Influence of Sampling Frequency on Estimation of Annual Total Phosphorus and Total Suspended Solids Loads1

Abstract: The determination of sediment and nutrient loads is typically based on the collection and analysis of grab samples. The frequency and regularity of traditional sampling may not provide representation of constituent loading, particularly in systems with flashy hydrology. At two sites in the Little Bear River, Utah, continuous, high‐frequency turbidity was used with surrogate relationships to generate estimates of total phosphorus and total suspended solids concentrations, which were paired with discharge to estimate annual loads. The high frequency records were randomly subsampled to represent hourly, daily, weekly, and monthly sampling frequencies and to examine the effects of timing, and resulting annual load estimates were compared to the reference loads. Higher frequency sampling resulted in load estimates that better approximated the reference loads. The degree of bias was greater at the more hydrologically responsive site in the upper watershed, which required a higher sampling frequency than the lower watershed site to achieve the same level of accuracy in estimating the reference load. The hour of day and day of week of sampling impacted load estimation, depending on site and hydrologic conditions. The effects of sampling frequency on the determination of compliance with a water quality criterion were also examined. These techniques can be helpful in determining necessary sampling frequency to meet the objectives of a water quality monitoring program.     

A STAGE DISTRIBUTION APPROACH TO ESTIMATING ICE RELATED FLOODING PROBABILITIES1

ABSTRACT: A procedure is presented for estimating flooding probabilities resulting from either open water or ice condition events. The methodology involves individually fitting a distribution function to water stages from open water and ice events and determining the composite probability of exceedence of any stage value. The parameters of the two distribution functions are estimated using censored maximum likelihood. The approach is evaluated with a Monte Carlo sampling program and is applied to estimate flooding probabilities on the Yukon River.     

Estimation of upper centile concentrations using historical atrazine monitoring data from community water systems

A survey sampling approach is presented for estimating upper centiles of aggregate distributions of surface water pesticide measurements obtained from datasets with large sample sizes but variable sampling frequency. It is applied to three atrazine monitoring programs of Community Water Systems (CWS) that used surface water as their drinking water source: the nationwide Safe Drinking Water Act (SDWA) data, the Syngenta Voluntary Monitoring Program (VMP), and the Atrazine Monitoring Program (AMP).The VMP/AMP CWS were selected on the basis of atrazine monitoring history (CWS having at least one annual average concentration from SDWA ≥ 1.6 ppb atrazine since 1997 in the AMP). Estimates of the raw water 95th, 99th, and 99.9th centile atrazine concentrations for the VMP/AMP CWS are 4.82, 11.85, and 34.00 ppb, respectively. The corresponding estimates are lower for the finished drinking water samples, with estimates of 2.75, 7.94, and 22.66 ppb, respectively. Finished water centile estimates for the VMP/AMP CWS using only the SDWA data for these sites are consistent with the results. Estimates are provided for the April through July period and for CWS based on surface water source type (static, flowing, or mixed). Requisite sample sizes are determined using statistical tolerance limits, relative SE, and the Woodruff interval sample size criterion. These analyses provide 99.9% confidence that the existing data include the 99.9th centile atrazine concentration for CWS raw and finished water in the Midwest atrazine high-use areas and in the nationwide SDWA dataset. The general validity of this approach is established by a simulation that shows estimates to be close to target quantities for weights based on sampling probabilities or time intervals between samples. Recommendations are given for suitable effective sample sizes to reliably determine interval estimates.     

THE EFFECT OF DATA INDEPENDENCE IN MODEL CALIBRATION AND MODEL TESTING1

ABSTRACT: With the increased use of models in hydrologic design, there is an immediate need for a comprehensive comparison of hydrologic models, especially those intended for use at ungaged locations (i.e., where measured data are either not available or inadequate for model calibration). But some past comparisons of hydrologic models have used the same data base for both calibration and testing of the different models or implied that the results of model calibration are indicative of the accuracy at ungaged locations. This practice was examined using both the regression equation approach to peak discharge estimation and a unit hydrograph model that was intended for use in urban areas. The results suggested that the lack of data independence in the calibration and testing of regression equations may lead to both biased results and misleading statements about prediction accuracy. Additionally, although split-sample testing is recognized as desirable, the split-samples should be selected using a systematic-random sampling scheme, rather than random sampling, because random sampling with small samples may lead to a testing sample that is not representative of the population. A systematic-random sampling technique should lead to more valid conclusions about model reliability. For models like a unit hydrograph model, which are more complex and for which calibration is a more involved process, data independence is not as critical because the data fitting error variation is not as dominant as the error variation due to the calibration process and the inability of the model structure to conform with data variability.     

