WINETECH Technical Yearbook 2019

WINETECH TECHNICAL YEARBOOK 2019

CONTENTS

Detection and quantification of black foot disease and crown and root rot pathogens in grapevine nurseries.......................................35 Know your seed for cover crop purposes.....38 What was the impact of the 2017-18 drought? – a case study using FruitLook data. ..............................................40 Rugose wood diseases and associated viruses............................................................45 OENOLOGY Organic yeast nutrients and Chardonnay aroma........................................50 Benchmarking of commercial tannins and mannoproteins.......................................53 Big Data and YAN analysis – new developments in A South African context. ...55 Chitin, chitinase, chitosan…. .........................58 Amino acids in South African grape musts (Part 1): Composition and profiles................61

Amino acids in South African grape musts (Part 2): Cultivar classification.......................64 Accumulation of volatile phenol glycoconjugates in grapes after bushfires.....67 Rapid and cost-effective quantification of YAN concentration and composition – vintage and cultivar effect.............................69 Thiol-releasing and low VA-forming hybrid wine yeasts.........................................72 Pectinases improve consistency in winemaking...................................................75 Pectin-lyase – the key wine enzyme?. ..........77 Mixed culture fermentations and Shiraz flavour.................................................79 Nitrogen and sulphur foliar fertilisation (Part 1): Background and context.................82 Nitrogen and sulphur foliar fertilisation (Part 2)...........................................................86 Nitrogen and sulphur foliar fertilisation (Part 3)...........................................................90

VITICULTURE Lessons to be learned from Israel’s

desert agriculture............................................7 Water footprint of grapes and wine. ............10 Strategy for the control of mealybug in South African vineyards and wine.................12 Detection and classification of grapevine stem pitting-associated virus. .......................14 A new era for wine grape yield estimation...17 Management and re-use of treated winery wastewater (Part 1): New Zealand wineries...................................20 Management and re-use of treated winery wastewater (Part 2): Dairy and municipal wastewaters in New Zealand........24 Management and re-use of treated winery wastewater (Part 3): Adelaide, Australia.........................................28 TerraClim – online terrain and climate spatial decision support systems. .................31

IMAGES COPYRIGHT: INDIVIDUAL AUTHORS, FLICKR, PIXABAY, SHUTTERSTOCK, UNSPLASH OR WOSA LIBRARY.

CONTENTS

Melatonin in wine protects your heart.........94 Sequential inoculation of Starmerella bacillaris and Saccharomyces cerevisiae – main effects...............................97 Chenin blanc and Pinotage fermentations with South African Torulaspora delbrueckii yeast isolates............................100 Focus on bentonite......................................103 Alternative theory for multiple sclerosis.....106 Stability of young Chenin blanc and Sauvignon blanc wine during storage (Part 1): Chemical evaluation......................109 (Part 2): Sensory aspects.............................113 When microbes get it wrong.......................116 Random oxidation and the role of bottle dimensions........................................119 IN THE VINEYARD Gen-Z cover crop demo plots – results and findings from the 2018 season............122 Stability of young Chenin blanc and Sauvignon blanc wine during storage

The role of leaf and soil analyses in vineyard management (Part 1)...................125 The role of soil and leaf analyses in vineyard management (Part 2)...................129 The role of soil and leaf analyses in vineyard management (Part 3)...................136 Grapevine lifetime – protection of a valuable asset.............................................................139 Looking at favourable vineyard locations in a warmer climate...........................................141 IN THE CELLAR The advantages and disadvantages of oxygen prior to bottling...............................144 The importance of oxygen control during bottling and the measurement of dissolved oxygen in wine. .......................147 New wine packaging for a changing consumer profile.........................................149 Winemaking with stems..............................152 Post-harvest practices in wine cellars.........154

The use of whisky barrels for wine maturation .........................................156 The role of ellagitannins in the oxidative stability of wine ...........................158 The role of tartaric acid in winemaking......160 The influence of different gasses on flotation.......................................................162 Lessons from Provence for the making of rosé wines...............................................163 The environmental awareness of cellars......................................................165 The efficiency of filtration prior to bottling....................................................167 GENERAL The positive contribution to transfer of knowledge...............................................170 Professionalism beckons for the wine industry...............................................172 Vineyard worker skills survey......................174 Get up to date with learning and development...............................................177

ASTRID BUICA Department of Viticulture and Oenology, Stellenbosch University abuica@sun.ac.za BERNARD MOCKE Private consultant bmocke@gmail.com CAREN JARMAIN Independent researcher cjarmain@gmail.com CARIEN COETZEE Basic Wine, Stellenbosch carien@basicwine.co.za CAITLYN MCCARTNEY Vinlab, Stellenbosch micro@vinlab.com CAROLYN HOWELL ARC Infruitec-Nietvoorbij, Stellenbosch howellc@arc.agric.za CHARL THERON Private consultant vinofino@mweb.co.za DARIUSZ GOSZCZYNSKI ARC Plant Protection Research Institute, Pretoria GoszczynskiD@arc.agric.za

HEINRICH DU PLESSIS ARC Infruitec-Nietvoorbij, Stellenbosch dplessishe@arc.agric.za HEINRICH SCHLOMS Vinpro, Paarl heinrich@vinpro.co.za JOHAN BURGER Department of Genetics, Stellenbosch University jtb@sun.ac.za JOHAN DE JAGER Vinpro, Paarl dejagerj@vinpro.co.za

