Remote sensing of primary production in the ocean: promise and fulfilment

Abstract

Remote sensing of ocean colour affords us our only window into the synoptic state of the pelagic ecosystem, and is likely to remain the only such option into the foreseeable future. Estimation of primary production from remotely sensed data on ocean colour is a research problem in two parts: (i) the construction of a local algorithm; and (ii) the development of a protocol for extrapolation. Good local algorithms exist but their proper implementation requires that certain parameters be specified. Protocols for extrapolation have to include procedures for the assignment of these parameters. One suitable approach is based on partition of the ocean into a suite of domains and provinces within which physical forcing, and the algal response to it, are distinct. This approach is still in its infancy, but is best developed for the North Atlantic. Using this method, and using the accumulated data from oceanographic expeditions, leads to an estimate for the annual primary production of the North Atlantic at the basin scale. Direct validation of the result is not possible in the absence of an independent calculation, but the potential errors involved may be assessed.