New Preservation Techniques for Nutrient Samples In Reply Refer To: August 5, 1994 Mail Stop 412 OFFICE OF WATER QUALITY TECHNICAL MEMORANDUM 94.16 Subject: New Preservation Techniques for Nutrient Samples SUMMARY Beginning October 1, 1994, the Water Resources Division will discontinue the use of mercuric chloride to preserve ambient samples collected for analysis of nutrients. Instead, the Division will preserve ambient water samples to be analyzed for nutrients by chilling only, with mailing within 3 days to the National Water Quality Laboratory (NWQL). Some Division projects may be collecting nutrient samples in cooperation with regulatory agencies wherein the samples are defined as compliance (collected for compliance with the Clean Water Act or the Safe Drinking Water Act) rather than ambient. The U.S. Environmental Protection Agency (EPA) and many State regulatory agencies require sulfuric acid preservation of such compliance nutrient samples. Therefore, on October 1, 1994, the NWQL will also begin analyzing sulfuric acid-preserved samples. This memorandum identifies a number of actions required to implement the noted changes. Table 1 summarizes the actions and associated implementation dates. Table 1. Actions and Implementation Dates Required to Change Nutrient Preservation Techniques _______________________________________________________________________ Implementation Date Action _______________________________________________________________________ October 1, 1994 o Discontinue use of mercuric chloride to preserve nutrient samples. o Adopt chilling only to preserve ambient nutrient samples. o NWQL begins analysis of chilled only nutrient samples.* o All schedules are changed and mercuric chloride-preserved NWQL codes are replaced with chilled only NWQL codes.* o NWQL also begins analysis of nutrient samples preserved with sulfuric acid collected for compliance purposes. o Begin 3-month grace period for NWQL analysis of nutrient samples preserved with mercuric chloride to assist Division projects that have special need for these data. December 31, 1994 o End 3-month grace period for analysis of nutrient samples preserved with mercuric chloride. January 1, 1995 o Lab codes for nutrient samples preserved with mercuric chloride become invalid and can no longer be requested. February 1, 1995 o Begin 1-month period in which the NWQL will accept unused mercuric chloride ampoules for disposal. February 28, 1995 o End 1-month period in which the NWQL will accept unused mercuric chloride ampoules for disposal. _____________________________________________________________________ *The term "chilled only" means chilling and maintaining all samples at 4 degrees Celsius (without freezing). Thus, for unfiltered nutrient constituents, field processing of chilled only samples includes filling and chilling the nutrient bottle. In contrast, for the so-called dissolved nutrient constituents, field processing of chilled only samples includes filtration through a 0.45-micrometer filter into a nutrient bottle followed by chilling. BASIS FOR THE DECISION In 1992-93, Charles Patton and Earl Truitt of the NWQL compared the effects of three preservation techniques--mercuric chloride plus chilling, sulfuric acid plus chilling, and chilling only--on nutrients in 11 surface-water and 3 ground-water samples from around the United States. Table 2 shows the ranges of nutrient constituent concentrations for the samples. Table 2. Constituent Concentration Ranges of Samples in the Nutrient Preservation Experiment ______________________________________________________________________ Constituent(s)* Range of Concentration (mg/L) ______________________________________________________________________ Surface Water Ground Water n = 11 n = 3 ______________________________________________________________________ Ammonia, filtered 0.02-5.5 0.02-0.04 Nitrate + nitrite, filtered 0.04-8.7 2.4 -56 Nitrite, filtered 0.00-1.9 0.00-0.01 Organic nitrogen + ammonia, filtered 0.1 -6.9 0.05-1.3 Organic nitrogen + ammonia, unfiltered 0.2 -9.8 0.04-1.4 Orthophosphate, filtered 0.01-3.9 0.00-0.03 Phosphorus, filtered 0.01-3.9 0.00-0.08 Phosphorus, unfiltered 0.03-4.8 0.00-0.2 ______________________________________________________________________ *The term "filtered" replaces the previously used term "dissolved" and "unfiltered" replaces "whole water recoverable." The data show that the samples encompassed a wide range of concentrations for each constituent, representing conditions from pristine to contaminated. The comparative preservation tests were run for a minimum of 31 days on each sample. The overall results indicated that within analytical and test precision, chilling only provides equivalent preservation to mercuric chloride plus chilling or to sulfuric acid plus chilling. As expected, the tests also showed that sulfuric acid systematically destroys nitrite in samples. In contrast to expectations, the 31-day stabilities of orthophosphate in all acid-preserved samples were comparable to those for mercuric chloride-preserved and chilled only samples. A summary of the nutrient preservation experiment is attached to the electronic copy of this technical memorandum. To access the memorandum and the attachment, one must be registered on QVARSA. After login, enter the command TECH_MEMO and follow the instructions given for the menu-driven system. In addition, Charles Patton is preparing a detailed report explaining the experimental design of the study and the results. District and Regional Water-Quality Specialists will be notified by EDOC when this document is available. In actual Division use, as in the NWQL experiment, all samples to be analyzed for filtered nutrient constituents (previously referred to as dissolved) are passed through a 0.45-micrometer filter at the collection site. This step is known to remove most bacteria and other microorganisms from filtrates. Therefore, microorganisms present in original samples should have negligible effect on the stability of filtered nutrient constituents. In the case of unfiltered samples, nitrogen and phosphorus metabolized by microorganisms will be released during the rigorous digestion step included in the analysis (with the exception of nitrate plus nitrite nitrogen). Thus, based on experimental results, plus knowledge of the effects of filtration and the analytical methods, the Division selected chilling only as the standard technique for preserving ambient samples for nutrient analysis. Chilling only was preferable to sulfuric acid plus chilling to (a) avoid widespread shipment, use, and disposal of sulfuric acid ampoules, (b) achieve the lowest possible costs, and (c) minimize the number of bottles that projects must fill and ship to the NWQL for each nutrient sample. With chilling only, two bottles are required for all nutrient species. With sulfuric acid plus chilling, a separate, third bottle, using chilling only, is required for the nitrite and orthophosphate determinations. THREE-MONTH GRACE PERIOD Some Division projects may be in the middle of long-term collection of nutrient data using mercuric chloride plus chilling for preservation. To assist these projects, the NWQL will continue to offer analysis of samples preserved with mercuric chloride through December 31, 1994. Projects can also use this period to send split nutrient samples to the NWQL (one preserved with mercuric chloride, the other chilled only) if a project- specific need exists for such a comparison. For split sample work, projects will be responsible for all analytical costs and interpretation of the resultant data. DISPOSAL OF UNUSED MERCURIC CHLORIDE AMPOULES The NWQL will accept unused mercuric chloride ampoules for disposal February 1-28, 1995. This timeframe allows projects about 6 months to locate and prepare unused ampoules for shipping. The NWQL must notify regulatory authorities in advance of the extra disposal of mercuric chloride during February. Therefore, no ampoules can be accepted before February 1, 1995. After February, the NWQL will not be permitted to dispose of waste water which has been treated to remove mercuric chloride. Therefore, unused ampoules received after February 28, 1995, will be returned to the senders who will have to arrange for disposal at a licensed disposal facility within their respective States. Mercuric chloride ampoules are a hazardous material and must be packed and shipped according to Department of Transportation (DOT) regulations. Unused ampoules MUST be shipped in their ORIGINAL cardboard