NEIL JOLLY ARC Infruitec-Nietvoorbij, Stellenbosch jollyn@arc.agric.za

PHILIP DU TOIT Inspection Services, Department of Agriculture, Forestry and Fisheries phillipdt@daff.gov.za

RUBEN GOUDRIAAN eLEAF, Fruitlook ruben.goudriaan@eleaf.com RODNEY HART ARC Infruitec-Nietvoorbij, Stellenbosch hartr@arc.agric.za SANDRINE LECOUR Department of Medicine, University of Cape Town sandrine.lecour@uct.ac.za

JOHN MOORE Institute for Wine Biotechnology Stellenbosch University moorejp@sun.ac.za

KACHNÉ ROSS Winetech, Paarl kross@winetech.co.za KARIEN O’KENNEDY Winetech, Paarl karien@winetech.co.za

SANTI BASSON Private consultant santib@mweb.co.za

SUSAN JANSE VAN RENSBURG Department of Health Science, Stellenbosch University sjvr@sun.ac.za TARA SOUTHEY Centre for Geographical Analysis, Department of Geography &

LIZEL MOSTERT Department of Plant Pathology, Stellenbosch University lmost@sun.ac.za

AUTHORS

FRANCOIS VILJOEN Vinpro, Paarl viljoenf@vinpro.co.za HANNO VAN SCHALKWYK Vinpro, Paarl hanno@vinpro.co.za

LUCINDA HEYNS Winetech, Paarl lucinda@winetech.co.za

Environmental Studies, Stellenbosch University tara@sun.ac.za

Winetech (Wine Industry Network of Expertise and Technology) has been the South African Wine industry’s research partner since 1999. Our main objective is to identify, prioritise, coordinate and fund relevant solution- driven research, development and innovation projects or initiatives that will directly contribute to strengthening the profitability and competitiveness of the industry. The past few years have seen several changes and new developments at Winetech that include increased focus on broader knowledge transfer engagements with the SA Wine Industry. We constantly strive to improve on existing knowledge transfer platforms, as well as to introduce exciting new initiatives. This Technical Yearbook combines all the technical articles from Winetech funded research, published in WineLand Magazine in 2019, in one electronic document for your convenience. It also showcases some of our ongoing learning and development initiatives for the industry. We will communicate on all the exciting changes that are taking place in our environment and how it will benefit the SA Wine Industry throughout 2020, so watch this space. Kind regards The Winetech Team

1 VITICULTURE

LESSONS TO BE LEARNED FROM ISRAEL’S DESERT AGRICULTURE

Netafim South Africa organised a technical tour to Israel for producers and industry role players. The purpose of the tour was to learn more about Israeli agriculture with regard to water management. BACKGROUND Geography and climate Israel is a Middle Eastern country bordered by the Mediterranean Sea between Egypt and Lebanon. Slightly larger than the Kruger National Park, the country spans 22 770 km 2 with a border length over 1 000 km. Israel has a Mediterranean climate in the north with average annual rainfall ranging from 500 to 700 mm per annum, whereas the south is arid with average annual rainfall ranging from 200 to 230 mm. Israel is home to over eight million people who consume nearly 1.91 billion m 3 of water each year. The only fresh water source is Lake Kinneret (the Sea of Galilee), which is fed by the Jordan River. Water status Mekorot, the national water company, is responsible for delivering drinking and i r r i ga t i on wa te r to t he who l e of I s rae l . Th i s i s ach i eved by more than 600 pumping stations and over 12 000 km of large diameter pipelines, known as the National Water Carrier. Lake Kinneret acts as the fresh water reservoir for the National Water Carrier and is supplemented by desalinated seawater

from the Mediterranean Sea. Desalinated water supplies two-thirds of Israel’s urban water use and is in some cases used for irrigation water. Treated municipal and industrial wastewater constitutes 32% of the irrigation water allocated to agriculture. The largest wastewater treatment facility, Shafdan in the Dan reg ion, recyc l es 380 000 m 3 of wastewater per day. It uses advanced mechanical-biological systems for purifying wastewater. The Negev receives 70% of its irrigation water from the Shafdan treatment plant. Treated wastewater is pumped into an infiltration basin as part of soil aquifer treatment. The aquifer is replenished with treated wastewater from the Shafdan. Water is then abstracted by means of boreholes and utilised in the Negev desert. The Shafdan plant is autonomous and powered by methane gas, generated by microbial reactors. The organic solid waste is dried and given cost-free to growers to use as organic fertiliser. ISRAELI AGRICULTURE Producing agricultural crops in Israel is quite expensive due to high labour, irrigation water and energy costs. Labour is often imported from foreign countries and minimal tillage is used to cut down on fuel expenditures. Although production costs run high in Israel, the growers are protected by extremely strict importation regulations, which ensures a higher degree of food security for the whole country.

GERT MALAN: ARC Infruitec-Nietvoorbij, Stellenbosch KEYWORDS: Desert agriculture, Israel, Netafim. JANUARY 2019

PHOTO 1. Ziv Charit, manager of Netafim’s Orchard and Vine Research Centre explaining wine grape production in Israel.

