My experience with Vivid Machines in 2024

 During the 2024 growing season I worked with Vivid Machines (https://www.vivid-machines.com/) using their XV3 camera and cloud platform to track apple crop load on a tree-level basis in two orchards across eight different orchard blocks including five varieties. In Massachusetts. What follows is my summary of the process of scanning the orchard blocks with the XV3 camera, transmitting the data to the cloud, and how the results are displayed and interpreted. There was no particular objective here other than to casually evaluate and demonstrate the Vivid Machines platform during the 2024 growing season.

First, the hardware and set-up. We received the XV3 camera and associated mounting hardware from VividMachines. (I will use VM from here on to stand for VividMachines). The camera just has a power button and is a rather robust, rugged looking, somewhat heavy piece of equipment. In other words, it is built. The camera was attached to our John Deere Gator – could be any ATV, tractor, etc. – using a rather floppy (but built), four-legged mount that was attached to the hood of the Gator using large suction cups. Getting the mount secured to the Gator and the camera attached and leveled to the mount was sometimes a bit of a coordination issue, but it generally worked OK. If you have watched Tales from the Loop, which I highly recommend – https://en.wikipedia.org/wiki/Tales_from_the_Loop – it kind of ended up looking like one of the robots therein. One has to watch the suction cups, however, to make sure they do not loosen up while scanning orchard rows. The XV camera can also be mounted more securely to a single vehicle using their rigid mount which I would recommend if it was only going to be used in one orchard and mounted to one vehicle. Next year maybe.

Once the XV3 camera is mounted securely, pushing a button powers it up. The XV3 camera is powered by an internal battery that can either be fully  charged by an AC adapter prior to using, or using a 12 volt DC cigarette-lighter style adapter while operating. The camera has high-level GPS accuracy (RTK?) and wi-fi built in, so it often took a minute or two to lock on the GPS signal and transmit wi-fi. Which is then used to connect to their Vivid Camera Control app on smartphone or tablet (iOS or Android). Camera Control is then used to control the XV3 camera while scanning orchard rows.

One more thing before starting scanning, is the outline of the orchard blocks needs to be mapped which is done using the Camera Control app by driving around the perimeter of the block and mapping corners so the platform knows where the block is, how big it is, etc. Seems to me there is some other information – tree spacing, variety, etc. – that needs to be inputted to the VM Dashboard – more on the Dashboard later – while doing this but that is a minor detail.

OK, once all of the above is in-hand, you can start scanning with the XV3 camera. In the Camera Control app, select the block you are scanning, push start scan (in the app) and start driving down any or all of the rows you want to scan. Maximum speed 5 mph, slower is better, we tried to do about 2-3 mph for our demonstration scans. Driving and scanning as we did is ideally a two-person job, but could be done by one operator once you get used to it. There are some nuances, camera positioning, etc. But overall it was relatively easy. Oh wait, one more thing: it’s called Calibration, which requires you to stop scanning, and measure some fruits (10 or so?) on a minimum of four trees (we typically did five). This is where having two people helps, although VM supplied a Bluetooth caliper so fruit measurements could go directly into the app. It worked, most of the time. It all goes into the CameraConnect app but it is a process that has to be followed for the most accurate result. Seemed a bit onerous at first, but once you get the hang of it, it typically took 10-15 minutes or so to scan 5 rows per block. On one side only, which seems to me was the recommended procedure.

Once scans are complete – and you can do as many blocks as you want, each called a “session” – the camera is detached from the mount, brought inside and attached to an internet-connected router (cabling and power supplied), powered up, and the data (scan images, etc.) are uploaded to the VM Tableau (https://www.tableau.com/) cloud-based Dashboard. It’s pretty much automatic, but can take a couple hours to upload a lot of image data, best to plug it in and do it overnight. VM would also occasionally push software updates to the camera.

Once fully uploaded, the data is available (next day typically) and displayed in the Dashboard. TABS – Home, Map, Scan Summary, Fruitlet Thinning, and Yield Potential – serve as the main navigation aid. I will briefly attempt to describe what is displayed in each tab, but you need to know within the Tabs many choices – Variety, Farm Block, Date, Stage, etc, – can be filtered for only the data you want to see. VM did supply some training and follow up support in using the Dashboard, as well as any problems or questions that arose when out scanning in the orchard. But I have to say it is not exactly for the timid or techno-phobe. Hint to VM here might be to simplify it a bit in terms of user interface?

