The goal of the approach is to provide natural resource managers with information to anticipate and respond to short and long-term climate conditions.

[3][4] This kind of analysis employs a Bayesian Network that provides a probabilistic graphical model of a set of parameters, and can accommodate unobserved variables.

A Bayesian structure is a probabilistic approach that is flexible for high-dimensional data, and allows ecologists to separate sources of uncertainty in their models.

[3][5] Forecasts can leverage Bayes' Theorem and be iteratively updated with new observations using a process called Data Assimilation.

"Forecasts of temperature, shown in the diagram at the right as colored dots, along the North Island of New Zealand in the austral summer of 2007.