2.0 Definition
The DOC content of seawater is defined as the total concentration of all non-volatile organic substances expressed as moles of C per kilogram of seawater. An alternate and equivalent definition for the DOC content of seawater is the number of moles of carbon dioxide produced when all of the non-volatile organic substances are fully oxidized. For example, if a sample contained 60 m mol of glucose per kilogram, then the DOC content would be 360 mmol C/kg.

3.0 Principle of analysis
This method of analysis is based upon the complete oxidation of organic compounds to carbon dioxide followed by quantitative measurement of the CO₂ produced by non-dispersive infra-red (NDIR) analysis. This technique was first attempted for seawater by Sharp (1973) upon modification of a procedure developed by Van Hall et al. (1963) for fresh water. Interferences from the particulate carbon and inorganic carbon in seawater are first removed by filtration through glass fiber filters and sparging with CO₂ -free gas after
acidification of the sample (Sharp and Peltzer, 1993).

The instrument response is calibrated by the method of standard additions. Known amounts of organic compounds are added to produce a series of solutions with consistently increasing concentrations of organic carbon. The slope of the regression line obtained when peak area is plotted against the amount of carbon added is the instrument response factor. Both distilled water and seawater solutions have been used for this calibration. The principle is the same although the calculations are slightly different. (See section 8.3 below).
The instrument blank is determined by injecting the identical volume used during sample analysis and measuring the peak area. The peak area represents the amount of CO₂ liberated from the catalyst/combustion tube upon injection of a liquid sample and so each injection must be corrected by subtraction of this amount. It is important that the water used for this purpose be as carbon-free as possible (otherwise over-correction will occur and the DOC concentration will be under-estimated) and that this measurement be  repeated throughout the analytical sequence to closely monitor the instrument blank which may vary over time and use. Until a universally available source of carbon-free seawater (CFSW) is developed, carbon-free distilled water (CFDW) is recommended.

4.0 Apparatus
4.1 Filtration apparatus: In cases where POC levels are high (>2 mmol C/kg), the samples need to be filtered to avoid interference with the DOC determination. Samples are filtered through a Whatman GF/F glass fiber filter using an in-line filter holder. Samples can be either gravity filtered directly from the Niskin bottle or pressured filtered  at < 3 psig. Samples should not be vacuum filtered as this often results in low
level contamination.

4.2 Sparging apparatus: After filtration and acidification, samples are sparged to remove > 99.95% of the inorganic carbon. For small volume samples (< 40 mL) samples can be sparged by bubbling CO₂ free gas (oxygen or nitrogen) through a fine teflon line (spaghetti tubing) placed directly in the sample to almost the vessel bottom. A flow-rate of 100-20 mL/min for 6-8 minutes is usually sufficient to remove all inorganic
carbon. For larger samples, a polyethylene frit on the end of a 3mm diameter teflon tube aids in the production of fine bubbles. For 80-100 mL samples a flowrate of 500 mL/min for 5-6 minutes is usually sufficient. Each investigator should check the effi-ciency of their sparging system by re-sparging several samples. A consistent decrease of > 1 mmol C/kg after resparging indicates insufficient sparging during the first time period.

4.3 DOC analyzer: Several versions of HTC/DI analyzers have been built, either commercially or “homemade”. Each of these consists of a furnace and gas processing stream containing the following essential components:

4.3.1 Source of CO₂ -free carrier gas (preferably oxygen although nitrogen has been used) delivered through a pressure regulator with a stainless steel diaphragm.

4.3.2 High temperature combustion furnace.

4.3.3 Syringe to inject the seawater sample.

4.3.4 Trap to remove HCl and SO₂ .

4.3.5 Aerosol filter.

4.3.6 NDIR CO₂ analyzer.

4.3.7 Peak area integrator

5.0 Reagents
5.1 Gases

5.1.1 Oxygen: Ultra-high purity or zero-grade oxygen may be used for sparging and as the carrier gas for the DOC analyzer. The gas may contain at most 1 ppm total hydrocarbons and 1 ppm CO₂ . Typically, the UHP gas is listed as >99.993%, the zero-grade gas as >99.6%—it contains some nitrogen. Both gases should be passed through a drying trap filled with ascarite for final removal of CO₂ immediately prior to use.