containers with the foam liner or in a cooler with the ampoules double bagged in trash bags with packing material. The original cardboard containers are DOT approved, but other cardboard boxes are not approved and cannot be used. The unused ampoules should be shipped via United Parcel Service (UPS) ground transportation. Federal Express and the Postal Service should not be used. Please write "ATTENTION CARLOS AROZARENA -- UNUSED HGCL2 AMPOULES" on each shipping container. Contact Will Lanier, NWQL, (WDLANIER), at telephone number (303) 467-8065, if you have problems either preparing the unused ampoules for shipping, or coordinating your shipment with the local UPS office. Before shipping, please send an EDOC to Carlos Arozarena (USERID - SAFETY) stating that a specified number of unused ampoules will be sent on a given date. This forewarning is required by UPS in case there is an accident. Please include the phone number of the NWQL guard [(303) 467-8064] in the shipping information so that if a problem occurs, the NWQL will be contacted and can then work to resolve the problem. IMPORTANT: Please note that the NWQL is offering disposal only of unused mercuric chloride ampoules. Do not send other hazardous materials for disposal. Other hazardous materials sent will be returned, or if return is illegal, the NWQL will bill the sender for the full cost of disposal. Elimination of mercuric chloride as a preservative is a change that has been widely sought for a number of years by Division personnel. To investigate options, the OWQ and NWQL expended a large amount of funds. Also, costs associated with handling and disposal of the unused ampoules will be borne by the NWQL. For these reasons, no refunds will be made for unused mercuric chloride ampoules sent to the NWQL. COORDINATION WITH THE U.S. ENVIRONMENTAL PROTECTION AGENCY The NWQL and OWQ have shared the experimental results of the tests with the EPA under the auspices of the Methods Comparability Council of the Intergovernmental Task Force on Monitoring Water Quality. We believe this effort will lead to EPA signing a memo acknowledging that chilling only is acceptable as the preservation technique for ambient water samples collected for nutrient analysis. The EPA will continue to require that sulfuric acid be used in conjunction with chilling to preserve nutrient samples collected for compliance with the Clean Water Act and the Safe Drinking Water Act. SULFURIC ACID-PRESERVED SAMPLES The NWQL will begin offering analysis of sulfuric acid-preserved nutrient samples on October 1, 1994. The purpose of offering analysis for sulfuric acid-preserved samples is to enable Division projects to meet regulatory agency requirements for chemical preservation of compliance samples. Sulfuric acid ampoules to preserve nutrient samples (in the standard 125-milliliter bottle) are available through the Ocala Quality of Water Service Unit and can be ordered through the Ocala Supply Orders ID (OCALAMAN). The cost is $35 per lot of twenty- four 1-milliliter ampoules containing 25-percent sulfuric acid. FIELD PREPARATION INSTRUCTIONS AND TURNAROUND TIMES FOR NUTRIENT SAMPLES Ambient Samples (Chill Preserved) As in the past, ambient samples to be analyzed for filtered nutrient constituents must be passed through a 0.45-micrometer filter into a brown 125-milliliter polyethylene bottle and immediately packed in ice at the collection site. Designate these samples as "FCC" to differentiate them from filtered, chilled samples preserved with mercuric chloride, which will continue to be designated "FC" until December 31, 1994. Omit the filtration step for ambient samples to be analyzed for unfiltered nutrient constituents. Designate these samples as "RCC" to differentiate them from raw chilled samples preserved with mercuric chloride, which will continue to be designated "RC" until December 31, 1994. Within 3 days of collection, ship the chilled samples to the NWQL with next-day-delivery priority. The NWQL will analyze nutrient samples within 14 days of receipt. These timeframes will insure that analytical determinations are performed well within the 31-day holding times supported by the experimental data. Compliance Samples (Sulfuric Acid Preserved) Beginning on October 1, 1994, the NWQL