Biogeochemistry

Nutrient and phytoplankton dynamics across the whole GBR region spans the range from highly eutrophic estuarine systems to extremely oligotrophic offshore reefs. These systems are not distributed evenly by area or latitude with estuaries and coastal waters (<30 m depth) accounting for 30%, mid- and outer shelf waters 60% of the area and offshore reefs contributing just 10% of the area (Furnas & Mitchell 1986). The shelf is widest (~120km) in the south which supports higher levels of productivity (up to 1g C m^-2) in inshore and offshore waters (Furnas et al., 2005).

Anthropogenic catchment loads enhance estuarine and inshore productivity particularly during the wet season and flood events. Nutrient rich plume waters support diatoms and larger phytoplankton growth modulated by high turbidity levels which can limit light available for photosynthesis. Coastal productivity is transported along shore (general northwards) and dispersed offshore by coastal currents and tidal mixing. This supports secondary productivity in the lagoon and inner reef including benthic deposit and filter feeders and reef communities. Inshore productivity is highest during the wet season (October – April) particularly in the central and northern GBR (Furnas et al., 2005).

On the outer reefs productivity is limited by very low nutrient supply. Typically the seasonal thermocline and associated nutricline intersects the shelf slope at ~100m which is below the depth of the outer shelf. New nutrients from offshore only arrive at the reef during episodic intrusions or upwelling events that are energetic enough to raise deep nutrient rich water onto the shelf. Episodic events have been observed throughout the year but are strongest in October – May (Furnas & Mitchell 1986; Andrews & Furnas 1986). At other times of the year primary production in the lagoon and outer shelf is maintained by efficient recycling of nutrients through the microbial loop. This phytoplankton community, consists of tiny pico- and nano-plankton including photosynthetic autotrophs, heterotrophs and microzooplankton, cycles nutrients with growth rates of 0-2 doublings per day (Furnas 1991).

In addition to land and oceanic nutrient loads, the atmospheric contribution of nutrients to the GBR, in rain water and dust deposition, is comparable to oceanic fluxes (see Table 1 & Furnas & Mitchell 1996). Nitrogen fixation by reef organisms and blue-green microalgae Trichodesmium also occurs throughout the region although the patchy distribution and cryptic life history of N fixing organisms makes this flux difficult to constrain (Furnas & Mitchell 1996).

The biogeochemical model has captured the foundational knowledge of the Queensland research community and synthesised it into a numerical hypothesis of biogeochemical inputs, transformations and losses on the shelf. The model is based on the CSIRO Environmental Modelling Suite (EMS) augmented with key tropical marine processes modules including coral and seagrass growth, carbon chemistry, Trichodesmiun nitrogen fixation, phytoplankton pigment synthesis and a spectral optical model for accurate representation of the in-water light field. Process enhancements undertaken in this study included the implementation of Trichodesmium, addition of carbon chemistry including reef photosynthesis & calcification, inclusion of in-water spectral irradiance, variable carbon to chlorophyll and diurnal autotrophic growth, atmospheric deposition of nutrients in rain, improved process models for seagrass, macroalgae and coral and parameterization of sub-grid scale reef substrate type and function.

The BGC model is organised as three ‘zones’: pelagic, epibenthic and sediment. The epibenthic zone overlaps with the lowest pelagic layer and shares the same dissolved and suspended particulate material fields. The sediment is modelled in multiple layers with a thin layer of easily resuspendable material overlying thicker layers of more consolidated sediment. Pelagic processes include phytoplankton and zooplankton growth and mortality, detritus remineralisation and fluxes of dissolved oxygen, nitrogen and phosphorus. Macroalgae and seagrass growth and mortality are included in the epibenthic zone whilst further phytoplankton mortality, microphytobenthos (benthic diatom) growth, detrital remineralisation and fluxes of dissolved substances are included in the sediment layer. Also included are the augmentation of nitrogen fixation, reef metabolism, filter feeders, radiative transfer, carbon chemistry including reef calcification and alkalinity models, air-sea exchange including atmospheric deposition, rain nutrients and dust, benthic biogeochemistry including links with substrate boulders reef etc., bioturbation and burial, seagrass (Figure 1).

Figure 1: Schematic diagram of the ecological model compartments, links and vertical layers.

The model is integrated over a three-dimensional hydrodynamic grid using the offline transport model that is computationally efficient and conserves mass. Initial biogeochemical model conditions and parameters were derived from climatologies, historical observations and literature values, with ocean boundary conditions scaled against density profiles for synchronisation of mesoscale forcing. Coastal nutrient and sediment loads were derived from catchment model loads for 22 major river systems along the Queensland coast and atmospheric nutrients were deposited into the upper ocean with rain. The transport model is forced with surface hourly or 24 hour mean short wave radiation from GBR4, which varies with latitude and cloud cover. Ambient photosynthetically active radiation (PAR) throughout the water column is calculated from incident surface radiation, attenuated by optically active substances in the water column.

Results from a 4 year hind-cast were compared with observations from a range of model platforms, sensors and analytical techniques to assess model skill. Observation sources include Reef Rescue monitoring data, data from Australian Institute of Marine Science and CSIRO cruises, data from the Integrated Marine Observing System, glider data and satellite observations.The model reproduced the observed nutrient climatology, the spatial gradients in ocean colour chlorophyll and in situ sensor and water quality observations with sufficient skill for the purpose of water quality (phytoplankton, nutrients, turbidity, oxygen) hind-cast, near real time and scenario simulation in the GBR World Heritage Area.