Farming by numbers: metrics that matterTechnology & Data
Jack Wilson, AgFirst Horticultural Consultant
Measuring and using orchard data to drive decision-making delivers decisions that are more objective, more transparent, more consistent and ultimately more effective, leading to improvements in efficiencies and long-term profitability.
But the ever-increasing number of tools available to measure and use data can be overwhelming, so where do you start?
Using data is not new. Growers have been using a range of easy to collect measures for time immemorial to understand how their orchard is performing year by year, block by block (or neighbour by neighbour) and to assist in making informed decisions.
Commonly-used measures, indicators or yardsticks for orchard comparison – known as ‘metrics’ in business jargon – include:
- Kilogram or bins per hectare
- Class 1 packout per hectare
- Bud counts per tree
What is new, is that the increasing sophistication of industry with access to new agtech, greater emphasis on quality and consistency in market requirements and facing pressure from rising labour and input costs, requires far more complex decision-making.
Development of new tools capable of the capture and evaluation of a greater range of data has also seen growing interest in more complex indicators that measure not only output but take into account costs to give measures of overall business performance, such as profit/hectare or profit/kg.
The metrics that matter to you in your decision-making may not be the same ones that matter to others, but there are some common indicators that can be used to guide orchard management to inform understanding and improvement of business performance.
This article will introduce some key orchard metrics that can be used to guide the winter pruning and thinning activities and support tools that can help guide your data-driven management decisions and help you achieve your business goals.
There are a range of tools that can help you capture and analyse your data.
The Australian apple and pear industry has two orchard management decision-support tools that can be used by growers and have built up 15 years of reference data.
Both can be used to create datasets that are regularly used as benchmark indicators with information shared via Future Orchards® Orchard Walks and from time-to-time through other APAL communications channels.
(i) OrchardNet – OrchardNet is a cloud-based orchard management tool that captures production, monitoring and financial metrics.
OrchardNet allows you to share data, see history, plan how to meet your goals for the following season. Its benchmarking capabilities allow you to compare your results year by year, block by block or between companies. OrchardNet is designed for pipfruit, kiwifruit, summerfruit, grapes and almonds. To illustrate how it works, I have set up the Australian Demo with this functionality fully operational. To go to the demo site, go to the OrchardNet login page http://www.hortwatch.com/orchardnet/ and
(ii) Orchard Business Analysis (OBA) program – As part of Future Orchards® project, the OBA is a modelling exercise that provides the industry’s best economic indicator of productivity and financial performance of orchard businesses. Year on year this provides a repeatable productivity and financial snapshot of the industry providing trends and benchmarking. Growers can use this to access relevant data related to new varieties, benchmarking production, costs and overall business performance (please refer to the end of this article for information on OBA access).
With the help of the key physiological metrics gathered in the season and the Orchard Business Analysis (OBA) program we can develop block profit reports using on-orchard and postharvest costs as well as robust industry returns. This allows us to analyse performance, comparing across time, between crops, blocks and locations, and between businesses.
The Orchard Business Analysis report can help you assess how your business’s performance sits against industry performance and help identify business strengths and weakness. If your income per ha and per kg is lower relative to industry averages, it may prompt investigation into why that is. Is your variety mix poor, your Class 1 packouts low or your fruit size below optimal?
Be bold and benchmark your costs of production using the OBA. For example, what is your total on orchard labour costs? If you are spending more than 55c per gross kg harvested, then you are above the Australian average. Then ask yourself why.
Consider the experience of one very established Australian grower who back in 2010 wouldn’t believe that he had one of the highest labour costs per kg of fruit grown in the country.
However once he was shown how his costs compared to the industry data as calculated in the Orchard Business Analysis (OBA), he went on a path of improvement. Workers were shifted from hourly rate to piece rate, the labour supply source changed and most importantly he lifted productivity, so the labour costs were spread over more kgs per ha.
Five years later he had one of the lowest costs of labour per kg and has been thriving ever since.
A data-based pruning plan
OrchardNet allows growers to use production goals such as desired yield, packout and fruit size to calculate target bud numbers to guide orchard tasks such as pruning, hand and chemical thinning strategies.
