The two countries are in the same hemisphere, and temperature latitudes but situated on either side of the prime meridian. In this context, the paper looks at some of the specific viticulture related aspects in different modalities, such as vector (point, contour), raster and text formats and then investigates into analysing the multimodal data collectively at a regional scale which is considered as appropriate for such a comparative study in this specific domain.
The commonly used major themes for modelling viticulture and wine production until to date have been; growing degree days (GDD), minimum/ maximum temperatures during berry ripening, frost days at budburst for the wine regions within a country or in the world, and are briefly outlined.
Meanwhile, at a relatively recent meso / micro scale (precision viticulture) modelling using grapevine vegetative growth and grape yield requires expensive equipment for multispectral satellite/ aerial borne imagery and yield data acquisition.
Following a brief outline on the use of contemporary technologies, such as GPS, and methodologies to analyse information integrated into GIS, the paper then elaborates on the results of a comparative study conducted on seven major wine regions of New Zealand and Chile using GIS based thematic mappings of terrain, topography, climatic conditions, grapevine varieties as well as wine quality, the latter represented by regional vintage ratings, sommelier comments and wine label ratings.
The results of one-way ANOVA tests show the difference across viticulture climate regimes of the seven regions as significant (95% confident).
However, between countries the difference is significant only for dew point in November and December, sea level pleasure in December, and total precipitation in December.
Keywords: viticulture, wine quality Legend Major lines of Latitude Tropic of Capricorn Tropic of Cancer Equator Prime Meridian Chile Colchagua Maipo Santiago New Zealand Hawke’s Bay Martinborough Martinborough Central Otago Figure 1. Chilean and NZ wine regions and viticulture climate regimes (base climate) source: http://www. weatherbase. com (T=Temperature) 1146 Shanmuganathan et al. , Analysing the Wine Regions of New Zealand and Chile: a GIS perspective . INTRODUCTION New Zealand and Chile are among the well-known “New World” wine producing countries. Incidentally, the major wine regions of the two nations are also in the same hemisphere, and temperature latitudes but on the opposite sides of the prime meridian. The rapid progress achieved in producing premium wine from these regions over the last few years have been described as remarkable hence a comparative study of this nature especially, on grapevine varieties cultivated and wine styles produced from the regions of the two countries would be appropriate.
It is also considered as a timely one because the wine regions of both old and new world countries are seen as highly vulnerable to the predicted global climate change in the near future (Jones, 2007, Web, 2006). In this context, the paper looks at some of the specific viticulture related aspects in different modalities, such as vector (point, contour), raster and text formats and then investigates into analysing the multimodal data collectively at a regional scale which is considered as appropriate for uch a comparative study and for analysing the effects of climate on viticulture and wine production. The paper presents an overview of 1) the base climate and the 2) a detailed description of modelling the seasonal variability in climate and on grapevine phenology as well as wine quality. Interestingly, until to date, the commonly used major themes for modelling viticulture and wine production have been; • • • growing degree days (GDD), minimum/ maximum temperatures during berry ripening, frost days for the wine regions in relation to varietal tolerance within a country/ in the world.
Meanwhile, modelling at micro and meso scales using modern precision viticulture (or PV) using grapevine vegetative growth and grape yield with finer details integrated into a GIS, requires expensive GPS fitted equipment for multispectral satellite/ aerial borne imagery and yield data acquisition. PV is about the use of GPS and GIS technologies (at the micro scale) to understand the impact of plant-soilwater dynamics at different phenological growth stages on vine physiology in order to achieve improved yield by introducing subtle changes to vineyard management practices.
Such a multipurpose integrated approach to mapping soil spatial properties illustrated in (Buss, et al. ,2005) as well described the success of the approch in generating irrigation management zones, evaluating the performance of the new irrigation schedules and the use of near continuous soil-water profile dynamics in irrigation scheduling, implementation and management strategies. The irrigation management strategies included were; regulated deficit irrigation (RDI) and partial root zone drying (PRD).