Evaluating the Illinois Stream Valley Segment Model as an Effective Management Tool

Stream habitat assessments are conducted to evaluate biological potential, determine anthropogenic impacts, and guide restoration projects. Utilizing these procedures, managers must first select a representative stream reach, which is typically selected based on several criteria. To develop a consistent and unbiased procedure for choosing sampling locations, the Illinois Department of Natural Resources and the Illinois Natural History Survey have proposed a technique by which watersheds are divided into homogeneous stream segments called valley segments. Valley segments are determined by GIS parameters including surficial geology, predicted flow, slope, and drainage area. To date, no research has been conducted to determine if the stream habitat within a valley segment is homogeneous and if different valley segments have varying habitat variables. Two abutting valley segments were randomly selected within 13 streams in the Embarras River watershed, located in east-central Illinois. One hundred meter reaches were randomly selected within each valley segment, and a transect method was used to quantify habitat characteristics of the stream channel. Habitat variables for each stream were combined through a principal components analysis (PCA) to measure environmental variation between abutting valley segments. A multivariate analysis of variance (MANOVA) was performed on PCA axes 1–3. The majority of abutting valley segments were significantly different from each other indicating that habitat variability within each valley segment was less than variability between valley segments (5.37 ≤ F ≤ 245.13; P ≤ 0.002). This comparison supports the use of the valley segment model as an effective management tool for identifying representative sampling locations and extrapolating reach-specific information.     

Surrogate Measures for Providing High Frequency Estimates of Total Suspended Solids and Total Phosphorus Concentrations1

Spackman Jones, Amber, David K. Stevens, Jeffery S. Horsburgh, and Nancy O. Mesner, 2010. Surrogate Measures for Providing High Frequency Estimates of Total Suspended Solids and Total Phosphorus Concentrations. Journal of the American Water Resources Association (JAWRA) 1‐15. DOI: 10.1111/j.1752‐1688.2010.00505.x Abstract: Surrogate measures like turbidity, which can be observed with high frequency in situ, have potential for generating high frequency estimates of total suspended solids (TSS) and total phosphorus (TP) concentrations. In the semiarid, snowmelt‐driven, and irrigation‐regulated Little Bear River watershed of northern Utah, high frequency in situ water quality measurements were recorded in conjunction with periodic chemistry sampling. Site‐specific relationships were developed using turbidity as a surrogate for TP and TSS at two monitoring locations. Methods are presented for employing censored data and for investigating categorical explanatory variables (e.g., hydrologic conditions). Turbidity was a significant explanatory variable for TP and TSS at both sites, which differ in hydrologic and water quality characteristics. The relationship between turbidity and TP was stronger at the upper watershed site where TP is predominantly particulate. At both sites, the relationships between turbidity and TP varied between spring snowmelt and base flow conditions while the relationships between TSS and turbidity were consistent across hydrological conditions. This approach enables the calculation of high frequency time series of TP and TSS concentrations previously unavailable using traditional monitoring approaches. These methods have broad application for situations that require accurate characterization of fluxes of these constituents over a range of hydrologic conditions.     

