Water quality is a term used to describe the condition or environmental health of a water body or resource. It is defined in the Clean Water Act as the standard of purity that is necessary for the protection of fish, shellfish and wildlife populations in the aquatic environment, and for recreational uses in and on the water. It is the shared responsibility of the United States Environmental Protection Agency (USEPA), states, and local governments to meet the goals of the Clean Water Act. Each state must ensure that its waters support beneficial uses that are important to its citizens, such as the primary contact recreation of swimming and fishing. The states also establish water quality criteria, which are levels of physical, chemical and biological variables that are required to meet beneficial uses (e.g. drinking water, recreational use, and support of aquatic life). Physical and chemical standards are set for maximum acceptable concentrations of pollutants, acceptable ranges for physical variables, and minimum and maximum values of other water quality parameters. Numeric biological criteria describe expected attainable community attributes and establish criteria based on measures such as species richness, presence or absence of indicator taxa, and distribution of classes of organisms. Narrative water quality criteria define, rather than quantify, conditions and attainable goals that must be maintained to support a designated use. These criteria establish a positive statement about aquatic characteristics expected to occur in a water body, and they may also describe conditions that are desired in a water body (National Research Council 1993).
Water quality monitoring involves taking measurements that provide information on conditions and allow scientists and managers to estimate trends. Monitoring provides the information needed for an assessment of the conditions of the water in relation to natural variability, human effects and intended uses (Chapman 1992). Although an assessment is a cumulative evaluation of overall system conditions, it is difficult to measure all the physical, chemical and biological properties of a water body. Instead, a few variables that provide general indications of environmental conditions are selected (Robertson and Davis 1993).
Many water quality variables are subject to large fluctuations in space and time. Understanding these fluctuations in the physical environment and determining whether such changes are natural or a result of anthropogenic influences can be a difficult problem. An ideal variable provides unambiguous information about the condition of the environment in relation to reference conditions and is relatively easy and inexpensive to measure (Wenner and Geist unpublished).
Variables being monitored in the ACE Basin by both the National Estuarine Research Reserve (NERR) and South Carolina Department of Health and Environmental Control (SCDHEC) programs include pH, conductivity (salinity), temperature, dissolved oxygen (DO), turbidity and water level. The importance of these variables as indicators of water quality has been well-documented (Chapman 1992). For example, dissolved oxygen levels are strongly influenced by point source discharges and for that reason the U.S. Environmental Protection Agency (EPA) uses DO in preliminary evaluation of in-stream water quality. Due to its sensitivity as an indicator of estuarine water quality, dissolved oxygen may provide reliable assessments of the efficacy of management efforts to control non-point source discharges and improve habitat conditions. Water quality conditions can deteriorate rapidly in response to non-point source pollutants that are often pulsed through estuarine systems by short-term episodic storm events. In urban areas, pollutants drain off of hard surfaces such as parking lots and streets, flow into storm drains and drainage ditches, and then into creeks and rivers. Pollutants from storm water runoff and point-source wastewater discharges contain organic materials and nutrients that contribute to consumption of dissolved oxygen. Dissolved oxygen concentrations below 4 milligrams per liter (mg/l) are considered to be unhealthy for many aquatic community inhabitants. When the level of dissolved oxygen falls to 2 mg/l, severe physiological stress to marine organisms occurs and death may result. Fish and other estuarine organisms also react poorly to rapid changes in dissolved oxygen levels. For these reasons, the amount of dissolved oxygen in water is a good indicator of its "quality". Some habitats such as tidal creeks undergo regular periods of low dissolved oxygen (<2 mg/l) but these instances are usually short-lived (2-3 hours) and are part of the natural fluctuation in dissolved oxygen.
The oxygen content of estuarine waters varies with temperature, salinity, turbulence, atmospheric pressure and the photosynthetic activity of algae and submerged plants (Chapman 1992). Temperature affects the amount of dissolved oxygen that water can hold. Solubility of gases such as oxygen, carbon dioxide, and nitrogen decreases as the temperature rises (See gas solubility ). Temperature also controls the rate at which planktonic organisms use oxygen. In some climates, the amount of dissolved oxygen in summer can be half of that found during the winter , but this is not solely caused by temperature and can be related to higher organic matter concentrations and increased biological activity. Temperature affects the rate of chemical reactions and increases evaporation and volatilization of substances. The metabolic rate of estuarine organisms is related to temperature so that in warm waters, respiration rates increase leading to increased rates of oxygen consumption and decomposition of organic matter. Growth rates of bacteria and phytoplankton increase in warmer temperatures, and can contribute to increased water turbidity, macrophytic growth and algal blooms (Chapman 1992). (See related section: Decomposers.)