Here are the TABS within the VM Dashboard:

• HOME – here there is a Farm Map, and Global Statistics (all varieties, all scan dates, etc. but can be filtered down to the one you want to see) can be displayed. Figure 1: VM-Home.
• MAP – produces a map down to the tree level of the scanned trees which includes counted fruit and fruit size. Along with a Histogram of Fruit Avg. Predicted Count. Figure 2: VM-Map.
• SCAN SUMMARY – gives you Avg. Predicted Count (per tree) and Avg. Predicted Size. Among various other scan statistics like Number of Trees scanned and Distance travelled. You can also Show Full Tree Data and Download it. Which is basically the individual tree raw scan data. It’s a lot to digest here, but it’s all there. Just waiting for you to dig into.
• FRUITLET THINNING – gives you a predicted (beta) Fruit Set Estimate expressed as Avg. Fruit Likely to Stay on Tree. Three successive scan dates once the fruitlets are in the 10 mm. size range to get this Estimate. If it can be verified, which is a work in progress, this could be very useful by greatly increasing sample size when trying to use the fruitlet growth rate model to predict fruit set and help make chemical thinning decisions.
• YIELD POTENTIAL – shows a Projected Fruit Size line chart (beginning with the first scan through the last scan) as well as Fruit Size Distribution in a bar chart. Also an Estimated Bin Count. 

Now, one can do a lot of data mining in any or all of the above as to what and how tree level data are manipulated and displayed. Frankly, it’s a lot of information. As I mentioned, I’ve given feedback to VM perhaps they could make it simpler? What to do, what to do? But where I do currently see “actionable” potential:

• Blossom scanning, clusters, not individual flowers, yes, starting at pink. Could be useful for precision pruning, and variable rate spraying of caustic bloom thinners.
• Fruitlet Thinning, using the fruitlet growth rate model (in some way, shape, or form), if it can be verified with further ground truthing, would be very useful for decision-making when it comes to chemical thinning. Doubt it’s ready for prime time. Yet. Plan is to double-down on it in 2025.
• Season-long scans to predict block-level Yield Potential and Fruit Size Distribution are handy for growers trying to manage harvest labor, bin needs, and sales desks. Could also be used to direct hand thinning needs in June. This is currently the best output, for wholesale operations that are vertically integrated, to use.

Although I mentioned we had no particular objective here other than to just evaluate the VM platform, but in the back of my mind I wanted  to compare VM vs. Outfield maps. (I already reported, similar to this, on Outfield here: https://jmcextman.blogspot.com/2024/11/fun-with-outfield-2024.html) Have not fully followed-through with that, might be the subject of another post. But I will say it’s kind of like comparing apples and oranges. Also, I need to compare the VM Fruitlet Thinning Fruit Set Estimate with some ground truth data we took using Malusim and/or FruitGrowth apps. Finally, I did have some help named Liam Oulette who was a CAFE summer scholar. I charged Liam for his project to do some VM ground truthing by comparing tree level VM fruit counts to actually tree counts by Liam (and me). As a side note, when two (or more) people count apples on the same tree, they never match (exactly). Sometimes, they are quite different, particularly when there are lots of apples on the tree(s). That being said, VM counts and hand counts by Liam (and me) were not exactly the same. Should we be surprised? Probably not. Like hand counts where two people differ, we saw where there were lots of apples on the tree, there was more difference in VM versus hand count, but where there were fewer (and larger apples) the VM count and hand count were pretty close. See Figure 3 for an example result. Although VM kind of suggested there might not be rationale for exactly comparing hand counts to their VM counts? Huh? Below (Figure 4) I have also included some tree pics as to what VM was seeing. So maybe you can see why they – hand counts vs. VM counts – did not agree exactly? Maybe.

OK, that is a lot. In practice, it’s not all that complicated and the potential is HUGE to use VM to better manage apple crop load. I think their technology is going to be evolutionary, and they are at the beginning. Consider yourself an early adopter if you choose to engage with them, and their technology is probably best suited to larger, wholesale apple growers. I am thinking this is kind of like when the original iPhone came out, and there was no App Store, you could only use what Steve Jobs chose to pre-install. That is where VM may be now, but you can see what might be coming?

Orchard Watch update...

 Orchard Watch – aka “How is the Orchard Feeling Today?” – is our nickname for a set of weather “stations” at the UMass Orchard in Belchertown, MA. Orchard Watch consists of:

• Two Onset Computer Corporation (https://www.onsetcomp.com/) RX 3000 “mother” stations that collect all the mote (see next) weather data as well as collecting their own: air temperature, humidity, rainfall, wetness, solar radiation, wind speed and direction, and soil temperature and moisture content.
• Seven “motes” that each measure: same, all of the aboved.