will analyze nutrient samples collected to assure compliance with requirements of the Federal Clean Water Act and the Safe Drinking Water Act. Field processing and maximum holding times prior to analysis for such samples are specified in Table II, paragraph 136.3, Volume 40 of the Code of Federal Regulations. Specifically, samples to be analyzed for filtered nutrient constituents must be passed through a 0.45-micrometer filter into a brown 125-milliliter polyethylene bottle, acidified to pH < 2 with sulfuric acid, and immediately packed in ice at the collection site. Designate these samples as "FCA" (filtered, chilled, acidified). Omit the filtration step for compliance samples to be analyzed for unfiltered nutrient constituents. Designate these samples as "RCA" (raw, chilled, acidified). Within 3 days of collection, ship the acidified, chilled samples to the NWQL with next-day-delivery priority. The NWQL will analyze nutrient samples within 14 days of receipt. These timeframes will insure that analytical determinations are performed well within the 28-day holding times specified in the Code of Federal Regulations. Compliance monitoring samples to be analyzed for nitrite and orthophosphate must be passed through a 0.45-micrometer filter into a brown 125-milliliter polyethylene bottle, and immediately packed in ice at the collection site. Sulfuric acid MAY NOT be added to these samples, which must be analyzed within 48 hours of collection. While field processing for "FCC" samples as described above is appropriate for such samples, the 48-hour holding time requirement is difficult to achieve when sampling locations are distant from the NWQL. Therefore, for compliance monitoring samples, it is most practical to arrange for analytical determinations of nitrite and orthophosphate at a laboratory near the collection site. In some cases, it may be possible to make special arrangements at the NWQL to determine nitrite and orthophosphate in compliance samples. Special arrangements for such samples would include MANDATORY, 1-week advance notice to affected NWQL service units and Monday, Tuesday, or Wednesday arrival days at the NWQL. To make special arrangements, contact the Chief, Inorganic Program (MSHOCKEY) at telephone number 303/467-8101. INFORMATION FOR SUBMITTING NUTRIENT SAMPLES TO THE NWQL To implement the described changes, the NWQL will change schedules, delete and create lab codes, and create new bottle types. Projects need to observe the upcoming changes to eliminate sample mix-ups and analytical errors. If District personnel wish to run comparison studies between chilled only and chilled plus chemically preserved samples, sample splits should be treated as duplicates and submitted on different Analytical Services Request (ASR) forms using different times. Unless this is done, the electronic data system will report only the second analytical value; the first will be lost. On October 1, 1994, mercuric chloride lab codes will be deleted from all schedules and replaced with new lab codes for the chilled only preservation technique. If mercuric chloride lab codes are needed during the 3-month grace period (October 1-December 31, 1994) they must be recorded as an "add" on the ASR form. On January 1, 1995, all mercuric chloride-related lab codes will become invalid and may not be requested. The NWQL will prepare new schedules for sulfuric acid-preserved samples. The schedule numbers will be released in September 1994. IMPORTANT: After February 28, 1995, the NWQL will no longer be able to dispose of mercuric chloride-preserved samples. Furthermore, the NWQL will not be running standards for mercuric chloride-preserved samples. Therefore, analytical results will be incorrect for samples designated "filtered-chilled only" ("FCC" bottle designation) or "raw-chilled only ("RCC" bottle designation) if they are actually preserved with mercuric chloride. Table 3 summarizes information that will be needed for personnel submitting samples to the NWQL for nutrient analyses beginning October 1, 1994. Table 3. Information for Submitting Samples to the NWQL for Nutrient Analysis Beginning on October 1, 1994 ______________________________________________________________________________ Standard Range Analyses ------------------------------------------------------------------------------- HgCl2 | Chilled Only** | H2SO4** NWQL M Bot | NWQL M Bot | NWQL M Bot CONSTITUENT(S)* Code PCode C Typ | Code PCode C Typ | Code PCode C Typ | | ------------------------------------------------------------------------------- Nitrogen, Ammonia, 0301 00608 B FC | 1976 00608 F FCC | 1991 00608 G FCA filtered | | Nitrogen, Nitrate + 0228 00631 B FC | 1975 00631 E FCC | 1990 00631 F FCA Nitrite, filtered | | Nitrogen, Nitrite, 0160 00613 B FC | 1973 00613 F FCC | NA filtered | | Nitrogen, Organic + 1687 00623 C FC | 1985 00623 D FCC | 1994 00623 E FCA Ammonia, filtered | | Nitrogen, Total 1570 00602 B FC | 1989 00602 C FCC | NA Species, filtered | | Nitrogen, Organic + 1688 00625 C RC | 1986 00625 D RCC | 1995 00625 E RCA Ammonia, unfiltered | | Phosphorus, 0162 00671 B FC | 1974 00671 H FCC | NA Orthophosphate, | | filtered | | Phosphorus, filtered 1685 00666 C FC | 1983 00666 D FCC | 1992 00666 E FCA Phosphorus, unfiltered 1686 00665 C RC | 1984 00665 D RCC | 1993 00665 E RCA Phosphorus, 0282 00678 A RC | 1987 00678 B RCC | NA Orthophosphate + | | hydrolyzable, | | unfiltered | | Phosphorus, 0279 00677 A FC | 1988 00677 B FCC | NA Orthophosphate + | | hydrolyzable, | | filtered | | ----------------------------------------------------------------------------- Low-Level Analyses ------------------------------------------------------------------------------ HgCl2 | Chilled Only** | H2SO4 NWQL M Bot | NWQL M Bot | NWQL M Bot CONSTITUENT(S)* Code PCode C Typ | Code PCode C Typ | Code PCode C Typ | | ------------------------------------------------------------------------------ Nitrogen, Ammonia, 0830 00608 A FC | 1980 00608 H FCC | NA filtered | | Nitrogen, Nitrite + 0826 00631 A FC | 1979 00631 G FCC | NA Nitrate, filtered | | Nitrogen, Nitrite 0827 00613 A FC | 1977 00613 H FCC | NA Phosphorus, filtered 0828 00671 A FC | 1978 00671 I FCC | NA Orthophosphate | | Phosphorus, filtered 0829 00666 A FC | 1981 00666 F FCC | NA Phosphorus, unfiltered 0837 00665 A RC | 1982 00665 F RCC | NA ------------------------------------------------------------------------------ Existing "Unpreserved" Lab Codes Which Will Be Deleted To Avoid Confusion ----------------------------------------------------------------------------- NWQL M Bot Code PCode C Typ ----------------------------------------------------------------------------- Nitrogen, Ammonia-LL 1278 00608 C FU Nitrogen, Nitrite + 1578 00631 C FU Nitrate Phosphorus-LL, 1277 00671 D FU ----------------------------------------------------------------------------- EXPLANATION OF SYMBOLS AND ABBREVIATIONS * - The term "filtered" replaces the previously used term "dissolved" and "unfiltered" replaces "whole water recoverable." ** - Lab codes for chilled only and sulfuric acid will not be initialized in the system until October 1, 1994 HgCl2 - Samples preserved by mercuric chloride plus chilling Chilled only - Samples preserved by chilling only H2SO4 - Samples preserved by sulfuric acid plus chilling NWQL Code - Same as Lab Code in the NWQL Laboratory Services Catalog PCode - Parameter Code MC - Method Code Bot Typ - Bottle Type NA - No analyses available BOTTLE TYPES FC - Filtered Chilled (Preserved with Mercuric Chloride) RC - Raw Chilled (Preserved with Mercuric Chloride) FCC - Filtered, Chilled Only (No Preservatives Added) RCC - Raw, Chilled Only (No Preservatives Added) FCA - Filtered, Chilled, (Preserved with Sulfuric Acid 25%) RCA - Raw, Chilled, (Preserved with Sulfuric Acid 25%) FU - Filtered, untreated ________________________________________________________________________________ IF YOU HAVE QUESTIONS ABOUT PRESERVATION OF AMBIENT SAMPLES Most nutrient samples collected by Division projects for submittal to the NWQL are categorized as ambient rather than compliance. As described in this memorandum, the Division requirement is for ambient samples to be preserved by chilling only, with sulfuric acid plus chilling to be used only for compliance samples. However, projects may occasionally need to collect ambient samples for nutrient analysis that have special characteristics that warrant concern regarding preservation. When such cases arise, please contact Charles Patton at the NWQL [(303) 467-8084] or your