5.1.2 Nitrogen: Ultra-high purity or zero-grade nitrogen may be used for pressure filtration. The gas should contain at most 1 ppm total hydrocarbons and 1 ppm CO₂ . Typically, the UHP gas is listed as > 99.998%. The gas is passed through a drying trap filled with ascarite for final removal of CO₂ immediately prior to use.

5.2 Dry chemicals

5.2.1 Ascarite: Thomas Scientific, Swedesboro, NJ.

5.2.2 Magnesium perchlorate (anhydrous): Fisher Chemical Co., Pittsburgh, PA.

5.2.3 Soda lime (4-8 mesh). Fisher Chemical Co.

5.3 Solutions

5.3.1 50% (w/w) phosphoric acid: Prepared by diluting the nominally 85% (w/w) concentrated acid (Fisher Chemical Co.) with CFDW.

5.3.2 AgNO₃ /H₃PO₄ : Mix 5 g of AgNO₃ (Fisher Chemical Co.) with 95 g 10% H₃PO₄ .

5.3.3 KHP stock solution: 4 mM potassium hydrogen phthalate (Aldrich Chemical Company, Milwaukee, WI) in CFDW.

5.3.4 30% (w/w) hydrogen peroxide: Fisher Chemical Co.

5.3.5 10% (w/v) sodium hydroxide: Mallinckrodt Specialty Chemicals Co., Paris, Kentucky.

5.3.6 0.1N hydrochloric acid: prepared from doubly distilled azeotrope.

6.0 Sampling
6.1 Sample bottle preparation

6.1.1 100 mL “Boston rounds”: Soak bottles overnight in room-temperature 10% NaOH. Drain, rinse three times with distilled water, three more times with 0.1N HCl and finally three times with distilled water. Oven dry overnight at 150°C. The green caps with integral teflon liners are cleaned by soaking for one hour or more in distilled water, rinsed with same then air dried. The removable teflon liners (which are added to the caps when dry) are cleaned by rinsing with distilled water, sonicating three times with acetone for fifteen
minutes followed by three more ultra-sonic treatments with dichloromethane. The liners are then rinsed with dichloromethane and oven dried at 150°C overnight.

6.1.2 40 mL “EPA vials”: Rinse each 40 mL vial three times with distilled water to remove dust and other fine particles. After air-drying, place in muffle furnace at 500°C overnight (12-16 hrs) then cool. Cap with green caps when cool. The green caps with integral teflon liners are cleaned by soaking for one hour
or more in distilled water, rinsed with same then air dried. The removable teflon liners (which are added to the caps when dry) are cleaned by rinsing with distilled water, sonicating three times with acetone for fifteen minutes followed by three more ultra-sonic treatments with dichloromethane. The liners
are then rinsed with dichloromethane and oven dried at 150°C overnight.

6.2 Niskin bottles: Use of “well-aged” Niskin bottles is recommended. Replace all O-rings with silicone ones and use either teflon coated stainless steel springs or heavy-walled silicone tubing. The stopcocks may be nylon, polypropylene or teflon but not PVC. The bottles should be free of oil and dirt and rinsed thoroughly with fresh water before the ship leaves port. At a test station or at the first station, the bottles should be well rinsed with seawater. Repeated lowerings and firings at 1-2000 m is recommended.

6.3 Drawing of samples: DOC samples are easily contaminated with organic compounds adsorbed from the air, from fingerprints or on the sampling ports. In order to keep the sampling ports as clean as possible for these samples, no Tygon Ò or phthalate containing tubing may be used in connection with the sampling ports prior to drawing the DOC samples. Ideally, DOC samples should be drawn first, and if not first, then immediately following the gas samples. The sample should be allowed to flow freely from the Niskin bottle for a few seconds to clean the port. No transfer tubing is necessary. The sample bottle should not be allowed to contact the sampling port, rather the sample should flow through a few cms of air before entering the bottle.
The bottles and caps are rinsed three times with a small volume of sample and then the bottle is immediately filled. Allow a sufficient headspace for sparging the sample.