Salinity or the total quantity of dissolved salts in water is a useful indicator of estuarine hydrography and habitat potential. It provides a direct measure of the relative influence of the sea and freshwater sources in an estuary. Salinity affects the distribution, abundance and composition of biological resources. A common misinterpretation is that much is known about the salinity structure of the nation's estuaries. In fact, data on salinity exist for only the most studied systems, while salinity distributions and variation in most estuaries have not been sampled (Orlando et al. 1994).
Conductivity is used to determine salinity and is actually a measure of the ability of water to conduct an electrical current. Conductivity is proportional to the concentrations of total dissolved solids and major ions, and its measurement is influenced by the amount of electrical charge on each ion, ion mobility and temperature of the water (Chapman 1992). Conductivity has proven to be useful in determining the extent of influence of run-off and effluent discharges in aquatic systems.
The pH of most healthy estuaries is between 6.0 and 8.5, ranging from slightly acidic to slightly basic. Most estuarine organisms are adapted to live within a narrow range of pH, so changes in pH can affect the population and distribution of the estuarine inhabitants. pH influences the availability and toxicity of contaminants. Marked changes in pH over time can indicate the presence of effluents and atmospheric deposition of acid-forming substances; however, diel variation in pH also occurs and can be caused by photosynthesis and respiration cycles of algae in eutrophic waters.
Turbidity, an indicator of sediment load and water clarity, is an important variable that generally ranges from 1 to 1000 NTU (Nephelometric Turbidity Units). Levels can be increased by the presence of organic matter, effluents, or run-off with high concentrations of suspended solids. Turbidity is highly variable on a temporal scale, with seasonal changes occurring due to biological activity and surface run-off. Heavy rainfall can also result in hourly variations in turbidity.
The SCDHEC shellfish monitoring program has selected water quality parameters which describe the sanitary conditions of the shellfish environment. This is necessary to prevent the harvest of animals unsuitable for consumption. Analysis for fecal coliform is the primary indicator used to determine shellfish water quality. Water is determined to be acceptable for shellfish harvest if sampling indicates the median number of fecal coliforms per 100 ml water sample does not exceed 14 and not more than 10 percent of the samples exceed 43/100 ml (SCDHEC 1995). The SCDHEC shellfish monitoring program also monitors salinity and temperature, as each of these variables affect the normal growth and behavior of shellfish.
Water samples collected at SCDHEC primary and secondary water quality stations are analyzed for indication of nonpoint and point-source pollutant loads to the watersheds. Nutrient and organic loadings are monitored in addition to pathogens, pH, salinity, temperature and dissolved oxygen.
The SCDHEC primary water quality stations monitor total ammonia, unionized ammonia, total Kjeldahl nitrogen (TKN), and nitrate/nitrite. Ammonia (NH3) occurs naturally in water bodies as a result of the breakdown of organic and inorganic matter in soil and water, excretion from biota, and reduction of atmospheric nitrogen by microorganisms. Also, some industrial processes discharge ammonia products. In an aqueous solution, un-ionized ammonia exists in equilibrium with the ammonium ion, the distribution of forms depending upon pH. Total ammonia is the sum of unionized and ammonium forms. Total ammonia concentrations in surface waters are typically less than 0.2 mg/l but may reach 2-3 mg/l (Chapman 1992). Higher concentrations could be an indicator of pollution such as domestic sewage, industrial waste, or fertilizer runoff. Seasonal fluctuations in ammonia concentrations are natural due to varying rates of organic loading and biological decay.
The nitrate ion (NO3) is the common form of nitrogen found in natural waters. It may be biochemically reduced to nitrite (NO2), usually under anaerobic conditions. The nitrite ion is rapidly oxidized to nitrate (Chapman, 1992). Natural levels of nitrate in surface waters seldom exceed 0.1 mg/l as N, but waters influenced by human activity normally contain up to 5 mg/l as N with levels over 5 mg/l as N indicating pollution by animal or human waste or fertilizer runoff. National drinking water standards for nitrates are 10 mg/l as N.
