Marine BGC DA Symposium

New Pathways to Understanding and Managing Marine Ecosystems: Quantifying Uncertainty and Risk Using Biophysical-Statistical Models of the Marine Environment


CSIRO Cutting Edge Science Symposia Series

Program Book and Abstracts available below

Symposium Aims and Objectives:

Marine biogeochemical models are used to support coastal management, especially with respect to management of environmental flows, and diffuse and point source loads, and more broadly to support prediction and management at ocean basin and global scales. These models are crucial for sustainable development of coastal and marine resources and industries because management scenarios often involve pushing systems outside the range of  loads or forcings for which we have observations, such as climate change impacts, so extrapolation is required. Also, managers are increasingly sophisticated and ask for indications of uncertainty around these predictions.


The aim of this symposium is bring together leaders in the marine modelling community with experts in the data assimilation (DA) and statistical science communities. These groups have complementary skills in that the modellers encapsulate their collective knowledge to form process models, where sources of error are often treated in an ad-hoc manner. The DA and statistics communities have robust methods to handle these sources of error. Through this symposium we seek to facilitate the sharing of ideas between the communities and progress the integration of models and observations. Furthermore, we hope to indentify new approaches that yield quantitative estimates of uncertainty that can be used in hind-casting, forecasting and long-term prediction in marine ecosystems.


Symposium Location: CSIRO Marine and Atmospheric Research, Hobart, Australia


Symposium Dates: Evening of the 27th of May – Afternoon of the 30th of May.

Attendance/Registration Cost: $0


CSIRO (OCE, WfO and CSS-TCP), has provided the seed funding for this event and therefore the venue, catering and conference dinner incur no cost, however, participants will need to cover their own travel expenses. We hope that this symposium can be seen as a “community” symposium that follows on closely from the themes discussed at the 2012 Ocean Sciences Special Session 041: Methods and Applications of Data Assimilation for Ocean Biogeochemistry. Our intention is to provide a relaxed atmosphere with plenty of time for informal discussions.

We have competitive funding available to partly fund the attendance of a small-number of early-career researchers (PhD or post-doc level).

We are hoping to keep this event relatively small and informal to facilitate plenty of discussion time. We are aiming to have between 40 – 50 participants, and at this stage we are seeking to gauge interest.



Broad Symposium Themes:


Incorporating Uncertainty in BGC Models:

The intention of this topic is to examine in some detail the trade-off between detailed process models that can be difficult to validate, versus simpler process models incorporating stochastic components to represent the fine detail.


Inference in Biophysical-Statistical Models:

This topic focuses on how we infer, quantify our knowledge of unknown parameters and make predictions of future behaviour using biophysical-statistical models. Of particular concern is how we combine process models with observations.


Computational Approaches:

This topic is concerned with the algorithms used in the inference procedure, including interpolation, filtering, variational and Monte Carlo approaches. Of interest are the strengths and limitations of these algorithms with regard to the treatment of uncertainty, their applicability to the various functional forms that a biophysical model may take, and their computational requirements.


For more information please contact Emlyn Jones.


Abstract Book

Day 1: Noel Cressie

Day 1: Yvette Spitz

Day 1: Katja Fennel

Day 1: Emlyn Jones


Day 2: Mike Dowd

Day 2: Yoichi Ishikawa

Day 2: Paul Mattern

Day 2: Brad Weir


Day 3: Arnaud Doucet

Day 3: Pierre Jacob


Endorsed by:





Legal Notice and Disclaimer | Copyright | Privacy