For more, Orchard Watch has a dedicated website: https://orchardwatch.wordpress.com/

Some interesting facts, in addition to the above, about Orchard Watch:

• Data from the motes is transmitted to the two mother stations by a proprietary wireless network called HOBOnet: https://www.onsetcomp.com/resources/product-overview-videos/hobonet-field-monitoring-system
• The HOBOnet “field monitoring system” at the UMass Orchard is composed of nine locations divided into “North” and “South” mother stations, North having 3 motes and South having 4 motes (in addition to an in-canopy wetness sensor). Figure 1. The motes are separated (at the maximum) by 3,000 feet horizontally and 180 feet vertically. Thus, nine micro-climate sites.
• Mother stations and motes log environmental (mostly “weather”) data every 5 minutes, from every sensor – air temperature and  humidity, wind speed/direction, rainfall, solar radiation, leaf wetness, soil temperature and moisture – and transmit the data (every 15 minutes) via cellular to an Onset website called Hobolink. On Hobolink, current conditions are displayed, charted, and the data is archived for later retrieval. Currently, this costs $700 a year ($350 for each mother station). Thanks to Massachusetts Fruit Growers's Association for paying the bill in 2024!
• How much data is there? The math: there are 104 sensors, each with a data log point (air temperature for example) every five minutes, times 12 (per hour) times 24 (per day) equals nearly 30,000 data points per day! Think 30,000 spreadsheet cells per day! Times 365 days per year equals nearly 11 million data points (spreadsheet cells). That’s a lot!

The original intent of Orchard Watch was to monitor microclimate to get a feel if it makes any difference when used in orchard pest management models – for example apple scab – that predict pest severity/incidence. In addition, there was some thought about a public-facing website “How is the Orchard Feeling Today?” Neither in particular have come to realization. There is, however, a lot of data mining that could be done. I have pulled together three minimal examples for consideration.

Temperature

How much did the temperature differ within Orchard Watch? Turns out not that much (with one caveat). I pulled the temperature data for May, 2024, charted in Figure 2. Consider, however, that I have found that on a clear, cool (cold?) night the temperature does indeed differ significantly. For example, most recently, on January 22,2025 the low temperature recorded differed by 7 degrees F. (Zero degrees in the higher elevation to -7 degrees F. in the lower elevation.) Figure 3. And not the first time that has happened: https://orchardwatch.wordpress.com/2020/10/18/cold-air-sinks-confirmed/ Why – all other things being equal – orchards should be planted on higher ground vs. lower to manage frost/freeze risk. Thus you see orchards on hills (most of the time).

Wetness (leaf presumably)

This is interesting. I buried a wetness sensor in a tree canopy so I could compare that to one of the wetness sensors out in the “open.” (Did I mention that the mothers and motes are all out in open areas, not in orchard blocks themselves.) Hypothesis being the wetness sensor in the canopy would be “wetter.” Wrong. Figure 4. Guessing the canopy shields the wetness sensor from dew (or light rainfall), thus it records less wetness vs. sensors out in the open that collect dew?

Apple scab

Did a quick comparison here on scab infection periods between the mother weather stations on NEWA – the motes are not on NEWA so do not have the capability to run the scab infection period model with those –  and also included another Onset and Rainwise station here at the UMass Orchard that are not part of Orchard Watch. For the month of May, 2024. Figure 5. Interesting, generally line up that would trigger fungicides sprays with the exceptions of May 1 and May 23-24. No idea why, probably a result of recording more wetting events/precipitation. Not convinced that overall, microclimate is going to make that big a difference in these models, however, certainly could use further investigation?

So, I got a lot of data, it is out there, free for the taking, I just need to find someone with some big data mining/analysis skills to make any of it actionable. Any takers out there? Or does anyone really care?

Figure 1 - Overview of OrchardWatch 'Hobonet' wireless network
at the UMass Orchard, Belchertown, MA

Figure 2 - Average, maximum, and minimum monthly temperature (May 2024)
at nine OrchardWatch locations

Figure 3 - Minimum air temperature (degrees F.) on 22-January, 2025
at nine Orchard Watch locations

Figure 4 - Wetness (leaf) outside vs. inside canopy, May 2024 (week 4)

Figure 5 - Apple scab infection periods (as predicted by NEWA) for May 2024 at two Orchard Watch  locations (OW-N and OW-S) and two other on-site weather stations (Belchertown and Belchertown-2)

It's cold out there...