Similarly, many more studies have reported on the complex and challenging issues relating to managing the dynamic relationship between site, soil, water and phenological stage, vine and wine quality within and among vineyards using a GIS and integrated data captured using wireless sensors (Fuentes, 2005). There is also research reported outside of Australia into climate and environmental factors integrated with the irrigation management in grapevines which are a traditionally nonirrigated crop (Cifre, et al. , 2005: Patakas, et al. , 2005: Ben-Asher, et al. 2006 : Guix-Hebrard, et al. , 2007). Recently remote sensing has led to the use of airborne multispectral and hyperspectral imagery incorporated into GIS for yield mapping integrated with soil or other properties such as soil spatial variability, vegetative growth, vulnerability to diseases (Ferreiro-Arm? an, et al. , 2006). Since the late 1990s, there has been significant progress in the use of PV with advanced GIS functions for monitoring yield and soil-water-plant dynamics with commercially available devices and technologies (Bramley, 2001).
However, yield mapping against vigour in vegetation over vintages is a very recent method, as far as Australian viticulture is concerned only three years old. Despite this recent introduction, it has been shown that a number of Australian wine grape growing areas could have grape yields in single management unites varying as much as 8 to 10 fold. The surveys also emphasised the need for more data within individual blocks on yield, fruitvine indices and soil properties to optimise yield, and to find the blocks that produce high yield, by overlaying the data on different thematic mappings in a GIS.
With that introduction to the use of contemporary technologies and methodologies for analysing information integrated into a GIS at the micro scale, the next section elaborates on the methods utilised for comparing and contrasting the wine regions of New Zealand and Chile using GIS climatic conditions, grapevine varieties as well as wine quality based on regional vintage ratings at the regional scale, and then sommelier comments and wine ratings at the vineyard level. 1147 Shanmuganathan et al. Analysing the Wine Regions of New Zealand and Chile: a GIS perspective 2. VITICULTURE AND THE CLIMATE Grapevine is among the most expensive cultivated and sensitive crops (Jones, 2007). Different grapevine varieties thrive under specific ideal climate conditions and niche local environmental settings, such as well drained soil, site aspect (compass direction of the slope). Due to this reason viticulturists undertake extensive investigations when selecting a site and the grape variety for a vineyard.
Of the many factors, base climate1 is the main factor used to determine a variety for the site. Once several blocks of vineyards have been established in a broader area, its climate patterns (or macro climate) becomes identified with certain grapevine clones of Vitis vinifer (or wine styles) for that area covering about 100 to 1000 miles, also referred to as the regional scale (Jones, et al. , 2003). Such base climate regimes and their varieties of seven wine regions (four from New Zealand and three from Chile) are shown in figures 1 and 2, and Table 1 respectively.
On the other hand the quality of vintage wines is determined by the seasonal weather patterns that ripened the grapes. In view of the above factors the analysis is conducted in three parts; 1. 2. Initially, the base climate and wine styles famously linked with the major wine region of Chile and New Zealand are outlined. Secondly, the regional ratings and monthly climate data from (Wine Enthusiast wine vintage chart, February 2011 p56-57) (http://www7. ncdc. noaa. gov) respectively are analysed to establish the correlations between the regional wine quality ratings and the climate data.
In the third part, quality of vintage wines are analysed to establish the correlations between wine descriptors extracted from sommelier comments (text) and their corresponding ratings (numeric) also provided by sommeliers. Chardonnay style Vintage wines from all seven regions are analysed individually and altogether to find the descriptors used for high and low rated vintages of this style. 3. 3. MODELLING THE CLIMATE CHANGE EFFECTS Observations and results of this three part analysis are discussed in this section. 3. 1.
Viticulture climate regimes of New Zealand and Chile The major wine regions of New Zealand and Chile (the viticulture climate regimes) as well as wine styles produced from the regions are listed in Figure 1 and Table 1. The viticulture climate regions of Chile seem to exhibit the extremes at both high and low temperatures (see Figure 1 graph). Furthermore, Pinot Grigio/ Gris are not grown in any of the Chilean regions. Of the all seven regions Colcahgua exhibits the lowest annual recorded temperature (-10oC) and second highest recorded annual high.
