INTERAGENCY FIELD MANUAL FOR THE COLLECTION OF WATER-QUALITY DATA

Collecting Water-Quality Samples

Sampling Methods
Clean Hands/Dirty Hands Techniques
Metal and Trace Element Sampling
Grab Sampling
Cross-Sectional Sampling
Single Vertical at Centroid of Flow Method
Equal-Discharge Increment Method
Equal-Width Increment Method
Processing Cross-Sectional Samples
Quality-Assurance Plans and Quality-Control Samples

3. Summary of grab sampling method and preservation, storage, and handling requirements
4. Summary of cross-sectional depth-integrated sampling method and preservation, storage, and handling requirements

The following sections describe grab sampling and cross-sectional sampling as currently implemented by various State and Federal water-quality monitoring agencies. The protocols most widely accepted at this time, especially when using the parts-per-billion analytical levels, require the use of clean sampling procedures. These sampling procedures help to reduce (to the extent feasible given current resources) the amount of contamination introduced when collecting water-quality samples in the field. "Clean" sampling procedures involve (1) using equipment that is constructed of noncontaminating materials and that has been cleaned rigorously before field work and between field sites; (2) handling equipment in a manner that minimizes contamination; (3) collecting, processing, and handling samples in a manner that prevents contamination; and (4) routinely collecting quality-control (QC) samples.

"Clean" sampling procedures, including CH/DH techniques, are required when collecting inorganic samples for metals and other trace elements. Clean sampling procedures are recommended for all other sampling, to the extent that is reasonable, but particularly when the target analyte could be subject to contamination from field or laboratory procedures at a level that could exceed DQOs for reporting and interpretation. CH/DH techniques separate field duties and dedicate one individual as "clean hands" to tasks related to direct contact with the sample. These techniques are summarized below:

1. CH/DH techniques require two or more people working together.
2. At the field site, one person is designated as "clean hands" (CH) and a second person as "dirty hands" (DH). Although specific tasks are assigned at the start to CH or DH, some tasks overlap and can be handled by either as long as contamination is not introduced into the samples.
3. Both CH and DH wear appropriate noncontaminating, disposable, powderless gloves during the entire sampling operation and change gloves frequently, usually with each change in task (wearing multiple layers of gloves allows rapid glove changes).
4. CH takes care of all operations that involve equipment that comes into contact with the sample; for example, CH:

• Handles the surface-water sample bottle.
• Handles the discharge end of the surface-water sample tube or line.
• Transfers the sample to the churn/cone splitter.
• Prepares a clean workspace (inside vehicle).
• Sets up the processing and preservation chambers.
• Sets the equipment (for example, the sample bottles and the filtration and preservation equipment) inside the chambers.
• Works exclusively inside the chambers during collection, processing, and preservation.
• Changes the chamber covers as needed.
• Sets up the field-cleaning equipment and cleans the equipment.

5. DH takes care of all operations that involve contact with potential sources of contamination; for example, DH:

• Works exclusively exterior to the processing and preservation chambers.
• Prepares and operates the sampling equipment, including the pumps and discrete samplers, peristaltic pump switch, pump controller, and manifold system.
• Operates the cranes, tripods, drill rigs, vehicles, or other support equipment.
• Handles the generator or other power supply for samplers.
• Handles the tools, such as hammers, wrenches, keys, locks, and sample-flow manifolds.
• Handles the single or multiparameter instruments for field measurements.
• Handles the churn carrier, including the outer protective bags.
• Handles the stream-gaging or water-level equipment.
• Sets up and calibrates the field-measurement instruments.
• Measures and records the water levels and field measurements.

The following field practices are recommended when sampling for metals and trace elements:

1. Think contamination: be aware of and record the potential sources of contamination at each field site.
2. Wear appropriate noncontaminating, disposable, powderless gloves.

• Change gloves before each new step during sample collection and processing.
• Avoid hand contact with contaminating surfaces (such as equipment, coins, and food).