Optimizing Liquid Effluent Monitoring at a Large Nuclear Complex

Effluent monitoring typically requires a large number of analytes and samples during the initial or startup phase of a facility. Once a baseline is established, the analyte list and sampling frequency may be reduced. Although there is a large body of literature relevant to the initial design, few, if any, published papers exist on updating established effluent monitoring programs. This paper statistically evaluates four years of baseline data to optimize the liquid effluent monitoring efficiency of a centralized waste treatment and disposal facility at a large defense nuclear complex. Specific objectives were to: (1) assess temporal variability in analyte concentrations, (2) determine operational factors contributing to waste stream variability, (3) assess the probability of exceeding permit limits, and (4) streamline the sampling and analysis regime. Results indicated that the probability of exceeding permit limits was one in a million under normal facility operating conditions, sampling frequency could be reduced, and several analytes could be eliminated. Furthermore, indicators such as gross alpha and gross beta measurements could be used in lieu of more expensive specific isotopic analyses (radium, cesium-137, and strontium-90) for routine monitoring. Study results were used by the state regulatory agency to modify monitoring requirements for a new discharge permit, resulting in an annual cost savings of US $223,000. This case study demonstrates that statistical evaluation of effluent contaminant variability coupled with process knowledge can help plant managers and regulators streamline analyte lists and sampling frequencies based on detection history and environmental risk.     

DISTRIBUTION AND STOCHASTIC GENERATION OF ANNUAL AND MONTHLY PRECIPITATION ON A MOUNTAINOUS WATERSHED IN SOUTHWEST IDAHO1

ABSTRACT: The 18-year precipitation record from the dense gage network on the Reynolds Creek Experimental Watershed located in southwest Idaho was used to determine the spatial distribution of annual and monthly precipitation on a mountainous watershed. Analyses of these data showed a linear relationship between annual amounts and elevation. This relationship was best when the gages were grouped into downwind and upwind sites. This grouping was appropriate because most of the winter storms moved over the watershed from the west and southwest, and the heaviest precipitation was on the west (downwind) side of the watershed. Gage sites along the western and southern watershed borders were most representative of the upwind gages on the east side, because they measured the precipitation from the air moving upwind onto the watershed. The maximum annual precipitation on the watershed was just leeward of the western watershed boundary. The monthly precipitation and elevation relationship was also best represented by grouping the gage sites into upwind and downwind sites. However, during the summer when there are only small amounts of pre cipitation and thunderstorms are the source of most precipitation, one equation can be used to represent the elevation relationship. This study also showed that the log-normal distribution could be used to generate the annual synthetic series, and the cube-root-normal distribution could be used to generate monthly synthetic series for all locations on the watershed.     

PREDICTION OF MEAN ANNUAL FLOWS IN NORTH AND CENTRAL FLORIDA1

ABSTRACT: A frequency analysis approach for the prediction of flow characteristics at ungaged locations is applied to a region of high annual precipitation and low topography in north and central Florida. Stationary time series of annual flows are fitted with the lognormal distribution and estimated parameters of the distribution are fitted by third order trend surfaces. These explain 65 and 74 percent of the observed variances in the mean and standard deviation, respectively. Predictions of parameters are then made for several locations previously unused in the study and they are used to estimate the return periods of various flows from the lognormal distribution. Application of the Kolmogorov-Smirnov goodness-of-fit test suggests that only one of the five test stations can be considered significantly different from the observed data, confirming the applicability of this technique.     

Optimal water quality monitoring network design for river systems

Typical tasks of a river monitoring network design include the selection of the water quality parameters, selection of sampling and measurement methods for these parameters, identification of the locations of sampling stations and determination of the sampling frequencies. These primary design considerations may require a variety of objectives, constraints and solutions. In this study we focus on the optimal river water quality monitoring network design aspect of the overall monitoring program and propose a novel methodology for the analysis of this problem. In the proposed analysis, the locations of sampling sites are determined such that the contaminant detection time is minimized for the river network while achieving maximum reliability for the monitoring system performance. Altamaha river system in the State of Georgia, USA is chosen as an example to demonstrate the proposed methodology. The results show that the proposed model can be effectively used for the optimal design of monitoring networks in river systems.     