6.4 Sample acidification: For open ocean seawater samples of 35ppt salinity, 5 mL of 50% H₃PO₄ should be added per mL of sample. The acid may be added immediately after the sample is drawn (if a clean environment for this work is available) or one may wait the 20-30 minutes required to sample the whole hydrocast, then acidify all the samples at the same time in the lab. Unless drawing the sample or acidifying, the bottles should be tightly capped at all times to avoid contamination of the samples from the ship's stack gases or fuel vapors.

6.5 Sample storage:

6.5.1 Refrigeration for short-term: Unless the samples will be analyzed immediately, they should be refrigerated at 2-4°C until analyzed immediately after acidification. This type of storage is acceptable for time periods ranging from a few hours to several months.

6.5.2 Freezing for long-term: If the samples are not to be analyzed during the course of the cruise, they should be frozen until time of analysis for best keeping. Immediately after acidifying, the samples should be placed in an aluminum block (specifically bored-out to maintain a tight fit with the sample vials) cooled to -20°C to achieve a rapid cooling of the samples. After one hour, the samples should be checked to see if they are frozen. Super-cooling often occurs. In this case a quick twist of the vial often encourages immediate
solidification of the sample with little or no brine formation. Once frozen, samples may be moved to a cardboard container for storage at -20°C. Samples should be kept frozen until analysis. Avoid thawing and slow refreezing of the samples as this encourages fractionation of the samples and brine formation.

7.0 Procedures
7.1 CFDW preparation: Carbon-free distilled water (CFDW) can be prepared by a variety of methods. However, no method is refined to the point that guarantees a DOC level below a certain limit. Thus it is imperative that the analyst continually check the quality of his blank water, maintain suitable quality control charts, and cross-check with other sources and analysts.

7.1.1 UV-H₂O₂ method: Low DOC water (<20 mMC)—either distilled, Milli-Q or reverse osmosis— is placed inside a one liter Quartz flask. One mL of 30% H₂O₂ is added and the solution tightly capped with a quartz stopper. The flask is then placed in direct sunlight on a cloudless day for 8-10 hours. This process is repeated 3-4 times, or until the instrument blank “levels-off”. Then the irradiation process is repeated once more without the additional H₂O₂ . After several days this solution becomes saturated with oxygen so one must be careful not to vigorously shake the solution. It is also a good idea to relieve the internal pressure from time to time.

7.1.2 Redistillation from persulfate: Very low DOC water (< 4 mMC, comparable to the UV-H₂O₂ oxidized CFDW) can be prepared by redistillation from per-sulfate. Milli-Q water is further purified by reverse osmosis then distilled in an all-glass still. This water is then re-distilled in 1L batches after addition of 1g K₂S₂O₈ and 1 mL 85% H₃PO₄ per liter of water (see Benner and Strom, 1993).

7.1.3 Milli-Q. Some Milli-Q systems are capable of achieving comparable quality water straight-away. However, this can only be verified by comparison against other sources and long-term reference solutions. Continual quality control is a must when this source of CFDW is used.

7.2 Standard preparation:

7.2.1 Distilled water standards: A series of reference solutions with a step-interval of approximately 32 mMC are prepared by sequential addition of the 4 mM KHP standard stock solution to 100 mL of distilled water. Add 0, 100, 200, 300, 400 and 500 mL of the standard stock solution to six 100 mL volumet-rics.
Fill to volume with the same CFDW used to make the reference water. To each add 500 mL of 50% H₃PO₄ . Seal and store at 4°C. The exact concentration of the standards can be calculated directly from the concentration of the stock solution:
            DOC ( mMC) = (( vol std * con.stock solution) / 100 ml)

7.2.2 A series of seawater based reference solutions with a step-interval of approximately 32 mMC are prepared by sequential addition of the 4 mM KHP standard stock solution to 100 mL aliquots of seawater water. It is best to use deep ocean seawater (> 1000m) or well filtered and aged surface water.
Weigh out the equivalent of 100 mL of seawater (mass = 100 mL * density at lab temperature—calculate density from measured salinity) into six 100 mL bottles. Add 0, 100, 200, 300, 400 and 500 mL of the standard stock solution DOC(mMC) vol std con. stock solution to the bottles in order. To each add 500 mL of 50% H₃PO₄ . Seal and store at 4°C. The exact concentration of the standards can be calculated from the concentration of the stock solution and the background DOC concentration as described below in section 8.3.2.