The Kjeldahl method for TKN tests for the presence of nitrogen in the tri-negative state. The subtraction of ammonia nitrogen from TKN gives "organic nitrogen". Organic nitrogen consists mainly of protein substances and their byproducts. Organic nitrogen is typically formed within the water column by phytoplankton and bacteria and cycled within the food chain, and is subject to seasonal fluctuations in the biological community (Chapman 1992). Concentrations of TKN are affected by both point and non-point sources.
Biological and Chemical Oxygen
Chemical oxygen demand (COD) is a measure of the oxygen equivalent of the organic matter in a water sample that is susceptible to oxidation by a strong chemical oxidant. COD is greater than BOD for any given sample and is typically less than 20 mg/l in unpolluted waters.
The ACE Basin study area includes coastal, estuarine, brackish and freshwater habitats within its watersheds. A majority of the water habitat is saline to some extent with measurable salinity 32.2 or more kilometers (20 miles) upstream in a number of the river systems (Eidson 1993). The mesotidal, low country of the study area experiences tidal influences 64.4 km (40 miles) upstream in some of the river systems. Most of the ACE Basin's naturally flowing water habitat is categorized as estuarine habitat and experiences daily tidal fluctuations.
Estuaries are dynamic zones of transition between purely marine waters and fresh water habitats. They have been described as an inlet of the sea reaching into a river valley as far as the upper limit of the tidal rise, normally divisible into three sectors: a marine or lower estuary in direct connection with the sea; a middle estuary subject to strong fresh and saltwater mixing; and an upper estuary characterized by freshwater but subject to daily tidal action (Kramer et al. 1994). The Venice system of estuarine classification segregates zones according to salinity concentrations: euhaline is 30 to 40 parts per thousand (ppt); polyhaline is 18-30 ppt; mesohaline is 5 to 18 ppt; oligohaline is 0.5 - 5 ppt. Tidal variation, fresh water flow and climatological changes result in significant temporal and spatial variation within an estuary. Specific conditions at any one point in space and time will always be different, providing a challenge to water quality monitoring programs. The ACE Basin has estuarine habitat located within a wide range of salinities. The temporal and spatial variability of estuarine water quality is discussed in a following section.
Cyclic variations in turbidity, salinity, suspended solids, and organic matter are common, due to resuspension of matter which has accumulated in the higher elevations of the intertidal region (salt marsh). The ideal strategy for long-term sampling would be to collect each sample at a mid-tide on a daily scale and at a mid point in the lunar cycle (midway between neap and spring tide) or to sample continuously (Kramer et al. 1994). The SCDHEC shellfish and primary/secondary station sampling protocols do not sample at a consistent tidal stage or lunar phase. This introduces additional variability into the sampling strategy and data.
The NERR Monitoring Program utilizes self-contained water quality monitoring units that sample basic water quality parameters every 30 minutes, twenty four hours a day. Data from this program show semidiurnal patterns in salinity, turbidity, pH, dissolved oxygen, and temperature corresponding to the tidal cycle. In addition, variations in salinity, turbidity and dissolved oxygen patterns may be correlated to spring tide conditions that facilitate resuspension of organic matter and sediments from upper intertidal regions.
Waste load is the flow-based calculation of pollutant mass discharged daily by a point or non-point source. Waste assimilation and mixing models developed by SCDHEC estimate the maximum daily pollutant load a section of stream may assimilate and still meet designated use criteria. Direct dischargers are issued permits with loading limitations based upon modeling results or existing regulations. Under the SCDHEC Watershed Water Quality Management Strategy, all permits within the project area are scheduled to be reissued on a five-year basis. All ACE Basin permits, except those issued for the Edisto River drainage basin, were issued/reissued in 1998. Edisto River dischargers had their permits issued/re-issued in 1999 (SCDHEC 1993).
Land coverage within the study area is primarily forest (56%), with a moderate amount of wetlands (17%) and agricultural land (12%) . Urban development is estimated to cover 2% of the area, but is likely to increase. Since 1978, Colleton County has experienced an average annual decrease in farm acreage of 5.7%. Much of this land has been converted back to timberland. The decrease in agricultural land area may have a measurable impact on nonpoint-source water quality parameters within the study area. (See related sections: Agriculture and Forestry.)
