2.0 Definition
2.1 Primary production is defined as the uptake of inorganic carbon into particulate matter as:
            Primary production = mg carbon / m³ / day

2.2 A vertical profile of production measurements can be integrated to yield a production rate per unit area in units of:
            Primary production = mg carbon / m²  / day

3.0 Principle of Analysis
The rate of carbon fixation (= primary production) by autotrophs in seawater is measured by tracing the uptake of radioactive ¹⁴C from the dissolved inorganic form to the particulate organic form. Radiocarbon is added at a known or assumed ratio to the total inorganic carbon content of the seawater sample. The uptake of radiocarbon by the particulate phytoplankton is converted to total carbon uptake by conversion using this
radiocarbon:total carbon ratio. Inorganic carbon uptake into particulate inorganic carbon is not measured as the samples are acidified before analysis. The method is easily expanded to include measurements of size-fractionated particulate production or the net production of radiolabelled dissolved organic carbon.

4.0 Apparatus
4.1 Scintillation Counter: The measurement of radioactivity is typically done by liquid scintillation counting. There are a large number of appropriate instruments, each of which has unique characteristics. As the use of radioisotopes usually involves some level of additional training and expertise in each research institution, it is assumed that the appropriate techniques for the use of the available scintillation counters is available.

4.2 Quench Corrections: Most scintillation counting techniques require the assessment of the amount of quenching of the scintillation signal by the scintillation cocktails and the particle and dissolved solutions. In some cases, an external gamma source is used to assess quenching of individual filter and liquid samples for conversion of counts per minute (CPMs) to disintegrations per minute (DPMs). Internal standard techniques are also available. Again, the investigator should become familiar with the appropriate quenching corrections for their individual applications.

5.0 Reagents and Supplies
5.1 Stock ¹⁴C sodium bicarbonate (aqueous, specific activity 5 mCi/ml, 5 mCi lots):available from a variety of vendors.

5.2 Teflon bottles for holding stock 14 C solutions (100 mls) and for preparing the stock solutions (500 mls).

5.3 Working Solution.: A sodium carbonate (anhydrous, Aldrich 20, 442-0) solution is prepared by dissolving 0.15 g in 500 ml Milli-Q water in a 500 ml, acid-cleaned teflon bottle. The 100 ml teflon bottle for holding the ¹⁴C stock solution is rinsed with the carbonate solution then 60 mls of the carbonate solution is transferred to the 100 ml teflon bottle. The ¹⁴C stock is added to the 60 mls of carbonate solution in the teflon bottle (the actual activity of the stock solution is often variable so the final specific activity is approximately 80 mC ml ^-1). The working solution is stored refrigerated (5°C) until use. Some labs recommend further purification of the stock solution to remove any residual trace metal contamination. Many labs make the stock solution in individual small aliquots so that a new aliquot can be used for each daily incubation or cruise. The stock solutions should always be stored in a non-contaminating container (preferably teflon, never glass).

5.4 Acid Cleaning Solution (0.5 N HCl; Baker Instra—Analyzed): prepared using Milli- Q water. A small aliquot of this solution can also be used for the filter acidification steps.

5.5 Ethanolamine (Sigma): used to prevent the radiolabelled inorganic CO₂ from escaping to the atmosphere. Other compounds are also acceptable.

5.6 Scintillation Cocktail: As with scintillation counters, there are a wide variety of scintillation cocktails available on the market. Some of the newer varieties are non-toxic. Each has different efficiencies and quench characteristics. It would be appropriate to compare the chosen cocktail with other labs and with commercially available ¹⁴ C standards.

5.7 Preparation of Reagents: Polyethylene gloves should be worn during handling of materials which come into contact with isotope solutions. Gloving precautions fulfill two roles, protecting the wearer from  contamination during handling of any materials that have been exposed to isotopes and protecting the living samples from contamination by human skin. Trace metal clean techniques should be used whereever
possible.

5.8 Incubation Bottles: Polycarbonate 0.25 l bottles are used for the productivity incubations. New bottles are soaked for 72 hours in a 5% solution of Micro detergent. Bottles are then rinsed thoroughly with deionized water, and subsequently soaked for 72 hours in the acid cleaning solution. The acid is discarded and the bottles rinsed 3 times with Milli-Q water and then soaked in Milli-Q for at least 48 hours. Once a new bottle has been cleaned as described above, then cleaning between cruises consists of soaking in the acid cleaning solution for several days and rinsing 3 times with Milli-Q. In some applications it may be appropriate to use smaller bottles, however, there is a general feeling (and some published papers) that suggests that larger
bottles are preferable. Large bottles have a smaller surface-volume ratio and thus minimize contamination and biological problems associated with the container walls. Larger bottles also result in much large volumes of radioactive waste. For investigations in any environment, investigators should conduct their own experi-ments
to determine the appropriate container volume. For the measurement of productivity on consecutive days (as on a long transect cruise), it may be advisable to have two or three complete sets of incubation bottles to allow for adequate washing of each set between incubations.

