The previously moribund field of DOC analysis was stimulated by
recognition that the high surface concentrations of dissolved organic
carbon (DOC) reported by Suzuki could form an important component of the
export of organic matter from surface waters. Biogeochemical models
incorporating DOC as an export term provided improved simulations of
nutrient distributions, for example by eliminating the trapping of high
phosphate concentrations under the equatorial upwelling [ Bacastow and
Meier-Reimer, 1991; Najjar et al., 1992]. These models gave revised
global new production estimates of 2.0--3.6 Mol C m
y
(8-15 GtC
y
) and lent strong support to the observations of a high level of
enhanced or `new' DOC in the surface ocean. In these models about
70-80% of the total export is assigned to the dissolved phase, but
it should be noted that this partitioning is defined in the model
code rather than being generated by the biogeochemical dynamics of
the model itself.
The new interest in DOC by biogeochemists and modelers stimulated a large amount of research directed toward refining and calibrating the high temperature catalytic oxidation (HTCO) technique for analyzing DOC in seawater [ Hedges and Farrington, 1993]. Benner and Strom [1993] showed that assessment of potentially large machine blanks associated with HTCO instruments was necessary for accurate interpretation of analytical data. The original observations of high DOC were withdrawn following interpretation of the blank problem [ Suzuki 1993; Hedges et al. 1993]. Comparative [ Hedges et al., 1993; Sharp et al., 1993] have now demonstrated that the high values were in error and that mean deepwater levels of DOC measured by HTCO techniques are consistent with earlier persulphate and other measurements. However the new techniques have resulted in much improved precision which enables resolution of vertical gradients in the upper ocean [ Sharp et al., 1993] as well as of temporal and spatial variability formerly interpretable only as analytical noise.
New analyses of the composition of oceanic dissolved organic matter (DOM) indicate that bulk C:N ratios range from 16-38 [ Hansell et al., 1993; Karl et al., 1993]. 22-33% of the total DOC is > 1000 molecular weight and composed largely of carbohydrate material, with C:N ratios of 15-22 [ Benner et al., 1992]. The synthesis of high C:N DOM might be one explanation for observations of non-Redfield utilization of inorganic carbon and nitrogen in both coastal and oceanic regions [ Karl et al., 1991; Sambrotto et al., 1993b]. Large-celled diatoms continue to fix carbon, in both particulate and dissolved forms following nitrogen depletion [ Goldman et al., 1992]. The reactive nature of this material suggests that it must play a role in supporting heterotrophic activity, especially bacterial respiration, but utilization is likely to be nitrogen- or phosphorus-limited [ Amon and Benner, 1994]. 20% of the DOC in the surface waters of the north Atlantic during the spring, 1989 phytoplankton bloom was removed by bacteria within 2-3 days [ Kirchman et al., 1991]. The utilization of the DOC was supported by inorganic nitrogen, suggesting that the biochemically labile fraction was deficient in nitrogen. The processes by which DOC is produced and utilized, and especially those which regulate its escape from complete utilization and subsequent export into the large oceanic DOC reservoir, are just beginning to be understood. Carlson et al. [1994] have showed that DOC export is potentially an important component of the carbon budget at Bermuda. At this point, the relative contributions of DOC and POC to the export of new production are not known within a factor of more than two.