Nutrient and sediment concentrations and corresponding loads during the historic June 2008 flooding in eastern Iowa

A combination of above-normal precipitation during the winter and spring of 2007-2008 and extensive rainfall during June 2008 led to severe flooding in many parts of the midwestern United States. This resulted in transport of substantial amounts of nutrients and sediment from Iowa basins into the Mississippi River. Water samples were collected from 31 sites on six large Iowa tributaries to the Mississippi River to characterize water quality and to quantify nutrient and sediment loads during this extreme discharge event. Each sample was analyzed for total nitrogen, dissolved nitrate plus nitrite nitrogen, dissolved ammonia as nitrogen, total phosphorus, orthophosphate, and suspended sediment. Concentrations measured near peak flow in June 2008 were compared with the corresponding mean concentrations from June 1979 to 2007 using a paired t test. While there was no consistent pattern in concentrations between historical samples and those from the 2008 flood, increased flow during the flood resulted in near-peak June 2008 flood daily loads that were statistically greater (p < 0.05) than the median June 1979 to 2007 daily loads for all constituents. Estimates of loads for the 16-d period during the flood were calculated for four major tributaries and totaled 4.95 x 10(7) kg of nitrogen (N) and 2.9 x 10(6) kg of phosphorus (P) leaving Iowa, which accounted for about 22 and 46% of the total average annual nutrient yield, respectively. This study demonstrates the importance of large flood events to the total annual nutrient load in both small streams and large rivers.     

SAMPLING FREQUENCY SELECTION FOR REGULATORY WATER QUALITY MONITORING1

ABSTRACT: The selection of sampling frequencies in order to achieve reasonably small and uniform confidence interval widths about annual sample means or sample geometric means of water quality constituents is suggested as a rational approach to regulatory monitoring network design. Methods are presented for predicting confidence interval widths at specified sampling frequencies while considering both seasonal variation and serial correlation of the quality time series. Deterministic annual cycles are isolated and serial dependence structures of the autoregressive, moving average type are identified through time series analysis of historic water quality records. The methods are applied to records for five quality constituents from a nine-station network in Illinois. Confidence interval widths about annual geometric means are computed over a range of sampling frequencies appropriate in regulatory monitoring. Results are compared with those obtained when a less rigorous approach, ignoring seasonal variation and serial correlation, is used. For a monthly sampling frequency the error created by ignoring both seasonal variation and serial correlation is approximately 8 percent. Finally, a simpler technique for evaluating serial correlation effects based on the assumption of AR(1) type dependence is examined. It is suggested that values of the parameter p₁, in the AR(1) model should range from 0.75 to 0.90 for the constituents and region studied.     

TOTAL SOLUTES LOADS AND GEOCHEMICAL YIELDS IN THE SASKATCHEWAN RIVER BASIN1

ABSTRACT: River solute loads have seldom been measured in very large, complex drainage basins, nor have the methods of calculating loads been critically examined. For sites in the Saskatchewan River Basin, Canada, rating curves were poor predictors of solute loads because correlations between discharge and total solutes concentration were weak (R² < 0.05 in most cases) and suffered from hysteresis. In contrast, the interval method produced reliable estimates in all seasons and sites tested, and was little affected by sampling schedule. The limit of precision (SE) for estimates of mean annual or seasonal solute load was 10–15 percent of the mean (5 percent in very small basins), reached with 10 years or more of data. Two-thirds or more of total annual solute load was transported during the open-water season, but the proportion carried during winter increased from 8 percent to 34 percent from the upstream to the downstream end of the basin, due to reservoirs retaining and mixing water. Annual loads of total solutes varied from 6.2 × 10⁴ tonnes in foothills tributaries to almost 4.0 × 10⁶ tonnes in the Saskatchewan River near the mouth. But, on an areal basis, the mountain and foothills region was the dominant solute source, producing 43–97 tonnes/km²/yr, compared with only 3–22 tonnes/km²/yr for prairie rivers. This difference is a consequence of greater rainfall and, hence, more rapid erosion in the mountains.     

Estimating upper percentiles of surface water monitoring data with sparse samples

The estimation of upper percentiles of chemical concentrations in surface water systems within sites and regions may be necessary for the assessment of potential risk to ecosystems and human health. Limited sample sizes at monitoring sites often limit the use of direct methods to estimate upper percentiles. In such cases, upper percentiles within regions within a time frame may be estimated by pooling data across sites and years, and then deriving percentile estimates from the pooled dataset. The method uses the observations resulting from either a known probability-sampling design or a sampling design treated like one because its observations come close to matching that of a probability-sample. These observations are then weighted to ensure that estimates are representative of a target population across all the sites within the region and the range of years in the time frame. This method of estimating upper percentiles of annual site concentration profiles is demonstrated using atrazine and validated using the monitoring data from both sparsely sampled and high-frequency water monitoring programs, where point and interval estimates of the 90th, 95th, and 99th pooled population percentiles are provided. This method shows that the pooled data from multiple sparse datasets can be used to provide estimates of near-peak concentrations with greater certainty, which are consistent with those generated by high-frequency sampling monitoring programs.     