5.9 Pipet Tips. In the system described here, all pipette samples are 0.25 ml and the entire operation can be accomplished with a single 0.25 ml Eppendorf style pipetter. Before use for inoculating the productivity samples, pipette tips are rinsed 3 times in acid cleaning solution followed by three rinses in Milli-Q water. Cleaned tips are stored in a plastic bag or polyethylene glove until use.

5.10 Trace metal clean water sampling system: The system for collecting the seawater should be capable of collecting an uncontaminated seawater sample. In the open ocean this is a non-trivial and usually impossible task. GoFlo style bottles are preferred because they are deployed in a closed configuration (they go through the dirty, air-sea interface without contaminating the inside of the bottle). They also lack internal mechanisms (e.g. springs). Traditionally, these bottles are deployed using cloth coated Kevlar hydrowire and plastic coated weights and messengers. Alternative systems can also be employed (e.g. the so-called trace-metal clean rosettes). The cleanliness of samples collected with all of these systems should be documented by a lab
qualified to measure trace metals at the appropriate concentrations. It is also good practice to occasionally or routinely collect samples for trace-metal analysis during the cruises to guard against contamination by sloppy handling during the cruise. The GoFlo bottles should be acid cleaned and precautions should be taken to ensure that the bottles do not become contaminated during a cruise.

6.0 Sampling
6.1 Shipboard sampling:

6.1.1 Sampling Depths. A set of 8 depths bracketing the entire euphotic zone (approximate light levels include 95% – 0.6%) should be selected. The selection process can vary depending on the application. Even spacing of samples between the surface and the 0.2-1.0% light depth is usually appropriate. Some investigators select depths based on the chlorophyll profile. It is important for the subsequent integrations of production data that the deepest depth be below the level of significant production (light bottle approximately equal to dark bottle).

6.1.2 Hydrocast. Before dawn, seawater samples should be obtained using the Go-Flo bottles deployed on a Kevlar line. The bottom weight on the line is wrapped in plastic. The line is lowered over a plastic-wrapped sheave, and bottles are triggered with plastic coated brass messengers. The hydrocast should be conducted in time to allow sample processing and deployment of the in situ array before dawn.

6.1.3 Dispensing Sample. Polyethylene gloves are worn at all times during handling of samples. Productivity flasks are filled directly from Go-Flos under low light conditions. Bottles are rinsed 3 times before filling. Five bottles are filled for each productivity measurement. In some labs, the entire sample is removed from the GoFlo bottle into an acid-cleaned carboy. This carboy is then transferred to the lab and all subsequent manipulations occur in a clean environment.

6.1.4 Isotope Inoculation. Under low light conditions, 0.25 ml of the ¹⁴C working solution (20 mC) is added to each bottle using an acid cleaned polypropylene pipet tip. One bottle is immediately filtered for a time zero control using the methods described below.

6.1.5 Dark Bottle. A dark bottle is made by wrapping one of the 5 inoculated bottles in aluminum foil and placing it in a black cloth bag with a velcro closure. If the dark production data are important beyond the minor correction of the light production data, they should also be replicated (triplicate).

6.2 In Situ Incubation Procedures

6.2.1 The method described here involves an in situ incubation of the productivity samples at the depths of collection. In situ incubations allow the samples to be exposed to the natural temperatures and light levels (both intensity and spectral quality). Deckboard incubators are also acceptable and in some instances (e.g. production in ice covered areas) are the only acceptable method. Neutral-density screens (e.g. perforated nickel) are usually employed (Lohrenz et al., 1992). Spectrally-corrected ‘blue’ screens have been recommended for ¹⁴C deck incubations (Laws et al., 1989). The most realistic conditions possible with regard to light quality and temperature are encouraged.

6.2.2 Preparations. The dark bottle and 3 light bottles are hooked together with an appropriate system for suspension on the in situ array. This can be a simple arrangement of plastic electrical cable ties or a complex plastic rack. The incubation bottles should be kept dark until deployment. The suspension apparatus should be tested for recovery under rough conditions.

6.2.3 Deployment. The productivity array should be deployed before sunrise. The bottom weight, attached to a premeasured polypropylene line, is lowered first. Each group of bottles is then secured to hooks attached to the line at the depth that the sample was originally collected. The entire productivity line is suspended from an orange plastic float, which is attached to a spar equipped with strobe flash and VHF radio beacon. Time and position of deployment are recorded.

6.2.4 Recovery. Approximately 0.5 hours after sunset, the productivity array is recovered. Sample bottles are detached from the line and placed in dark plastic bags until filtration.Filtrations should be carried out as soon as possible since respiration and grazing continue once the bottles are onboard. Time and position of recovery are recorded.

7.0 Procedures
7.1 Sample analysis

7.1.1 Total Radioactivity. A 0.25 ml aliquot for counting total added ¹⁴C activity is removed from each incubation bottle with a 0.25 ml pipet and placed in a scintillation vial (vial size depends on the scintillation counter, here assumed as 20 ml vial) containing 0.25 ml ethanolamine (Sigma). The mixture is held at room temperature until subsequent liquid scintillation analysis.

