Abstract—The paper elaborates upon a hybrid approach
consisting of data mining and statistical methods, to modelling
seasonal climate effects, i.e., arising from year-to-year
variability in weather conditions, on grape crop of three
different varieties cultivated in northern New Zealand. Recent
research using an iterative χ2 method based approach to
modelling climate effects on “high” and “low” yearly yields (of
perennial crops) with data at the regional (macro) and grape
yield from different vineyards, with climate data at macro scale,
are briefly outlined. The grape varieties studied are
Chardonnay, Pinot Noir and Pinot Gris. The results show
interesting patterns in the nexuses between extreme daily
weather conditions and grape crop data in terms of daily
maximum, temperature observed for “low” and “high” yields,
and within the macro and meso scale data, covering a period of
less than ten years.
Index Terms—Year-to-year variability, seasonal patterns, extreme weather conditions.
Subana Shanmuganathan and Philip Sallis are with Geoinformatics Research Centre, Auckland University of Technology, Private Bag 92006, Auckland 1142, New Zealand (e-mail: email@example.com; firstname.lastname@example.org).
Ajit.Narayanan is with School Computing and Mathematical Sciences, Auckland University of Technology, Private Bag 92006, Auckland 1142, New Zealand (e-mail: email@example.com).
Cite:S. Shanmuganathan, P. Sallis, and A. Narayanan, "Data Mining and χ2 Test Based Hybrid Approach to Modelling Climate Effects on Grape Crop in Varieties of Kumeu, New Zealand," International Journal of Machine Learning and Computing vol.2, no. 4, pp. 521-525, 2012.