Link to USGS home

USGS-WRD SEDIMENT LABORATORY CHIEFS WORKSHOP

March 23-26, 1998
USGS District Office, Baton Rouge, LA
Meeting Summary (final 10/5/98)

Contact John R. Gray email: jrgray@usgs.gov
LeRoy Schroder email: schroder@usgs.gov

A workshop for chiefs of sediment laboratories operated and/or used by the U.S. Geological Survey (USGS), Water Resources Division (WRD) was hosted by the USGS District Office in Baton Rouge, Louisiana, on March 23-26, 1998. This is a summary of the meeting's significant outcomes gleaned from the minutes of the meeting and from discussions, both formal and informal, that took place during and after the meeting.

The purposes of the workshop were to:

  1. Fulfill the requirements of OSW Technical Memorandum 98.05 and update guidelines as a new memorandum (OSW Technical Memorandum 99.04), which requires that each District or contract laboratory chief attend periodic meetings of the laboratory chiefs,
  2. Present the results of the Sediment Laboratory Quality Assurance (SLQA) project and proposed performance standards in the SLQA that will effect laboratories in the future.
  3. Increase communication and enhance consistency in methods, instruments, and equipment between the sediment laboratories,
  4. Accomplish proficiency training,
  5. Articulate problems, needs, accomplishments and new directions,
  6. Set goals to be addressed before the next sediment laboratory chiefs meeting (approximately 3 years) and recommend a mechanism to achieve those goals.
Jack Kramer, Heidelberg College, along with the following USGS employees participated in most or all of the meeting: Arthur Adams (LA), Dave Funderburg (NM), John George (CO, retired), Dan Gooding (CVO), John Gordon (OWQ-BQS), John Gray (OSW), Bill Johnson (CVO), Marlon Johnson (LA), Cheryl Joseph (LA), Kathy Love (MO), Scott McEwen (CR), Von Miller (IA), Allan Mlodnosky (CA), Raphael Pena (PR), LeRoy Schroder (OWQ-BQS) Clyde Sholar (NR), Libby Shreve (KY), Pam Smith (IA), and Lucille Wright (LA). Of the eleven production sediment laboratories servicing the USGS, all but the USGS laboratories in Montana and Hawaii were represented.

Ed Martin, District Chief, LA, welcomed the group to Baton Rouge, and Wayne O'Neal, Chief, Federal Interagency Sedimentation Project (FISP), MS, provided a briefing on the products and activities of the FISP. The remainder of the workshop was spent in discussions and deliberations regarding sediment laboratory operations with particular emphasis on quality assurance and safety.

Libby Shreve did an outstanding job in keeping meeting minutes. Following is a summary of discussions, recommendations, and conclusions from the meeting.

Sediment Laboratory Safety (principal contact - Sholar): Sediment laboratory safety needs have expanded in recent years. In addition to standard safety requirements consisting of relatively traditional concerns such as chemical storage, proper hood venting, emergency shower facilities, and proper grounding of electrical circuits, current concerns focus on biocontaminants in water samples (such as Phiesteria), and development of a mechanism to notify sediment laboratories of the potential presence of pathogens in samples received by a laboratory.

The workshop's agenda did not permit an exhaustive evaluation of the status of sediment laboratory safety, which administratively falls under the purview of the District and Regional Safety Officers. However, because of its importance, Clyde Sholar, NE Region Safety Officer and former Chief of the Kentucky Sediment Laboratory, agreed to develop a generic plan on newly emerging safety issues to be incorporated in the safety plans of each sediment laboratory.

Workshop Outcome: Clyde Sholar will develop a generic sediment laboratory safety chapter dealing with newly emerging safety issues, and will work with the Division's safety committee to provide the guidelines needed by the laboratories to address emerging safety issues.

Sediment Laboratory Environmental Data System (SLEDS) (principal contact - Gooding): A large majority of workshop attendees favor adopting SLEDS as the WRD's sole, centrally supported sediment laboratory software. This conclusion was based on several factors, including: Concerns were voiced over program "bugs"; costs associated with installing and maintaining SLEDS; and assurance for its long-term support.

