Table 9221 Water Quality Parameters Sampled In Urban Drainage Studies


Detection Limits

Precision Level (absolute or relative)

Study Objectives or Observations

Common Constituents and Indicators


Water temperature Conductivity (at 20°C)



Kjeldahl nitrogen Total phosphorus Ammonia

Nitrites and nitrates

Organic Indicators

Trace Elements


Zinc and other heavy metals


VSS (at 550°C) Settleable solids

Bacterial Indicators

Total coliforms

Fecal coliforms

Special Parameters

Persistent toxic substances (PTS) such as organochloride pesticides Polyaromatic hydrocarbons Chlorinated benzenes


Impact of salts used for deicing Cross-connections; parasites in waters

Changes during runoff, monitoring and control Rainfall quality analysis Sediment transport

Impact on receiving waters Eutrophication process Impacts on detention basins with recreational purposes Cross-connections

Impact on receiving waters by oxygen depletion Cross-connections

Impact on receiving waters; toxics accumulation in sediments

Turbidity, oxygen reduction, transport of toxics; increase of hydraulic roughness Organic part, oxygen depletion Maintenance problems in sewers and detention basins in recreational areas

Impact on receiving waters with recreational use Detection of cross-connections

Impact on receiving waters Pollution of receiving waters sediments Bioaccumulation in food chains

^Depending on the instrument and/or analysis method. tDepending on the substance analyzed.

tential remains quite high because of variations in pollutant concentrations during runoff events.

Two basic methods can provide estimates of pollutant loading during a storm event:

1. To determine total pollution loading during a storm event, a flow-weighted composite method is adequate. In these methods, either aliquot volume or time between aliquots is varied to construct a truly flow-weighted composite from many samples. Analyzing the composite sample and using synoptic flow data allow computation of an accurate estimate of runoff pollution loads, if the intervals between samples are short.

2. When, in addition to total pollution loading, it is necessary to investigate load variations during a storm event, the sequential discrete procedure must be used. A

series of samples is retrieved during a monitored runoff event. Following laboratory analysis of each sample and analysis of synoptic flow data, the runoff hydrograph and a curve of pollutant concentration or loading as a function of time may be plotted as shown in Figure 9.22.1. By determining the area under the curve, an accurate estimate of the total pollutant load for an event may be determined.

The interval between sample collection for the above procedures depends on the response time and duration of the storm. In general, at least four samples on the rising limbs and six samples on the recessing limbs should be collected for proper resolution of nonpoint source pollution loads in urban areas.

Samples may be collected either manually or by automatic samplers. Table 9.22.2 shows a matrix of advantages and disadvantages related to each sampling technique. A summary of methods used in urban stormwater sampling and comments on each was prepared by Shelley (Shelley and Kirkpatrick 1975).

Experimental results show sediment distribution in a stream cross section flowing at 5 ft/sec. An analysis of water quality constituents in the stream cross-section should be made to determine the distribution across the width and from top to the bottom of the stream. Samples should be tested for a suspended parameter (such as TSS) and a soluble parameter (such as orthophosphate). The testing should be carried out at a small runoff event and a moderate-to-high flow event if possible. Vertical sampling should be done using depth samplers (such as Kenmeyer bottles) or closeable bottles if the stream is more than 4 to 5 ft (1.2 to 1.5 m) deep. This factor should be considered in designing manual and automatic sampling procedures.

0 0

Post a comment