PHOTO 3. Almond trees grown hydroponically in lysimeters. Concentrations are in ppm.

keep the grapevine under moderate stress to reduce vegetative growth, which in turn lowers transpiration. It is also suggested that RDI enhances wine characteristics. Interestingly, they only control weeds on the grapevine row to reduce herbicide costs (Photo 2). The Agricultural Research Organisation (ARO) has six research institutes and four research stations spread throughout Israel. The Gilat Research Station is situated in Israel’s northern Negev at approximately 150 m above sea level. This research station comprises three main research units for Plant Sciences, Soil and Water, as well as Plant Protection, respectively. During the visit, the tour members had the opportunity

to learn more about current research projects and trials. One of these trials is an almond nutrition and physiology study. Almond trees are grown hydroponically, i.e. without soil by applying a nutrient solution over the course of the day. The aim of the study is to evaluate the effect of nitrogen, phosphorous and potassium levels on plant physiology. Although the trial is almost two years ongoing, the differences in growth habit are evident (Photo 3). The tour also included a visit to the Ministry of Agriculture’s research institute in the Jordan Valley. At this particular institute, the local research focuses on soil, irrigation and netting, as well as technology transfer. This region produces table grapes, vegetables

PHOTO 2. Wine grapes irrigated by subsurface drip from the centre of the work row. Note that herbicides are only applied on the vine row.

(Photo 1 and 2). The reason for this is to minimise damage to the pipes caused by implements and animals. It must be noted that subsurface drip is not considered as a means to reduce evaporation losses. The Israeli growers implement regulated deficit irrigation (RDI) by means of the pressure chamber. They apply irrigation when the mid-day stem water potential reaches -1.2 to -1.4 MPa. The rationale behind RDI is to

RESEARCH CENTRES VISITED A number of research stations were visited throughout Israel. The Netafim Orchard and Vineyard Research Centre focuses on delivering knowledge about drip irrigation technology. The centre also offers training on subsurface drip irrigation, maintenance, irrigation control and fertigation. Wine grapes are irrigated by means of subsurface drip, which is 30 cm below the soil surface

and even bananas. In spite of the extremely high salt content of the soil in the Jordan Valley, i.e. around 9 dS/m, table grape yields average 25 ton/ha. However, this is 50% of the production around Tel Aviv. Dan-ben-Hannah is a favourite and is commonly cultivated with drip irrigation under 50% shade net. The high soil salinity also requires a high leaching fraction, which causes producers to irrigate around 1 200 mm every year.

PRODUCTION SITES VISITED Kibbutz Hatzerim is one of the largest jojoba production sites in Israel. Jojoba oil was originally produced as an industrial lubricant. However, it soon became popular with the cosmetics industry due to its anti- inflammatory properties. The production of jojoba in the deserts of Israel is unique. Due to the mechanical harvesting procedures, the top soil in the tree row and work row is compacted. This is permissible since

PHOTO 5. Jojoba plantation that is irrigated by means of subsurface drip.

the rainfall is extremely low and the trees are irrigated by means of subsurface drip (Photo 4 and 5). This particular jojoba pl antat ion i s i r r i gated wi th recyc l ed municipal wastewater from the nearby city. Wastewater is often rich in minerals, such as nitrogen, potassium, chlorine and sodium. As for many of the sites visited, the crop is sufficiently supplied of nitrogen from the irrigation water without additional fertilisers. The subsurface drip also reduces the risk of damage caused by wild animals in the area. CONCLUSION Israel’s agriculture is dynamic and strongly supported by their government, as well as their private sector. Although many challenges exist, agronomical decisions are based on the specific growing conditions at

the time. These decisions are often backed by research carried out by specialists in the field. There are many unused water sources in South Africa, which can become increasingly important as the demand for fresh water increases. Although there are many difficulties when using unconventional water sources for irrigation, sustainable use can be achieved with proper guidelines. ACKNOWLEDGEMENTS SATI and Winetech for funding my tour. Drs Carolyn Howell and Philip Myburgh for initiating my participation of the tour. ARC Infruitec-Nietvoorbij for allowing me the opportunity to gain overseas experience. * This article was published in the November 2018 edition of the SATI Technical Bulletin.

PHOTO 4. Tour group being addressed by jojoba production manager. For public safety, purple pipes indicate recycled waste water is being used for irrigation. Fertigation tanks are buried subsurface to reduce theft.

– For more information, contact Carolyn Howell at howellc@arc.agric.za.

CAREN JARMAIN: Independent researcher KEYWORDS: Grapes, wine, water footprint. JANUARY 2019 WATER FOOTPRINT OF GRAPES AND WINE

During the past season, with many people living in the Western Cape trying to survive with a mere 40 to 50 L of water per person per day, many figures were quoted reminding us just how much water is commonly used in our daily living. For example, that about 19 L of water is used for a 90-second shower and that one flush of your toilet will use 9 L ( www.thesouthafrican.com ). Many people were also surprised to realise just how much water they “consume” through eating their daily food. For example that it takes 132 L of water to produce a single cup of coffee (125 ml), 196 L for a large (60 g) egg and 18 L of water for a slice of bread of 30 g. For supper, your 200 g steak will cost you 3 083 L of water and your small (125 mL) glass of wine 109 L ( www.project-platforms.com ). But the question is: what do these values really mean and do they matter?

These values also referred to as a “water footprint (WF)”, provide a measure of the amount of water used to produce each of the goods and services we use – whether your favourite fruit, a glass of wine or pair of jeans. Because of the nature of goods and services, a WF can be expressed in different ways, for example, a litre of water used per kg of crop produced (L/kg), but also a litre of water used per unit of currency derived (L/R). WFs considers both direct and indirect water uses and are sometimes expressed in its colour components: green, blue and grey. Let me explain. For example, the 109 L of water used to produce your 125 mL of wine will take into account all water used right from the production of grapes on the farm (e.g. rainfall and irrigation) all the way to the cellar (e.g. harvest container washing, winemaking processes, cooling and cleaning), but even beyond that, all the way up to the recycling of your wine

bottle. Indirect water uses considered will include the water used in the production of electricity, fuel, the wine bottle, labels and others. The colour classification refers to the origin or impact of the WF. For example, the green water component will represent the rainfall which ends up in the soil and is used to grow the grapes. The blue water again will represent all “physical” water used (for example for irrigating the crops) and which is extracted from streams or groundwater. The grey water component represents something very different to that which we consider grey water at home. The grey water fraction here is the volume of water required to dilute “polluted” water back to an acceptable standard for an area. Grey water is typically the result of on-farm chemical application or waste water generated in a process, such as making wine.