7.3 Blank determination: It is essential that all peak area measurements are corrected for the instrument blank. In order to do this, a CFDW sample is injected at regular intervals throughout the day's analysis run (see section 7.5). Typically, three injections of the blank water sample are made at a regular time interval (usually 4-5 mins). This water is acidified and sparged in the same fashion as the samples.

7.4 Response factor determination: There are two ways to determine the instrument response factor. The first involves running the complete set of standard solutions. Generally, this method is used only when a few or no samples are to be run that day due to its time-consuming nature. The second involves running only two standards (high and low) spanning the range of concentrations expected for that days run. Typically,
this method is used when a large number of unknown samples are to be run that day. The standards are then run both at the beginning and the end of the days run (see section 7.5).

7.4.1 Standard addition series: After running 3 or 4 warm-up samples (three injections of each) and a CFDW blank, the complete set of the standard addition series is run—again, three injections of each. Finally, a CFDW blank is run. The response factor is calculated as per the method in section 8.3.1 for dis-tilled
water or 8.3.2 for seawater based standards.

7.4.2 Two-point determination: When a large number of samples are to be run, a two-point calibration is practical. The two standards should bracket the extremes of that day’s runs. There should be a difference in concentration between the two of 60-120 mMC for typical open ocean samples. The two standards should be bracketed by CFDW samples to observe and correct for any change in instrument blank. This calibration is done twice: Once at the beginning of the day’s run and once at the end. By repeating the calibration at
both the beginning and end of the day’s run it is possible to tell if the instrument response factor was drifting during the day and to correct for any drift observed.

7.4.3 CO₂ gas standard calibration: Both of the proceeding methods assume that complete oxidation of the added standard is occurring. In order to verify this, one can by-pass the uncertainty of the oxidation step by injecting CO₂ in air standards. These should be obtained from a reliable source (e.g. in the U.S.,
NIST) with the concentration known to a precision of ±< 1 ppm. Calibrate the instrument response by injecting (in triplicate) a series of volumes then plot mean peak areas versus moles of CO₂ injected divided by your nominal injection volume. Remember that CO₂ is not an ideal gas so the Van der Waals equation of state must be used to calculate the number of moles injected from the observed volume and room temperature and pressure. The slope of this line should be identical with your normal calibration.

7.5 Analytical protocol: A typical day’s run consists of 3-4 warm-up seawater samples, a CFDW blank, a calibration set, a series of samples run in groups of 4-6 with CFDW blanks interdispersed, a CFDW blank, a second calibration set and a CFDW blank. The warm-up seawater samples are run to minimize and stabilize the instrument background/blank. The same sample is run repeatedly so it will be possible to see if the instrument blank has stabilized. If the instrument is still drifting after 4 samples, run a few more until a repeatable signal is obtained for the warm-up sample before beginning the high-low calibration set.

7.6 Sample injection: All samples (warm-up, CFDW, calibration, or unknown) are injected in triplicate. Samples are first sparged with CO₂ -free gas (see section 4.2) then the syringe is filled. First, rinse the syringe three times with sample, discarding each rinse, then over-fill the syringe. Invert to expel all air bubbles and express excess sample. The sample is then injected into the furnace. Different instruments have different procedures but a common thread is the injection of samples at regular time intervals to minimize instrument background/blank variation. While making one run, sparge the sample for the next analysis. All NDIR data is digitized and recorded by computer.

7.7 Post-Analysis: Following the sample analysis runs, a recalibration sequence and CFDW blanks must be done. Finally, the CFDW used for the day’s run is compared with the long-term standard to check for drift/contamination. The data are reprocessed according to the equations in section 8.

