Abstract
Current methods for representing uncertainty in hydrographic products typically describe the type or completeness of data collected, but not the risk of transit or certainty about the data presented. This paper proposes a model for risk based on the background information, own-ship information, and how much is unknown about the area. Based on a model of underkeel clearance, dangers to navigation and other factors, the method predicts a scalar risk for an area, projected track or total voyage. The paper describes the risk assessment methodology, and illustrates it in track planning, area-based risk assessment and real-time decision support. These use cases highlight the benefits and flexibility of the model, and the concomitant requirement for calibration of input data.
Acknowledgements
I would like to thank W. H. F. Smith for the suggestion that eventually became the MSPP description of unobserved objects, and the anonymous reviewers for their insight.
Funding
This work was supported by NOAA grants NA05NOS4001153 and NA10NOS4000073, which are gratefully acknowledged.