Is Total Primary Production in Shallow Coastal Marine Waters Stimulated by Nitrogen Loading?

Abstract

The global nitrogen cycle has been extensively modified by human activity to the extent that more N is fixed annually by human-driven than by natural processes (Vitousek 1994). This alteration influences production and species composition of terrestrial ecosystems (Tilman 1987) and contributes to acidification and forest dieback (Schulze 1989). The influence of nitrogen on eutrophication of coastal marine ecosystems is even stronger with profound effects on plant communities and food webs (Nixon et al. 1986). The total primary production of marine plants in coastal areas (as organic carbon or dry matter produced annually per unit of surface area within the ecosystem) is generally assumed to increase with increasing loading of nutrients from land (e.g. Boynton et al. 1982, Nixon et al. 1986, Paerl 1993). The effects of widespread eutrophication, such as oxygen deficiency and mass mortality of benthic invertebrates and fish, have alarmed the public and are considered to emerge from the enhanced oxygen consumption required to mineralize the increasing amounts of organic matter produced within the ecosystem. Much of our current analysis and understanding of energy flow and carbon and oxygen dynamics in the coastal marine environment presuppose that anthropogenic loading and total primary production are directly positively related. In the following we will argue, however, that the total primary production of shallow coastal areas, i.e. the combined production of microand macroscopic plants living in the water and at the bottom, does not change systematically by nutrient enrichment. Our analysis is based on a comparison of published rates of total primary production and nitrogen loading from land using the same approach as Boynton et al. (1982) and Nixon et al. (1986) to analyse the dependence of phytoplankton production on nitrogen loading. The data originate from different temperate coastal ecosystems with very different morphometry and hydraulics. The quality of the data (in terms of suitability of methods and number of measurements) also varies considerably among the systems but does not systematically influence the patterns reported. We have tried to ensure a fair representation of systems which are most likely to respond to enhanced nitrogen loading by including relatively deep (mean depth down to 40 m) and clearly phytoplankton dominated ecosystems and by omitting data we suspect overestimate benthic primary production. Thereby, we obtain a balanced analysis of our hypothesis.