 But not too cold? How cold is too cold? It depends😀

Arctic high pressure invaded the country this week. Dangerous wind chills from the Midwest to the East coast and a record snowtorm in New Orleans, LA where they got 10 inches of snow. (That's more than we have had here all season.) Snow from Houston to the Florida panhandle. Meanwhile, here in New York and New England it was just plain cold. Nowhere near record, but flirting with -10 degrees F. where I might expect to see some damage to tender fruit (peaches) buds. How cold did it get across Massachusetts NEWA weather stations which are mostly on-farm? -17 to 16 degrees F., the former (Richmond) being in the Berkshires, the latter (Hyannis) down in the Cape Cod banana belt. See what I have to deal with?

Now, microclimate within site I know by fact had something to do with these temperature variations. For example, I have a set (too many, OrchardWatch) of weather stations at the UMass Orchard in Belchertown, MA and the low temperature last night (this morning, Janaury 22, 2025) there varied from zero to -7 degrees, high to low elevation, with about a 200 foot vertical height difference (maybe a couple thousand feet horizontal difference). What do they say about real estate? Location, location, location?

It has been cold up to this point with no significant warm-up, so in most, if not all locations, we will have a peach crop. But check back with me in May...

IFTASAfrica2024

 Three nights in Cape Town, six nights in Stellenbosch, three nights in Makhanda (eastern Cape). Plus two nights flying between Newark and Cape Town = two weeks gone on my South African venture. Thanks to an International Fruit Tree Association Study Tour plus optional Safari all capably honchoed by Onward Travel. I'll try to highlight some take-homes from each South African orchard we visited and a bit about my Safari -- it was wonderful -- at the end. But first, some general observations about South Africa:

• It seemed big, with lots of jutting rocks (mountains). Rugged terrain, largely uninhabited. Mountain ranges traversed by hairpin turns and a long(ish) tunnel.
• The disparity in income was evidenced by homes surrounded by razor wire and electric fence interspersed by metal "shanty" towns. Unemployment is 40%. One must be watchful when out and about in Cape Town, however, the University town Stellenbosch seemed relatively "safe."
• Tourism has become the economy of choice along the "Garden Route" of the southeastern South African coast, although it extended towards western Cape Town with many wineries and other tourist attractions. Think whale watching, zip lines, foodie-oriented venues, horseback riding, fat-tire bike dune riding, and of course Safaris.
• In general, the service was excellent, I suspect people covet their jobs, and the USDollar was strong compared to the South African Rand so most everything seemed like a good deal.
• South African fruit production -- despite it all being exported at relatively low margins -- was quite advanced in infrastructure and attention to quality and sustainability. At least that is what we were shown.

OK, enough of that, on to the orchard tours which I will break down by day.

Days 2 and 3 (Day 1 was arrival in Cape Town) - Well, not an orchard tour but a couple days in Cape Town and vicinity after a long day of travel for many. Including an aerial cable car ride up Table Mountain, a visit to the Company's Garden, and a guided tour of the Cape Peninsula, including Cape of Good Hope and penguins!

Cape Town from Table Mountain

The oldest cultivated pear tree in South Africa at Company's Garden, Cape Town

Day 4 - Le Roux Group, Sandrivier Estate, Wellington. Plums. Covered production to keep fruit unblemished (wind and sunburn), and a brand new packhouse featuring tate-of-the-art post-harvest cooling and a brand new packing line (although it was not running at the time). Commitment to self-energy production and sustainability throughout the process. Pretty much a WOW! introduction to South Africa fruit growing. Even though I am not really into plums. Because they are hard to grow here! OK, it's time to say there are more pics in my IFTA SAfrica 2024 Flick photo album.

Espalier? plums at Le Roux Group, Sandrivier Estate

Day 5 - Dutoit Agri, outside of Ceres (kind of an economic hub). Three hours out of Stellenbosh, including those mountain ranges/hairpin turns/long tunnel. Felt rather remote. Big country. Apples. More apples. Cherries. All pretty much covered production to prevent sunburn and fruit blemishes. Fruit surface needs to be nearly perfect for export. Covers included drape-net style, some not fully deployed (yet) and whole orchard covers. Before I forget, somewhat depending on soil types, which were quite variable, a lot of berming up orchard rows. Cherries were quite lovely, some approaching harvest. These guys were pretty much on top of all things over their 2,500 acres of production. Here we learned M7 was/is the common apple rootstock, also MM109. Yea, kind of semi-vigorous so growth control -- ringing, root pruning -- was a topic of discussion. Some interest in Geneva rootstocks, but this is arguably "ranching" and the semi-vigorous rootstocks are still working for them.