8.0 Calculation and Expression of Results
8.1 Peak Screening : Before calculating the mean corrected peak area for each sample, it is imperative that the peak integration be verified. Check that the integration baseline is correct—intercepting the middle of the baseline noise at both the beginning and end of each peak. Reject peak areas (or re-integrate peaks) where improper baseline is observed, poor or irregular peak shape is observed or there is other indications of a
bad injection. Average all acceptable peaks for each sample or blank run.

8.2 Blank Correction: Early in the lifetime of the combustion tube, the instrument blank tends to slowly decrease. In these cases, interpolate the instrument blank between CFDW runs to blank correct the sample runs. Use a simple linear interpolation. Later in the combustion tube lifetime, the instrument blank can be stable. On these days, average the instrument blank over the course of the days run. Calculate the mean
corrected peak area by subtracting the appropriate instrument blank.

8.3 Response factor determination

8.3.1 Distilled water standard addition series: Plot the mean corrected peak area as a function of the concentration of the distilled water standard. Fit a linear regression to the points. The slope of the line is the instrument response factor in area units per micromole.

8.3.2 Seawater based standard addition series: Because the seawater used to make the seawater-based standard addition series contains DOC, one must do the calculation twice. The first pass determines the background DOC level, the second pass to determine the concentration of each standard. First plot the
mean corrected peak areas vs. the amount of DOC added calculated by the following formula:
            DOC - add(mMC) = ( vol. std * conc.stock solution) / (( mass of sea water/density) + vol. std. + vol. acid)
Fit a linear regression to the points. The slope of the line is the instrument response factor in area units per micromole. The DOC background can be calculated from the y-intercept:
            Background DOC = y - intercept / slope
Now the exact concentration of each standard can be calculated taking into account the DOC background and the acid+std. dilution effect:
            DOC(mMC) = (( vol. std * conc.stock solution)  + (bkgrd * mass of sea water / density)) / (( mass of sea water/density) + vol. std. + vol. acid)

Now re-plot the mean corrected peak areas vs. the actual concentration of the standard solutions. Fit a linear regression to the points. The slope of the line is the instrument response factor in area units per micromole. Note that this slope includes an adjustment for the amount of acid added. To accurately determine the sample concentrations, they will need to be corrected for the amount of acid added (see section 8.4.4).

8.3.3 Two-point determination: After running the two standards, correct their mean areas for the instrument blank, then calculate the instrument response factor:
            slope =( mean net area ( high - standard) - mean net area ( low - standard)) / (conc ( high - standard) - conc ( low - standard))
This calibration is done twice daily. Differences between the morning and afternoon calibrations greater than 3% of the mean calibration mean that the instrument calibration is drifting and the response factor must be interpolated for that day’s run (Section 8.4.2, below). Differences of less than 3% are most likely due to noise. Calculate the average of the two response factors.

8.4 Sample analysis

8.4.1 Blank determination: Plot the mean area for each of the day’s CFDW runs (in area units) versus run number. If no trend is apparent, then the mean of that day’s CFDW runs should be calculated. Otherwise, to determine the blank, a simple linear interpolation is generally sufficient. For example, find the dif-ference
between two successive blanks, count the number of runs in between and divide the difference by this count plus one. The quotient is the step difference in the blank for successive runs.

8.4.2 Response factor interpolation: When the difference between the morning and afternoon calibrations is greater than 3% of the mean response factor, it is necessary to interpolate the response factor for calculation of sample concentrations measured during the day. A simple linear interpolation is used. To find the step difference in the calibration factor, find the difference between the two calibrations and divide by the number of intervening runs plus one.

8.4.3 Zero water adjustment: The CFDW used to make instrument blank measurements throughout the day is often > 0 mMC. When this area is subtracted from the sample peak areas, it results in an over-correction and an under-estimation of the DOC concentration. Thus it is important to adjust the blank correction. This is done by adding the concentration of DOC in the CFDW back to the sample. (For example see sections 8.4.4 and 8.4.5.) The DOC concentration of the CFDW is measured by comparing it to a “primary” DOC free distilled water which has very low DOC and has been set aside for this purpose. It is (by definition) the CFDW that gives the smallest apparent instrument blank.