3. Use equipment constructed of materials that are relatively inert with respect to targeted analytes. Metal samplers must be epoxy-coated to prevent trace element contamination.
4. Use only equipment that has been cleaned according to prescribed procedures. See the equipment cleaning procedures in the Preparing for Water-Quality Sampling section of this manual. Sample processing equipment should be kept covered (when not dispensing samples).
5. Field-rinse the equipment only as directed. Some equipment for some analytes should not be field-rinsed.
6. Use correct sample-handling procedures:

• Minimize the number of sample-handling steps.
• Use CH/DH techniques as required for parts-per-billion trace element sampling.
• Adapt CH/DH techniques for all sample types, as required to obtain data of known quality.
• Train for and practice field techniques under supervision before collecting water samples on your own.

7. Collect and process samples in a clean enclosure such as a dedicated water-quality field vehicle or field processing chamber. Metallic objects, dirt, oil residue, engine exhaust, and food can all be sources of contamination. Simple processing chambers can be fashioned from a polyvinyl chloride (PVC) framework and a clear plastic bag.
8. Filter samples for dissolved trace elements and metals as soon as practical after collection. Use a disposable, tortuous path, capsule filter (effective pore size of 0.45 µm). The USEPA uses a Gelman Supor, model no. 12175 (15-mm diameter or larger), or an equivalent model. A variable speed, battery-operated pump fitted with a peristaltic pump head that forces the sample through Tygon^TM or Teflon^TM tubing is recommended. Filtered samples should be preserved with (1+1) ultra-pure nitric acid (HNO₃) to a pH of 2.0 or less. Normally, 3 mL of (1+1) acid per liter should be sufficient to preserve the sample. Ultra-pure nitric acid is available in 2-mL polypropylene vials.
9. Collect a sufficient number of appropriate types of QC samples. QC samples should be reviewed to determine if cleaning procedures are sufficient and contamination has been minimized.
10. Follow a prescribed order of sample collection and processing.

Specific details regarding grab and depth-integrated sample-collection methods, preservation, storage, and handling requirements are summarized in tables 3 and 4. The basic techniques of obtaining the water samples are outlined in the grab sampling and cross-sectional sampling sections that follow.

Grab samples are indicated in table 2 for DQOs I, II, and III. A grab sample, as explained in an earlier section, is collected in an open container from a single point at or near the stream/river/lake/reservoir surface. Grab samples can be collected with a suspended or hand-held polypropylene (Nalgene^TM) 5-gal container, disposable bailer, or narrow, open-mouth bottle. If the grab sample is collected by hand-held methods, the sample collector should wade to where the sample will be collected (preferably at the centroid of flow or mid-channel) and immerse a hand-held narrow-mouth bottle. Water samples should be collected before any other work is done at a site. If other work (for example, sediment-sample collection, flow measurement, or biological/habitat assessment) is done before the collection of water samples, a representative sample will be difficult to collect from a disturbed stream. The sample collector should stand downstream of the bottle while it is being filled. Care must be taken to avoid collecting particulates that are resuspended as the result of wading. Various examples of grab samples include dip, discrete, and pump samples. Dip samples typically are collected by dipping the collection container (of appropriate noncontaminating material) into the upper layer of the water body. Discrete or point samples are collected by either (1) lowering a sampler to a specified depth and then collecting a sample by opening and closing the sampler or (2) using a single-stage sampler, which fills when stream stage rises to a predetermined height. Thief-type samplers and some pumps are the samplers most often used to collect samples by method 1. Although these samplers are designed primarily to sample still waters, they can be adapted to slowly flowing water. Single-stage samplers used in method 2 include the U-59 and are useful at stations on flashy streams or at other locations where it is difficult to reach a station to manually collect samples. Pump samples typically are collected with suction lift or submersible pump systems designed to collect water-quality samples. Pump systems can be portable or permanently installed and automated for sampling.

For a routine water-quality sample where near-surface water is representative of the water mass, a water sample can be collected by directly immersing the container beneath the water surface to a depth of 1 ft. A bucket can be used to collect a sample if the mixed surface layer is very shallow or accessible only from a bridge. If a bucket is used, extreme care should be taken to avoid contaminating the sample with debris from the rope and bridge. Care must be taken also to rinse the bucket between stations. In slow-moving rivers, reservoirs, and estuaries, the depth of the mixed layer can be determined from field measurements by locating the thermocline or an abrupt change in specific conductance. In tidally-influenced water bodies, the mixed surface layer is defined as that part of the water column from the surface to the depth at which the specific conductance is 6,000 µS/cm greater than the specific conductance at the surface. For mixed surface-layer samples (depth greater than 1 ft), pre-rinse one of the following sampling devices at least once with native water before using: submersible pump tube, Kemmerer, or Van Dorn. A minimum volume of 3 L should be collected from each site. Sample containers do not have to be rinsed with site water. Care should be taken at all times during sample collection, handling, and transport to prevent exposure of the sample to direct sunlight. Information on sample processing, preservation, and holding times for the grab sampling method, including information on dissolved metals in water, is listed in table 3.