Willie Kotze all about apples at Dutoit Agri. Note covers to prevent sunburn, defects.

Nice, young sweet cherry orchard at Dutoit. Once, again, note covered production.

Day 6 - a half-day visit to Chiltern Farms in the Vyeboom Valley. Some interesting stuff going on here, mostly looked at apples, ringing (to control growth), whole tree branch grafting to convert a vigorous apple orchard to another producing variety in one year, innovative way to ameliorate blind wood issues (see Flickr album), and BigBucks Gala 💰 LOL. (BigBucks is marketed as Flash GalaTM). CEO/managing director interestingly said, despite labor being cheap, it's still an economic issue with goal to reduce by 2% year-to-year. More dwarfing rootstocks, M9, might be the answer to that with increased efficiency. Did I mention there is absolutely no fire blight in South Africa. Huh?

'Bigbucks' Gala apple trees at Chiltern farms

Day 8 (Day 7 was a day trip to Hermanus where I went sea kaying) - Two-A-Day (TAD) Group, Oewerzicht Farm near Greyton. In a river valley, and a bit of an escape "cottage-resort" in addition to the orchard. Again, a lot going on here, including rootstock trials sponsored by Provar, and their own 'TAD' research orchard. Interesting to note a big flood several years back had standing water in the orchard for a month but the trees seemed not too weary for the wear. This is a hot growing area, and lack of winter chill is an issue that is getting worse. Not sure where it was said, but climate change (global warming) is one of their biggest growing concerns in South Africa.

'Drapenet'-like covered trees at TAD, however, I believe it was Chinese-made netting

Geneva 757 rootstock being trialed at TAD experimental orchard

Day 9 - Oudenwagendrift Farm in the Nuy Valley. Yea, stone fruit instead of more apples. Once again, whole orchard covers to keep wind down, and most nectarines being grown to an in-row 'V' vs. across-row 'Perpendicular-V'. Might have better light distrubition in the former vs. latter? I guess my take-home is we should just give up growing stone fruit here in the Northeast as this was a pretty great place to grow the trees and fruit. Or maybe we should do all our stone fruit under covers of some way-shape-form? Now there is an idea! This half-day morning orchard tour was capped off with lunch at Nuy in the Hill restaurant and a stone-fruit tasting hosted by Provar, independent cultivar evaluation, using Culteva.

Bearing nectarine orchard at Oudenwagendrift Farm

Peach/nectarine cultivar evaluation using Provar 'Culteva' app

After five days of orchard tours, some of them kind of hot and sunny, and interspersed of course with some good food and sight-seeing, I ended up spending three nights at Woodbury Tented Camp near Makhandra in the Eastern Cape. Can't speak highly enough of it, including twice daily safari excursions, seeing most of the 'Big 5' and great food and service. Can't recommend it enough.

White rhinos at Amakhala Game Preserve/Woodbury Tented Camp

Post-safari evening dinner at Woodbury Tented Camp with new friends from Stemilt (Lyndsey and Alex) and Okaw Valley Orchard (Jennifer and master donut maker Mike)

Thanks of course to IFTA, Education Director Greg Lang, and the Onward couple Molly and Kat for putting this all together and pulling it off logistic-wise w/o any major incidents in what is quite the far-away foreign country. Well at least English was generally spoken. Should be on everyone's bucket list as long as you are in for the long plane ride!

More pics: IFTA SAfrica Flickr photo album

IFTA SAfrica group photo, Chiltern Farms, Day 6, December 7, 2024. Photo courtesy TJ Mullinax, Good Fruit Grower

'Fun' with Outfield 2024

 My second year working with Oli Hilbourne and Jim McDougall from Outfield using my DJI Air 2S UAV (drone) to do blossom density, fruit variability, and – although not officially sanctioned – yield estimate ‘heat’ maps. (See my 2023 Outfield post.) My intent here is to just highlight some of my observations using this particular PACMAN (Precision Apple Cropload MANagement) tool in 2024. Although, I did have a specific applied research objective too, that is/was to compare the Outfield product to Vivid-Machines results, who I also worked with in 2024, but that will be a separate post.