8.4.4 DOC calculation: Use the following formula to calculate the DOC concentration
of a sample:
        DOC = | mean sample area - blank ( CFDW) / reaponse factor + DOC( CFDW) | * dil. factor
Where:
        mean area sample = mean peak area (in mV-secs) for three injections of the sample
        blank (CFDW) = peak area (in mV-secs) for instrument blank, either the daily mean or the interpolated value
        response factor = instrument slope as appropriate - either the daily mean or the interpolated value (mV-secs/mMC)
        DOC (CFDW) = apparent DOC concentration of the CFDW used to measure the instrument blank that day
        dil. factor = dilution factor: Vol (sample)/[Vol (sample) + Vol (acid)];
        use only if seawater standards are used to calibrate slope
8.4.5 Sample spreadsheet calculation:

9.0 Quality control/quality assessment
9.1 QC charts: In order to have tight quality control over the analyses, plot the following on a daily basis. Instrument drift or bad blanks will be readily apparent from any trends in the data.

9.1.1 Daily blanks (mean with range in mMC units): Each day plot the mean and the range of all CFDW blanks. A spurious blank will be readily apparent as an anomalously high value; the range should decrease as the combustion tube ages. Note that range = high and low not ± one standard deviation. Also plot the value of the reference CFDW used to check the bottle of CFDW prepared/ used daily.

9.1.2 Daily response factors: Each day plot the mean and the range of both calibration tests. Units = area per unit concentration = milli-volt-secs/micro-molar carbon.

9.2 Quality assurance: Although the HTC/DI-DOC analytical method has begun to develop some acceptance within the marine chemical community, it is imperative that each investigator demonstrate the validity of their own analyses. This may be accomplished via several mechanisms: (1) oxidation of recalcitrant compounds, (2) CRM analyses, (3) comparison with a referee method, and (4) shipboard reference material.

9.2.1 Hard-to-oxidize standards: The simplest means of determining the “completeness”  of oxidation of any particular technique is to analyze a set of sea-water samples spiked with a variety of “recalcitrant” organic compounds. Percent yield of CO₂ based on the amount of each standard added is a direct measure of the efficiency of oxidation of the particular method. Suitable test compounds are: alginic acid, caffeine, EDTA, fulvic and humic acids, soluble starch, urea, 2,2'-dipyridyl, and oxalic acid.

9.2.2 Certified Reference Material (CRM) analysis: Alternatively, if a certified reference seawater were available, then one could check for completeness of oxidation directly. Unfortunately, such a material is not available at this time but may become available in the future.

9.2.3 Referee analysis: Two mechanisms exist for comparison with a “referee” method. First, is the often tried inter-lab comparison exercises. While these are useful in determining relative accuracy, they often fail to demonstrate whether any of the methods involved achieved truly complete combustion. The second method is to compare the HTC/DI-DOC technique to sealed-tube combustion. Wangersky (1975, 1993) and others have cited this technique as being the most likely candidate for achieving complete oxidation of all the organic carbon in a sample. A direct comparison of samples analyzed by both methods will give an estimate of the “completeness of oxidation” of an individual technique.

9.2.4 Shipboard reference analysis: In the absence of a CRM-seawater standard, it is possible to simulate one over the course of a cruise. Collect a large volume (>1L) sample at the test station or the first hydro-station from >2000m. The DOC in this sample should be old and relatively stable and recalcitrant. Careful
storage at 4 ° C should preserve it for the course of most normal cruises. Analysis of this sample from time-to-time throughout the cruise will serve as a reliable reference material.

10.0 Notes
10.1 General precautions: DOC is the most easy to contaminate substance to be measured in oceanographic samples. As such, stringent anti-contamination protocols must be adhered to at all times. Most important to observe is what others around you may be doing which could adversely affect your samples. A general rule of thumb for DOC contamination is: if you can smell it, then it is probably trouble.