Cross-sectional sampling (used in this manual to denote cross sectionally integrated, flow-weighted composite) is achieved by using depth-integrating, nozzled samplers that fill isokinetically. An isokinetic sampler operates in such a way that the water-sediment mixture moves into the sampler with no change in speed and direction (velocity) as the water enters the sampler intake. This sampling method ensures that the sediment concentration in the water-sediment mixture in the sampler and the sediment concentration in the stream are equal. An isokinetic sampler is lowered through a vertical (the center of each increment or part of the stream cross section) at a predetermined transit rate. The transit rate (the rate at which the sampler is lowered and raised) is mainly a function of the sampler nozzle diameter, volume of sampler container, stream velocity, and sampling depth. The transit rate must be kept constant during sampler descent and ascent through a vertical. Three different sampling methods based on the use of isokinetic samplers are the single vertical at centroid of flow (VCF) method, the equal-discharge increment (EDI) method, and the equal-width increment (EWI) method. These methods will be summarized in this section. The centroid of flow is the point in the increment at which discharge is equal on both sides. The number of increments needed to collect a discharge-weighted sample at a site is related primarily to how stringent the DQOs are and how well mixed or homogenous the stream is with respect to the physical, chemical, and biological characteristics (variation) of the cross section. Information on sample processing, preservation, and holding times for the cross-sectional depth-integrated sampling method is listed in table 4.

The VCF method uses one centroid of flow for the stream cross section, and therefore, only one vertical is sampled. Consequently for appropriate use of this method, the cross section should be well mixed vertically and laterally with respect to concentrations of target constituents. The following steps should be followed to use the VCF method:

1. Measure discharge along the cross section to be sampled. Space the verticals to provide about 25 to 30 subsections or increments.
2. Locate the centroid of flow (the point in the stream where the discharge is equal on both sides) directly from the discharge measurement sheet OR
3. Construct an equal-discharge increment (EDI) curve by plotting cumulative discharge or cumulative percentage of discharge against cross-section width (in feet from right streambank). The centroid of flow (as determined by the EDI curve) would be located at the cross-section width where 50 percent of the discharge crosses the plotted line.
1. If the stream channel is stable at the cross section to be sampled, then EDI curves of cumulative discharge at various stages can be based on historical discharge measurements. The location of centroids can be determined from these EDI curves so that discharge measurements do not have to be made before each sampling. EDI curves require occasional verification by comparison to recent discharge measurements.
2. Examine the cross section for uniformity of appearance.
4. Measure the cross-sectional variation of field measurements (temperature, pH, DO, and specific conductance) at sites that have sampling history. Generally these field parameters should be measured at no less than 10 evenly spaced increments about 1 ft below the water surface. Record and review variations along the cross section. If values of field measurements differ by less than 5 percent and show that the stream is well mixed both along the cross section and from top to bottom, a single vertical of flow can be used.
5. Evaluate data from steps 1-4 to decide if the VCF method is appropriate. If discharge, field-measurement, or chemical-analysis data do not confirm that the cross section is well mixed vertically and laterally, use either the EDI or the EWI method.
6. If the VCF method is appropriate, go to the "Equal-Discharge Increment Method" section and follow step 3 to select the transit rate and step 4 to collect samples.

The objective of the EDI method is to collect a discharge-weighted sample that represents the entire flow passing through a cross section by obtaining a series of samples, each representing equal volumes of stream discharge. The EDI method requires that flow in the cross section be divided into increments of equal discharge. Equal-volume samples are collected at the centroid of each of the EDIs along the cross section. Samples are collected by passing the sampler through a vertical located at the centroid of each EDI. The sample collector should (1) use isokinetic, depth-integrating sampling equipment (Appendix D), (2) use the same size sampler and nozzle at each of the verticals, (3) use the same constant transit rate during each ascent and descent in verticals, and (4) composite the samples from all verticals with the appropriate compositing equipment.