Workshop Recommendations:
  1. Adopt SLEDS as WRD's sediment laboratory software if issues concerning bugs, installation and maintenance, and long-term support can be resolved to the satisfaction of the participating sediment laboratories. October 1, 1998, was considered a realistic goal for announcing adoption of SLEDS in an OSW Technical Memorandum (Gray).
  2. A formal SLEDS training session for lab personnel should be held, ideally in FY99.
  3. The OSW should propose a mechanism for funding SLEDS annual support needs, understanding that support of the program is beneficial to the integrity of sediment data collected by the WRD (Gray).
Office of Surface Water Technical Memorandum 98.05, "A National Sediment Laboratory Quality Assurance Program" (principal contacts - Gray, Gordon): The concept of a national sediment laboratory quality assurance program was favorably received by workshop participants.

Clarification: Use of the double-blind standard reference samples by sediment laboratory chiefs for internal laboratory checks is optional.

Resolutions Needed:

  1. A means for providing travel costs for selected sediment laboratory chiefs to participate as understudies in a sediment laboratory review in conjunction with a triennial surface-water review needs to be identified (Gray).
  2. Criteria for designating a sediment laboratory as being "out of control" need to be finalized and circulated (either 2 or 3 f-pseudosigma) (Gordon, Schroder).
  3. Ascertain whether or not constitution of sediment-water mixtures by field personnel as part of the double-blind standard reference sample program will result in large errors unattributable to sediment laboratory performance (Gordon in concert with Lab Chiefs).
  4. Determine how Districts with laboratories would prefer to 'store' funds for payment of travel expenses of a certified sediment laboratory reviewer.

    One suggestion -- starting in FY1999, have the OSW collect from each District with a sediment laboratory about $300 annually (roughly a third of the cost of the certified lab reviewer's travel) regardless of the timing of a District's sediment laboratory review. The OSW, in turn, would pay travel expenses for the certified lab reviewer. There are two advantages here: It spreads the cost of travel over a 3-year period, and it effectively results in "cost equalization." It was suggested to also include costs of SLEDS support in the annual assessment if those costs are not otherwise covered (Gray).
Recommendations:
  1. QC data analyses for a given laboratory will be kept confidential until the sample size is considered adequate, after which all QC data obtained under the single- and double-blind standard reference sample programs will be made public and identified by originating laboratory (Gordon).
  2. The OSW and BQS need to market this national program to other Federal Agencies, and, if appropriate, to foreign sediment laboratories perhaps via Fact Sheet (Gordon, Gray).
Action: Alan Mlodnosky, Chief, California sediment laboratory, and John Gordon, Chief, National Sediment Laboratory Quality Assurance Program, will collaborate on an initial test of the double-blind reference-sample project, and provide insights on potential problems in this program.

Update on Quality Assurance and Quality Control Recommendations from the 1994 Sediment Laboratory Chief's Workshop (principal contacts - Schroder, Shreve): The 1994 recommendations on standard operating procedures (SOP's) were evaluated and updated as part of three breakout groups, and are attached as appendix A.

Libby Shreve provided insights on SOP's -- "Everything else one needs to know that isn't included in the sediment laboratory QA plan."

Recommendation: Formalize the SOP's from the workgroup and those provided by Libby and others, and make the information available to each laboratory for inclusion in the sediment laboratory quality assurance plan, or as an attachment to the plan (Shreve).

World Wide Web Sites for Sediment Laboratories (principal contacts - McEwen, Gordon, Gray): The participants agreed to develop Web sites for their respective laboratories with links to other sediment laboratory web sites and to the OSW's, which is public (the OSW has set up a sediment laboratory Web site at http://water.usgs.gov/osw/technqiues/sedimentlabs.html). Although the Web sites would be unique for each laboratory, a single format would be used to convey a sense of commonality to laboratory customers.

John Gordon suggested that the web page be used as a tool to facilitate communication between labs regarding subtle nuances in methods that could influence data quality and as a way to "horse-trade" for needed supplies. Labs with an excess supply of a certain item could post a notice to the page offering to trade for a supply they need which another lab may have in excess, or otherwise offer the item for another lab's use.