10.1.1 Sampling: No amount of post-analysis mathematical manipulation can salvage poorly drawn or contaminated samples. Every precaution should be taken to collect samples in the cleanest environment possible. DOC samples should be drawn first to avoid contamination from the tubing used as transfer
lines in the collection of most gas samples. Tygon Ò is especially trouble-some. Most troublesome is the rosette interloper. Watch-out for someone who wants to just hop ahead for one sample. Their technique is generally poor and their presence is especially erratic making any problems they cause intermittent. Above all else, keep you fingers out of the samples. Do not trust rubber/plastic gloves to do anything except keep your hands from getting salty.

10.1.2 Sample storage: DOC samples are prone to contamination at this stage as well. Avoid storing samples in refrigerator/freezers which contain copious amounts of organic material, especially fresh (and not-so-fresh) fish. Check-out the reliability of the sample storage bottles carefully and well in advance of when the samples are to be collected. Caps and cap liners are often the cause of inadvertent and highly variable contamination. Do not ever ship sample containers filled with strong acids or bases to clean them while in
transport.

10.1.3 Lab-space requirements: Just as sample storage space must be odor free, so must the analytical space be free of organic vapors and heavy dust loads. Good ventilation with clean outside air free of organic solvent vapors is a must.

10.2 Possible modifications:

10.2.1 Blank water: Presently, CFDW serves as an adequate instrument blank checking material. However, in terms of good laboratory practices and a rigorous analytical chemical approach, carbon-free seawater is the unquestionably superior material for measuring the instrument blank. Development of a process to produce this material quickly, reliably, easily and cheaply is a priority.

10.2.2 Standard solutions: Several standard compounds (glucose, KHP, etc.) are used as a calibration material as well as both distilled and seawater. Ideally, a single organic compound in a single matrix should be used by the entire community. This protocol recommends KHP in seawater—either deep (>2000m) ocean water or filtered and well-aged coastal seawater. Analytically speaking, one should use the same matrix for blanks and standards as in the samples.

10.3 Backward compatibility: It is now apparent that a fair degree of correspondence exists between the historical analyses and the newer HTC/DI-DOC methods. Although there is some evidence that the HTC/DI-DOC technique achieves a higher degree of oxidation efficiency, this increase appears to be small: 10-20%. Three obstacles to a direct comparison of present analyses to the data in the literature exist:
Temporal variability, spatial variability and precision of analysis. There is little the analysts can do to avoid the first two; indeed, studying these is one of the objectives of oceanography. However, the third needs considerable attention.

10.3.1 Precision problems: Historically, DOC concentrations were regarded as both relatively uniform and invariant, in part, due to the relatively poor precision of the analyses. The uncertainties in these older methods were on the order of 10-25% of the DOC and 10-50% of the gradients. Thus much of the oceanographic
information was lost to the imprecision of the methods. By achieving a precision of ±1 mMC, this situation can be greatly improved and a much more adequate picture of the oceanic organic carbon cycle will be revealed.
This level of precision (± 1-2%) can be achieved and should be the goal of each and every analyst.

10.3.2 Deepwater reference: One of the more analytically useful features of DOC is that the deep oceanic concentrations of DOC are relatively low, virtually invariant in time and with extremely shallow gradients. The deep water DOC serves as a natural CRM for controlling the quality of the DOC analyses. Thus, each and every cruise where DOC is measured an effort should be made to collect and analyze samples from >2-3000m as a check against consistency. It will be on the basis of these analyses that we can best compare the results of the newer analytical techniques to the historical database.

10.4 Volatile organic carbon: By virtue of the nature of the analytical protocol there is little  that a DOC analyst can say regarding the presence or distribution of volatile organic compounds as these were stripped from the samples during the sparging step. For most of the oceanic samples this is of little consequence as these compounds comprise only a tiny fraction of the total DOC pool. However, in certain environments (e.g., sediments, trapped bottom water/fjords, arctic basins, coastal waters and estuaries), this may not be the case and analysts using this technique in these areas should be aware of the potential possibility for analytical artifacts due to the presence or variable distribution of volatiles.