The following steps summarize the EDI method:

1. Prepare the field site.
1. Upon arrival, set out safety equipment such as traffic cones and signs. Park the vehicle in a location and direction that will prevent sample contamination from vehicle emissions.
2. Assemble the equipment needed and set up a clean workspace.

• To collect organic compound samples, use fluorocarbon polymer, glass, or metal for equipment components that are in direct contact with samples to be analyzed for organic compounds. Do not use plastics other than fluorocarbon polymers.
• To collect inorganic constituent samples, use fluorocarbon polymer or other relatively inert and uncolored plastics or glass for equipment components that are in direct contact with samples to be analyzed for inorganic constituents. Do not use metal or rubber components for trace element sampling.
• Collect bacteria samples (if required) using the filtration method described in the Field Folders/Field Measurements section of this manual.
• Calibrate field instruments as recommended in instrument operations manuals.

2. Select the number and distribution of EDIs.

   The number of EDIs selected for a sampling site is governed by factors described in the following steps and should not be determined arbitrarily.

1. Visually inspect the stream from bank to bank and longitudinally, observing velocity, width, depth distribution, and apparent sediment and aquatic biota distribution along the cross section. Note and document the locations of stagnant water, eddies, backwater, reverse flows, areas of faster-than-normal flow, and piers or other obstructions along the cross section.
2. Determine the stream width from a tagline or from distance markings on a bridge railing or cableway.
3. At sites that have little sampling history: measure, record, and review the cross-sectional variation of field measurements (temperature, pH, DO, and specific conductance). Review the magnitude of the variations along the cross section.
4. Measure discharge along the cross section for sampling or use existing EDI curves drawn from historical discharge measurements (if available).
5. Determine the volume of discharge that will be represented in each EDI on the basis of DQOs for the study; the variation in discharge, field measurements, and stream-channel characteristics along the cross section; and the volume of sample required for analyses of target constituents.
6. Divide the cross section into EDIs.

• When determining the number of increments to be sampled, remember that the sample collected at the centroid of flow within each EDI must represent the mean discharge measured for that increment. If the mean discharge for that increment is not represented, the number of increments should be increased by decreasing the volume of discharge in each EDI until the mean discharge is represented.
• As a guide, a minimum of four sampling increments is recommended; the maximum number of increments is nine.
7. Locate of the centroid of flow within each EDI from discharge measurements as follows (the centroid of flow also can be located from previously constructed curves, as described later in the TECHNICAL NOTE):
• Construct a curve by plotting cumulative discharge or cumulative percentage of discharge against cross-section width, or
• Determine EDI locations using the discharge measurement sheet (produced in step 2d above) by following steps i-iv.
• Example:
1. If the stream cross section will be divided into five increments, divide stream discharge (determined in step 2d above) by 5 to determine the EDIs. For example, each increment equals 20 percent of discharge (100 percent/5 = 20 percent), thus equal increments of discharge.
2. Locate the centroid of flow (the point at which discharge is equal on both sides) of the initial increment where cumulative discharge equals one-half the EDI (for example, 20 percent/2 = 10 percent). This also is the location of the first vertical at which a sample is collected.
3. Locate each remaining (four in this example) centroid of flow by adding the next incremental discharge to the discharge at the previously sampled centroid and determining where that cumulative discharge occurs along the cross section.
4. The EDI centroids will correspond to locations along the cross section where calculations in iii above indicate 10, 30, 50, 70, and 90 percent of the cumulative discharge along the cross section.

TECHNICAL NOTE: If the stream channel is stable at the cross section to be sampled, graphs of cumulative discharge or percentage of cumulative discharge at various stages can be based on historical discharge measurements. Location of EDI centroids can be determined from these graphs so that discharge measurements do not have to be made before each sampling. Linear interpolation based on discharge can be made between EDI curves. However, these EDI curves require occasional verification by comparison to recent discharge measurements.

3. Select the transit rate.
1. Determine the sampling depth and the mean stream velocity at the centroid of each EDI.
2. Determine the actual transit rate for each centroid that will yield samples of approximately the same volume, using sampling depth, mean stream velocity, and information in Appendix F.