Forestry is the largest industry within the project area. Some major forestry companies have voluntarily begun to implement best management practices on managed lands. Best management practices will hopefully have a positive impact on nonpoint stream pollutant loadings.
Several organizations such as the Lowcountry Open Land Trust, Edisto Island Open Land Trust , and The Nature Conservancy provide incentives to private land owners in the ACE Basin to protect their lands from future development. Additionally, state and federal holdings within the project area protect a significant land area. Participation in land trust and conservation easement programs may have future beneficial impacts on nonpoint source pollutant concentrations. (See related section: Protected Lands.)
Effects of urban development are likely to increase in the characterization project area as they have in nearly every area of the world. Construction of roads (e.g. the widening of highway 17) and the conversion of forest, wetlands, and agricultural land to housing and commercial establishments within the project area encourages additional development and increases nonpoint source pollutant loadings. Furthermore, lack of a comprehensive land use planning strategy can produce a landscape that is least suitable for controlling nonpoint source runoff. Eidson (1993) found increasing trends in nitrate/nitrite concentrations associated with urbanized sections of watersheds upstream of the study area on the Edisto River. (See related section: Urban Areas.)
The Clean Water Act grants the states authority to set their own water quality standards but requires that the state's beneficial uses, numeric and descriptive criteria comply with national goals. The South Carolina Department of Health and Environmental Control (SCDHEC) has the primary responsibility for setting and enforcing standards needed to attain water quality goals within the state of South Carolina.
The South Carolina Department of Health and Environmental Control
SCDHEC data analyzed within this section was obtained from the EPA STORET (STOrage and RETreival) database maintained by the USEPA. This database serves as a central repository for environmental data collected by federal, state, local and private entities nationwide. SCDHEC provides water quality monitoring data to STORET. All evaluations made within this section are based on data from the period from 1985 through 1995.
National Estuarine Research Reserve
Collecting long-term trend data that capture natural variability requires adequate temporal and spatial coverage. To address the temporal coverage, the NERR system-wide monitoring project employs YSI TM model 6000 sondes to collect water quality data. These data loggers record at 30-minute intervals, relay measurements to internal memory, run unattended for weeks at a time, and can operate in depths of a few centimeters. The salinity and temperature-compensated dissolved oxygen sensor provides accurate readings with little drift for extended periods, although deployment duration may differ during the year due to seasonal fouling.
In terms of spatial coverage, two data loggers are deployed in tributaries of the South Edisto River. One is located on a tidal marsh creek off Big Bay Creek. Surrounded by residential and commercial development and likely subject to non-point source pollution, this station was selected to monitor anthropogenic impacts. There are several boat ramps, a marina, and commercial seafood docks. Big Bay Creek is subject to heavy boat traffic. The creek is closed to shellfish harvesting because fecal coliform concentrations exceed state guidelines for shellfish waters. In contrast, there is little development or agriculture near the St. Pierre Creek deployment site, although new development on adjacent Bailey's Island may have long term impacts on the site. This station is surrounded by a wide expanse of Spartina alterniflora marsh. Upland areas are dominated by maritime forest. The creek is subject to light boat traffic.
To augment the NERR water quality data, an automated weather station, located at Bennetts Point, provides continuous information on meteorological conditions that influence water quality variables. As pointed out above, rainfall influences salinity in estuaries and can increase runoff of sediment and organic material that in turn may influence dissolved oxygen and turbidity levels as well.
Sampling results at the SCDHEC primary and secondary monitoring stations were compared to current state or federal water quality criteria for each parameter subject to guidelines. (See Water Quality Means and Frequency with which Samples Exceeded Standards ).)South Carolina has classified all waters in the project area into the following use categories:
Each use category has a set of water quality criteria which define compliance with the proposed use.
Eleven primary and secondary monitoring station water samples were tested for noncompliance with SCDHEC standards at time of sampling. Stations were located on the Combahee, Ashepoo, S. Edisto, N. Edisto, and Coosaw Rivers (Monitoring stations ). In general, all samples were in compliance over 90% of the time except for those at the two stations on the Ashepoo River and two of the three stations on the Combahee Rivers. Noncompliance at the two Ashepoo River stations was near 40%. The Combahee River stations had excursion rates of 0%, 12% and 26%. The stations collecting samples with excessive rates of noncompliance are located in areas of high natural organic input and limited tidal water exchange. These low DO concentrations are most likely naturally occurring events which would be expected during periods when low photosynthetic activity relative to respiration occurs in combination with high natural organic loadings.
