DocServer

Modelling and simulation of seasonal rainfall using checkerboard copulas of maximum entropy

Borwein, Jonathan M. and Howlett, Phil and Piantadosi, Julia (2015) Modelling and simulation of seasonal rainfall using checkerboard copulas of maximum entropy.

[img] PDF
Download (1010Kb)

    Abstract

    We propose and justify a model for seasonal rainfall at a single site using a checkerboard copula of maximum entropy for the joint probability distri- bution and a set of two-parameter gamma distributions for the marginal monthly distributions. The model allows correlation between individual months and thereby enables an improved model for seasonal variation. A central theme is the principle of maximum entropy which we use—subject to clearly stated assumptions—to find the most parsimonious representation for the underlying distributions required to model the relevant statistical char- acteristics. We used observations over 109 years from 1905 to 2013 to con- struct joint distributions for monthly rainfall during the wet season January- February-March-April at three locations—Cairns, Brisbane and Sydney—on the east coast of Australia. Empirical evidence from extensive simulations is used to validate each application. Simulated seasonal rainfall over 109 years exhibits a high degree of statistical variation at each site. We conclude that caution should be exercised when using observed records to study rainfall trends as records are typically too short-lived to capture long-term trends.

    Item Type: Article
    Subjects: UNSPECIFIED
    Faculty: UNSPECIFIED
    Depositing User: Mrs Naghmana Tehseen
    Date Deposited: 28 Mar 2016 13:10
    Last Modified: 28 Mar 2016 13:10
    URI: https://docserver.carma.newcastle.edu.au/id/eprint/1694

    Actions (login required)

    View Item