Guidelines for selecting the EDI-sampling transit rate

• Collect samples of equal volumes at each centroid. This is required for the EDI method if the sample will be composited. Generally, to collect equal volumes, transit rates will vary from centroid to centroid.
• Keep the transit rate equal and directionally constant and within the isokinetic transit-rate range of the sampler (Appendix F) when collecting isokinetic samples at each centroid.
• DO NOT EXCEED the maximum transit rate. The maximum transit rate will have been exceeded if the minimum sample volume for the selected nozzle, bottle, and stream velocity is not collected. Exceeding the maximum transit rate will affect the concentration of particulates greater than or equal to 0.062 mm.
• Use the slowest transit rate possible, but make sure that the minimum transit rate is sufficiently rapid to keep the sampler from overfilling. (The sampler is overfilled when the water level in the bottle or bag is above the bottom of the nozzle when the sampler is held in the horizontal sampling position.)

4. Collect samples (the procedures are the same whether the sample collector is wading or using a reel-and-cable suspension method).

• Collect samples at EDI centroids as many times as necessary to ensure collection of sufficient sample volume for analysis if the sample is to be composited. For composite samples, care must be taken to obtain the same total volume from the vertical at each EDI centroid if more than one traverse is made at each centroid. This ensures that the composited cross-sectional sample will be proportional to flow at the time of sampling.
• Remember to stay within the isokinetic transit-rate range of the sampler at each vertical. If stream velocity is less than the isokinetic transit-rate range of the sampler, a discharge-weighted sample still can be obtained by collecting equal volumes at each vertical; however, this sample will not be isokinetic.
1. Read and record the starting gage height from the staff gage or wire weight if available. Instantaneous discharge (relative to gage height at streamflow-gaging stations) routinely is stored with water-quality data to calculate loads. Move the sampling and support equipment to the centroid of the first increment to be sampled. Field-rinse the sampling equipment and record the sampling start time.
2. Lower the sampler at the predetermined constant transit rate until a slight contact is made with streambed.
• Do not pause upon contact with the streambed. Raise the sampler immediately at a constant transit rate to complete the vertical traverse. The descending transit rate does not have to equal the ascending transit rate, but each rate must be equal and directionally constant and within the isokinetic transit-rate range of the sampler.
• Take care not to disturb the streambed by bumping the sampler onto it; bed material might enter the nozzle, resulting in erroneous data.
• Ensure that the sampler is not overfilled. Overfilling results in enrichment of the sample with heavy particulates because of secondary circulation of water through the sampler (from the nozzle through the air exhaust). This enrichment could result in an artificially increased suspended-sediment concentration and bias particle-size distribution toward heavier and larger particulates.
3. Inspect each bottle or bag-type sampler, looking for overfilling and the presence of anomalously large amounts of particulates that might have been captured because of excessive streambed disturbance during sample collection. If either of these conditions is noted, discard the sample (make sure no residual particulates are left in the container) and resample.
4. Ensure that the sampler is not underfilled. Underfilling results in a sample that is not isokinetically collected because the maximum transit rate has been exceeded. The minimum volume must be collected to ensure that the sample is isokinetic.
5. Depending on DQOs, process and (or) analyze the samples collected at each vertical as either a separate sample, as a composite with other subsamples collected along the cross section, or split the samples for further processing. If the total volume of samples to be collected will exceed the operational capacity of the churn splitter, decrease the number of increments or use an appropriate sampler with a smaller bottle or with a bag with a smaller nozzle. When transferring the sample from the sampler bottle or bag to the splitter, ensure that all particulates in the sampler are transferred with the water.
• Remove the sampler cap.
• Swirl the sample gently to keep the particulates suspended.
• Quickly pour the sample into the sample splitter.
6. Move the equipment to the next vertical.
• Determine the transit rate for this vertical that will yield the same volume as for the previous EDI centroid.
• Repeat the procedures in steps 4b through 4e.
• Repeat this procedure at the remaining verticals along the cross section.
7. Record the following information after all samples have been collected:
• Sampling end time.
• Ending gage height.
• All field observations and any deviations from standard sampling procedures.

5. Process the samples as appropriate. This step includes removing necessary volumes of water from the compositing device for filtering or dispensing into the required sampler bottles. After filling the bottles specified for the scheduled analyses, add the necessary preservatives and package the bottles appropriately for shipment to the laboratory.
6. Clean the equipment as appropriate.