Regional Water-Quality Specialist to discuss how to preserve the samples. David A. Rickert Chief, Office of Water Quality (Attachment in electronic OWQ Technical Memorandum system but not included with hard copy) This memorandum supersedes Office of Water Quality Technical Memorandums 80.26 and 85.07 and modifies Office of Water Quality Technical Memorandums 92.08 and 92.11. Key Words: Mercuric chloride, sample preservation, nutrients, ambient sample, compliance sample Distribution: A, B, S, FO, PO, AH ATTACHMENT Draft Summary of OWQ/NWQL Nutrient Preservation Experiment What follows is a draft that outlines: 1) questions raised in the planning phase of this work about the ability of three widely applied field treatments to stabilize nutrient concentrations in samples prior to laboratory analysis; 2) the experimental design that evolved to answer these questions; 3) differences between this study and some previous ones; 4) field and laboratory sample processing protocols; 5) preliminary interpretation of data, and; 6) a list of data analysis tasks that still need to be completed. Ed Gilroy performed most of the statistical analysis on the data set and he will be a co-author on reports and journal articles resulting from this study. ============================================================ DRAFT DRAFT QUESTIONS ADDRESSED 1. How do field treatments -- * addition of 40 mg mercury(II)/L and chilling (USGS) * addition of sulfuric acid (reduce sample pH to < 2) and chilling (USEPA) * filtration and chilling, or chilling alone --affect stability of nutrient concentrations in natural water samples over a 30-day storage period? 2. Does water type (ground, surface, municipal supply) affect stability of nutrient concentrations in samples over a 30-day storage period for any or all of the three field treatment protocols? 3. Are particular analytes -- phosphorus and Kjeldahl nitrogen (total and dissolved); ammonium, orthophosphate, nitrate + nitrite, nitrite (dissolved only) -- more or less stable over a 30-day storage period following any of the three field treatment protocols? 4. Following any of the three field treatment protocols, are samples with relatively "high" nutrient concentrations more stable over a 30-day storage period than those with relatively low ones? 5. Is there a relationship between the amount of suspended sediment in samples and the stability of nutrient concentrations in samples over a 30-day storage period? (DELETE, NO DATA) 6. For each of the three field treatments, how does the 30-day stability of nutrient concentrations dissolved in a natural matrix (real samples) compare with that of nutrient concentrations dissolved in a deionized water matrix (control samples)? 7. Are there any statistically significant bottle effects? 8. Is there a clear-cut relationship between the temporal stability of nutrient concentrations in a particular sample and the biological activity in that sample for any of the three treatments? HOW THIS EXPERIMENT DIFFERED FROM SOME PREVIOUS ONES Unpolluted surface and ground water samples collected from 15 stations throughout the continental United States were used in the Office of Water Quality nutrient preservation experiment performed at the National Water Quality Laboratory by Patton and Truitt (April - July, 1992), hereinafter referred to as "THIS STUDY." In contrast, highly polluted samples collected from 12 sewage outfalls in the Metropolitan Chicago area were used in the Chicago U.S. Environmental Protection Agency nutrient preservation experiments performed by Carter and Jirka (unpublished work, circa 1974 -- hereinafter referred to as "USEPA STUDY") that ultimately resulted in nutrient sample preservation protocols mandated (40 CFR, paragraph 136.3, Table II) for National Pollution Discharge Elimination System (NPDES) compliance monitoring. In THIS STUDY, sample processing (splitting, filtration -- 0.45 micrometer pore size --, addition of preservatives) was performed in the field immediately following sample collection according to protocols described below. Samples arrived at the NWQL within 24 hours of collection and were immediately analyzed in quadruplicate for dissolved and total nutrients. The only sample processing operation performed at the NWQL entailed