Scott McEwen, CR Computer Specialist, provided a preliminary Web page prototype that is available at http://water.usgs.gov/osw/technqiues/sedimentlabs.html.

Action: Each sediment laboratory will develop a Web site and place it on their District home page. Each lab will create links to the other labs and to the OSW sediment laboratory Web site (McEwen).

Recommendation: In a consistent format, each laboratory Web site should list: When considered ready for release, the Web site should contain the laboratory's statistics from the Sediment Laboratory Quality Assurance Program along with a link to the BQS Web site which contains the Program's national statistics.

Certification Training (principal contacts - Schroder, Gray): The two forms of certification training discussed in the meeting are based on:

  1. A requirement for sediment laboratory chiefs to attend a "training period under a certified sediment laboratory chief, and ... periodic meetings of sediment laboratory chiefs" (OSW Technical Memorandum 98.05).
  2. The OSW's need to train and maintain a cadre of certified sediment laboratory reviewers.
Fortunately, training needs for these purposes are complementary and can often be achieved simultaneously through mentoring by a certified laboratory reviewer on a triennial sediment laboratory review.

As of March 1998, the following individuals are considered certified laboratory reviewers: John George, Vernon Norman, LeRoy Schroder, Russ Ludlow, Dan Gooding, Al Onions, and Clyde Sholar. There is a disparity in availability and time since direct involvement in day-to-day sediment laboratory activities associated with those listed. For example, Dan Gooding is the only current laboratory chief in the group. On the other extreme, Vernon Norman has not been involved with sediment laboratories for years, is retired, and has given no indication of interest in his future involvement in this capacity. A larger cadre of certified sediment laboratory reviewers with recent laboratory experience needs to be trained, and soon.

Recommendation: Annually send 1-3 laboratory chiefs not certified as sediment laboratory reviewers on sediment lab reviews until all in need of this training have participated. We may opt to provide certification on a method-by-method basis (Gray).

Status of Wet Sieving (principal contacts - Schroder, Love): A non-standard wet-sieving technique that uses higher-than-standard water-rinse velocities to wash sediments in the sieve appear to provide less-biased data than those analyzed by the traditional method of washing with a hand-held rinse bottle.

Action: Shortly following the workshop, Kathy Love collaborated with Alan Mlodnosky to develop a proposal to test the non-standard rinse technique to determine if it should become a SOP.

Recommendation: The OSW fund a variation of the proposal recommended by LeRoy Schroder and John Gray.

Standardization (principal contact - Schroder): It was recognized that some minor variations inevitably exist in techniques and equipment used by the laboratories. For example, there is no standard design for the "J" tube siphon used to remove water from water-sediment mixtures. Also, the sediment laboratory review checklist needs to be updated.

Recommendations:
  1. Follow through on recommendation for lab chiefs to visit other laboratories (Gray).
  2. Have all laboratory chiefs review the TWRI Book 5, Chapter C1, and OFR 92-499 (which supersedes the TWRI on suspended-sediment concentration determinations) and document any deviations from the techniques described therein in a memorandum to the BQS.
Technology (principal contact - Gooding, Gray): Several technologies developed for purposes other than sediment laboratories show promise in either replacing some laboratory instruments, or in automating some aspects of laboratory procedures.

Recommendation: Dan Gooding and John Gray are encouraged to continue efforts toward bringing robotics to sediment laboratories, and more promising technologies to the test phase. They are encouraged to seek funding from the OSW and from other entities with a potential interest in these technologies.

Criteria for Evaluating QC Data

John Gordon gave a presentation on a proposed set of minimum QC standards that would apply to all of the laboratories that produce physical sediment data for the WRD. The key tenants of this proposal are nonparamentrically based warning and control limits and corrective actions when warranted. The rational for nonparametric-based warning and control limits and the guidelines specifying the level of follow-up that may result depending on laboratory performance was described. The group expressed support for this method of evaluating and using SLQA data.

John Gordon then gave a presentation discussing the results of the Sediment Laboratory Quality Assurance Project (OSW Technical Memorandum No. 96.11) in 1996-97. Only those samples reported as not having lost water or sediment during transit were included in this analysis, i.e., the effects of commonly used shipping procedures was not a part of this presentation. Warning limits on John Gordon's control charts were set at ±2 f-Pseudosigma values. Control limits were set at ±3 f-Pseudosigma values. The benchmark for performance is for all of reported data for each analysis to be "in control."