• If the sampler will not be reused during a field trip, rinse the components with DIW before they dry and place the sampler in a plastic bag for transporting back to the office laboratory for cleaning.
• If the sampler will be reused, clean it at the sampling site while it is still wet.
• Collect a field blank, if required, after the sampling equipment has been cleaned in the field.

For the EWI method, the stream cross section is divided into a number of EWIs. Samples are collected by lowering and raising the sampler through the water column at the center of each EWI. The combination of the same constant transit rate used to sample at each vertical and the isokinetic characteristic of the sampler results in a discharge-weighted sample that is proportional to total streamflow.

1. Prepare the field site
1. Upon arrival, set out safety equipment such as traffic cones and signs. Park the vehicle in a location and direction that will prevent sample contamination from vehicle emissions.
2. Assemble the equipment and set up a clean workspace.

• To collect organic compound samples, use fluorocarbon polymer, glass, or metal for equipment components that are in direct contact with samples to be analyzed for organic compounds. Do not use plastics other than fluorocarbon polymers.
• To collect inorganic constituent samples, use fluorocarbon polymer or other relatively inert and uncolored plastics or glass for equipment components that are in direct contact with samples to be analyzed for inorganic constituents. Do not use metal or rubber components for trace element sampling.
• Collect bacteria samples (if necessary) using the filtration method described in the Field Folders/Field Measurements section of this manual.
• Calibrate field instruments as recommended in instrument operations manuals.

2. Select the number and width of EWIs.
1. Visually inspect the stream from bank to bank and longitudinally, observing velocity, width, depth distribution, and apparent sediment and aquatic biota distribution in the cross section. Note and document the locations of stagnant water, eddies, backwater, reverse flows, areas of faster-than-normal flow, and piers or other obstructions along the cross section.
2. Determine the stream width from a tagline or from distance markings on a bridge railing or cableway.
3. At sites that have little sampling history, measure, record, and review the cross-sectional variation of field measurements (temperature, pH, DO, and specific conductance). Review the magnitude of the variations along the cross section.
4. Determine the width of the increment or the distance between verticals. To determine the width of each equal increment, divide the stream width by the number of verticals necessary to collect a representative sample. The number of increments must be a whole number. Increment width is based on DQOs for the study and on the variation in discharge, field measurements, and stream-channel characteristics along the cross section. For all but very wide and shallow streams, a minimum of 10 and a maximum of 20 verticals usually is sufficient.

• Collect the sample at the center of each EWI (the vertical). This sample should approximately represent the mean discharge for that increment.
• If the sample does not appear to represent the mean discharge for that increment, decrease the width of the increment until the mean discharge is approximate. This will increase the number of increments sampled.
• Locate the verticals at the center of each remaining EWI along the cross section.
• Make slight adjustments to sampling locations, if necessary, to avoid sampling where the flow is affected by a pier or some other obstruction.
• Example:
1. If a stream 56 ft wide has been divided into 14 increments of 4 ft each, the first sampling vertical would be 2 ft from the water's edge and subsequent verticals would be at 6 ft, 10 ft, 14 ft, and so forth.
2. Even if streamflow is divided, as in a braided channel, EWIs must be identical from channel to channel, and the same constant transit rate must be used at each vertical.

3. Select the transit rate.
1. Refer to Appendix F for guidelines for determining the isokinetic transit rates for collecting isokinetic, depth-integrated samples. Unless the mean velocity is actually determined, use the trial-and-error method to determine the minimum transit rate.
2. Locate the EWI that has the largest discharge (the largest product of depth times velocity) by sounding for depth and either measuring or estimating velocity. At the vertical for this increment, use of the minimum transit rate results in the maximum allowable filling of the sampler bottle or bag during one vertical traverse.
3. Determine the minimum transit rate at this vertical for the size of the sampler nozzle and bottle or bag that will be used.
4. Approximate the mean velocity (in feet per second) of the vertical by timing a marker as it travels a known distance. Divide the distance (in feet) by the time (in seconds) and multiply by 0.86.
5. Make sure that the transit rate does not exceed the maximum allowable transit rate to be used at any of the remaining verticals along the cross section. This can be determined by sampling the slowest and shallowest increment with any substantial flow. If the minimum sample volume is not collected at this vertical, then the EWI method cannot be used at this cross section to collect a discharge-weighted sample.

