David Maxwell

Absolutely!  I do not have nearly enough water.  And if I were doing it over again, I would somehow wangle a lot more water storage, (but I still want some space to grow plants...), OR ,(and I think this might have been a better choice in the first -place), a GAHT system, which is both more efficient at capturing heat rapidly, and stores 100 times more energy per cu. ft. of storage, (because of the way it works, depending on the heat of evaporation rather than simple thermal storage).

Not sure whether I am understanding correctly.  My own 16 X 12 ft greenhouse has above ground water tank storage against the (insulated) back wall, 16" deep X 28" high X 12 ft. , with plywood covers which we use for seed starting in spring. This does not store enough BTUs to maintain the air temp at anything better than 5-10 degrees C at night, with occasional dips to -2 degrees C.  (But the soil temp never fell lower than +8 C all winter.  The water circulates up to car radiators in the peak when the air temp exceeds the water temp, (during the day), and drains down when the pumps turn off with falling air temp.  (So no concerns about freezing in radiators or supply lines). In addition, the water circulates through the grow beds continuously, (separate set of pumps, running 24/7).  The temperature of the water returning from the radiators runs, on average, ~ 2 degrees C higher that the water going up. The temperature of the soil in the grow beds may run up to 2 degrees C lower than the water temperature after prolonged periods of dull weather, but did not fall below 8 degrees at any point during this past winter.  (FWIW, during the coldest period, as long as we got a day of sun, it rose to ~ 12-13 degrees), slowly falling back over the succeeding several dull days to a nadir of 8 degrees.)  The whole affair is passive solar - no supplementary heat source, (but does use 12-V power for the pumps, fans, and controllers.)

In practice, the limiting factor for growing lettuce through the winter has not been the temperatures, but the lack of light.  (Lettuce needs a minimum of 10 hrs. of light daily, the so-called Persephone Point.)  Swiss chard grew right through the winter.  (And we are not into arugola or other cold-tolerant greens.)

David Maxwell, Zone 5b, Nova Scotia, Canada

Not as simple as this.  Indeed, water is more efficient at heat storage per unit mass than rock.  BUT 1) A major limitation is getting the heat both into and out of the storage, at least in terms of air temperature. (Using the soil in the grow beds as thermal mass, and circulating heated water from the water storage, through buried pipes, (as I did, albeit not deliberately planned), provides good control of soil temperatures.  (And lettuce in particular is remarkably tolerant of air temperatures significantly below freezing - it goes limp and looks terrible until the air warms back up, when it comes right back to crisp, bright life.)  2) One needs an awful lot of water to store the heat - I read a figure something like 3 to 5 Gal of water per square foot of area, at the time I was building my own greenhouse, (but cannot put my hands on the reference today).  I was unable to provide anywhere near this, (above ground), without encroaching on growing space excessively.   3) Large volumes of water rapidly stratify heat, reducing significantly the efficiency of heat storage, unless acively stirred.  4) Much more to the point, GAHT heat storage, (Ground to Air Heat Transfer, with perforated pipes buried in coarse rock, through which air is blown) is based on an entirely different principle:  the heat in the air is stored by condensation of moisture into water, on the cool rock surfaces, and released by evaporation.  The energy stored and released by the phase change is roughly 100 times greater than raising the temperature of liquid water 1 degree.

I am in zone 5b, on the east coast of Canada. I have something like 1Gal of water/ sq ft - quite inadequate.  I have added active heat capture, by blowing the heated air in the greenhouse through  salvaged  car radiators, with small solar pumps to circulate the water, (on differential thermostat).  But when I added circulation of the water from my storage tanks through buried pipes in the grow beds, I realised an approx. doubling of heat storage.

I have been using, for the past 5 years, a multi-channel monitoring system from a company in Ontario, Rogersdini, who sell through eBay.  They sell a dongle which plugs into the USB port of any Windows machine, (can be 20 years old!).  Comes with program with instructions in Chinese, but I have been able to set it up with occasional help from seller, (very helpful and responsive).  Can go up to 64 temperature probes if you really want to go to ridiculous extremes.  (I am currently running 6 probes).  4-channel version costs $59 US (plus shipping);  increase to 8 channel is $115.  (This is exactly the same unit, but with a bunch of splitters and extra probes.)  Logs temperatures of all probes at set interval, stores lots of data, displays graphs of readings directly, but also downloads readings as comma delimited file which can be loaded directly into Excel.  (And if you go this route, I have a little Excel routine to group the readings which I am very willing to share with anybody interested.)  Basically, very cost-effective multi-channel temp monitor, seems to be reliable, (5 years so far).  Only requirement is a Windows computer, (which can be an old one no longer in use for anything else.)  And a heck of a lot cheaper than a HOBO, (although HOBOs are a lot fancier).  eBay listing for 4-channel: https://www.ebay.ca/itm/4-Channel-Temperature-Chart-Recorder-Data-Logger-analyzing-Alarm-by-Computer-PC/112652870901?hash=item1a3aa264f5:g:bZkAAOxyzGlQ8Wr~  (More channels can be located from this listing if you need/want them)