Exceedance of control limits is cause for initiating corrective action as described in general terms in OSW Technical Memorandum 98.05.

Corrective action for a laboratory may also be triggered by:

  1. A run of all QC data of a given type on one side of the median with a length consisting of the majority of samples.
  2. A continued rise or fall (trend) within a given type of data.
  3. Periodicity in QC data, or data that "hug" the ±3 f-Pseudosigma limits.
Control charts were displayed for the different sample sizes used in the study. The samples sent to the laboratories for analysis were in 3 different concentration ranges:

A known amount of sand-size material was added to each sample. The mass of sand was between 10 and 30 percent of the mass of the fine material. The volume of water was carefully measured using volumetric flasks and was between 200 and 500 mL for each sample. A synopsis of results for the 3 concentration ranges follows:

50 - 100 mg base:

There were some notable differences in laboratory performance. One laboratory reported 6 concentration values that exceeded the control limits. The next highest number of values outside of the control limits for single laboratory in this concentration range was two.

100 - 300 mg base

One of the laboratories had four values outside the control limits. No others had more than one reported value in this concentration range that exceeded the control limits.

> 2000 mg samples

The results for the highest concentration samples were the best for the same laboratory that struggled to meet the control limits on the smaller concentration ranges. Two of the reported values in this size class for this laboratory were within 1 percent of the actual sediment concentration.

Summary

In general the participating laboratories did very well in these QC tests, with some notable differences between them. While a few of the laboratories showed considerable improvement in their results over time, others consistently performed extremely well. In addition, some labs that did very well on sediment concentration were provided marginally acceptable data on percentages of sand versus finer material (sand-fine splits). The reverse was also true--some of the labs that were among the best at determining the sand and fine fractions did not always excel at concentration analyses.

Other Observations and Outcomes from the Sediment Laboratory Chiefs Workshop (principal contact - all):

Quart Glass Bottle: The mayo sample bottle is being discontinued, according to FISP Chief Wayne O'Neal. Input on this needs to be directed to Wayne.

Sample Evaporation: Von Miller agreed to summarize his work on evaporation from samples in a dark, temperature-controlled room and share them with OSW, BQS, and laboratory chiefs. He noted considerable evaporation (10-30 percent) around some caps (Miller).

"J" Tube Specifications: Von Miller agreed to provide the specifications of the Iowa Laboratory "J" tube, and perhaps to fabricate several for distribution as-needed.

Sediment Laboratory Review Check List: Clyde Sholar agreed to take the lead in updating the checklist with help from Von Miller and Cheryl Smith. Dave Funderburg provided a copy of the current checklist.

Appendix A:

March, 1998, Update of "Quality Assurance and Quality Control Recommendations" from the 1994 Sediment Laboratory Chiefs meeting at Cascades Volcano Observatory.

Balances, Desiccators, and Ovens

The work group analyzing quality control for balances, desiccators, and ovens has identified a number of quality control activities and quality control data acquisition requirements. Several quality control activities require the laboratory chief to be proactive.

Balances

I. Balance Quality Control (QC)
  1. Identify the balances used in the laboratory.
    1. Prepare a calibration, maintenance, and service log.
      1. Log can be either electronic or hard copy.
      2. Log should contain at least the following information:
        Date, Identification of Balance, Name of Analyst, Time, Action taken, and Data.
      3. A separate logbook will be maintained for each balance.

  2. Balance use QC
    1. Each laboratory should have a heavy shock-proof table.
    2. If shock-proof table is unavailable, the laboratory will measure the drift and variability twice a week and determine if the data from the balance meet the manufacturer's specifications.
    3. Each analyst will be trained to use all the laboratory balances by the laboratory chief or certified trainer.
    4. The laboratory chief will go through the laboratory review form with each analyst to ensure that the analyst is aware of the minimum QC expected by outside reviewers.