• The descending and ascending transit rates must be constant in each direction and must be the same at each vertical along the cross section.
• The maximum transit rate must not be exceeded at any of the verticals.
• The minimum transit rate must be sufficiently rapid to keep the sampler from overfilling. (The sampler is overfilled when the water level in the bottle or bag is above the bottom of the nozzle when the sampler is held in the horizontal sampling position.)
• The same size sampler nozzle and bottle must be used at all verticals along the cross section.
• The total volume collected must not exceed the recommended volume for the churn splitter used to composite the sample.
• If the transit rate exceeds the maximum allowable rate, use the EDI method instead of the EWI method.

4. Collect samples (the procedures are the same whether the sample collector is wading or using the reel-and-cable suspension method; the same sampler bottle or bag can be used for all verticals in the cross section).
1. Move to the first vertical (the midpoint of the first EWI near the edge of the water) and field-rinse the equipment.
2. Record the start time and gage height.
3. Lower the field-rinsed sampler at the predetermined constant transit rate until a slight contact is made with the streambed.

• Do not pause upon contact with the streambed. Raise the sampler immediately at a constant transit rate to complete the vertical traverse.
• Take care not to disturb the streambed by bumping the sampler onto it; bed material might enter the nozzle, resulting in erroneous data.
• Ensure that the sampler is not overfilled. Overfilling results in a sample that is not isokinetic and that could be enriched with heavy particulates because of secondary circulation of water through the sampler (from the nozzle through the air exhaust). This enrichment could result in an artificially increased suspended-sediment concentration and bias particle-size distribution toward heavier and larger particulates.
• Ensure that the sampler is not underfilled. Underfilling results in a sample that is not isokinetically collected because the maximum transit rate has been exceeded. The minimum volume must be collected to ensure that the sample is isokinetic.
• If the required volume cannot be collected, use the EDI method for discharge-weighted samples.
4. Inspect each bottle or bag-type sampler, looking for overfilling and the presence of anomalously large amounts of particulates that might have been captured because of excessive streambed disturbance during sample collection. If either of these conditions is noted, discard the sample (make sure no residual particulates are left in the container) and resample.
5. Move the sampling equipment to the next vertical. Maintain the established transit rate. The sample volume can vary considerably among verticals. Samples can be collected at several verticals before emptying the sampler, as long as the maximum sample volume in a bottle or bag-type sampler has not been exceeded.
6. Continue to the next vertical until no more samples can be collected without overfilling the sampler. Empty the sampler into a field-rinsed compositor or splitting device and repeat the sample collection procedure until samples have been collected at all verticals.
• If the total sample volume to be collected will exceed the operational capacity of the churn splitter, use the EDI method.
• When transferring the sample from the sampler bottle or bag to the splitter, ensure that all particulates in the sampler are transferred with the water. Remove the sampler cap. Swirl the sample gently to keep the particulates suspended and quickly pour the sample into the sample splitter.
• Collect as many samples at the EWI verticals as necessary to ensure that an adequate sample volume is collected as required for analysis, but sample at each vertical an equal number of times. (The composite cross-sectional sample will remain proportional to flow at the time of sampling.)
7. Record the following information after all samples have been collected:
• Sampling end time.
• Ending gage height.
• All field observations and any deviations from standard sampling procedures.

5. Process the samples as appropriate.
6. Clean the equipment as appropriate.

• If the sampler will not be reused during a field trip, rinse the components with DIW before they dry and place the sampler in a plastic bag for transporting back to the office laboratory for cleaning.
• If the sampler will be reused, clean it at the sampling site while it is still wet
• Collect a field blank, if required, after the sampling equipment has been cleaned in the field.

Cross-sectional samples typically are composited to produce a water-quality sample that is representative of the total stream discharge at the sampling station. This compositing can be accomplished by depositing each sample volume from each vertical into a churn splitter -- a polyethylene container that slowly stirs the composited sample with a polypropylene disk. Because the churn splitter requires 3 to 8 L of composited water, verticals in a narrow stream may need to be sampled more than once. All verticals should be sampled the same number of times to retain the representativeness of the sample to the stream. It is also important to churn while drawing sample aliquots from the splitter. Do not use a churn splitter to composite samples collected for VOCs, organic carbon, oil and grease, pesticides, herbicides, or bacteria, because the plastic components have the potential for adsorbing and contaminating the samples. Instead use glass containers for sampling these constituents with the grab-sampling method.