So, regarding Outfield:

• I/we actually mapped a total of thirteen different apple blocks across five orchards in three different states (MA, NJ, and NY). Wow. Not all orchard blocks were mapped for all three Outfield products – blossom density, fruit variability, and yield estimates – but still, wow. Going forward, however, I am not going to bring up yield estimates because Outfield is not ready to roll this feature out yet. It’s kind of in Beta testing still, TBD?
• Two orchards in Massachusetts, the UMass Orchard in Belchertown and Tougas Family Farm in Northboro were included in flyovers and mapping. I also had the opportunity to demonstrate Outfield at a meeting of the New Jersey State Horticultural Society at Wightman’s Orchard in Morristown, NJ in June. See this Fruit Notes article. Plus in August did a demo at Crist Brothers Orchards in Walden, NY and Once Munsee Orchard in Wallkill, NY.
• When I say we, I mean I had help, Liam Ouelette who was a UMass CAFE summer scholar I was lucky to have help me out. Brain works much faster in young folks (Liam) than old folks (me), and he was a natural at flying the drone (Figure 1). As part of his summer scholar work, Liam produced a nice poster (Figure 2).
• In general, using the drone along with the Outfield Dashboard and Litchi app was pretty easy. File management got a bit tricky, however,  as the number of blocks we flew over increased. Flights typically averaged five minutes or so to complete, but it depended a bit on block size, which ranged from one-half an acre (< 3 minutes) to approximately nine acres (10+ minutes). One notable ‘incident’ occurred when Jon did not check a particular block perimeter closely, so the drone and an oak tree had a ‘meeting’ which resulted in the drone getting stuck about 30 feet up in the tree canopy. Fortunately, someone with an adept throwing arm was able to knock it out of the tree using – what else? – apples as the dislodging projectile. The drone was undamaged, amazing.
• Uploading picture files taken by the drone during the largely hands-off, programmed block flyovers resulted – within 24 hours usually – the Outfield ‘heat’ maps in their Dashboard. Examples of bloom density and fruit variability maps, as well as sample region/image locations are pictured in Figures 3 - 6. They are pretty self-explanatory.

OK, the “so what?”

• As I alluded, I was hoping to eventually compare the Outfield ‘heat’ maps to those produced by Vivid-Machines. I will have to dig into that more extensively, but it’s kind of like comparing apples to oranges so I will have to re-think that.
• Anyone can do this, i.e., buy a $1,000 drone, get your FAA Part 107 license (preferably), pay Outfield for their Dashboard and flight plans (rack rate circa $100 an acre in 2024), outline your orchard blocks of interest, and fly and make maps until your heart's content. Typically during full bloom (blossom density map), then once the fruits reach one inch diameter in mid-late June (fruit variability map), again in July, again in August (fruit variability maps).
• What’s actionable? Ideally variable rate sprayer to apply bloom and fruitlet thinners, or maybe you can just have the map in your tractor and speed up/slow down? The whole variable rate sprayer thing is coming, trust me. Hand thinning, give a map to the crew supervisor to make sure time is well spent, hand thinning where most needed? Pre-plan harvest operations, but really need a yield estimate from Outfield for that, hint 🙂. What about variable rate ReTain spraying? What am I missing? Please leave a comment!

The Fruitlet Size Distribution (FSD) Model, my 2024 experience…

As an introduction – why reinvent the wheel? – per LAURA HILLMANN AND TODD EINHORN, Michigan State University:

“Fruit set prediction models aim to produce timely estimates of fruitlet abscission after thinner applications to guide precision crop load management. The time to generate a prediction after an application is important to facilitate grower decisions to re-apply thinners while they are still efficacious, avoiding expensive hand thinning operations. The fruitlet growth rate (FGR) model, developed by Dr. Duane Greene, is a powerful tool that can accurately predict the percentage of fruitlets that will set in an orchard. Although an Excel data template and App are available to run the FGR model via computer and smartphone, respectively, adoption has been limited by the measurement-intensive procedure. A new approach, termed the ‘Fruitlet Size Distribution (FSD) Model’, described herein, was developed to produce predictions of apple fruit set comparable to the FGR model but achievable with less time investment. The principle underlying both models is the same: the relative growth rate or size of a fruitlet is compared to the most rapidly growing or largest fruitlet within the sample date to determine if it will abscise. Most predictions can be made within 8 days from thinner applications, though the duration of time depends on climatic, biological and horticultural factors. To optimize the FSD model we suggest beginning the model three days after the average fruitlet diameter of the orchard is 6 mm. Thus, the model partners well with thinning applications between bloom and 6 mm. For example, if a prediction can be achieved by 8 days, assuming an average growth of ~0.8 mm per day, then fruitlets will be ~ 12 mm if another application is needed; 12 mm fruitlets are very sensitive to many thinning chemistries.”

https://pacman.extension.org/2024/03/28/the-fruitlet-size-distribution-fsd-model-a-how-to-guide-2024-update/ Including a complete HOW-TO guide.