The NOAA National Estuarine Eutrophication Survey has classified water quality samples as anoxic if they are 0 mg/l, hypoxic if they are below 2 mg/l, and stressed if they are between 2 and 5 mg/l (NOAA 1996). Samples at the eleven stations evaluated for compliance with SCDHEC standards were evaluated to determine their classification according to these criteria. Anoxic conditions were not seen in any samples but hypoxic conditions were reported 3% of the time at one Ashepoo River station and one Combahee River station, and one percent of the time at one S. Edisto River station. A large proportion of samples from all eleven stations within the ACE Basin fell within the "stressed" classification. The Ashepoo River stations fell within stressed conditions 80 and 59 percent of the time; the Combahee River stations fell within the stressed classification 45, 25 and 32 percent of the time; the S. Edisto River stations fell within the stressed classification 28 and 8 percent of the time; the Coosaw River station was stressed 48% of the time; while the N. Edisto and Dawhoo River stations were stressed 100, 9 and 30 percent of the time.
Excursion rates at the 53 stations analyzed ranged from 0% to 97%. The Ashepoo River downstream of Walterboro had the highest rates of noncompliance. Average fecal coliform concentrations at this station were also the highest within the study area. Numerous point and and non-point sources associated with the Walterboro area are the likely cause of the high excursion rates.
The Big Bay and Fishing Creek estuaries located adjacent to the town of Edisto Beach exhibited high rates of fecal coliform noncompliance. Average sample concentrations for the seven shellfish stations in the area ranged from 23 to 117 per 100 ml. Possible sources of contamination include expansive residential and commercial development and associated runoff, failing septic systems, area marinas and boat docks, and runoff/infiltration from the town's no-discharge wastewater treatment facility. Treated, chlorinated/dechlorinated, wastewater from this facility is used for spray irrigation on the island's Fairfield Golf Course.
The six South Edisto River stations located downstream of Highway 17 were found to have relatively high mean values of fecal coliform concentrations and excursion rates ranging from 4% to 67%. The general rural development of Edisto Island, with associated livestock and agricultural practices, are likely explanations for these high values. The Dawhoo Creek stations located on the north side of Edisto Island detected similar high average fecal coliform concentrations and excursion rates ranging from 2-57%.
One additional area of frequent fecal coliform excursions was at station CSTL 98 located on the Combahee River at Highway 17. New development in the area with associated septic systems and a boat landing may contribute to these excursions.
Only one of the eleven stations evaluated displayed excursion rates exceeding 10% and all stations averaged less than 0.2 mg/l. Station MD 119 on the South Edisto River at Highway 17 exceeded the guideline value 26.5 % of the time. Possible sources of inorganic nitrogen in this area are not known, although the sampling station is located at Jacksonboro, a small roadside community with several fuel stations and restaurants.
ACE Basin water quality data from nine primary and secondary fixed stations, and 29 shellfish stations were analyzed for historical trends (Trends in water quality ). Parameters analyzed included:
Spatial trend analyses were performed on fecal coliform concentrations for the three major watersheds in the project area: the South Edisto, Ashepoo and the Combahee Rivers, plus the North Edisto River and the Coosaw River estuarine systems. An upstream/downstream concentration gradient was tested for significant trend using Kendall's Tau test and an alpha value of 0.05.
Since passage of the Clean Water Act in 1972, municipal loads of BOD to the nation's waterways have been reduced at least 46% while industrial loadings have been reduced 71% (Smith et al. 1987). Additionally, a general decreasing national trend in fecal coliform, and total phosphorus was noted. Trends in these parameters are most strongly associated with point source reductions, while nonpoint-source contributions remain unidentified. Nitrate concentrations within the nation's rivers have generally been on the increase due to nonpoint sources such as agriculture, animal feed lots, atmospheric deposition, and urbanization (Smith et al. 1987). Changes in nitrate concentrations may well reflect land-use changes within an area of study.
Trends in Fecal Coliform
The three additional stations (Big Bay Creek, Fishing Creek and S. Edisto River) with significant increasing fecal coliform trends were located adjacent to the town of Edisto Beach. Continued development of the area, increased recreational boating and dockage, and failure of existing septic tank installations are possible explanations for the trend.