One could possibly argue that WF “numbers” do not matter much where water availability is sufficient, sustainable and of good quality. But under water-constrained conditions like the semi-arid regions of South Africa, including the Western Cape, these WF numbers and their meaning should matter. But just how much do we know about the WF of for example the wine or table grapes produced in the Western Cape of South Africa? Likely less than we think. The Water Footprint Network (WFN) is a good source of numbers for a wide range of crops and products for different countries and regions and can be viewed on waterfootprint.org. According to the WFN, the average WF of wine grapes produced in South Africa is 603 m 3 /ton (1 m 3 is equal to 1 000 L) of grapes, compared to the global average of 869 m 3 /ton. Interestingly, the WF of fresh grapes is 422 m 3 /ton of grapes, compared to the global average of 608 m 3 /ton. These values were generated using generic crop growth models and long- term climatic data for the period 1996- 2005. Just considering these values, it would suggest that South Africa is fairing quite well compared to the global averages. Unfortunately, since no contextualisation of the values is given, this conclusion cannot be drawn.

To shed more light on WFs, a new study jointly funded by the Water Research Commission ( www.wrc.org.za ) and Winetech aims at determining the water footprints for table and wine grape production, as a sustainability indicator. This project aims at generating more than just WF numbers, but placing these into the context of the environmental, societal and economic conditions. The research, under the leadership of Dr. Caren Jarmain, will consider the WF of both the wine and table grape industries. Although both the wine and table grape industries revolve around the Vitis vinifera plant, the processes involved in producing table grapes and wine differs significantly. In this study, the WF of wine (in m 3 of water per L of wine) will represent the water used in all the processes, right from the vineyard block up to the wine cellar, but before bottling. Similarly, for table grapes, the WF (in m 3 of water per kg of grapes) will consider all processes from the vineyard block up to packing the grapes in the packhouse, but before final cooling. This approach should allow the researchers to separate the WF into the part up to the farm gate from that beyond. This research is conducted in the Western Cape Province and a number of table grape

pack houses and co-operative cellars from the different production regions are already participating. This means that the wide range in production conditions which the Western Cape is known for will be accounted for. To provide anonymity to the participants of this study, the final WFs will be expressed per region and not individual pack house or co-operative cellar. What makes this study unique is the use of satellite-derived vineyard water use data and the integration thereof with available grape and wine production, climatic and other datasets (e.g. cellar water use and soil water content). The fact that block specific production and vineyard water use data are available, means that the WF of wine (of co- operative cellars from a specific region) or table grapes (for pack houses from a region) can be expressed as a range in WF values, rather than a single value. This WF range will allow for interpretation of the variation, which will lead us to understand the drivers (production or other conditions) behind low or high WFs. Finally and at the end of this project (2020), this new knowledge should empower both the wine and table grape industries to make the most of these numbers. Whether by addressing poor (high) WFs through specific changes at farm, cellar or packhouse level

and subsequently improving the impact these sectors have on water. Or, by utilising this knowledge in promoting South African products and produce on the basis of its relatively small WFs. TEAM MEMBERS Caren Jarmain, independent researcher, Centre for Geographical Analysis (CGA), Stellenbosch University, Pretoria University, Free State University, WineMS. FUNDERS Water Research Commi s s i on (WRC) , Winetech. REFERENCES https://www.thesouthafrican.com/cape- town-dayzero-what-can-you-do-with-50- litres-of-water-a-day/ . http://www.project-platforms.com/files/ productgallery-new.php . Mekonnen, M.M. & Hoekstra, A.Y., 2011. The green, blue and grey water footprint of crops and derived crop products. Hydrology and Earth System Sciences 15(5): 1577- 1600. Refer to: https://waterfootprint.org/ en/resources/waterstat/product-water- footprint-statistics/ . Water Research Commission Knowledge Review 2017/18. Refer to: http://www.wrc. org.za .

– For more information, contact Dr. Caren Jarmain at cjarmain@gmail.com.

Mealybug and scale insects are currently the only proven vectors of leafroll virus. All producers do not have the financial means to remove and replace leafroll infected grapevines. New grapevines from certified virus-free plant material are consequently being established amongst infected blocks. HANNES VAN RENSBURG 1 , PETRIE DE KOCK 2 AND PIETER LE ROUX 3 : 1 Private consultant, 2 BASF, 3 Villa Crop Protection KEYWORDS: Mealybug, scale insects, leafroll virus. FEBRUARY 2019 OF MEALYBUG IN SOUTH AFRICAN VINEYARDS AND WINE STRATEGY FOR THE CONTROL

HOW TO DETERMINE THE PRESENCE AND POPULATION OF MEALYBUG • Physical monitoring. • Pheromone traps. • Degree day model. The following control measures may be applied for mealybug: NB: Only active ingredients are listed – consult your agricultural chemical advisor. Read and follow the prescriptions and safety periods on the label. Spray judiciously – only sections where infection occurs. Withholding period is indicated in brackets. ANT CONTROL This is the first step in the control of mealybug. Ants and mealybug live in symbiosis with each other – ants eat the honeydew secreted by mealybugs. The following chemicals may be used: • Alpha-cypermethrin (28 days). • Chlorpyrifos (28 days). • Fipronil + Lambda Cyhalothrin (28 days). BIOLOGICAL CONTROL This entails the release of natural predators that parasitise the mealybug. Locally bred predators are available for this purpose: • Anagyrus pseudococci.