Let’s see if I followed the instructions?

✔ At the UMass Orchard in Belchertown, MA, in 2024 I tried the FSD model in two tall-spindle apple blocks, one Ambrosia, the other Cripps Pink ‘Maslin’ cv. Both were on G.41 rootstock. I flagged 8 representative trees in each block at bloom. 

✔ Marked five representative flowering spurs (clusters) per tree for a total of 40 clusters per block.

✔ Oh yea, I counted the total number of flower clusters on five of the marked trees, which ranged from 50 to 80 in Ambrosia, 90 to 170 (a lot) in Cripps Pink. 

✔ Over the course of three “sample” dates – May 22, 26, and 30 – in each block I collected – by either snipping, or pulling off, fruitlet stem included, see Figure ? – the fruitlets from each cluster, brought them inside, and weighed them, sitting down, nice. More on that in a minute.

✔ Data went right into the FSD Excel spreadsheet – can download from above link, have to enable Macros, it’s easier to do on a Mac than a PC, but that is another story  – and Voila! %FRUIT SET predicted in a nice chart! Figure 1 and 2.

Figure 1 - FSD prediction results for Cripps Pink

Figure 2 - FSD prediction results for Ambrosia

Humph. In terms of so “what good is it?,” if assuming a good bloom, I want the fruit set to be in the 10% range, equivalent to one apple for every two clusters? Right? That is up to you to set that target fruit set. From what I could see, we were on the right track, but I could also see for Ambrosia at least (Figure 2) it might be wise to put another chemical thinner spray on? Not sure if that happened, both blocks had a good (too many?) apples at the end. Yea, my follow through could use help. 🙂

Now for the minute I mentioned above, I was kind of impressed, particularly no measuring fruitlets in the orchard in the hot sun and bugs, just collect, and bring them into the comfort of air conditioning (no bugs if you keep the doors closed) where you individually (sitting down) throw them on an accurate scale (down to 0.01 grams accuracy), record, and hit ‘Start Calculations’ in the XLS FSD spreadsheet. Now, you can do this manually, but I had some money so I bought an Ohaus 'Scout' (STX222 AM) scale, and the USB interface, about $550. After some serious digging around in the PC (no Mac here, sorry) system and application settings, I got it to work! I had a little help from Laura in Einhorn’s MSU lab too. Put the fruitlet on scale, hit the scale enter button, and automatically record it in the spreadsheet. Figures 3 & 4. How cool! I was pretty psyched!

Figure 3 - Collected fruitlets (note with stems), scale, USB interface, and computer for fast weighing and recording of fruitlets

Figure 4 - Close up of OHAUS scale and Excel spreadsheet data entry

“Less time investment?” Yea, maybe, between collecting the fruitlets and weighing them, probably took 30 minutes. But I could do it by myself, two people not needed as with measuring and recording, so in terms of people hours, yes, definitely less. Did I mention the personal discomfort index was less too! That’s important to me… 🙂

So, I proposed MARKUSIM previously and now FSD to achieve one level – precision pruning, precision chemical thinning actually USING the FSD model, and hand thinning being other levels – of precision apple crop load management (PACMAN) for mere mortals! Me, not being one to leave it alone of course, and have to try it myself, also worked with OUTFIELD and VIVID-MACHINES in 2024, that’s upcoming. If you are thinking I will tie it all together at the end, dream on…I will leave it up to you to decide what you can do in terms of precision apple cropload management, one size does not fit all.

MARKUSIM - Precision chemical apple fruit thinning for the rest of us?

Chemical thinning of apples is one of the most important sprays growers make all season. Final fruit set affects profitability greatly. Thus, being able to predict fruit set while making chemical thinning sprays is an important decision-making step. Although there are precision tools, including the apple fruitlet growth rate model, to help here, they are admittedly time consuming and onerous, particularly in orchards with many different blocks and varieties which may respond differently to chemical thinning sprays. Here, I will propose using a relatively simple method, “MARKUSIM,” for helping growers predict apple fruit set and determine need for chemical thinning sprays across their orchard blocks.

First, I have to give credit 100% to Mark Rusell and Jill Mackenzie, Two of Clubs Orchard, LLC, Appleton NY for “MARKUSIM” – get it, a play on “Malusim” – to come up with this method for predicting apple fruit set. I have, however, as described here, abbreviated their procedure somewhat. I have only done this one year and really need to do it for another year or two to get a better feel for its “accuracy.” My initial impression, however, is that it forces you at the very least to look at a defined sample of fruitlet clusters and give you great intuition quickly as to what is going on and how the chemical thinning sprays are working (or not).

