Link or sink: a modelling interpretation of the open Baltic biogeochemistry

Abstract

Abstract. A 1-D model system, consisting of the 1-D version of the Princeton Ocean Model (POM) coupled with the European Regional Seas Ecosystem Model (ERSEM) has been applied to a sub-basin of the Baltic Proper, the Bornholm basin. The model has been forced with 3h meteorological data for the period 1979-1990, producing a 12-year hindcast validated with datasets from the Baltic Environmental Database for the same period. The model results demonstrate the model to hindcast the time-evolution of the physical structure very well, confirming the view of the open Baltic water column as a three layer system of surface, intermediate and bottom waters. Comparative analyses of modelled hydrochemical components with respect to the independent data have shown that the long-term system behaviour of the model is within the observed ranges. Also primary production processes, deduced from oxygen (over)saturation are hindcast correctly over the entire period and the annual net primary production is within the observed range. The largest mismatch with observations is found in simulating the biogeochemistry of the Baltic intermediate waters. Modifications in the structure of the model (addition of fast-sinking detritus and polysaccharide dynamics) have shown that the nutrient dynamics is linked to the quality and dimensions of the organic matter produced in the euphotic zone, highlighting the importance of the residence time of the organic matter within the microbial foodweb in the intermediate waters. Experiments with different scenarios of riverine nutrient loads, assessed in the limits of a 1-D setup, have shown that the external input of organic matter makes the open Baltic model more heterotrophic. The characteristics of the inputs also drive the dynamics of nitrogen in the bottom layers leading either to nitrate accumulation (when the external sources are inorganic), or to coupled nitrification-denitrification (under strong organic inputs). The model indicates the permanent stratification to be the main feature of the system as regulator of carbon and nutrient budgets. The model predicts that most of the carbon produced in the euphotic zone is also consumed in the water column and this enhances the importance of heterotrophic benthic processes as final closures of carbon and nutrient cycles in the open Baltic.

A generalized model of pelagic biogeochemistry for the global ocean ecosystem. Part I: Theory

“…Reduction equivalents represent all the reduced ions produced under anaerobic conditions. This variable was originally used only in the benthic nutrient regeneration module of ERSEM (Ruardij and Van Raaphorst, 1995) but was extended to the water column in Vichi et al (2004).…”

“…The bacteria LFG comprises 3 CFFs for the C, N and P content, with 3 dynamical equations that have been recently formulated by (Vichi et al, 2003b) and extended to include denitrification and anaerobic processes in Vichi et al (2004). Bacteria are heterotrophs and their production rates depend directly from the availability of organic substrate.…”

“…Nevertheless, these processes are important for the sulfur cycle and for the fate of exported carbon in the meso-and bathypelagic layers of the ocean where bacteria are the major drivers of these processes. To account for hypoxic and anoxic remineralization in the water, the original ERSEM parameterization of anaerobic processes in the sediments proposed by Ruardij and Van Raaphorst (1995) was extended to the pelagic system by Vichi et al (2004). The state variable "reduction equivalents" N (6) ( Table 1 and Fig.…”

“…The basic constituent is indicated with the letter R because this variable account for all the reduced biochemical products, although it should be mostly regarded as sulphur S. Reduction equivalents are produced as a result of bacterial anoxic respiration and are partly used for the parameterization of denitrification processes and partly for direct sulphide production. We refer to Vichi et al (2004) for major details on the parameterization and for an application to the Baltic Sea.…”

“…On the other hand several implementations of this model have shown the skill of this approach, both in coastal areas with large land-derived inputs but also in the oligotrophic Mediterranean regions and in the subtropical Atlantic Ocean (Ruardij et al, 1997;Allen et al, 1998;Vichi et al, 1998;Zavatarelli et al, 2000;Obernosterer et al, 2001;Allen et al, 2001;Petihakis et al, 2002;Vichi et al, 2004;Raick et al, 2005). The same approach has also been used in the context of climate studies, particularly to capture and analyze climate variability in the North Sea and in the Adriatic Sea (Taylor et al, 2002;Vichi et al, 2003a).…”