Biogeochemical models are sensitive to the physical framework in which they are imbedded. To make quantitative model intercomparisons, then, different models have to be run under the same physical conditions. For this reason few intercomparison studies have so far been carried out. To promote such studies, the U.S. JGOFS SMP has formed a working group tasked with providing 1-D "test beds," which allow investigators to run a variety of biogeochemical models in a specific physical context. This website is an exploratory effort to develop such a testbed for the central Arabian Sea at 15.5^oN, 61.5^oE, the site of the WHOI mooring. Our approach is to use environmental fields obtained from a 3-D physical model to run biogeochemical models "offline" in 1-D.

In the following, we first describe the physical fields available at the website, providing both required and optional fields as downloadable files. These fields are obtained from the layer model described in McCreary et al. (2000). Then, we discuss two examples of 1-D biological models (both a layer and a level system), which illustrate how to use the physical fields. Finally, we provide available observational data for model validation. We hope to expand the website in the future, including environmental fields from other physical models, examples of other biological models, and additional data.

The physical fields are obtained from the 4½-layer physical model of McCreary et al. (2000), which extends throughout the entire Indian Ocean. For our 1-D purpose, we extract the needed fields from the grid points nearest the WHOI mooring site.

The model consists of 4 upper-ocean layers with velocities v_i =(u_i , v_i), layer thicknesses h_i, temperatures T_i, and salinities S_i (i=1,2,3,4 is a layer index), overlying a deep inert ocean with temperature T_d and salinity S_d. The layers correspond to either distinct oceanic regions or water-mass types, namely the surface mixed layer (layer 1), the diurnal thermocline (layer 2), the seasonal thermocline (layer 3), and the main thermocline (layer 4). The diurnal-thermocline layer allows the system to "remember" physical and biological variables when the mixed layer thins during the day, and thereby helps to prevent spurious vertical mixing between the mixed layer and the deeper ocean. Fluid is allowed to transfer between the layers with velocities w₁, w₂, and w₃, and system is thermodynamically active in that T_i and S_i vary horizontally in response to surface heat and buoyancy fluxes, horizontal advection, entrainment and detrainment. Finally, h₁, h₂, and h₃ are not allowed to become thinner than minimum values h_1min=10m, h_2min=1m, and h_3min=10m. These minima are necessary to keep the model numerically stable, solutions are not sensitive to their values provided they are sufficiently small.

After an initial spin-up period of several years, the model is forced by daily-mean data from the WHOI mooring for the period from October 15, 1994, through October, 1995. These fields include wind stress, scalar wind, air temperature, specific humidity, and solar radiation Q. Diurnal forcing is included by modifying Q to have a realistic diurnal cycle.

Variables

The physical fields that are necessary for running a 1-D version of the McCreary et al. (2000) biological layer model are the layer thickness fields h_i, the exchange velocities across the layer interfaces w_i, and the surface light intensity Q. Salinity S_i and temperature T_i fields are "optional," in that they are used in some, but not all, biological models.

The following table provides these 1-D environmental fields for a solution to the 3-D layer model with diurnal forcing. Data is stored every half hour for the time period from October, 1994, through October, 1995.

Interpolation

In order to be usable in a level model (the typical type of 1-d biological model), the environmental variables from the layer model must be interpolated onto a level grid. Our interpolation grid encompasses the depth range -400m < z < 0m with a vertical resolution of 1m. Details of the interpolation scheme are located in interpolation.pdf.

The fields that are needed for a biological level model are the mixed-layer thickness h₁, the vertical velocity field w(z), and Q.

Below are the program codes which can be used to obtain the physical field data on a level model interpolation grid.

Two examples of biological models are given below: a layer model and level model. Each section provides the necessary code and data to run each of the model types.

Layer model

The 1-D biological layer model is a straightforward adaption of the 3-D model described in McCreary et al. (1996). It determines nitrogen concentrations in four compartments: inorganic nitrogen N (that is, nitrate, nitrite and ammonia), phytoplankton P, zooplankton Z, and detritus D. In this model, hyperbolic saturation functions are used to describe how the autotrophic production rate is related to photosynthetically active radiation (PAR) and nutrient concentration, and how zooplankton growth rate is related to food supply. Vertical exchanges of nitrogen are specified via entrainment and detrainment from layers, background diffusion, and sinking. Details can be found in biolayermodel.pdf.

The following table provides the information necessary to run the model, and to reproduce the solution shown below. It includes the model code (row 1) and the input data (row 2), as well as initial conditions for biological variables q_i (row 3) and advective fields v·Ñq_i for the diurnal solution (row 4)from McCreary et al. (2000). The initial conditions and advective terms are optional, but when they are used the resulting solutions are identical to the response of the 3-D model at the WHOI grid point.

The level-model formulation provided below is an offline version of 6-compartment ecosystem model described in Hood et al. (2000). It is similar to the McCreary et al. (1996) model in that it is a nitrogen-based system, includes N, P, Z and D compartments, and uses hyperbolic saturation functions to describe autotrophic growth and zooplankton grazing. However, it differs in that it includes two additional state variables, one for dissolved organic nitrogen (DON) and the diazotrophic cyanobacterium, Trichodesmium. In addition, the zooplankton compartment in this model is replaced by a "heterotroph" compartment, which is intended to represent both bacteria and microzooplankton. Vertical exchanges are specified via advection, diffusion and sinking of detritus. For details see biolevelmodel.pdf.

The following table provides the information necessary to run the biological level model. It includes the model code (row 1), as well as program codes to interpolate the various fields. Note: These interpolation codes are identical to those given in the Interpolation section above. To interpolate the input data fields, download the file layercode_data.asc described in the Physical Fields-Variables section above and use it as input into the interpolation program l2l.f. Similarly, download the appropriate data files from the biological layer model section above and use the corresponding interpolation program to create data for the biological level model.

Mixed layer depth at the mooring site (A. Fischer / R. Weller):

mldepths.asc.Z

mldepths.zip

The upper panel of the figure shows the mixed-layer depth at the WHOI mooring, defined to be the depth at which the ocean temperature is 0.02oC less than SST. The lower panel shows the mixed-layer depth from the physical model (the h1 field). The two fields compare well in their overall structure. The largest differences occur during Nov-Dec 1994 and Aug 1995, times at which mesoscale eddies were known to pass through the mooring location.

Biological data

Productivity and chlorophyll data at the mooring site (J. Marra et al., 1998):	marra_testbed.asc

The upper panel of the figure shows the comparison of chlorophyll concentration between the solution from the biological layer model and data from J. Marra et al. (1998). The lower panel shows a similar comparison for primary production.

Marra, J., T. D. Dickey, C. Ho, C. S. Kinkade, D. E. Sigurdson, R. A. Weller, and R. T. Barber, 1998: Variability in primary production as observed from moored sensors in the central Arabian Sea in 1995. Deep-Sea Res., 45:2253-2267.

McCreary, J. P., P. K. Kundu, and R. L. Molinari, 1993: A numerical investigation of dynamics, thermodynamics and mixed-layer processes in the Indian Ocean. Prog. Oceanogr., 31: 181-244.

McCreary, J. P., K. E. Kohler, R. R. Hood, and D. B. Olson, 1996: A four-component ecosystem model of biological activity in the Arabian Sea. Prog. Oceanogr., 37: 193-240.

McCreary, J. P., K. E. Kohler, R. R. Hood, S. Smith, J. Kindle, A. Fischer, and R. A. Weller (2000) Influences of diurnal and interseasonal forcing on mixed-layer and biological variability in the central Arabian Sea. J. Geophys. Res.. Accepted, in revision.