VOL. 82, NO. 31 JOURNAL OF GEOPHYSICAL RESEARCH OCTOBER 20, 1977 Methane Production in the Waters off Walvis Bay MARY I. SCRANTON • AND JOHN W. FARRINGTON Department of Chemistry,Woods Hole Oceanographic Institution, Woods Hole, Massachusetts 02.543 Nine stations were occupied in the vicinity of Walvis Bay, Namibia, during a detailed study of the distribution of methane in this highly productivecoastal environment. The principal featuresof the observed coastal methane distribution included (1) excess methane in the mixed layer of from 2 times to greaterthan 300 timessolubility equilibrium with the atmosphere, (2) a subsurface maximum,located in the top of the pycnocline, at whichconcentrations ranged from 2.6 to 440 times solubility equilibrium, (3) an intermediate depth minimum, where concentrations were comparableto those offshoreat similar depths and which we attribute to the influence of onshore movement of subsurface offshore water, and (4) a bottom maximum, which we attribute to input of methane to the water column from the anoxic sediments in the Wal¾is Bay area. An attempt was made to identify the relative importance for methane supplyto the coastal mixed layer of in situ biological production and of eddy diffusive and advective transport of methane-rich water which has been in contact with the bottom at the coast.Calculations suggest that both in situ productionand physical processes are major sources of excess methane for the highly productivecoastal surfacewaters. However, the complicated circulation patterns make quan- tiffcation extrem. ely difficult. Upwelling areas,suchas the one near Walvis Bay, Namibia (formerly South-West Africa), are known to be regions of extremely high biological productivity.In Walvis Bay the high surface productivityis accompanied by rapid accumulation of organic matter in the sediments [Boonet al., 1975;Bremner, 1974], which thus becomeanoxic. Methane is produced in anoxic environments asa finalstep in the anaerobic degradationof organic matter [ Wolfe, 1971]. It may alsobe produced in oxygenated ocean waters, probably as a by-product of algal metabolism and/or anaerobic activity in the gutsof zooplanktonor higher marine animals[Scranton and Brewer, 1977]. The occurrence of both reducing sediments andhigh primary productivity near Walvis Bay suggested that a study of this area might allow us to determinethe relative importance of methane input from coastal sediments and of in situ methaneproduction in controllingthe methanedistribu- tion in th e watercolumn of a productive coastal area. The data which will be discussed in thispaper were collected on cruise 93 of the R/V AtlantisH (AII 93) to the Walvis Bay region during late December 1975 and early January 1976. Relevant station locationsare shown in Figure 1. METHODS Temperature, salinity, oxygen, phosphate, and methane measurements were made at all stationsdiscussed. (See Fig- ures2-7. Tabulated data from water samples taken on All 93- 3 can beobtained from J. W. Fartington, Woods HoleOcea- nographic Institution.) Temperatures were obtainedfrom re- versing thermometers and from expendable bathythermo- graphs (XBT's) at the start of each station. It was surmised that observed differences between XBT and thermometer tem- peratures were due to ship drift in areasof stronghorizontal temperature gradients and that the reversing thermometer measurements, taken at the time at which the water samples Werebeing collected,were more appropriate for comparison with nutrient and methane measurements. XBT's were cali- brated by shifting the traces to agree with the thermometer values.Salinity was determined with a conductive salinometer at WoOds Hole about 6 weeks after sample collection. • Now at U.S. Naval Research Laboratory, ChemicalOceanography Branch, Washington, D.C. 20375. Copyright ¸ 1977 bythe American Geophysical Union. Oxygen concentrations were determined by a modification of the Winklet method [Carpenter, 1965]. Phosphate concen- trations were determined by the molybdenum blue method of Murphy and Riley [1962]. Temperature, oxygen, phosphate, and methane were measured within 24 hours of collection on unfiltered samples. Methane measurements were made by a modificationof the technique developed by Swinnerton et al. [1962a,b] and Swin- nerton and Linnenbom [1967] as describedby Scranton and Brewer [1977]. Briefly, this method involves transfer of a sampleunder methane-free helium from a ground-glass stop- pered bottle into a glassstripper. Purified helium gas is bub- bled through the water in the stripper for 20 min, and the dissolved gases are thus extracted. The helium containing the extracted gases is dried by passage through a magnesium per- chlorate drying tube, and the methane is then removed by passage of the gas through an activatedcharcoaltrap main- tained at dry ice-acetonetemperatures. Once the methane from the sample has been completely trapped, the trap is heated for I min by using a hand-held hair dryer.The methane is injected into a l-in. OD silica gel(60-80 mesh) columnin the oven of an HP 5410 gas chromatograph equipped with dual- flame ionization detectors. Measurements of atmospheric methaneconcentrations were obtained while the ship was steaming betweenstations. Sam- ples werecollected by sucking air (using a vacuum pumpin the main lab) throughabout 150ft (45.72 m)of copper tubingand then througha calibrated air sample loop (115 ml) attached to a six-way gas sample valve. The air inletwaspositioned on the top deck of the ship forward of the smokestack. Air samples were only taken while the ship was under way. For each sample, air was sucked throughthe tubingfor about 10 min at 250 ml/min. Then the pump was shutoff, and the gassample valve wasquickly switched, allowing carrier gas to flush the•air sample from thesample loopinto a charcoal trap cooled to dry ice-acetone temperatures. The trapping andall following steps are exactly thoseusedin water sampleanalysis. DISCUSSION Circulationin Walt)is Bay Area The W alvis Bay region has .been studied by a numberof workers [Stander, 1964; Visser, 1969 ß Calt)ert and Price, 1971' Paper number 7C0421. 4947