  3. Daily Balance QC
    1. Check that all weights are traceable to the NIST or NBS.
    2. Set the minimum acceptable differences allowed from the known weights.
      1. Difference between the measured weight and known weight that are 1 gram or less must be less than 1 percent.
      2. Difference between the measured weight and known weight that are greater than 1 gram must be less than 0.75 percent.
    3. At the beginning of the weighing session, analyst weighs two weights that are representative of the weights to be made that day and determines that the weights are within 1 milligram for analytical balances and 0.5 grams for macro balances. Results will be documented in logbook.
      1. If the weights do not meet these criteria, the analyst will re-calibrate the balance before continuing the analysis.
      2. If the analyst cannot get the balance to operate correctly, the laboratory chief will check the balance and tell the analyst how to fix the problem.
      3. If the laboratory chief cannot correct the problem, the manufacturer's representative will be called.
      4. The analyst will not continue to analyze sample if the balance is not within manufacturer's specifications.
    4. Laboratory chief will determine how to examine QC data. If plotting is used then:
      1. The plot will be the measured value versus the data measured. Each plot will denote the analyst who made the measurement.
    5. Each analyst will be trained to readjust the balance to zero every ten measurements.

  4. Monthly Balance QC
    1. Laboratory chief will determine that the daily quality control checks have been performed.
    2. A person other than the normal analyst will make one daily balance check.
      1. If the balance fails this check, the laboratory chief will determine if the balance is functioning within manufacturer's specifications.
      2. The monthly check data will be stored in the balance log.
    3. It is recommended (not mandatory) that the laboratory chief evaluate the test data. If plots are used, the following is recommended:
      1. If a pattern--6 or more values of high or low measured weights are found, the laboratory chief will service the balance.

  5. Yearly Balance QC
    1. The balances will be serviced and calibrated by a manufacturer's representative at least once per year. Record of servicing will be documented in the balance logbook.
    2. Weights need certification of calibration either in-house or at the service representative facility.

Desiccators

Desiccators will be used to store crucibles and evaporating dishes after removal from the drying oven to prevent absorption of moisture. This includes both the empty glassware prior to its initial weighing, and the glassware with the dried sediment.

  1. Cleaning
    Desiccators should be kept clean, and thoroughly cleaned at 6-month intervals. Desiccators requiring use of sealant grease must receive extra care to avoid excess grease accidentally being transferred to the crucibles or evaporating dishes. All maintenance of desiccators should be recorded in a logbook.

  2. Desiccant
    Calcium chloride or silica gel may be used as a desiccant. A color indicator must be included in the calcium chloride desiccant. Manufacturer's recommendations should be followed for the regeneration of the specific desiccant used. In addition to using a color-indicating desiccant, a humidity meter should be used with each desiccator as a secondary means of monitoring moisture. It is a good practice to replace desiccant at least once per year. Some desiccant cannot be adequately recharged after repeated use. Records of desiccant regeneration/replacement should be maintained in a logbook.

  3. Glassware storage time
    Crucibles and evaporating dishes should be cooled in the desiccator for a minimum of 3 hours prior to weighing.

Ovens

Ovens are used to dry the glassware and sediment samples. Convection-type ovens will be used to allow for even temperature distribution within the oven. Ovens will be vented to the outside for the safety of laboratory personnel.

  1. Temperature control
    Temperatures between 85 and 95°C should be maintained when water is in the dishes, and increased to 103°C (+ or - 2 degrees) after all visible water has evaporated. These are actual temperatures, and not necessarily the control knob setting, as the control knob may not be calibrated to the oven temperature. If the oven has a high-temperature limiter, set it at 106°C.

  2. Temperature checks
    Temperature checks will be made at least twice during the drying cycle. Results of temperature checks will be documented in the oven logbook. When documenting temperature checks, include both the thermometer reading and the oven setting.

  3. Thermometer checks
    The oven thermometer will be checked against a calibration thermometer at least every 6 months. The oven thermometers will be replaced if they differ by more than 2°C from the calibration thermometer. Thermometer checks and action taken will be recorded in the oven logbook.