• Cryptolaemus montrouzieri. • Coccidoxenoides perminutus.

process not only mealybug predators, but also other beneficial insects are killed: • Acetamiprid (45 days) – table grapes only. • Borax + Orange oil – no withholding period. • Carbaryl (14 days). • Chlorpyrifos (28 days). • Dichlorvos (7 days). • Dimethoate (28 days). • Methidathion (8 days) – only one application. • Mevinphos (7 days). • Spirotetramat – not after pea berry stage. • Sulfoxaflor (28 days). CONCLUSION We have the knowledge and the means to control ants and mealybug. If mealybug is absent, the leafroll virus cannot spread to healthy vines. The only distribution can then take place by the vector(s) from already infected plant material. With thanks to Petrie de Kock and Pieter le Roux for their technical input.

CHEMICAL CONTROL AT ROOT LEVEL IN THE SOIL • Imidacloprid (112 days). • Clothianidin (90 days). • Thiamethoxam (application one month after harvest). Soil application at the time of establishment of new blocks. Some registrations are available for autumn application. WINTER AND SUMMER CHEMI CAL CONTROL ON SECTIONS OF THE PLANT THAT ARE ABOVE THE SOIL WINTER This is a good option considering the fact that the population of natural enemies during winter is low: • Chlorpyrifos – before budburst. • Prothiophos – before budburst. • Profenofos – application during dormancy of grapevine. SUMMER NB: This is the last resort, seeing that in the

– For more information, contact Lucinda Heyns at lucinda@winetech.co.za.

ILANI MOSTERT 1 , JOHAN BURGER 1 & HANO MAREE 1,2 1 Department of Genetics, Stellenbosch University, Stellenbosch; 2 Citrus Research International, Stellenbosch University, Stellenbosch KEYWORDS: Grapevine rupestris stem pitting-associated virus, GRSPaV. MARCH 2019 DETECTION AND CLASSIFICATION OF GRAPEVINE STEM PITTING- ASSOCIATED VIRUS

The aim of this study was to investigate the genetic diversity of GRSPaV in South Africa and to evaluate the success of biological indexing as a detection tool for this virus. Grapevine rupestris stem pitting-associated virus (GRSPaV) is a ubiquitous grapevine virus commonly detected in cultivated vines. It has been reported to be associated with rupestris stem pitting (RSP) and possibly Syrah decline (SD) (Photo 1) (Meng & Gonsalves, 2007). Its only known means of transmission is through vegetative propagation or grafting, and there have been no reports of an insect vector capable

of spreading GRSPaV (Meng & Gonsalves, 2007). Hardwood indexing is the primary method of diagnosis used to monitor the GRSPaV status for the fulfilment of sanitary requirements for candidate mother plants. Currently, six different variant groups, also referred to as strains, of GRSPaV are recognised (groups I, IIa, IIb, IIc, III and IV). Each variant group is represented by the complete genome sequence of one or more isolates of which 18 are available on an online database of nucleotide sequences (GenBank). Sequences of two parts of the virus genome, the coat protein and the replicase polyprotein, are used to detect and classify GRSPaV variants (Glasa, et al ., 2017). The symptoms caused by GRSPaV vary greatly among virus variants and depend largely on the cultivar infected. Infection of indicator ‘St. George’ grapevines ( V. rupestris Scheele) with group IIa variants does not elicit symptoms associated with RSP, but is strongly associated with grapevine vein necrosis (GVN), while variants of group IIc cause RSP. Furthermore, infection with group IIb causes mild or no symptoms, whereas group I is closely associated with Syrah decline (Habili, et al ., 2006; Lima, et al ., 2006; Meng, et al ., 2005). Inconsistencies in symptom expression may be explained by simultaneous infection by multiple viruses or multiple sequence variants of GRSPaV (Nakaune, et al ., 2008).

(IMAGE: VITIS LAB, DEPARTMENT OF GENETICS, STELLENBOSCH UNIVERSITY.) PHOTO 1. Pitting and grooves on the stem of V. rupestris St. George, the indicator for rupestris stem pitting.

regions, a section encompassing the coat protein gene (‘ CPreg ’) and a part of the replicase gene (‘ pREP ’), were determined and used to determine which GRSPaV variant(s) were present in infected plants. To investigate the impact of recombination on variant classification and the success of the GRSPaV coat protein and replicase doma i ns to co r rec t l y c l a s s i f y v i r us variants, the full genome sequences of the 18 representative GRSPaV isolates from GenBank were used. The sequences were screened for the presence of recombination using a recombination detection programme (RDP4) (Martin, et al ., 2015). The 18 isolates were then classified both by their full genomes, and by considering only the CPreg or pREP areas of their genomes. While a low incidence of GRSPaV was observed in the MB survey (6.25%), the prevalence of GRSPaV was much higher in the OV vineyards (29.06%), indicating a decrease of GRSPaV in plants screened using biological indexing. Isolates from five of the six currently recognised variant groups (I, IIa, IIb, IIc and III) were detected. A distinct variant group that is not represented by any of the full genome representative sequences was detected and subsequently labelled group IId. In most cases, the classification of variants from a single sample based on both the CPreg and pREP regions was consistent. The majority of isolates from the surveys clustered with groups IIa, IIb, IIc and IId.