Fecal coliform concentrations were tested for correlation to river mile (as measured from river mouth upstream in the center of the river) for the three major river systems (Ashepoo, Combahee and S. Edisto Rivers) and two estuarine systems (N. Edisto and Coosaw Rivers) within the study area (See fecal coliform correlations ). The Combahee and S. Edisto Rivers were found to have a significant increasing trend in fecal coliform proceeding upstream (Kendall-Tau correlation for Mile vs. Coliform ). Ashepoo River data suggested the same trend, but the relationship was not found to be significant at the 95% confidence level. The two estuarine systems displayed no significant upstream trend. Fecal coliform sources are human and animal discharges into otherwise fecal coliform-free waters. Ocean waters, which are characteristically free of appreciable concentrations of fecal coliforms, dilute river waters that have become contaminated with fecal coliforms as they pass through zones inhabited humans and livestock. Dilution of estuarine sections of rivers by ocean tides most likely produces the positive correlation of fecal coliforms with upstream river mile.
Trends in Biochemical Oxygen Demand
Trends in Total Kjeldahl Nitrogen
Trends in Total
Trends in Temperature
Spatial Trends in
Monitoring instruments have been deployed by National Estuarine Research Reserve (NERR) staff since March 1995 in St. Pierre and Big Bay Creeks. Data from these sites and others nationwide are available over the Internet at http://inlet.geol.sc.edu/nerrscdmo.html along with graphs and summary statistics. Automated data loggers recording at short time intervals over extended periods provide a unique means of capturing the episodic as well as cyclical nature of water quality variables. Having a time series of data available for the ACE Basin is important in further understanding underlying periodicities in the data. The following discussion of trends in water quality at the NERR monitoring stations is based on data collected from March 1995 through December 1997. Hourly means of this data are available in Excel spreadsheets for each creek and year: St. Pierre 1995 ( sp95hr.xls), St. Pierre 1996 ( sp96hr.xls), St. Pierre 1997 ( sp97hr.xls), Big Bay 1995 ( bb95hr.xls), Big Bay 1996 ( bb96hr.xls), Big Bay 1997 ( bb97hr.xls). [Note: Clicking on these files will launch Excel; return to browser using the back button. The water quality parameter column headings for each table are defined as: Meandep=depth in meters, Meandomg=dissolved oxygen in mg/l, Meandosa=dissolved oxygen in % saturation, Meancond=conductivity in microsiemens /cm, Meanturb=turbidity in nephelometric turbidity units (NTU), Meansal=salinity in parts per thousand (ppt), Meantemp=temperature in degrees C.]
An hourly plot of mean salinity for a selected period in August 1995 and January 1996 revealed that salinity was strongly influenced by tidal stage (indicated by water depth). As expected, highest values of salinity occurred on the flood tide (See SP hourly salinity ). Spectral analysis which is a type of time series analysis that is useful for exploring data and determining if periodicity is present, confirmed that a sinusoidal period occurred at ~ 12 h, representing tidal periodicity . This tidal periodicity was also supported by spectral analysis of water depth in a 500 h record.
Another statistical technique, periodic regression analysis, was used to determine whether significant seasonal, tidal and diurnal components were present in the data (Lorda and Saila 1986). This regression analysis hypothesizes a known period and removes its component from the data if found to be significant. By detrending hourly salinity data using this method, it was revealed that seasonal periodicity was significant and accounted for most of the variation in salinity at both sites ( Periodic Regression ). Tidal periodicity was also significant but accounted for less than 5% of the total variation, while the diurnal component accounted for less than 1% of the total variation although the slope was significant.
Marked variations in pH can also occur diurnally due to changes in primary production and respiration. The excursions in pH found at the NERR monitoring sites in the ACE Basin appear to be attributable to natural conditions.
At St. Pierre Creek, a simple plot of dissolved oxygen concentrations and water depth showed that DO tracked depth for part of the cycle, with lowest values occurring on low tides. Afternoon and evening values were not as low as early morning values, suggesting a diurnal component (See hourly DO at St. Pierre ). By applying periodic regression analysis (Lorda and Saila 1986), significant seasonal, tidal and diurnal components in dissolved oxygen were confirmed ( Periodic Regression ). The first step in the analysis was to quantify seasonal variations at each site. A quadratic model was fit to the data and it was determined that there were seasonal trends in dissolved oxygen. Similarly, diel and tidal variation were quantified with results finding both components significant; diel variation, however, accounted for a greater portion of the total variance in the model than tidal variation. It appears then that most of the variation in dissolved oxygen is explained by seasonal periodicity.