  4. Drying time
    The clean crucibles with filters should be dried at 103°C for a minimum of 2 hours. Evaporating dishes should be dried for a minimum of 1 hour. After the sediment has been placed into the crucibles and evaporating dishes, a minimum of 2 hours should be allowed for drying after all visible moisture has evaporated. Additional drying time may be necessary for large amounts of sediment in either the crucibles or evaporating dishes.

  5. Sample handling
    Samples should be removed from the oven and immediately placed in the desiccator within 10 minutes after the oven is turned off. Dried samples should not be stored in the oven. Ovens should be loaded from the top down, and unloaded from the bottom up to prevent particles from falling into the samples.

  6. Oven cleaning
    All ovens should be kept clean at all times. Racks should be inspected and replaced or restored if there is danger of particles falling into the samples.

Decanting, Filtration, and Glassware

Following are the recommendations from the Decanting, Filtration, and Glassware work group.

  1. Decanting
    The variety of fluvial material in sediment samples requires that the sample be allowed to settle with a minimum of sample movement before decanting the supernatant water. When decanting large volumes do not increase the decant rate; instead, build additional decant time into the price structure.

    1. The rate of removal of supernatant water must not disturb the solids.
      1. A needle valve can be installed in the vacuum line to remove water at an ideal rate of 1 liter/minute, but no more than 1.5 liter/minute.
      2. A much lower decant rate or no decanting should be used for samples containing sediments of relatively low density, such as coal particles.
    2. The decanting process:
      1. Do not tip the sample bottle when decanting.
      2. Keep the J-tube against the inner rim of the bottle mouth while lowering the J-tube at a smooth, even rate.
        1. Avoid stirring motions or sweeping motions with the J-tube.
      3. Keep the suction tip of the J-tube near the water surface while decanting.
      4. Do not lower the J-tube tip closer than 1 inch from the surface of the sediment layer. Continue decanting until the water level drops to the height of the J-tube tip.
      5. If at any time sediments begin to rise from the bottom of the sample bottle, immediately remove the J-tube and stop decanting the sample.
    3. Decanting suggestion:
      1. For labs that use Erlenmeyer filtration flasks, use a filtering Erlenmeyer flask (Kimex heavy wall, Pyrex safety coated, or equivalent), stopper with single hole, fitted with a 1/8" - 1/4" tube long enough to rest below the flask hose barb, pass through the stopper, and connect to the J-tube hose.
    4. Determinations of sediment concentrations in decanted water is not required. However, each laboratory is encouraged to measure sediment concentrations in the water decanted from selected samples, with particular emphasis on supernatant from samples containing relatively low-density sediments. For laboratories with fixed filtration racks, the quality-control procedure follows:

      Set up a crucible holder on either end of the rack that can both securely hold a crucible, but function as a quick release, fit the J-tube hose with a #7 single hole stopper, and a 1/8" - 1/4" tube, and fix the stopper to a properly prepared crucible. Decanted water passes through the crucible as would a sample. At the end of the set, rinse the stopper and tube into the crucible, and handle the crucible as a typical sample.

  2. Filtration
    The process used by the laboratories has remained essentially the same for the last 30 years. The only major changes were to standardize the filter, which is used for concentration analysis, and the use of fritted-glass crucibles by some laboratories. The specific recommendations for filtering are:

    1. A consistent vacuum pressure should be used by all laboratories.
    2. The WRD should, in a cost-effective manner, standardize the crucible used by the sediment laboratories.
      1. Physical and quality control data need to be obtained and evaluated before the decision is made.
    3. The filter or porous part of the fritted-glass crucible should be examined after every use and the crucible discarded if the frit appears to be plugged.
    4. There is a need to determine the loss through the 934-4H filters.

  3. Glassware care
    Both the care of bottles and evaporation dishes were considered by the work group. Also, one possible safety problem was considered.

    1. Sample bottles
      1. If a laboratory has a dishwasher, the bottles should be cleaned 1 cycle in dishwasher.
      2. Check the bottle for chips around the mouth of the bottle. If there are chips around the mouth of the bottle, the bottle should be discarded.
      3. Manual washing with an electric bottle-brush allows laboratory personnel to clean and examine the bottles in one step.
      4. Bottles should be air-dried.
      5. Examine one percent of the bottles for changes in the tare value and for dissolved-solids residue. Laboratory chief should determine to clean or dispose of bottles. The tare values should be recorded.
      6. After bottles have been cleaned and allowed to air dry, bottles are capped and stored for future use.