7.1.2 Filtration. Maintaining low light conditions, an aliquot is withdrawn from each productivity bottle using a plastic syringe. In most environments, a 50 ml aliquot is adequate. In some environments, a smaller volume may be appropriate if the filter clogs before 50 mls has been filtered. Some investigators filter the entire sample volume to ensure that large, rare algae are included. The aliquot is filtered onto a 25 mm Whatman GF/F glass fiber filter maintaining vacuum levels of 70 mm Hg or less. The filter is not rinsed (though this is also a debated point). The filter is placed in a 20 ml glass scintillation vial, covered with 0.25 ml 0.5 N HCl (to remove the inorganic carbon), and held at room temperature until subsequent processing.

7.1.3 Filter Processing. The productivity sample vials are uncapped in a fume hood, and allowed to dry overnight. This procedure insures complete removal of unfixed inorganic ¹⁴C. A 10 ml aliquot of liquid scintillation cocktail is added to the dried filters.

7.1.4 Total Radioactivity Sample. 10 ml of liquid scintillation cocktail plus 2.5 ml Milli- Q water are added to the vials containing the 0.25 ml sample and 0.25 ml ethanolamine (see above). The mixture is shaken vigorously. This method produces a uniform jell with Aquasol and some other cocktails. However, each cocktail is different in the way it handles large amounts of aqueous solution and an alternative mixture might be required.

8.0 Calculation and expression of results
Rate Calculations. DPM values are converted to daily productivity rates using the following equation:
        Production (mg Cm ^-3d ^-1 ) =((SDPM/V) * (W * 0.25 x 10 ^-3 )/TDPM) *(1.05/T)
where:
        SDPM = DPMs in filtered sample
        V = volume of filtered sample (liters)
        TDPM = Total ¹⁴C DPMs (in 0.25 ml)
        W = DIC concentration in samples (approx 25000 mg Cm ^-3 ; should be measured for non-oceanic habitats)
        0.25 x 10 ^-3 = conversion of pipette volume to liters
        1.05 = correction for the lower uptake of 14 C compared to 12 C
        T = time (days)

8.1 This calculation is made for each light bottle, and the triplicate values are averaged. A similar calculation is made for time zero and dark bottle samples. All values should be reported separately. In some applications, the dark bottle rate is subtracted from the mean rate for the light bottles to correct for non-photoautotrophic carbon fixation or adsorption. At the bottom of a profile, dark bottle values are often equal to light bottle values and some (very small) negative production rates can occur by subtracting dark from light values.

8.2 Integrated Water Column Production. The individual depth measurements of daily production are used to calculate water column integrated production (mgCm ^-2d ^-1 ) by trapezoidal integration. The rate nearest the surface is assumed to be constant up to 0 m, and a zero rate is assumed for an arbitrarily deep depth (e.g. 200 m). The production at each pair of depths is averaged, then multiplied by the difference between the two depths to get a total production in that depth interval. These depth interval values are then summed over the entire depth range to get the integrated production rate.

9.0 Quality Control
The measurement of primary production generally has no independent method for calibration. Intercomparrison of techniques is also difficult without explicit activities on the same ship or same station. Data are generally evaluated for “reasonableness” in the context of other measurements in the area or other measurements by that lab group. The coefficient of variation for replicate variations should be < =  10% (Richardson 1991).

10.0 Notes
Safety Precautions and Regulations. The use of radioisotopes is more carefully controlled in most countries than other analytical compounds used in oceanography. Each investigator will have to follow the specific guidelines appropriate for their situation. Issues like waste disposal and the required documentation and training vary widely. It is imperative that people who use isotopes are familiar with the safety issues associated with the use of each isotopes and with general practices for safe handling and disposal of
isotopes.

It is important to avoid exposure of productivity samples to high light. This is most important for samples collected from deep in the euphotic zone that are photo-adapted to very low light levels. Short-term exposure to high light can both enhance (provide more light for photosynthesis) or degrade (light shock) the photosynthetic performance of the phytoplankton. As stressed above, it is extremely important to avoid even trace levels of contamination by metals. Collaborations and interactions between biological and trace-metal chemists help greatly in the development of the appropriate “trace-metal clean awareness” by the
biologists.

11.0 References
Fitzwater, S.E., G.A. Knauer and J.H. Martin (1982) Metal contamination and its effects on primary production measurements. Limnol. Oceanogr. 27: 544-551.
Joint, I., A. Pomroy, G. Savidge and P. Boyd (1993). Size-fractionated primary productivity in the northeast Atlantic in May-July 1989. Deep-Sea Res. II 40: 423-440.
Laws, E. A., G.R. DiTullio, P.R. Betzer, D.M. Karl and K.L Carder (1989). Autotrophic production and elemental fluxes at 26 o N, 155 o W in the North Pacific subtropical gyre.Deep-Sea Res. 36: 103-120.
Lohrenz, S.E., D.A. Wiesenburg, C.R. Rein, R.A. Arnone, C.T. Taylor, G.A. Knauer and A.H. Knap.(1992). A comparison of in situ and simulated in situ methods for estimating oceanic primary production. J. Plankton Res. 14:201-221.
Richardson (1991). Comparison of 14 C primary production determinations made by different laboratories. Mar. Ecol. Prog. Ser. 72:189-201.