The choice of using the parts-per-million/milligrams-per-liter protocol or the parts-per-billion/micrograms-per-liter protocol is based on the minimum reporting level (MRL) required in the sampling plan. Most analytical work and most Federal drinking-water regulations are reported in micrograms per liter. The parts-per-billion protocol with its extremely low MRLs requires certain field practices, such as the CH/DH technique, appropriate cleaning/handling of sampling equipment, and rigorous and appropriate QC sampling and planning to ensure minimum contamination occurs during the entire water-quality sampling and processing procedure.

Some constituents require different sampling methods than others. Always check with the laboratory that will perform the analyses about container type and volume, preservation, and holding times. Clear descriptions of selected constituents, preservatives, and holding times should be included in the sampling plan before samples are collected. The current properties and constituents, preservatives, and holding times for both grab and depth-integrated sampling methods are listed in tables 3 and 4.

Samples should be wrapped and packed in containers so that they are received unbroken at the laboratory. Record all identification numbers on the field notes or field logbook. Communicate with the laboratory about the sample delivery time and method of shipment. Follow necessary procedure, complete the required forms, and retain copies so that samples can be tracked after shipment.

Assuring the validity of environmental data or QA is accomplished by collecting QC samples, reviewing and analyzing QC data, and making adjustments to data-collection procedures on the basis of the results. Because environmental sampling is done in a relatively uncontrolled manner subject to many variables, large sources of error may exist. Therefore, measures of control over the collection of water-quality samples are crucial for answering any questions regarding the validity of the data. QA plans should contain planned and systematic procedures necessary to provide confidence that the data will satisfy established requirements for quality. Detailed information about QA/QC samples can be found throughout the literature and within specific agencies responsible for water-quality data collection. This section gives a review of the types of QC samples commonly required in QA plans.

Blank solutions or "blanks" are solutions (DIW or rinse water) that are laboratory certified to have target constituent concentrations that are less that the method detection limits. These solutions are used to develop specific types of blank samples. Equipment blanks are blank solutions that are processed through all equipment used for collecting and processing an environmental sample. Once analyzed, the blanks indicate the effectiveness of cleaning of field equipment. Equipment blanks should be collected after sampling the station with the highest contamination. Field blanks are containers of DIW that are subjected to the same aspects of sample collection, field processing and preservation, transportation, and laboratory handling as environmental samples. Field blanks are used to check for contamination in the laboratory and for cross contamination during the collection and shipment of samples. Trip or travel blanks are containers of DIW that are put in the same type of bottle as used for environmental samples and are kept with the set of sample bottles both before and after sample collection. These blanks are used to detect contamination that occurs during sample transport or storage or in the laboratory. Replicate samples are used to assess the performance in those parts of the procedure that are replicated. Replicate samples are collected in a manner such that the samples are thought to be essentially identical in composition. Many types of replicate samples are possible, and each might yield slightly different results in a dynamic hydrologic setting. Split replicates sent to separate laboratories are used to ensure that results from the different laboratories are comparable. Split replicates sent to the same laboratory are used to measure the variability of the laboratory. Concurrent or sequential replicates are used to ensure that samples collected with different samplers or by different sampling methods are comparable. Spike samples are samples that have known concentrations of specific constituents added in such a manner as to minimize the change in the matrix of the original samples. Spike samples are used to verify the method performance for either accuracy or recovery. Accuracy data reflect the best results that can be expected at the time the samples were analyzed, and recovery data reflect bias from an environmental sample matrix.

Most QA plans and QC samples take into account the following:

• Equipment blanks should be collected annually or whenever new equipment is used.
• Field conditions could require blanks or other QC samples not previously planned for.
• All field QC samples should be collected on the same day that environmental samples are collected. The same equipment should be used for both types of samples.
• Preservatives from the same lot number should be used for the environmental and associated samples. The preservative lot number should be recorded.
• QC samples should be labeled appropriately.
• Results of QC analyses should be stored in a separate database.

Contact Dee Lurry for more information