The PROCEDURE, in as simple terms as I can describe it:

1. FLAG (with flagging tape, don't use white, D’oh!) ten representative trees per block/variety at the pink to bloom bud stage.
2. Attach numbered CLOTHESPINS to five flowering/fruiting clusters per tree at the pink to bloom bud stage. Colored (fluorescent orange spray painted) and numbered (Sharpie permanent marker) clothespins are easier to find. Or you can use fluorescent orange flagging tape alone or in conjunction with the clothespins. Bottom line is the clusters need to be easy to find over the next few weeks. And they should be in the TOPS of trees too! Figure 1 for example.
3. Beginning when FRUITLETS exceed 5 to 6 mm in size, go out and RECORD the NUMBER of apple fruitlets you THINK are growing and persisting. At first it will appear most are doing this. No need to measure! Just count the number of fruitlets that appear to be staying! RECORD! I used a simple Gsheet as in Figure 2.
4. After the first CHEMICAL thinning spray, 3 or 4 days later, again RECORD the NUMBER of fruitlets that appear to be GROWING. It will start to become quite obvious which are which, GROWING or NOT GROWING and likely to abscise. RECORD!
5. REPEAT two or three or four times to PREDICT fruit set and determine need (or not) for more thinning sprays. TYPICALLY, you want one apple remaining per two fruit clusters = 10% fruit set assuming bloom is good (no frost, etc.) and you have pruned trees to roughly a desired number of flowering/fruit spurs (precision pruning). 

OK, I did this (roughly) at the UMass Orchard in Belchertown in 2024. In a Honeycrisp and Macoun block, both grown to tall-spindle. I had some interesting results, mostly on account of the fact MARKUSIM was predicting a very low fruit set, far below 10% by the final date I counted. In fact, more like 2% for both Honeycrisp and McIntosh, as can be seen in Figure 2 for Honeycrisp. Also note Figure 3, depicting a final fruit set predicted to be less than 0.5 apples per cluster which would be 10% set. Needless to say, I was concerned. There was a frost/freeze event here in late April when trees were late tight cluster to pink bud stage, so I kind of figured that might be an issue. I don’t have chemical thinning details, but the presumption was petal fall and approximately 10 millimeter chemical thinning sprays went on. I did count the final fruit set (number of apples on tree) just pre-harvest, and in reality, fruit set was approximately 28 apples per tree in the Honeycrisp, assuming my target was about 60 apples per tree, that is 5% fruit set (assuming about 120 flower clustes per tree at bloom). Whoaa, what went slightly wrong with MARKUSIM? Not sure? Maybe I did not clothespin enough spurs in the tops of trees? (Note to self, ladder needed next year.) Maybe I picked errant behaving spurs? Maybe I was not a good judge of what appeared to be growing (or not) fruitlets? See Figures 4-6 for examples. I will say, however, and this may be fodder for another Fruit Notes article, but the other predicting fruit set tool I used (Malusim) were more on-target compared to what I ended up with MARKUSIM? Although Malusim predicted pretty low fruit set on Macoun. Anecdotally, here at the Orchard, we all kind of felt/agreed that the final fruit set at harvest was much better than expected in late May where we kind of thought excess thinning might have occurred. Crops in bottom of trees were definitely affected by the frost, whereas tops were still over-cropped. Damn weather!

So, I did tell you MARKUSIM needs a couple more seasons of evaluation before I can wholeheartedly recommend, but it was kind of fun and a heck of a lot easier than measuring fruitlets. Maybe it’s just “semi-precision?” Which is better than “no precision,” aka seat-of -the-pants? You decide.

Figure 1 - Example spur marked with numbered clothespin. Darn, thought I had a better pic but you get the idea!

Figure 2 - Gsheet used to record MARKUSIM data. Rows 2-51 are number of apples on each cluster determined to be growing (vs. starting potential of 5 apples). Sum is a calculation; % set is a calculation based on sum and potential (250 = 50 clusters times 5 apples per cluster); avg. no. apples is a calculation.

Figure 3 - Honeycrisp average number of apples per cluster. By 6/2 it should be about 0.5 apples per cluster, not still the nose bleed dive seen here.

Figure 4 - How many fruitlets growing/staying here? I'd say 4?

Figure 5 - How many fruitlets growing/staying here? 6? 5? 3? It's not always clear cut is it?

Figure 6 - How many fruitlets growing/staying here? Easy-peasy, ONE!