Category: About the project

Why there are more and bigger bushfires?: Smoke Taint in Berries and Wine, What can we do about it?

Non-invasive smoke-taint detection in berries from grapevine (Vitis vinifera L.) using near infrared spectroscopy and machine learning models

I have been spending some whole nighters working on this topic due to recent news about extensive bush fires in Chile and in New South Wales, in January and February 2017 respectively. Both of these events have been named by the media as the biggest bush fire events in their respective histories. In Chile, the majority of the bush fires were in the central part of Chile, coincidentally to the majority of grapevine plantations. We do know that smoke taint has the biggest effect on berry contamination after veraison (7 days after on-set), which was about the timing of bushfires for a few cultivars for both Chile and NSW.



A recent report (February 2017) from the Victorian Government – Australia, has concluded that bush fires will increase severity and the window of opportunity due to climate change, specifically due to increases in temperature, increased frequency and intensity of heat waves and drought events.


So, what can we do about it?

As said before, I started to dig up some data from an ARC – Linkage project in which I worked as part of a team from The University of Adelaide. In that project we smoked artificially a number of cultivars to see the physiological effects of smoke contamination. From the canopy perspective and stomatal conductance, to be more specific, this effect can be explained through the chemical reaction between the main compounds found in smoke: carbon dioxide (CO2) and carbon monoxide (CO) and water. When getting in contact with stomata, smoke gases they can pass to the sub-stomata cavity, which is at 100% relative humidity. These compounds are then mixed with water forming carbonic acid (H2CO3) which reduced pH (acidic) hence close stomata.

This effect was reported in a poster in which I did a model to detect this effect on stomata conductance to discriminate canopies that have been contaminated or not with smoke. See posting by clicking here. The models worked really well for all the cultivars studied but Sauvignon blanc. I did attributed this effect to the morphology of leaves for this cultivar, which have high pubescence in the abaxial side. My hypothesis was that this offers a barrier to smoke, which can explain the inefficiency of the model based on the lack of stomatal conductance reduction.

I am currently working in the development of these models considering the top of canopies to apply them using infrared thermal imagery from Unmanned Aerial Vehicles (UAV), which can map a whole vineyard in the days after a bush fire event.

After the bush fires in Chile, I have revisited the Near Infrared (NIR) data from berries to see whether I was able to generate machine learning models to detect smoke taint in berries triggering the instrument around the skin and then measuring also in halved berries. The instrument that we used was a ASD FieldSpec® 3, Analytical Spectral Devices, Boulder, Colorado, USA. Which measures in the range of 350 – 1880 nm.


I have decided to concentrate the models in the700 – 1100 nm range using the second derivative of data, since it is the range of inexpensive NIR instrumentation. Also, after analysis of the rest of the wavelength range, the improvements of models obtained did not justified the jump in price of instrumentation from around $2,000 – $3000 to $38,000 – $65,000.

I did obtain three interesting models, the first was to detect whether the berries measured in a bunch have been contaminated or not using Artificial Neural Networks. I tried with data from whole and half berries and surprisingly I got better results with full berries. This is important since it renders the methodology as non-invasive. And makes sense when reading reports which found that the majority of glycocongugates are found in the skin of berries, which are higher than the pulp and higher that those found in seeds.


Then I tried models using machine learning fitting algorithms to see whether I could predict the levels of glycocongugates in the berries (whole):


And finally, whether I was able to predict smoke taint compounds in the wine made with contaminated and non contaminated berries, such as guaiacol, Syringol and cresols:



These are very exciting preliminary results which I am in the process of writing up for a peer review publication. The best thing is that if measurements made in the field are associated to GPS tagging, this could produce a contamination map using simple kriging interpolation techniques. This tool can support the decision making process towards differential harvest to avoid contaminating the whole production and salvaging fruit that has not been contaminated to the levels of spoilage. It offers also alternatives for winemaking to avoid excessive crushing and fermentation with skins that could contribute to the increase of smoke taint compounds in the final wine. Decisions can also be made by assessing the timing of ageing in barrels and whether it is required at all. And obviously, making non-contaminated wine using non-contaminated fruit. As can be seen in my preliminary results, the model for glycocongugates rendered a 10% error, which by quantifying the levels of smoke taint compounds does not make much difference in the final wine and it may contribute even to increased organoleptic characteristics with tones of leather, wood, bacon, etc

Something to look forward after all these tragedies.

