8226

• UCREFRP Catalog Number: 8226
• Author: Shelton, L. R.
• Title: Field Guide for Collecting and Processing Stream-Water Samples for the National Water-Quality Assessment Program.
• USFW Year: 1994.
• USFW - Doc Type: 94-445,
• Copyright Material: NO

COLLECTION METHODS

Proper sampling techniques must be used to ensure that a sample is representative of the flow in the
cross section. A discussion of sampling techniques is presented in reports by Edwards and Glysson (1988)
and Ward and Hair (1990). Some aspects of sampling also are included in other USGS TWRIs, OWQ
technical memorandums (see list of references, appendix A), and in the recommended methods for water-
data acquisition (U.S. Geological Survey, 1978). A discharge measurement should be made prior to
sampling if a rated discharge is not available. .

Collect samples at the same cross section throughout the period of record, if possible. This will
eliminate many of the potential problems that might arise during the interpretation of water-quality data.
For example, measuring streamflow in a pool and sampling in a nearby riffle might prevent use of the
hydraulic information to compute constituent transport. Sand may move through the pool as bedload and
through the riffle as suspended load. This does not mean that the same section used during the low-water
wading stage must be used during higher stages that require the use of a bridge or cableway. However,
the flow characteristics at the different cross sections can result in incomparable data if the cross sections
are not located near each other or in the same flow regime.

The number of verticals sampled at a site should be based primarily on the requirement to collect a
sample representative of cross-sectional chemistry and secondarily to obtain the volume of the sample
required. Samples usually should be collected using a standard multivertical depth-integrating method to
obtain the most representative isokinetic sample possible. However, abbreviated sampling methods (that
is, weighted-bottle or dip sample) are sometimes the best procedures for collecting a sample representative
of the stream chemistry. Single vertical, dip, or other point-sampling methods can be used when the
cross-sectional transport characteristics of the site are documented adequately or extreme flow conditions
exist that preclude the use of standard methods. The Telfon bag sample might not provide a sample
representative of stream hydrodynamics; however, it can collect a representative noncontaminated sample
in deep or fast moving streams. Considering the limits of the other samplers, the D-77 TM might be the
most appropriate sampler under many conditions, even when used beyond its limits. All samples collected
by nonstandard methods should be checked periodically against standard cross-sectional samples to
develop correction coefficients for the data.

Prior to initial sampling at a site, and again 3 to 4 times per year, obtain a stream profile of field
measurements (velocities, specific conductance, temperature, pH, CUld dissolved oxygen). Record
observations from several verticals and depths in the cross section to determine the uniformity of these
characteristics. These measurements should be used as guides in selecting an adequate number of verticals
for obtaining a representative sample.

The vertical transit rate and operational depth of each sampler is a function of the stream velocity,
sample-container volume, and nozzle size. The following chart gives the recommended vertical transit
rates and the maximum depths for isokinetic sampling based on samplers and nozzles. Specific limitations
of the samplers are in OSW technical memorandum 94.05 (see appendix A).

Nozzle Depth
Sampler diameter Ratio
(inches) (feet)
DH-81 1/4 0.4 9
DH-81 5/16 0.4 6
D-77 TM 1/4 0.1 15
D-77 TM 5/16 0.2 15

Stream velocity x ratio = maximum vertical transit rate

14 Collecting and Processing Stream-Water Samples for the Naflonal Water-Quality Assessment Program