Time series analysis of climatic variables
(J Boland and B Ridley with L McArthur, M Luther, P Lauret and B Webby)
The modelling of the climate variables is two-fold. There is the time series analysis of variables such as ambient temperature and solar irradiation for locations where there are extensive data available.
The object of this exercise is to determine the characteristics of the two components of these variables, the deterministic and the stochastic. Briefly, if the deterministic component were the only one of significance, the weather on any one day of the year would be the same no matter what year it was. The stochastic component gives the fluctuations about this deterministic component which give the day-to-day weather variation.
If the characteristics of these variables are known, then one can construct synthetic weather sequences for use in testing models of physical systems whose performance is affected by the weather. This activity is made more difficult for locations where only gross weather statistics such as monthly daily average solar irradiation are known.
Publications
Boland J. (2008) Time series and statistical modelling of solar radiation, Recent Advances in Solar Radiation Modelling, Viorel Badescu (Ed.), Springer-Verlag, pp. 283-312.
Boland J. and Ridley B. (2008) Modelling diffuse radiation, Recent Advances in Solar Radiation Modelling, Viorel Badescu (Ed.), Springer-Verlag, pp. 193-220.
Barbara Ridley, John Boland and Philippe Lauret, Modelling of diffuse solar fraction with multiple predictors, Renewable Energy, (in press), 2009.
Magnano L., Boland J. and Hyndman R. (2008) Generation of synthetic sequences of half-hourly temperature, Environmetrics, 19, pp. 818-835.
External Collaborators
L McArthur, M Luther (Deakin), Philippe Lauret (Universite de la Reunion) and Brian Webby (UniSA Whyalla).