Infections by more than one GRSPaV variant per plant were detected in both surveys. Genetic diversity was lowest in the MB survey and all but one positive sample originated from the same vineyard, which could be the result of failed elimination of the virus from source material. Isolates from this block fall within group IIa or IIb, which are known to elicit mild or no symptoms on indicator plants (Meng, et al ., 2005). This could indicate that hardwood indexing is less effective than RT- PCR at detecting asymptomatic or mildly symptomatic GRSPaV variants. Recombination was detected in three of the 18 full genome sequences used to classify GRSPaV isolates found in the field. Most notably, classification of reference isolates for variant group IIc based on the CPreg and pREP regions was not consistent. Because a recombination event occurred between the CPreg and pREP regions of these isolates, variants from group IIc were classified as group III based on sequences of the CPreg area. These results were reflected in the survey, where isolates from the same samples for which the pREP sequences fall within group IIc, seemed to cluster with group III based on the sequence of their coat protein. This demonstrates the possibility of incorrect classification if only one genomic region is considered. This study illustrates the broad genetic diversity of GRSPaV in South Africa. A

Parental variant 1

Parental variant 2

Recombination event

Recombinant sequence

Breakpoint position

FIGURE 1. Graphical representation of a recombination event between two virus variants resulting in a recombinant sequence containing fragments from both parental variants. Breakpoint position indicated with a red arrow.

Recombination is a process during which one virus variant exchanges parts of its genome with that of another variant. The resulting virus, also referred to as the ‘recombinant,’ is a ‘hybrid’ of the two parental variants and may possess a mixture of the parental traits. Recombination in viruses is a mechanism to generate genetic variability and has been linked to an increase in virulence, evasion of host immunity, and an increase in resistance to antiviral agents (Figure 1) (Simon-Loriere & Holmes, 2011). If a recombination event occurred between or inside the regions used to categorise variants, classification may be incorrect. It is therefore important to take recombination into consideration when classifying virus variants.

The influence of recombination and the use of two different genomic regions for classification of virus variants were also assessed. A survey (OV) was carried out on South African vineyards that were established prior to the implementation of current sanitary protocols. A second survey (MB) was conducted on mother blocks that, based on results from biological indexing, previously conformed to South African certification requirements, but are no longer used for vine propagation. Diagnostic reverse transcription polymerase chain reactions (RT-PCR) were used to identify plants infected with GRSPaV. The nucleotide sequences of two genomic

previously unclassified variant group of GRSPaV, IId, has been detected and remains to be further investigated. A noticeable difference in the distribution of virus variants was observed between previously certified and uncertified plants. The presence of mildly symptomatic or asymptomatic variants of GRSPaV in plants that tested negative by hardwood indexing indicates the need for development of improved detection and disease control methods, as non-symptom- causing variants may still play a role in virus pathogenesis (Alabi, et al ., 2010). Sequence data generated in this study can assist in the advancement of such methods. Although the results imply that the replicase area of the genome might be superior to the coat protein at discerning between virus variants, the ability of recombination to influence variant classification stresses the need to take these factors into account during assay design. SUMMARY Grapevine rupestris stem pitting-associated virus (GRSPaV) is linked to several grapevine d i seases . The ex i stence of mu l t i p l e strains of GRSPaV coupled with variable symptom expression and the occurrence of recombination has led to difficulties in detecting and classifying GRSPaV isolates.

Two surveys were conducted to investigate the genetic diversity of GRSPaV in South Africa, to compare the effectiveness of two genomic domains to classify virus variants and to investigate the impact of recombination on variant classification. GRSPaV variants identified in the surveys clustered into five of the six currently recognised lineages, and a seventh, previously unclassified lineage was detected. A correlation was observed between the detection of recombinant GRSPaV sequences and inconsistencies in classification when using different genome regions for analysis. Results from this study indicate that visual diagnosis on an indicator host is not always sensitive enough to detect mildly symptomatic or asymptomatic variants of GRSPaV, stressing the need for development of more sensitive detection assays to fulfil sanitary requirements. ACKNOWLEDGEMENTS The authors would like to acknowledge Winetech for project funding (Grant number: GenUS 15/2). We also want to thank the owners of the wine estates for permission to sample vineyards and for providing information on the ages and cultivars of sampled vines, the National Research Foundation of South Africa for personal funding, and Dr. Rachelle Bester and Dirk

Aldrich for their assistance in performing some of the lab work for this project. REFERENCES Meng, B. & Gonsalves, D., 2007. Grapevine rupestris stem pitting-associated virus: A decade of research and future perspectives. Plant Viruses 1: 52-62. Glasa, M., Predajňa, L. & Šoltys, K. et al. , 2017. Analysis of Grapevine rupestris stem pitting-associated virus in Slovakia reveals differences in intra-host population diversity and naturally occurring recombination events. Plant Pathology Journal 33(1): 34-42. doi:10.5423/PPJ.OA.07.2016.0158. Habili, N., Farrokhi, N., Lima, M.F., Nicholas, P. & Randles, J.W., 2006. Distribution of Rupestris stem pitting-associated virus variants in two Australian vineyards showing different symptoms. Annals of Applied Biology 148(1): 91-96. doi:10.1111/j.1744- 7348.2006.00041.x. Lima, M.F., Alkowni, R., Uyemoto, J.K., Golino, D., Osman, F. & Rowhani, A., 2006. Molecular analysis of a California strain of Rupestris stem pitting-associated virus isolated from declining Syrah grapevines. Archives of Virology 151(9): 1889-1894. doi:10.1007/s00705-006-0742-y. Meng, B., Li, C., Wang, W., Goszczynski, D.