Dr Sigfredo Fuentes

The University of Melbourne

New Interview: The Vineyard of The Future. Italian Blog


How much Australian wine growers spend fornew technologies like UAV’s, sensors, robots?

There is very scattered information in this aspect and there are no formal studies about formal spending of consumers, specifically from winegrowers in Australia. Maybe Wine Australia will have some more information in this aspect, specially for funding money spent on research and applications for the wine industry in Australia. The latter, since Australian Grape growers pay a levy every year, which Wine Australia administer for research that benefits the industry. These funds are given in a competitive basis to research institutions. In 2015 there was a specific call called Digital Viticulture, which addressed all these topics. This in a sense can serve as a scope that we are in early stages in the formal application of this technology for commercial purposes. This beside what you can find in advertisements from private companies offering services using drones and robots for management either in the irrigation, fertilization, pest and disease control, etc.

Link to the full interview: CLICK HERE

New Interview: Digital Vineyards: building technology so farms can think for themselves


Dr Sigfredo Fuentes is a plant physiologist and engineer at the University of Melbourne whose end game is the smart farm.

In his mind, it won’t be restricted to a single crop – maybe it’ll grow peaches and nectarines and vines as well – all of which are mapped out digitally.

And at the heart of the operation will be the drone, autonomous and equipped to collect all the data it needs to monitor the entire farm down to the last plant or tree.

And it’s fast – it can cover 2500 hectares in a day, which is about 2000 football fields.

Link to full article: CLICK HERE

New Article:Development of a robotic and computer vision method to assess foam quality in sparkling wines.

Quality assessment of food products and beverages might be performed by the human senses of smell, taste, sound and touch. Likewise, sparkling wines and carbonated beverages are fundamentally assessed by sensory evaluation. Computer vision is an emerging technique that has been applied in the food industry to objectively assist quality and process control. However, publications describing the application of this novel technology to carbonated beverages are scarce, as the methodology requires tailored techniques to address the presence of carbonation and foamability. Here we present a robotic pourer (FIZZeyeRobot), which normalizes the variability of foam and bubble development during pouring into a vessel. It is coupled with video capture to assess several parameters of foam quality, including foamability (the ability of the foam to form) drainability (the ability of the foam to resist drainage) and bubble count and allometry. The foam parameters investigated were analyzed in combination to the wines scores, chemical parameters obtained from laboratory analysis and manual measurements for validation purposes. Results showed that higher quality scores from trained panelists were positively correlated with foam stability and negatively correlated with the velocity of foam dissipation and the height of the collar. Significant correlations were observed between the wine quality measurements of total protein, titratable acidity, pH and foam expansion. The percentage of the wine in the foam was found to promote the formation of smaller bubbles and to reduce foamability, while drainability was negatively correlated to foam stability and positively correlated with the duration of the collar. Finally, wines were grouped according to their foam and bubble characteristics, quality scores and chemical parameters. The technique developed in this study objectively assessed foam characteristics of sparkling wines using image analysis whilst maintaining a cost-effective, fast, repeatable and reliable robotic method. Relationships between wine composition, bubble and foam parameters obtained automatically, might assist in unraveling factors contributing to wine quality and directions for further research.

New Article:Assessment of an automated digital method to estimate leaf area index (LAI) in cherry trees

A study was carried out during two growing seasons to evaluate the performance of a digital photography method to estimate the LAI (LAID). The trial consisted in 10 ‘Bing’ and 10 ‘Sweetheart’ trees where actual LAI (LAIA) was obtained by allometric relations. Estimations of LAID were obtained by the batch processing of images of the canopy of the same trees which were obtained by a conventional digital RGB camera. Comparisons of averages LAIA and LAID resulted in a good level of agreement for ‘Sweetheart’ for the two growing seasons (mean absolute percent error: MAE% = 10.4 %). For ‘Bing’, LAID was accurate in the first growing season (MAE% = 17.7 %) but underestimated by 44% (MAE%) in the second growing season, presumably due to differences observed in the clumping index and the light extinction coefficient. Results evidenced the robustness of this simple method for determining the LAI of cherry trees.