2015-03 Uncertainty Evaluation of Computer Simulation Models (Draft Report)

Deterministic agricultural and agro-ecosystem systems models have an important role worldwide. Although these models are mathematically deterministic, there are many possible sources of uncertainty that can propagate through the model. The development of tools to evaluate uncertainty in deterministic models is an active area of research. There are a number of options for model uncertainty evaluation that have been discussed in the literature. The objectives of this review paper are threefold: 1. To describe a formalised state-space framework within which to describe the types and sources of uncertainty that arise in computer simulation models. 2. To provide a framework to carry out a robust uncertainty evaluation of a computer simulation model. 3. To summarise a selection of relevant sampling and analysis techniques for the uncertainty evaluation of computer simulation models.

Esther Meenken, Chris Triggs et al.

Download

 

2015-02 New Zealand’s Climate Data in R – An Introduction to clifro

The National Climate Database contains data collected from around
6,500 weather stations throughout New Zealand, providing a necessary
repository for any study involving New Zealand’s weather or climate.
Although data from this database can be accessed via a web portal (CliFlo),
the vast amount of information is not readily consumable. The clifro
R package provides the necessary functions to easily integrate the
National Climate Database with R, and provides a variety of elegant plotting
methods for data exploration. Also, by supplying functions to access the
database via CliFlo programmatically, clifro allows for transparent and
reproducible research when using these data.

Blake M. Seers and Nick T. Shears

Download