& Gonsalves, D., 2005. Complete genome sequences of two new variants of Grapevine rupestris stem pitting-associated virus and comparative analyses. Journal of General Virology 86(Pt 5): 1555-1560. doi:10.1099/ vir.0.80815-0. Nakaune, R., Inoue, K. & Nasu, H. et al. , 2008. Detection of viruses associated with rugose wood in Japanese grapevines and analysis of genomic variability of Rupestris stem pitting- associated virus. Journal of General Plant Pathology. http://agris.fao.org/agris-search/ search.do?recordID=US201300878498 . Accessed July 5, 2017. Simon-Loriere, E. & Holmes, E.C., 2011. Why do RNA viruses recombine? Nature Reviews Microbiology 9(8): 617-626. doi:10.1038/ nrmicro2614. Martin, D.P., Murrell, B., Golden, M., Khoosal, A. & Muhire, B., 2015. RDP4: Detection and analysis of recombination patterns in virus genomes. Virus Evolution 1(1). doi:10.1093/ ve/vev003. Alabi, O.J., Martin, R.R. & Naidu, R.A., 2010. Sequence diversity, population genetics and potential recombination events in Grapevine rupestris stem pitting-associated virus in Pacific North-West vineyards. Journal of General Virology 91(Pt 1): 265-276. doi:10.1099/vir.0.014423-0.

– For more information, contact Johan Burger at jtb@sun.ac.za or Hano Maree at hjmaree@sun.ac.za.

BERNARD MOCKE: Private consultant KEYWORDS: Wine grape yield, machine learning, FruitLook. APRIL 2019 A NEW ERA FOR WINE GRAPE YIELD ESTIMATION

With technology at the fingertips of researchers and producers alike, the possibility exists to not only generate and study vast datasets, but also to find re l at ionships between them through machine learning and statistical analysis. Such relationships could provide valuable information on yield estimation, véraison and harvest date. According to the link https://bit.ly/2K0TV2Q , machine learning (ML) is a branch of artificial intelligence based on the idea that systems can learn from data, identify patterns and make decisions with minimal human intervention. But why is this important? An incredible amount of data has been captured within the wine grape industry in the Western Cape over the years and put to good use, but all of this information could be even more valuable if relationships between datasets could ultimately allow the producer to more accurately estimate yield.

In addition, data capturing can be done not only by traditional means, but also through remote sensing (RS) by satellite through initiatives like FruitLook. This article discusses research funded by Winetech which investigates the use of available “big datasets” for wine grape crop estimation. The research was led by Dr Caren Jarmain and other researchers from Stellenbosch University, with support from WineMS and Vinpro. WHAT WERE THE AIMS OF THIS STUDY? Two specific aspects were investigated: • Whether relationships existed between the spatial datasets available through FruitLook (FL) and vineyard block yield information when statistical analysis and machine learning (ML) approaches were applied. • If these relationships could be used to build a wine grape yield and possibly a harvest date model.

PHOTO 1. Illustration of machine learning.

(October until April for 2011/12 to 2015/16) at a spatial resolution of 20 m. • Crop production data – actual vineyard block production information (quality and quantity) for the above-mentioned five production seasons obtained through the industry. • Block boundaries – vineyard block field boundary information obtained through the industry, indicating the geographical locations and extent of the fields considered. Additional datasets were derived and all datasets were joined at block level to study the relationship between crop yield and RS derived FL data, using a combination of statistical modelling and ML. Initially the ML involved manually changing target variables, but subsequently a brute force approach was applied. This involved a near- exhaustive set of experiments, each using a different permutation of target variables, input variables and geographical areas. WHAT CONCLUSIONS CAN BE MADE FROM THIS STUDY? Although crop yield per block is frequently recorded, the associated hectares and other associated information are not regularly updated. The graphically delineating information (block boundaries) too are not readily available in electronic format. This study was a first attempt to model grape yield using RS variables. The initial

TABLE 1. Many variables were considered in this study. Spatial data FL weekly data from 2011/12 to 2015/16, including biomass production, actual evapotranspiration, evaporation deficit, water use efficiency, leaf area index, normalised difference vegetation index, plant nitrogen, nitrogen in top leaf layer and FL various derivatives also at monthly and seasonal time steps (made available by eLEAF and the Western Cape Department of Agriculture).

WHAT ARE THE CURRENT WINE GRAPE YIELD ESTIMATION CHALLENGES? S i mp l y pu t , t h e ye a r l y und e r - o r overestimation of vine yield by producers leads to under- or overestimated wine volumes which in turn could lead to a loss of money. It is estimated that the expected error could be up to 20% for forecasts based on bunch counts in spring, 10-15% for forecasts based on berry counts at fruit set and 5% on harvesting segments close to the harvest. The estimated yield has a direct effect on cellar capacity, chemical usage and marketing activities. For growers, yield directly impacts planning of the harvesting processes and the scheduling of labour and machinery. Current yield prediction largely relies on historical yield data and weather indices combined with manual vineyard measurements and sampling, but this approach is inaccurate and time consuming. WHAT MEASURING PARAMETERS AND SOURCES WERE USED IN THIS STUDY? Selected production areas with the required “big datasets” (block specific and remote sensing derived) were included in the study and extended over the Coastal, Breede and Olifants Wine of Origin regions of the Western Cape. Three main data sources were considered: • FruitLook data – weekly FL spatial datasets on crop growth, water use and nitrogen content for five production seasons

Block specific data

Grape yield in ton/ha or ton/block for 2 104 fields (obtained through industry).