To use this function, you will need to know:
- Estimate yield per tree or hectare
- Class 1 packout per hectare
- Buds required per fruit
- Trunk cross-sectional area (TCA)
- Pickout (The percentage of fruit after thinning you expect to get into the bin (after fruit drop, field grading etc.) – 90 per cent is the default however this number can be variety and block specific.
Step 1: Set your production target by block.
For a more accurate estimate, it is worth looking at least three years of historical data including yield, average fruit size and packout. Good historical data removes the need to rely on memory. With that data you can make an informed decision on what would be an optimum target.
Having an estimate of your production then allows you to develop a plan to achieve your production target. Calculating and collecting that metric allows you to develop a pruning strategy that can be quantified through databases such as OrchardNet.
Figure 1 shows the metrics from an old Royal Gala block. This data enables you to identify the block’s potential and any obvious trends that have occurred within the last three seasons.
A graph (Figure 2) can be generated in OrchardNet showing the performance of a block making it easy to visualise a production estimate for the coming season.
The Royal Gala block in Figure 1 shows a slight biennial bearing trend, so we can expect an increase in tonnage in 2022.
Setting our target tonnage to 95 t/ha with the help of other biennial reduction strategies such as early crop removal will help us avoid exceeding the optimal crop load and causing the biennial tendencies to continue. Setting the right fruit size for the destination market is another key metric, in this case we are setting an average fruit weight of 160 grams and entering both estimates in the input tab (Figure 3).
Step 2: Setting your crop load target
Once you have established your appropriate crop estimate it is time to formulate a pruning plan.
A pruning plan opens up a range of metrics which again begin to make decisions easier with the guidance of hard numbers. OrchardNet’s ‘metrics’ tab offers a range of different quantitative metrics for use at a block level. Winter pruning metrics that are required to create a pruning plan include:
- Trunk cross-sectional area (TCA). A metric commonly used in young trees as a way of regulating crop load depending on the size of the tree.
- In OrchardNet this can be entered as a trunk diameter (cm) or area (cm2)
- Pickout percentage
- The percentage of fruit after thinning you expect to get into the bin (after fruit drop, field grading etc.) – 90 per cent is the default however this number can be variety and block specific.
- Buds required per fruit (refer to step 3)*
- How many fruit buds to leave for each desired piece of harvested fruit
- Typically, this ranges between 1-3 depending on goal and the variety.
- The lower the number the more vigour induced, the higher the less vigour.
Step 3: Setting an appropriate bud target per fruit
Knowing the appropriate target winter bud number required per fruit will be foreign to many growers who have traditionally relied on their experienced eye to judge the quantum of wood retained. Building up your own data over time will help fine tune the accuracy of your bud target per tree. This data gives the business valuable knowledge to predict the buds required to meet the production target in the coming season.
The pruning report (Figure 4) shows three previous years’ data on the actual bud density per fruit experienced with each crop (narrow range of 1.6–1.8 actual buds/fruit). This data enables a more accurate prediction for the coming season. If your business is new to this approach, broad guideline targets are shown in Table 1. Note there is a wide range based on variety, vigour and chemical thinning planning.
After entering this data, you can generate a “pruning report” in OrchardNet (Figure 4) which indicates an objective bud target (number of fruit buds for each tree).
The bud target per tree in practical terms can help you assess if you or your team’s pruning performance is on track to deliver your bud target per fruit, shown in the ‘bud monitor prune’ column in Figure 4.
Step 4: Implementing the pruning plan
With these numbers the manager can start developing practical pruning rules following a simple process shown below:
1.Conduct pre-prune counts.
- Ideally it is best to do this after the necessary structural cuts that ‘need to happen’
- This value will allow you to compare your pre-pruning bud numbers to your target bud numbers developed in the pruning plan above.
- This will give you an idea as to whether the bud per tree pre-prune is higher or lower that the bud targets
2. Prune trees as desired with bud target/budget in mind
- Keep in mind the result of your pre-prune count and prune with the appropriate detail, or lack of.
3. Post-prune count
- If this is on target; you can develop rules to instruct your pruners to get to the same sort of tree with simple rules.
Farming by quantifiable metrics including the counting of winter buds becomes extremely important and worth the investment as it provides certainty that your trees are going to be set up for optimal balance.
As the industry evolves, the need for optimal performance and hence precision in everything we do increases.