Spectral analysis was also used to indicate periodicity in dissolved oxygen. Two sinusoidal periods for DO were found in a 100 h record at both sites and in a 500 h record at Big Bay Creek (See spectral analysis of DO ). The large peak at 12 hours represents tidal periodicity, and the peak at 24 hours represents diel periodicity. The size of the peaks is a measure of their relative importance in explaining variance in dissolved oxygen (Wenner et al. 1998).
Monthly mean dissolved oxygen varied with season. Highest mean values occurred in winter, while lowest mean values occurred in summer at both sites (See monthly DO ) Monthly mean plots also indicated differences among years in % saturation at the two creeks. Scatter plots of hourly dissolved oxygen values at the two sites further illustrated annual differences . Hypoxic events, as indicated by points below the reference line at 28% saturation (~2 mg/l), occurred most frequently in summer, although few hypoxic events occurred at either site in 1995. Over all years, hypoxic conditions were observed in every season but summer was clearly the season with the greatest percent of time that DO was # 28%. There were clearly differences among the sites with hypoxic conditions occurring over a greater percent of time at St. Pierre Creek . July and August were the months when hypoxia occurred >10% of the time at St. Pierre Creek, while Big Bay had hypoxic conditions for 17% of the time in September (See seasonal differences in dissolved oxygen levels ). These differences among creeks most likely reflect physiographic and circulation differences among the two sites and are not likely due to anthropogenic influences.
Why hypoxia was common in some years and not others is not clear. Many investigators have suggested that natural factors (e.g. degree of stratification, circulation patterns) and human development (e.g. nutrient loadings) contribute to the development and maintenance of low DO events. Further evaluation of long-term water quality monitoring data for the NERRS sites in conjunction with information on weather patterns, rainfall, sea-level rise, nutrient loadings, habitat differences, and land cover will be required to understand long-term trends and year-to-year differences in low DO.
Long-term water quality monitoring in the ACE Basin provides a unique opportunity to increase our understanding of how various environmental factors influence estuarine processes. Frequent measurements are important if one is to determine variability and the temporal dynamics of water quality, with the eventual goal of partitioning natural variability from that influenced by human activities.
Thus far, there are few indications of major problems in water quality within the Reserve portion of the ACE Basin. Dissolved oxygen concentrations often dropped below 4 mg/l at stations along the Ashepoo and Combahee Rivers, but these excursions are thought to be natural phenomena. Dissolved oxygen at the two NERR continuous monitoring sites also drops below 4 mg/l, a critical threshold for growth, reproduction and survival of living resources (Diaz and Rosenberg 1995). As coastal watersheds surrounding the ACE Basin develop, the frequency and duration of low DO events are expected to increase due to increased nutrient loadings.
Monitoring stations on the Ashepoo River downstream of Walterboro, SC; adjacent to Edisto Beach; and adjacent to Edisto Island on the S. Edisto River and Dawhoo Creek all had high instances of non-compliance with fecal coliform standards. The Coosaw River and Big Bay Creek adjacent to Edisto Beach also had statistically significant increasing concentrations of fecal coliform. Stations indicating fecal coliform problems were located in areas of human development where septic systems are the primary domestic wastewater treatment schemes and sources of animal waste, urban runoff and land application of wastewater are found. The continued rapid growth of the Beaufort area, and expansion of Edisto Beach and Walterboro will certainly affect ACE Basin water quality in the future.
The concentrations of nitrates/nitrites and phosphorus are generally stable or decreasing. Values of these nutrients are generally below levels of concern for nutrient enrichment, except in those areas of natural phosphate deposits. Improvements in forestry and agricultural management practices, the conversion of agricultural land to natural cover, and improvements in the treatment of point source discharges have contributed to these findings. As urban development continues to spread on the outskirts and within the project area, these trends will likely be reversed and concerns about eutrophication may increase.
