Data assimilation is a crucial aspect of modern oceanography. It allows the future forecasting and backward smoothing of ocean state from the noisy observations. Statistical methods are employed to perform these tasks which assumes that the locations associated with observations are known with certainty. This is reasonable for typical oceanographic measurement methods. Recently, however an alternative and abundant source of data comes from the deployment of ocean sensors on marine animals. This source of data has some attractive properties: unlike traditional oceanographic collection platforms, it is relatively cheap to collect, plentiful, has multiple scientific uses and users, and samples areas of the ocean that are often difficult of costly to sample. However, inherent uncertainty in the location of the observations is a barrier to full utilisation of animal-borne sensor data in data-assimilation schemes.
I have worked with statisticians from CSIRO, Hobart, on a project where our goal was to adapt the statistical methods to incorporate the location uncertainty that is associated with the observations.
The research is published as:
Sengupta, A., Foster, S. D., Patterson, T. A., and Bravington, M. (2012)
Accounting for Location Error in Kalman Filters: Integrating Animal Borne Sensor Data into Assimilation Schemes. PLoS ONE 7(8): e42093. doi:10.1371/journal.pone.0042093
Interested audience can find the article
here
In this article we have examined this issue of erratic locations, and suggest a simple approximation to explicitly incorporate the location uncertainty. The approximation stems from a Taylor-series approximation to elements of the updating equation.