Differential Assessment of Designations of Wetland Status Using Two Delineation Methods

Two different methods are commonly used to delineate and characterize wetlands. The U.S. Army Corps of Engineers (ACOE) delineation method uses field observation of hydrology, soils, and vegetation. The U.S. Fish and Wildlife Service’s National Wetland Inventory Program (NWI) relies on remote sensing and photointerpretation. This study compared designations of wetland status at selected study sites using both methods. Twenty wetlands from the Wetland Boundaries Map of the Ausable–Boquet River Basin (created using the revised NWI method) in the Ausable River watershed in Essex and Clinton Counties, NY, were selected for this study. Sampling sites within and beyond the NWI wetland boundaries were selected. During the summers of 2008 and 2009, wetland hydrology, soils, and vegetation were examined for wetland indicators following the methods described in the ACOE delineation manual. The study shows that the two methods agree at 78 % of the sampling sites and disagree at 22 % of the sites. Ninety percent of the sampling locations within the wetland boundaries on the NWI maps were categorized as ACOE wetlands with all three ACOE wetland indicators present. A binary linear logistic regression model analyzed the relationship between the designations of the two methods. The outcome of the model indicates that 83 % of the time, the two wetland designation methods agree. When discrepancies are found, it is the presence or absence of wetland hydrology and vegetation that causes the differences in delineation.     

Temporal and spatial distributions of sediment total organic carbon in an estuary river

Understanding temporal and spatial distributions of naturally occurring total organic carbon (TOC) in sediments is critical because TOC is an important feature of surface water quality. This study investigated temporal and spatial distributions of sediment TOC and its relationships to sediment contaminants in the Cedar and Ortega Rivers, Florida, USA, using three-dimensional kriging analysis and field measurement. Analysis of field data showed that large temporal changes in sediment TOC concentrations occurred in the rivers, which reflected changes in the characteristics and magnitude of inputs into the rivers during approximately the last 100 yr. The average concentration of TOC in sediments from the Cedar and Ortega Rivers was 12.7% with a maximum of 22.6% and a minimum of 2.3%. In general, more TOC accumulated at the upper 1.0 m of the sediment in the southern part of the Ortega River although the TOC sedimentation varied with locations and depths. In contrast, high concentrations of sediment contaminants, that is, total polycyclic aromatic hydrocarbons (PAHs) and polychlorinated biphenyls (PCBs), were found in sediments from the Cedar River. There was no correlation between TOC and PAHs or PCBs in these river sediments. This finding is in contradiction to some other studies which reported that the sorption of hydrocarbons is highly related to the organic matter content of sediments. This discrepancy occurred because of the differences in TOC and hydrocarbon source input locations. It was found that more TOC loaded into the southern part of the Ortega River, while almost all of the hydrocarbons entered into the Cedar River. This study suggested that the locations of their input sources as well as the land use patterns should also be considered when relating hydrocarbons to sediment TOC.     

A GIS Approach to Model Sediment Reduction Susceptibility of Mixed Sand and Gravel Beaches