Continuous, long-term data sets for estuarine water quality, like those collected for the ACE Basin NERR, will be required to establish baseline conditions against which the degree and magnitude of changes can be measured. While point data are useful in evaluating water quality, they do not provide information on short-term temporal variability. Estuarine water quality naturally varies on many spatial and temporal scales (e.g. tidal, diel, lunar, seasonal, annual, by region and by creek). Alterations to estuarine water quality that are associated with development of coastal watersheds affect the amplitude and type of periodicity as much as it alters the mean of observed values (Wenner et al. in press). With increasing urbanization, water quality in the ACE basin may decline unless sound management principles are applied.
E. Wenner, SCDNR Marine Resources Research Institute
M. Thompson, SCDNR Marine Resources Research Institute
D. Sanger, SCDNR Marine Resources Research Institute
Chapman, D. 1992. Water quality assessments. Chapman and Hall, London, UK.
Diaz, R. J. and R. Rosenberg. 1995. Marine benthic hypoxia: A review of its ecological effects and the behavioural responses of benthic macrofauna. Oceanography and Marine Biology: An Annual Review 33:245-305.
Eidson, J. P. 1993. Water quality. In: Assessing change in the Edisto River basin, an ecological characterization. South Carolina Water Resources Commission Report 177.
Funderburk, S. L. (ed.). 1991. Habitat requirements for Chesapeake Bay living resources. Habitat Objectives Workgroup, Chesapeake Bay Program, U.S. Environmental Protection Agency. Annapolis, MD.
Kramer, J. M., U. H. Brockman, and R. M. Warwick. 1994. Tidal estuaries: Manual of sampling and analytical procedures. Balkema Publishing, Brookfield, VT.
Lorda, E. and S. B. Saila. 1986. A statistical technique for analysis of environmental data containing periodic variance components. Ecological Modeling 32:59-69.
National Oceanic and Atmospheric Administration. 1996. NOAA's eutrophication survey. Vol. 1: South Atlantic region. Office of Ocean Resources Conservation Assessment, Silver Spring, MD.
National Research Council. 1993. Managing Wastewaster in Coastal Urban Areas. National Academy Press, Washington, DC.
Orlando, S. P., P. H. Wendt, C. J. Klein, M. E. Pattilo, K. C. Dennis, and G. H. Ward. 1994. Salinity characteristics of the south Atlantic estuaries. Office of Ocean Resources Conservation and Assessments, National Oceanic and Atmospheric Administration, Silver Spring, Maryland.
Robertson, A. and W. Davis. 1993. The selection and use of water quality indicators. p. 119-128. In: Water Environment Foundation (ed.). Challenges facing environmental laboratories: methods, quality, media, and liability. Specialty Conference Series, Santa Clara, CA. Water Environment Foundation, Alexandria, VA.
Smith, R. A., R. B. Alexander, and M. Wolman. 1987. Water quality trends in the nation's rivers. Science 235:1607-1615.
South Carolina Department of Health and Environmental Control. 1998. South Carolina General Assembly, Document 2218, Chapter 61. http://www.lpitr.state.sc.us/regs/2218.htm. Accessed September 1998.
South Carolina Department of Health and Environmental Control. 1998. Water classifications and standards. Bureau of Water, Columbia, SC.
South Carolina Department of Health and Environmental Control. 1993. Watershed water quality management strategy Savannah-Salkehatchie watershed. Technical Report 002-93. Bureau of Water Pollution Control.
South Carolina Department of Health and Environmental Control. 1996. The state of South Carolina water quality assessment pursuant to section 305(b) of the federal Clean Water Act. Bureau of Water Pollution Control.
South Carolina Department of Health and Environmental Control. 1995. Watershed water quality management strategy Saluda and Edisto River basins. Technical Report 003-95. Bureau of Water.
South Carolina Department of Health and Environmental Control. 1995. State of South Carolina monitoring strategy for fiscal year 1996. Technical Report 002-95. Bureau of Water Pollution Control.
Wenner, E. L, A. F. Holland, and D. Sanger. 1998. Assessing short-term variability in dissolved oxygen and other water quality variables in shallow water estuarine habitats. In Marine Technology Society (ed.). Proceedings of the Ocean Community Conference, November 1998, Baltimore, MD. Marine Technology Society, Washington, DC.
Wenner, E. L. and M. Geist. unpublished. Reserved for the future: A national system to preserve estuarine waters.