      NOTE: All manually operated electric bottle brushes must be connected to a ground-interrupt circuit.

    2. Evaporation dishes
      1. Any dish that is chipped or cracked should be discarded. Heating and handling chipped or cracked dishes is also a safety hazard.
      2. The weights of clean, dry, cooled evaporating dishes will be measured and recorded before every use.

Use of Blank Samples

Blank samples are suggested for the three determinations: 1) suspended-sediment concentration, 2) sand/silt splits, and 3) particle size by dry sieve. The determination of particle size by dry sieving has not been monitored using blank samples; so, to obtain some quality control data, the work group suggests a re-analysis procedure.

  1. Suspended-Sediment Concentration
    A blank is defined as 300 mL of deionized water that is filtered through a preweighed crucible and filter. The crucible and filter is dried in the oven, cooled in a desiccator, and weighed. So the blank sample is processed as a natural sample except the blank sample neither comes in contact with a J-tube nor is decanted.

    1. Laboratories preparing an evaporating-dish blank sample should pour 200 mL of deionized water into a preweighed dish, evaporate the water and dry the dish in an oven, cool the sample in a desiccator, and measure the weight.
      1. A good blank should have a mass of 0.0005 g or less.
      2. Blanks with a mass greater than 0.5 g could indicate a problem in the analytical process.
    2. A database should be developed by measuring a blank every tenth sample. After the database is established, the number of blank samples can be reduced to 20 samples per month with a minimum of 200 blanks per year.
    3. Each new laboratory employee should analyze a blank sample every tenth sample during the training period of about 8 weeks.
    4. All blank-sample data should be stored in a data base in a quality control file with hard copies of the data in a notebook.

  2. Sand/Fine Split
    A blank is defined as 300 mL of deionized water that is poured through a 0.062 mm mesh sieve. Then, a second blank is defined as 300 mL of deionized water that is poured through the inverted 0.062 mm sieve.

    1. Both blank waters should be filtered through separate crucibles and filters. Both crucibles are processed as normal samples. (Note: Could be combined into a single blank without the loss of information?)
      1. A good blank should have a mass of 0.0005 g or less.
      2. Blanks with a mass greater than 0.0005 g could indicate a problem in the cleaning of or recovery of material from the sieve.
    2. A database should be developed by preparing the pair of blanks every tenth sample. After the database is established, 5 to 10 percent of the samples should be blanks.
    3. Each new laboratory employee should analyze the dual blank every tenth sample through the training period.

  3. Dry Sieve
    No blank was defined for the dry-sieve process; so, to obtain some quality control data, samples analyzed by dry sieving can be analyzed twice (rerun).

    1. After recording that mass of the sample retained on each sieve, the sample should be re-combined.
    2. Sieve the sample as second time and record the mass of the sample retained on each sieve.
    3. Compare the two sets of data.
      1. If the two results from any pair of data differ by more than 5 percent, the present and previous sample should be rerun and the results compared. If results from any pair of data differ by more than 5 percent, stop analyzing samples and determine the problem in the sample analysis.
    4. Experience indicates that most of the errors are caused by inaccurate weighing or in calculating the sums.
    5. All dual analysis data should be stored in a quality control database.
      1. The laboratory chief should evaluate the data every 2-3 months.

Appendix B:

Informal Notes on Safety from the 1998 Sediment Laboratory Chiefs Workshop, Baton Rouge, LA:

Safety Plans:

Personal Protective Equipment:

Personal Hygiene

Medical Monitoring

Immunizations (WRD Memorandum 98.06: Safety - WRD Immunization Program)

Back Injury Protection:

Electrical Safety:

Laboratory Ventilation:

Special Safety Concerns:

Respiratory Protection:

Other Sediment Laboratory Safety Issues or Questions?

The following bullets were developed from a presentation by Alan Mlodnosky, who led the Workgroup on Safety that included Von Miller, Dave Funderburg, and Clyde Sholar.

[an error occurred while processing this directive]