Other block attributes including quality indicators (acid, pH and sugar), block number, block area (ha), cultivar, grape type, plant year, rootstock, trellis and irrigation method.

Field boundary data for 2 104 fields.

Geographical delineation of wine production wards, districts and regions and climatic zones.

Other data like production season.

regression analyses to investigate the strongest relationships between individual FL variables and yield generally yielded poor results (R 2 < 0.3), but the models improved when individual cultivars, specific regions and seasons were considered. For example, the 2014/15 season’s weekly (non-aggregated) FL variables yielded a strong (R 2 = 0.83) model for Pinotage in the Coastal region. Other ML findings, especially from the Olifants River region, were very encouraging. When RF (Random Forest), an artificial intelligence classifier, was applied to the 2015/16 seasonal (aggregated) FL

variables in this region, an overall accuracy of 85% was achieved when all the cultivars were considered. Similar results were also observed for the 2014/15, 2011/12 and 2013/14 seasons, confirming the consistently strong relationship between all the FL variables and yield for this region. More data and work is needed regarding a harvest date model. The data that was used was obtained from less than 30 blocks, but yielded promising results. Ultimately, based on the regression and ML experiments the following was concluded:

Figuur: Illustrasie van ML

Machine learning algorithm

• No individual variable can be used to model wine grape yield. • Using RS data for wine grape yield modelling is very complex as other (non- remote sensing) factors often have a (more) substantial impact on yield. • The accuracy of the models are strongly driven by cultivar (with Chenin blanc and Colombar being the most successful) and by region (with the Olifants River region being the most successful). • Weekly FL variables generally produced stronger models than (aggregated) monthly and seasonal variables. WHAT FUTURE RESEARCH IS RECOM- MENDED? The research team recommends that the wine industry develops and maintains a standardised geographical database of vineyards and their related attributes, seeing that data inaccuracy or completeness were

barriers to research. Sawis records, as well as the “Fly-over” database from the Western Cape Provincial Department of Agriculture, would serve as a good starting point. More research on wine grape modelling is strongly recommended and to that end other raw satellite data (e.g. from Sentinel-2 and Landsat-8), in addition to the FL data, should be considered. The area considered in the research should also be expanded. Finally, more research into modelling harvest date should be done with specific focus on the use of ML and a factor classification approach. This will define which FL dataset drive the positive results obtained. The current dataset considered served the exploratory work well, although it was too limited to be split into a training and test set for ML while still retaining the seasonal variability aspect, as well as differences between sites and cultivars that also impact on phenology modelling.

Un-labelled databases

Other related data

Labelled samples (training set)

Labelled databases

Human knowledge (or observed data)

– For more information, contact Dr Caren Jarmain at cjarmain@gmail.com.

CAROLYN HOWELL: ARC Infruitec-Nietvoorbij, Stellenbosch KEYWORDS: Treated winery wastewater, New Zealand wineries. MAY 2019 MANAGEMENT AND RE-USE OF TREATED WINERY WASTEWATER (PART 1): NEW ZEALAND WINERIES

Carolyn went to New Zealand to participate in a technical tour focussing on the management and re-use of treated winery wastewater, with particular reference to wine grape irrigation. She reports… The Western Cape is experiencing one of its worst droughts to date. Urban users are currently limited to 105 L of water per person per day and water is a precious, scarce resource. Approximately three years of good rainfall is needed to recover from the drought. This implies that even if the 2018 winter rainfall is normal, the region will still feel the negative consequences for some time thereafter. Therefore, alternative sources of irrigation water for vineyards, e.g. using treated wastewaters, will become more important under the above-mentioned conditions or if climate change reduces long- term winter rainfall.

THE GEOGRAPHY AND CLIMATE OF NEW ZEALAND New Zealand is an island country located approximately 2 200 km from Sydney in the south-western Pacific Ocean. It consists of two main islands, namely the North Island and the South Island separated by the Cook Strait. The climate is mostly temperate, with January and February being the warmest months and July the coldest. Most settled, lowland areas of the country have between 600 and 1 600 mm of rainfall. The west coast of the South Island has the highest rainfall, while the east coast of the South Island and interior basins are the driest. WINERIES VISITED IN THE WAIRARAPA REGION The Wairarapa wine producing region, of which Martinborough is a sub-region, is a

PHOTO 1. The windbreak on the right-hand side where the winery wastewater at Martinborough Vineyard is irrigated. Note that during winter the irrigation line is stored on the vineyard irrigation line. The wastewater irrigation line is purple, as per convention to indicate wastewater.

small grape producing region and produces 1% of New Zealand’s wine grape harvest. Martinborough Vineyard is a small winery in Martinborough where approximately 200 tonnes of grapes are crushed per year. Seventy percent of the grapes crushed at Martinborough Vineyard are Pinot noir. Most

of the wastewater in the winery is generated by cleaning tanks and washing floors. The winery has stopped using sodium hydroxide (NaOH) as a cleaning agent. The wastewater is pumped through a coarse sieve and stored in two 3 000 L underground tanks. With the exception of pH adjustment, the wastewater