The morphological form of mixed sand and gravel beaches is distinct, and the process/response system and complex dynamics of these beaches are not well understood. Process response models developed for pure sand or gravel beaches cannot be directly applied to these beaches. The Canterbury Bight coastline is apparently abundantly supplied with sediments from large rivers and coastal alluvial cliffs, but a large part of this coastline is experiencing long-term erosion. Sediment budget models provide little evidence to suggest sediments are stored within this system. Current sediment budget models inadequately quantify and account for the processes responsible for the patterns of erosion and accretion of this coastline. We outline a new method to extrapolate from laboratory experiments to the field using a geographical information system approach to model sediment reduction susceptibility for the Canterbury Bight. Sediment samples from ten representative sites were tumbled in a concrete mixer for an equivalent distance of 40 km. From the textural mixture and weight loss over 40 km tumbling, we applied regression techniques to generate a predictive equation for Sediment Reduction Susceptibility (SRS). We used Inverse Distance Weighting (IDW) to extrapolate the results from fifty-five sites with data on textural sediment composition to field locations with no data along the Canterbury Bight, creating a continuous sediment reductions susceptibility surface. Isolines of regular SRS intervals were then derived from the continuous surface to create a contour map of sediment reductions susceptibility for the Canterbury Bight. Results highlighted the variability in SRS along this coastline.     

Surface water quality and its control in a river with intensive human impacts–a case study of the Xiangjiang River,China

Surface water quality and its natural and anthropogenic controls in the Xiangjiang River were investigated using multivariate statistical approaches and a comprehensive observation dataset collected from 2004 to 2008. Cluster analysis (CA) grouped the 15 different sampling stations into five clusters with similar hydrochemistry characteristics and pollution levels. Four principal components (PCs), nutrients, heavy metals, natural components, and organic components, were extracted from the entire dataset. Comparison of the different regional characteristics of these four PCs revealed a decreasing trend for heavy metals and an increasing trend for organic factor on an annual scale, and the seasonal trend was only observed for natural factor. We also conducted analysis of variance (ANOVA) in combination with principal component analysis (PCA) to quantify the relative contribution of spatial and temporal variations to each of the four PCs. The results revealed that 62% of the contributions from the spatial sites were responsible for variations in heavy metals, while 83% of the contributions from the sampling time were responsible for natural variations observed. However, no significant spatial or temporal contributions were found to be responsible for the nutrient and organic variations. Finally, some suggestions regarding water management were put forward based on the current status and future trends of surface water quality in the Xiangjiang River.     

NUTRIENT LOAD CHARACTERIZATION FROM INTEGRATED SOURCE DATA FOR THE LOWER MISSISSIPPI RWER1

ABSTRACT: Nutrient data from all available sources for the lower Mississippi River were examined for potential differences among sampling agencies and geographic locations for the period between 1960 and 1998. Monthly means grouped by parameter, sampling location and agency, were calculated and compared as paired sets, excluding those months where data were not available for both sets. Some significant differences were found between various agencies collecting nutrient data on the river, as well as between various stretches of river, especially in the case of phosphorus nutrient data. Results were used to synthesize data sets from which a history of nutrient loading in the Mississippi River was determined. General trends in nitrate+nitrite, total Kjeldahl nitrogen, orthophosphate, total phosphorus and silica loads, as well as changes in nutrient proportions and the specific limiting nutrient (by month) are reported. This study provides a useful summary of contemporary and historical nutrient data that may assist in the evaluation of Mississippi River water quality and its potential effect on the Gulf of Mexico.     

Relations of fish community composition to environmental variables in streams of central Nebraska,USA

Nine sites on streams in the Platte River Basin in central Nebraska were sampled as part of the US Geological Survey's National Water-Quality Assessment Program during 1993–1994. A combination of canonical correspondence analysis and an index of biotic integrity determined from fish community data produced complementary evaluations of water-quality conditions. Results of the canonical correspondence analysis were useful in showing which environmental variables were significant in differentiating fish communities at the nine sites. Five environmental variables were statistically significant in the analysis. Median specific conductance of water samples collected at a site accounted for the largest amount of variability in the species data. Although the percentage of the basin as cropland was not the first variable chosen in a forward selection process, it was the most strongly correlated with the first ordination axis. A rangeland-dominated site was distinguished from all others along that axis. Median orthophosphate concentration of samples collected in the year up to the time of fish sampling was most strongly correlated with the second ordination axis. The index of biotic integrity produced results that could be interpreted in terms of the relative water quality between sites. Sites draining nearly 100% cropland had the lowest scores for two individual metrics of the index of biotic integrity that were related to species tolerance. Effective monitoring of water quality could be achieved by coupling methods that address both the ecological components of fish communities and their statistical relationships to environmental factors.