splitting one of the two, 1 liter samples from each treatment group (FC and RC mercury, FC and RC sulfuric acid, FC and RC water [control]) into ten 125 milliliter bottles as described below. In a previous USGS nutrient preservation experiment performed by Fishman, Schroeder, and Shockey (J. Environmental Studies, v. 26, p. 231-238, 1986) -- hereinafter referred to as "FISHMAN STUDY", three, 19 liter surface water samples were collected from the metropolitan Denver area in 5 gallon, plastic carboys. These three bulk samples were returned to the laboratory, processed through 10 micrometer filters, and then stored in a refrigerator at 4 degrees Celsius. Additionally, bulk samples of NWQL tap water and deionized water were also collected. These two samples along with one of the three environmental samples were spiked with organic and inorganic nitrogen and phosphorus compounds. One to two days elapsed before each of the 5 bulk samples was spiked (see above) and split into sixty-two 250 mL bottles. Sample splits were dosed with various preservatives (sulfuric acid, mercuric chloride, chloroform), three of which were immediately analyzed in triplicate for dissolved and total nutrients. On each of three other occasions (days 4, 8, and 16), three splits were selected and again analyzed for dissolved and total nutrients. In THIS STUDY, control samples with certified concentration estimates (USEPA "Nutrient Concentration 1" and "Nutrient Concentration 2"), which were prepared in deionized water containing the appropriate preservative, were analyzed in quadruplicate each time real samples were analyzed. Daily concentration variations in these pristine matrix control samples could be compared with concentration variations in natural water samples to estimate the contribution of instrumental variability to the over all variability in the experiment. Pristine matrix control samples were not included in the FISHMAN STUDY. Data from the EPA Study are no longer in existence. At the end of THIS STUDY microbiological activity of samples from each station was assessed by tritiated adenine uptake experiments and fluorescent staining techniques. Samples in the Fishman study were not assessed for microbiological activity. Data from the EPA Study are no longer in existence. EXPERIMENTAL DESIGN Samples from fifteen (15) stations around the United States were selected for this experiment. Eleven stations were NASQAN sites (surface water), three were wells (ground water), and one was tap water (municipal supply). In THIS STUDY the acid preservation protocol used was that mandated by the U.S. Environmental Protection Agency for NPDES compliance monitoring as defined in 40 CFR, paragraph 136.3, Table II (Mark Carter and Andrea Jirka, unpublished work, circa 1974). The mercury preservation protocol stemmed from work at the U.S. Geological Survey as described by Marvin Fishman, LeRoy Schroeder, and Merle Shockey (J. Environmental Studies, v. 26, p. 231-238, 1986). Six (6) field treatments were applied to each sample collected for determination of digested nutrients (total and dissolved Kjeldahl nitrogen, total and dissolved phosphorus): 1. Addition of deionized water (ampule effect control), and chilling. (whole water, "RC", subsample) 2. Addition of mercury (II) (approximately 40 mg per liter of sample), and chilling to 4 degrees Centigrade. (whole water, "RC", subsample) 3. Addition of sulfuric acid (sufficient to reduce sample pH to less than 2), and chilling to 4 degrees Centigrade. (whole water, "RC", subsample) 4. Filtration, addition of deionized water (ampule effect control), and chilling to 4 degrees Centigrade. (filtered water, "FC", subsample) 5. Filtration, addition of mercury (II) (approximately 40 mg per liter of sample), and chilling to 4 degrees Centigrade. (filtered water, "FC", subsample) 6. Filtration, addition of sulfuric acid (sufficient to reduce sample pH to less than 2), and chilling to 4 degrees Centigrade. (filtered water, "FC", subsample) Three (3) field treatments were applied to each sample collected for determination of dissolved nutrients (ammonia, orthophosphate, nitrate + nitrite, nitrite): 1. Filtration, addition of deionized water (ampule effect control), and chilling to 4 degrees Centigrade. 