11.0 Intercomparison
11.1 Other methods: MacKinnon (1978) and Gershey et al. (1979) were among the first to try a direct comparison between methods. Although their sealed-tube measurements were not as easy to perform as the newer HTC/DI-DOC technique, they do provide a similar picture when compared with both the UV and persulfate techniques. The slightly higher yields of the sealed-tube analyses preceded the current HTC/DI-DOC revolution by many years, but the lower precision of the competing analyses did not warrant a significant investment of time nor resources due to the limited statistical reliability regarding this difference.

11.2 Recent HTC comparisons:

11.2.1 Seattle Workshop: In the late spring of 1991, a community-wide international workshop on the analysis of DOC by various methods —principally by HTC/ DI-DOC — was held in Seattle. The results of this workshop are now published (Mar. Chem., 41(1-3) (1993). The reader is referred to this report for essential reading regarding the development of the method. While the community failed to achieve an acceptable level of agreement between analyses on common samples, considerable progress to resolving these differences
was made and many recommendations for future modifications and improvements are included.

11.2.2 Bermuda paper: Sharp et al. (1994) have published a comparison of several of the commercially available HTC/DI-DOC analyzers. While the data contained in this report is somewhat limited due to the time and logistical constraints imposed, there is some useful information in this report regarding modifications (both realized and potential) to these various instruments.

11.2.3 EqPac comparison: Sharp et al (submitted) have compared several HTC/DI-DOC methods with the modified persulfate technique on a large suite of samples collected during two of the U.S. JGOFS EqPac cruises in 1992. This comparison is unique in the large number of samples involved and the high degree of correlation between several of the analysts. The greater precision of the HTC/DI-DOC analysis versus the modified persulfate technique is also apparent. This paper stands in direct contrast to the Seattle Workshop where values of 30- >300 mMC were reported for a single sample. In this report, typical variations between analysts were on the order of a few mMC.

11.2.4 2 nd community-wide comparison: A second, community-wide, international DOC comparison is in progress (see Sharp et al., 1994). The first stage involved the shipping of blank water, low DOC seawater and spiked seawater to the analysts. The samples were identified to the analysts so they could see how well they were doing relative to a given standard. The second stage will consist of a set of blank water, known standards and several unknown samples. Results will be reported with the analysts identified at a future date.

12.0 References
Benner, R. and M. Strom. (1993). A critical evaluation of the analytical blank associated with DOC measurements by high-temperature catalytic oxidation. In: Measurement of Dissolved Organic Carbon and Nitrogen in Natural Waters (eds. Hedges and Lee). 41(1-3):153-160.
Gershey, R.M., M.D. MacKinnon, P.J. Le B. Williams and R.M. Moore (1979). Comparison of three oxidation methods used for the analysis of the dissolved organic carbon in seawater. Mar. Chem., 7:298-306.
MacKinnon, M.D. (1978). A dry oxidation method for the analysis of the TOC in seawater. Mar. Chem. 7:17-37.
Sharp, J. (1973). Total organic carbon in seawater—Comparison of measurements using persulfate oxidation and high temperature combustion. Mar. Chem., 1:211-229.
Sharp, J. and E.T. Peltzer (1993). Procedures subgroup report. Mar. Chem. 41:37-49
Sharp et al. (1994). Bermuda instrument comparison. Jour. Mar. Res.
Sharp, J.H., R. Benner, L. Bennett, C.A. Carlson, S.E. Fitzwater, E.T. Peltzer and L. Tupas (submitted). Dissolved organic carbon: Intercalibration of analyses with Equatorial Pacific samples.
Van Hall, C.E., J. Safranko, and V.A. Stenger (1963). Rapid combustion method for the determination of organic substances in aqueous solution. Anal. Chem. 35:315-319.
Wangersky, P. J. (1975). Measurement of organic carbon in seawater. pp. 148-162, In: Analytical Methods In Oceanography, T.R.P. GIBB (ed.), American Chemical Society, Washington, DC.
Wangersky, P. J. (1993). Dissolved organic carbon methods: A critical review. In: Measurement of Dissolved Organic Carbon and Nitrogen in Natural Waters (eds. Hedges and Lee). 41(1-3):61-74.