Monitoring and analysing your vines helps to predict grape yield and when to harvest your grapes to get the highest quality.
How many vines should I monitor?
First you need to create a sampling plan which is clear and easy to manage – making sure it’s actually doable is the most important element. If it’s too complicated you are likely to lose patience with it! Research papers recommend you sample 3% of your vineyard, this is the minimum researchers recommend you need to do to get a representative sample for predicting your grape yield. In our experience many commercial grape growers sample closer to 1-2% of the vineyard and can still get very good predictions — the key is to make sure you are regularly going out and looking and adjusting any predictions based on what you see. It is after all an art and a science!
Which vines should I monitor?
We recommend proportional stratified sampling: in basic terms this means you need to split your vineyard in smaller somewhat homogeneous blocks. Maybe your vineyard is already naturally broken down, for example you may plant different varieties or clones in blocks in the vineyard, that makes it simple! If you have a large-ish area of vines that significantly under or over performs in comparison to other area you could also define this area as its own block.
If you don’t have anything in place already, don’t worry, think about your number of rows and the number of bays on each row, and draw a simple map of the vineyard, then outline the different blocks of vines that are similar. When using Sectormentor it’s also good to think about what do you need this information for? For example, often (especially initially) vintners manage all clones of a specific variety in the same way. However, they will still setup Sectormentor to monitor based on blocks of clones, that is because they need to do separate yield predictions for each clone for the winery and therefore to get that level of information they need to ensure they are sampling enough vines within each clone.
Some good questions to ask yourself when determining where each block should be are: when harvesting, what do you measure the yield for, each variety? Each clone? And when doing a specific activity, or yield prediction, does it vary widely depending on variety? Or clone?
How do I choose sample sites?
Once you have your blocks defined, you can decide which bays to sample within each block. It is suggested you sample 1-3% of bays in each block using a systematic method of selecting random samples (systematic sampling). You should always start a few rows in to avoid getting skewed results from the rows on the edge of the vineyard.
When you use Sectormentor For Vines to record data you scan an RFID tag which immediately identifies the sample site you are working in. RFID tags make it simple, because you can attach them directly to the vine supports, and simply scan them when you want to collect data at their location. Each sample site can have it’s own RFID tag. You can also use the GPS function on the Sectormentor for Vines app to set the location of your RFID tags, so you can view them all on a map and find them easily out on the field.
Some vintners choose to sample “randomly” and go to a set number of vines, but at ‘random’ locations each time. On the largest vineyards this is sometimes more convenient, but it’s often not the most accurate method because we humans aren’t very good at choosing ‘random’ locations on the fly. It is also possible to setup Sectormentor this way, if you are interested just let us know and we can set this up for you.
What is an example of systematic sampling?
Let’s look at an example setup. Let’s say you have a vineyard split into 2 blocks, based on 2 different clones planted. Each of these blocks is relatively homogeneous. Consider one block, if a block has 30 rows with 20 bays in each row, and each bay has 5 vines. This means there are 100 vines/row. So that is 3000 vines in that block (see diagram below). You plan to sample close to 3% of the vines, then you have to sample 90 vines.
Using systematic sampling, you could pick every 5th row to sample, excluding edges. And if each row has 20 bays, you could have a system across the vineyard of always sampling from the 3rd, 9th and 15th bays. These bays are where you put the RFID tag and start the sampling from.
For efficiency we suggest sampling multiple bays at each location, so in this example you would always sample 3 bays from the tag, so if the tag was at bay 3, then you would use that tag to sample bays 3, 4 and 5.
Alternatively you could do systematic sampling across the entire vineyard for example every 4th vine on every 6th row. So you go to every 6th row and sample vines 4, 8, 12, 16, 18, 22 etc along that row. Although this method could potentially be more comprehensive, it can take longer to complete, and may not be manageable for your vineyard. For this method, we recommend putting one RFID tag at the end of the row, so you have sample rows instead of bays. When you scan the RFID tag the Sectormentor For Vines app will automatically remember which row you are on and save the data for all vines along that row.
One North Eastern US study states: “Crop estimation helps ensure consistent production of high-quality fruit over multiple years in our variable climate. A grape grower who is unable to invest in, or elects to ignore, developing operational competence in crop estimation is likely to be at a competitive disadvantage in tightening markets.”