Data-driven chemical thinning
Now that we have successfully set up the trees with the appropriate pruning strategy, we start to think about thinning and how we might go about using quantitative data to get the fruit number down as efficiently as we can. Generally, your chemical thinning strategies rely on historical data. People often say chemical thinning is more of an art than a science because of how much you rely on past experience and your judgement on forecasted weather conditions. However, although your decisions may never be fully science driven, we can use a range of metrics to fill the gaps in our knowledge allowing us to be reasonably well informed.
Step 1: Gather details from your previous chemical thinning activities.
Recording metrics to measure business performance does not always have to be quantitative.
In terms of chemical thinning, by knowing the history of applications and the qualitative outcome of those strategies you can be confident you have the appropriate information in front of you in order to succeed again.
To make a well-informed decision it is vital that you have a full dataset.
A list of your chemical thinning applications detailing:
- Chemical types used and their rates
- Water rates
- timing/date of application
- date of crop stage
- area of canopy targeted.
- weather conditions
- other factors that influenced your decisions
- e.g. warm nights, hot days, dry conditions, waterlogged soil.
Couple this with a detailed set of notes of your satisfaction of the flower/fruitlet response to each spray and you have a very powerful tool to refine chemical thinning strategies at the block level and minimise the chance of repeating a less-than ideal result.
Step 2: Analyse historical results
OrchardNet has a built-in chemical thinning planning and recording tool that allows this data to be captured to an individual block level.
Once you have a record of your previous strategy; seasonal reviews allow for adjustments to be made to optimise your intended outcome.
Step 3: Make adjustments to your chemical thinning plan based on season reviews to optimise outcomes.
Where previous chemical thinning strategies have provided good results for your business’s risk profile, target spend and desired result, there is little need to change. Some tweaking of rates to account for seasonal conditions (e.g. warm weather, waterlogged soil, weak bloom) may be necessary but with a full dataset of chemical thinning
strategies, conditions and results these small tweaks can be planned and, once the season is complete, confirmed to provide a better result.
Where past season results are less than optimal, a considered approach is the best course of action. Do rates need to be increased/decreased? Could another thinning strategy be applied to the problem block? What other factors might be influencing the result?
By collecting a quality dataset each year, your process is recorded and over time you will be able to be more confident in your chemical thinning approach and ensure continual optimisation of your strategies.
Thinning and harvest labour planning
When entering the part of the season where labour demand is high, orchardists need to be planning well in advance, especially where, as with the current Covid-19 travel restrictions, supply is further constrained. Having captured a range of metrics through the winter months, labour-related decisions become based on accurate numbers. OrchardNet contains a labour planning tool (Figure 6) for the hand thinning period. The hand thinning tool considers your estimated production data and calculates the
number of thinners needed assuming a person can thin 320 kg of harvested fruit per hour (industry average). It will calculate, over a select period, how many thinners are required assuming a 5-day or 6-day working week. Metrics you need to calculate your thinner requirement include:
- the estimated production values per block or variety (a metric we calculated in the winter)
- the proposed start and finish date of your thinning period.
From an orchard physiological point of view, orchard metrics continue to develop alongside new technology. New canopy systems such as FOPS (Future Orchard Planting System) and 2D have dramatically changed the physical structure of wood in a repeatable manner meaning we can start to measure elements such as crop loading with
more precision using Branch Cross Sectional Area (BCA) or fruit per linear meter.
In both the Australian and NZ industries, water has become the new gold, which means our quantification of what our trees actually need will need to become more precise.
With normalised difference vegetation index (NDVI) technology we are quickly realising the extent of variability within our orchards. Metrics around the severity of variability in order to minimise the effects may become a normal operation to better our margins.
These are all new metrics based on new technology that will help growers navigate their way through farm business. Staying on top of these technologies will put us in a preferable position moving forward allowing us to continue to farm precisely and efficiently. Like the industry, OrchardNet is a forever developing beast and with
the help of leading industry minds, it will continue to provide supportive services based on new technology and trends.
The OBA report and OrchardNet are both available through the Future Orchards® program. The OBA Analysis Report is available exclusively to levy-paying growers. To request a copy please contact APAL by emailing or phone 03 9329 3511.
If you want to try OrchardNet, either contact your local Front Line Advisor or contact firstname.lastname@example.org for set-up assistance.