Wednesday, January 2, 2019

NC-140 rootstocks gone rogue?

2014 NC-140 Honeycrisp planting at UMass Cold Spring Orchard, 13-May, 2018
While recently working with colleagues on a rather comprehensive article (for Fruit Notes and Horticultural News, soon to be published) on data results from the 2014 NC-140 Honeycrisp planting at the UMass Cold Spring Orchard in Belchertown, MA I was shaking my head. Yea, we got lots of nice large tables with lots of numbers via measurements we took in the field, and they are all nice and statistically separated, but how does the average (non-scientist?) person (apple grower?) sort through it all? And what is the take-home message?

So, I may be going rogue here -- or to put it another way "some NC-140 results for dummies" -- but here is how I/we can look at it in a way I hope may be more useful to the average grower thinking about using any of these rootstocks. (If you really want more in-depth, be sure to visit the NC-140 website.)

First, let me point out a few details of the planting: tree spacing is app. 3 feet by 16 feet, planted in 2014, data collection started in 2014 (tree size only) and then beginning in 2015 including fruit yield. You will see what rootstocks are in the planting shortly, but Honeycrisp is the variety, and there are 10 replications of the rootstocks, i.e., 10 trees of each rootstock that are randomized down the row. This NC-140 planting objective is to look at some as-of-yet unreleased Vineland (V.) rootstocks compared to some common commercial rootstocks (M.9, M.26, and B.9) as well as including most of the recently released Geneva (G.) rootstocks.

Let's start with tree size, which of course is of inherent interest in these NC-140 apple rootstock plantings. Tree size (trunk diameter) is measured at app. 12 inch height above the graft union every year at the end of the growing season. Every tree, and then averaged across the individual rootstocks. At the end of 2018, here is what it looks like:
Trunk diameter (in inches) at end of 2018 growing season,
NC-140 Honeycrisp planting at UMass Cold Spring Orchard
OK, you can see which trees are larger -- G.890, V.6, V.5, V.7 and G.30, and they need to be planted at a lower density, let's say 6 feet apart, app. 520 trees per acre. Then tree size kind of breaks at V.1, G.969, G.214, M.26, G.935, and G.41. I figure these trees need to be planted at about 4.5 feet apart, app. 800 trees per acre. And then there are the smallest trees on G.11, M.9, and G.202. These could be planted 3 feet apart at circa 1,320 trees per acre. I want to note here that G.202 should be bigger, I think there was something wrong(?) with these trees in this planting, so I would discount here whatever you see on G.202.

Now let's look at apple yield. The following chart shows fruit yield per tree in 2018 and cumulative fruit yield per tree from 2015 to 2018. Rootstocks are ordered top to bottom by decreasing apple yield (in pounds) in 2018. Now, consider some trees are larger (G.890) and therefore are going to have more apples compared to smaller trees like G.11. Note G.969 and G.30 (and arguably G.890, but it's a bigger tree!) stand out in cumulative yield. Interesting. The rest of this story later, and you might surmise where we are headed here with this already?
Fruit yield and cumulative yield (in lbs.) per tree at end of 2018 growing season,
NC-140 Honeycrisp planting at UMass Cold Spring Orchard
But before we get there, let's look at a measure of productivity regardless of tree size, where it's a fact that generally the bigger tree, the more fruit yield, and vice-versa (smaller tree = less fruit per tree). This correction factor is called yield efficiency, a measure of fruit yield per unit of trunk area. Next up we have a chart of yield efficiency and cumulative yield efficiency. Yield efficiency here is measured by lbs. of fruit produced per square inch trunk area (per tree). Once again, from top to bottom, rootstocks are ordered by decreasing 2018 yield efficiency. In general, higher yield efficiency is better. Note how the order is more-or less reversed from the absolute per-tree yield chart above. This is because, typically the more dwarfing rootstocks produce more apples per unit of trunk area. They grow fruit, not wood, that is why we like them. G.969 again looks like a standout. But G.11, M.9, G.214, G.935, G.41, and G.30 look good too. (G.30 mostly because cum. yield efficiency looks very good.)
2018 yield efficiency and 2015-18 cumulative yield efficiency (in lb. per sq. in. trunk area),
NC-140 Honeycrisp planting at UMass Cold Spring Orchard

One more thing before we get to the gold mine. It's useful to also look at the number of apples per unit of trunk area to assess productivity. Here, because we typically view a range of 4 to 8 apples per square centimeter of trunk area as 'ideal' I am going to have to give you the results in metric units (number of apples per square centimeter of trunk area). For Honeycrisp, I like to see about 5 to 6 apples per square centimeter trunk area for optimum fruit quality and size, as well as to help prevent biennial bearing. You can see belwo what rootstocks were in this crop load range in 2018. G.969 at 10 apples per square centimeter trunk area was over-cropped and apples were notably green in skin color on these trees. Most else was good, although once we start dropping much below five apples per sqare centimeter trunk area one could argue the trees are under-cropped. (Note to self here: have not fully explored biennial bearing tendence of these rootstocks, would not be hard to do, just need to do it. Maybe an update to this blog post someday?)
Number of apples per sq. cm. trunk area at end of 2018 growing season,
NC-140 Honeycrisp planting at UMass Cold Spring Orchard

OK, now just one more thing. Really what matters is how much money you can make, i.e., what is the production going to be worth given tree spacing (trees per acre) and how much each rootstock has produced through the 5th-leaf growing season? Well, that is easy to do -- simply multiply tree density (either 520, 800, or 1,320 trees per acre per above discussion) by the cumulative yield in pounds per tree, and assign a dollar value per pound. To make this super-easy, let's use $1.00 per pound value of Honeycrisp apples across the board, certainly doable, a little high for some, quite low for some others. So what do we come up with in terms of production value per acre over the five years? Humph. See chart below. I like G.41 and G.11. A lot. But consider it's going to cost more per acre to put in a G.11 orchard because of closer tree spacing. G.969 looks real good, but I did not like the lack of color on those fruit in 2018. (I assumed 100% pack-out for this analysis, why would I want to grow anything less? G.969 would not have given me good pack-out at that crop load, fruit too green.) G.935, G.214, and G.30 deserve consideration too. I want to say year-in, year-out G.30 has looked real good when I harvested it, but it root suckers profusely. With the exception of V.1, the Vineland roostocks are simply too big and not very yield efficient. (But consider they may have other attributes.) I already mentioned G.202 seems to be an anamoly in this planting. M.9 looks good, but who wants to lose trees to fire blight? Ditto for M.26, that one needs to be buried (literally). And now you have the rest of the story...
Predicted cumulative dollar value of appled produced 2015-18 in the NC-140 Honeycrisp planting
at UMass Cold Spring Orchard