Petite Sirah style vine is produced only from Maipo Valley region. Casablanca and Hawkes’s Bay have the mildest conditions in the Chilean and New Zealand regions respectively. Wine styles Chardonnay, Gewurztraminer and Pinot Noir are produced from all seven regions hence could be described as more tolerant varieties. 1 Figure 2. Major wine regions of Chile (left) and New Zealand (right) with the terrain elevation of the regions. Base climate reflects the weather conditions experienced over a longer time period i. e. , 3-5 decades. 148 Shanmuganathan et al. , Analysing the Wine Regions of New Zealand and Chile: a GIS perspective Table 1. Major wine regions of Chile and New Zealand along with the wine (styles) produced from the seven regions during 1990-2009. (Source: http://buyingguide. winemag. com) 3. 2. Seasonal climate change effects on the quality of Chilean and New Zealand wines Seasonal weather is the major influencing factor in determining the quality of vintage wines apart from those arising from grapevine varietal and winemaker experience/ talent.
The influence exerted by the seasonal weather as 50% and this can be seen in literature of traditional grapevine growing and winemaking as well as recent research findings (Shanmuganathan, et al. , 2010). In view of this, monthly average, maximum and minimum temperatures, dew point and total precipitation are analysed along with regional wine ratings (figure 3). The monthly averages were calculated from the daily weather data extracted from (www. ncdc. noaa. gov) for the closest stations for each of the seven regions for this work. Meanwhile the regional wine ratings for the seven regions were obtained from 2011 vintage chart (www. inemag. com). The graphs (figures 3 and 4) show the years of high and low vintage ratings at the regional scale and the average variability in weather conditions experienced in the regions between 1990-2009. ANOVA test results One-way ANOVA test results of weather data (monthly averages of daily minimum and maximum temperatures, sea level pressure (slp) and total precipitation over the growing season September-April (19902009, figure 4), confirm the difference in the seven regions as significant (at 95 % confidence) except for monthly average sea level pressure in February.
Meanwhile, One-way ANOVA test conducted to see the difference across countries showed monthly averages of dew point for November, December, average sea level pressure and total precipitation, both for the month of December significant (at 95% confident level). Figure 3. Graphs showing the regional wine ratings of the Chilean (left) and New Zealand (right) wine regions from 1990 to 2009 analysed to establish the correlations between macro climate and regional wine quality.
Data source: http://www. winemag. com/PDFs/Vintage_Chart_022011. pdf 1149 Shanmuganathan et al. , Analysing the Wine Regions of New Zealand and Chile: a GIS perspective Figure 4. Seasonal climate variability (averaged for 1990-2009) observed in Chilean and New Zealand regions. Central Otago monthly averages of temperature, dew point and total rainfall are the lowest even though Colchagua base climate graph (figure 1 bottom) exhibits the lowest of all regions.
Classification rules WEKA (J48) Rules created with monthly averages of dew point for November, December, average sea level pressure and total precipitation for December (significant by ANOVA test results) are listed in Table 2. 3. 3. Modelling the wine vintages and sommelier comments using ratings Wine styles Chardonnay, Gewurztraminer and Pinot Noir are produced from all seven regions (see table 1), hence Chardonnay wine comments from all seven regions were studied to investigate the correlations between wine descriptors and ratings specified by sommeliers.
The sommelier comments extracted from (http://buyingguide. winemag. com) for 556 wines, were programmatically converted into a matrix of 190 wine descriptors and 556 weights (of 350 Chilean and 206 NZ vintages) for this vector space model text mining approach as applied in (Shanmuganathan, et al. , 2009). The vintages used in the sample were from 1996-2010. Of these 190 wine descriptors only 55 were found to be significant by a one-way ANOVA test ran for both Chilean and NZ together and separately.
Using this 556 x 55 wine descriptor weight table, rules were generates with C5, CRT (Clementine), JRip and J48 (the latter two WEKA algorithms). For this analysis wine ratings were converted into a binary rate with one (=87). The descriptors found to be correlated with these two ratings are presented in tables3. Furthermore, it could be noticed that a few descriptors used exclusively for describing wines of a particular region and either for one/two rating, for example blanc (for Sauvignon Blanc) is used for Marlborough region (table 4).
👋 Hi! I’m your smart assistant Amy!
Don’t know where to start? Type your requirements and I’ll connect you to an academic expert within 3 minutes.get help with your assignment