2. Filtration, addition of mercury (II) (approximately 40 mg per liter of sample), and chilling to 4 degrees Centigrade. 3. Filtration, addition of sulfuric acid (sufficient to reduce sample pH to less than 2), and chilling to 4 degrees Centigrade. Samples were shipped to the NWQL for first nutrient determinations within 24 hours of collection. FIELD AND LABORATORY SAMPLE PROCESSING PROTOCOLS At each of 15 stations, approximately 14 liters of water were composited in a churn splitter. The composite sample was processed in the field into twelve, prelabeled, 1 liter bottles as follows. With constant churning, six (6) whole water samples were collected in bottles marked "RC". Then churning was discontinued and six (6) filtered water samples (0.45 micrometer filter) were collected in bottles marked "FC". The twelve bottles were then sorted into three treatment groups according to color coded labels. Each group consisted of two (2) "FC" bottles and two (2) "RC" bottles. One (1) ampule containing mercuric chloride preservative solution was added to each of the four bottles labeled "MERCURY" (green color code). One (1) ampule containing sulfuric acid preservative solution was added to each of the four bottles labeled "ACID" (red color code). One (1) ampule containing deionized water preservative solution was added to each of the four bottles labeled "WATER" (blue color code). The twelve treated samples were then packed in ice and shipped to the National water Quality Laboratory in Arvada by overnight express. With one exception, samples were received at the NWQL, processed further, and analyzed for eight nutrient parameters within 24 hours of their collection, as described below. Immediately upon receipt at the NWQL, two bottles (one "FC" and one "RC") from each of the three treatment groups were shaken, poured into the appropriate ten-port cone splitter, and collected in ten, prelabeled, 125 mL, brown polyethylene bottles. Head space in 125 mL bottles was minimized by squeezing their sides gently to bring the sample level to the top of their necks before caps were secured. A different cone splitter was used exclusively for each treatment group throughout the experiment. Just prior to use, each cone splitter was rinsed copiously with 5 percent v/v hydrochloric acid and deionized water. In each treatment group, filtered samples ("FC" bottles) were always split before unfiltered samples ("RC" bottles). Samples from each of the 15 stations in the experiment were analyzed in quadruplicate for eight nutrient parameters -- dissolved ammonia, dissolved orthophosphate, dissolved nitrate + nitrite, dissolved nitrite, dissolved Kjeldahl nitrogen, dissolved phosphorus, total Kjeldahl nitrogen, and total phosphorus -- the day of arrival at the NWQL, and five additional times thereafter. CONCLUSIONS 1. Field preservation treatments had little effect on temporal variation in nutrient concentrations in the 2 to 5 reporting limit regime. Concentration changes of nutrients observed in THIS STUDY (with the exception of nitrite in acidified samples) do not decrease or increase linearly. That is, a concentration gain followed by a loss, or the reverse, was the rule rather than the exception in THIS STUDY. NOTE AGAIN THAT THESE CONCENTRATION CHANGES WERE IN THE 2 TO 5 REPORTING LIMIT REGIME. 2. No statistically significant differences in storage stability between surface and ground water were observed in THIS STUDY for any of the preservation treatments. 3. With the exception of nitrite in acidified samples, stability of nutrient concentrations was statistically equivalent for all preservation protocols. 4. Initial concentrations of nutrients in samples did not appear to exert a major effect on storage stability. 5. Data to assess the affect of suspended sediment concentration on storage stability apparently does not exist. 6. On the basis of BQA regression equations for standard reference waters, nutrient concentrations (with the exception of nitrite in acid preserved samples) in samples from all 15 stations remained within statistical control limits for 30 days. 7. No statistically significant bottle effects were observed in this study. 8. Temporal variation nutrient concentrations cannot be unambiguously attributed to biological activity in samples.