site 1). The drainage for site 2 is mostly concrete armored and the drainage for site 1 is a straight channel with dense vegetation along the banks for most of its length. This again gives turbidity information on two different types of drainages that are roughly the same length. Background levels at site 1 were only similar at 36 hours after the initial precipitation, while Site 2 reached its background in roughly 8 hours.
            The other proposed question was whether concrete armored drainages recorded a greater spike in turbidity through the duration of the rain event.  To test this, sites 7&4 and sites 2&1 were again compared.  Samples were taken at equal intervals and resulted in a far greater peak in NTU’s among the concrete armored drainages. Figure 6 shows that site 1 (natural drainage) reached a gradual peak of 11 NTU’s within roughly 9 hours, while site 2 (concrete armored) quickly peaked at 80 NTU’s in just over 2 hours. Testing sites 4 and 7 resulted in a similar trend.  Figure 7 shows site 7 (concrete armored) had a rapid peak and decline in turbidity downstream at site 4 (gabion lined), turbidity was relatively lower and declined gradually.  Testing both sites proved the presumption that the concrete armored drainages regain background levels quicker and have a relatively greater spike than more natural drainages.  The reason for the difference in times between the two drainages is a result from different discharge rates.  According to (Jones 2004), concrete channels have increased stream velocity compared to vegetative drainages, which act to retard water flow and can be explained as the difference in the roughness coefficient


            One question that was proposed from this study was whether the selected tributaries complied with ADEM regulations; which was not to exceed 50 NTU’s above background.  The answer is yes, however ADEM requires sites to be sampled 72 hours after a rain event.  This does not effectively address the issue of reducing the overall human induced turbidity.  This is where regulations meet reality due to the engineering marvels, which allow drainages to rapidly move run-off downstream.  This is the case for sites 2 and 7 that potentially could release exceeding amounts of sediment but still comply with ADEM regulations.  Evidence of this was observed during the study when site 2 went from a background of 2.6 to 80 NTU’s within about 2 hours and then dropped off to 4.2 NTU’s in under 5 hours.  Data from site 6 revealed that the spike in turbidity at site 2 had little overall influence downstream.  However, keep in mind that data was collected during a .34 in. rain event that lasted 2 hours and effects could be exponential given a characteristic Mobile deluge of 2+ inches. Continuing research could build on this idea by studying the effects greater amounts of precipitation have on varying drainages.


            The differences in drainages observed in this study point to one obvious conclusion; drainages with vegetation have less turbidity for longer durations and have less dramatic reactions to a rain event. “Natural-like” streams such as Montlimar Canal that have dense vegetation along the banks act as buffers, which filter sediment entering streams. Even overgrown gabion-lined drainages such as site 4 acted to slow and reduce turbidity. However, in one aspect these drainages may not necessarily be better for the aquatic environment. Drainages that release somewhat turbid waters for several hours may impede water quality as much as drainages that spike high turbidity in short durations.
The problems and transformations facing an urban watershed like Dog River are many, and some are ultimately unavoidable.  However, restoration of a stream’s health can be attained if proper policies and standards are met. Further, some regulations may need to be addressed again by setting standards for specific tributaries based on discharge rates.  Researching the specific problem of increased turbidity due to rainfall events could provide standards for various drainages, which could aid in pinpointing siltation problems.

Cited Sources

ADEM.  Alabama Department of Environmental Management (1994) A Survey of the Dog River Watershed Mobile.


Caplinger, Jessica A. and David N. Mott  Water Quality Assessment of Springs and Perennial Streams within the watershed of Buffalo National River.  National Park Service Sep. 10, 2002.


Ford, David Wesley.  Dredging and Dog River Department of Earth Sciences, University South Alabama.  http://www.usouthal.edu/geography/fearn/480page/dogriver.html Accessed Apr. 13 2004.


Jones, R. Phillip.  Arguments Against Concrete Armored Channels in Dog River Watershed.  Department of Earth Sciences, University of South Alabama. Accessed Apr. 13 2004.  http://www.usouthal.edu/geography/fearn/480page/dogriver.html


McIndoe, James, M. Alabama Department of Environmental Management Water Division-Water Quality Program Chapter 335-6-10 Water Quality Program Sep 2002.


www.fivecreeks.org/monitor/turbidity.html  Water Quality Monitoring:  Turbidity 2004.


www.waterontheweb.org/under/waterquality/turbidity.html  Turbidity 2004.