This session is for all papers that lay outside of the scope of the proposed sessions. Papers submitted here will be reviewed in the same way as papers submitted to regular sessions. Where appropriate the scientific program co-chairs will suggest movement of some of these papers into regular sessions.

Organisers: E Ebert, Bureau of Meteorology (E.Ebert@bom.gov.au)
Cedric Griffiths, CSIRO (cedric.griffiths@csiro.au)

Numerical forecasting or prediction is being attempted today in a wide variety of disciplines, including meteorology, geology, economics, biology, engineering, hydrology, and social sciences. Most attempts at numerical prediction are faced with the comparing the predictions with observations either as part of an inverse scheme, or to demonstrate veracity. This spatio-temporal verification problem demands new cross-disciplinary approaches. The session aims to bring together experiences and problems from a wide range of disciplines to share knowledge, data sets and solutions.

Organisers:Dr Samantha Low Choy (s.lowchoy@qut.edu.au)
Prof Kerrie Mengersen (k.mengersen@qut.edu.au)

Expert knowledge provides valuable input to modelling in a wide range of ways. Expert elicitation is a multidisciplinary field that addresses the rigorous and repeatable process of quantifying expert knowledge for input into models. In the field of Bayesian statistical modelling, elicitation formally quantifies expert knowledge into prior distributions. In risk assessments and reliability analysis, expert knowledge is quantified as parameters describing probabilities of events. At a more fundamental level, researchers in cognitive psychology have been investigating the way in which non-mathematicians conceptualize probabilities and probabilistic statements. This session aims to take advantage of MODSIM's ability to bring together modellers from a variety of disciplines to provide a multidisciplinary forum for discussing Expert Elicitation. To help promote multidisciplinary cross-communication, we encourage presenters to at least: briefly define the expertise sought and the experts consulted, outline the model, define the encoding method used to quantify expert knowledge as model parameters, refer to computational technique, in addition to the usual presentation of interesting results and conclusions.

Organisers: Carmel A. Pollino, ANU (carmel.pollino@anu.edu.au)
Colette Thomas, CSIRO (Colette.Thomas@csiro.au)

Recognition of the value of models in risk-based decision-making is steadily growing. Models assist in risk-based decision-making by facilitating a process for assessing complex problems in an integrated and systematic way, enabling decisions that are understood and acceptable to a broad range of stakeholder groups.

In modern risk assessment and management, models play a central role in:

• providing focus on critical parts of a problem,
• offering an evidence-based framework to assist in prioritising actions to ensure that high quality and practical decisions can be made;
• supporting how we conceptualize, evaluate and cope with uncertainty in the making decisions and setting priorities; and
• assessing the balance between risks and rewards, upon which strategies to minimise risk and maximise benefits can be formulated.

In this session, we welcome papers that discuss tools and their application in risk analysis; techniques for handling uncertainties in models; and models for risk management and policy. Papers from this session will be invited to contribute to a special issue of Human and Ecological Risk Assessment.

Organiser: Mick Hartcher, CSIRO, Land and Water (Mick.Hartcher@csiro.au)

As data capture technology improves, and the range of data types and volume of available data increases, the development of formalised data management systems within an organisation becomes increasingly important. In addition, modellers require data to be accessible, logically structured, secure, properly archived, legally obtained, and to have demonstrable integrity. This session invites papers describing aspects of data management systems including:

• the development of and adherence to a data management policy;
• the provision of appropriate data management tools such as a metadata catalogue, data storage and delivery methods;
• data publishing and data discovery;
• data custodianship;
• data licensing processes, including defining intellectual property rights and developing inter-agency data sharing agreements;
• the development of workflow processes for generating reports, creation of audit trails for checking reported results and for regenerating data if necessary; and
• the analysis of data uncertainty and quality control and assurance.

This session provides a forum for the exchange of ideas and methods for the provision of robust data management solutions, which are integral support platforms for science delivery and the long-term maintenance of information. Abstracts addressing these topics, or closely related issues, are welcome.