I realised that I never posted a follow-up with my experiences.  When I was posting previously I had only my two tanks of water, heated by air-to-water heat exchangers (car radiators), controlled by a differential thermostat.  Since then I added black plastic pipe buried about 8" deep in the grow beds with the water from the storage tanks circulated through them continuously.  I did not anticipate how effective this would be.  The temp in the water storage does not increase nearly as much through a sunny day as it it used to, but nor does it drop down at night the way it did.  I reckon I have effectively more than doubled my heat storage capacity.  The soil used to get down to freezing on really cold nights, (didn't seem to bother the greens...); now the soil remains consistently ~ 10 degrees C +/-. (Lowest soil temp so far this winter 8 degrees).  Air temps still drop to just under 0 C on really cold nights.  I reckon the only way to modify this is to add movable thermal curtains of some sort over the glazing.

A number of comments:
1) If you rip out your existing trees and try to plant new ones, with the apples at least, you may have problems with replant disease.  (This will stunt your new trees.  Commercial growers fumigate the soil to sterilise it to prevent this)
2) You can certainly prune your existing trees to a fruiting wall form, by lopping off all branches which intrude into the alleyways, (and if necessary, fostering the growth of new branches in the plane of the tree row)
3) Commercial apple orchards are now almost all planted with relatively dwarfing rootstocks, supported on trellises, and trained to a flat fruiting wall.  Interestingly, these are treated as "crops", with an expectation that they will last 20 years or so, at which time they will be ripped out, (the soil fumigated), and replanted.  If you have larger rootstocks, they will last far longer - 100 years for full size roots.
4) In passing, peaches: yes peach trees are relatively short-lived, much shorter than apples.
5) If you do not prune/train to a fruiting wall, a very effective understory management plan is to spread cardboard under the trees, covered by a thick mulch of wood chips, (ideally with "ramial" wood chips, branches up to 2 3/4" in diameter, (see research done at Laval University,  in the 80's or 90's)
6) Espalier, properly, involves a lot more than training to a relatively flat plane.  It involves both training to a specific geometric pattern and regular summer pruning to encourage development of fruit spurs rather than vegetative growth.  (This is not intuitively obvious, but a fruit bud has 5 leaves for each bud, while a vegetative bud has only one.  So while the tree is markedly controlled in shape and size, the photosynthesis area is maintained.)
7) If, after all this, you still want to "start afresh", you could consider sawing off the existing trees a couple of feet off the ground, and grafting in new scion wood, (with bark grafts) in the spring. (Place as many scions as will fit comfortably on each stub - that way even if some fail, you are likely to have at least one succeed on each tree.  Cut out extras the following year, when you are sure your graft is healthy and solid.)  Then carefully prune and train your new growth into whatever shape you fancy.  You could also take the opportunity to introduce new cultivars into your orchard.

From what I have read, the main heat loss in small-scale greenhouses occurs at the edges of the curtain, which needs to be sealed.  So, multiple smaller curtains suffer from the problem of multiple edges. Hence a single wide blanket is preferable..  
How best to seal the edges?  Neatest way I have seen is to fasten magnetic tape to the sides of the curtain, and build ledges along the outer walls to support the edges, also equipped wth magnetic tape.  The two magnetic tapes will automatically align themselves, avoiding the issue of the roll getting off-centre.  The long section of curtain is supported by wire cable stretched parallel with the joists, (metal clothesline has been suggested).  The designs I have seen urge the incorporation of  weights to maintain an equal tension on the wires so they don't sag as the temperature rises, but I am not convinced an 8 ft. length of wire is going to change in length that much.  Have any of you any experience here?

Just noticed the line about getting only one chance to get the graft perfect when you move from practice to actual surgery.  This is not so;  if your first cut isn't right, you just make another cut a little farther back.  As long as you still have wood, you can keep hacking it back.  What the poster may be referring to is the idea that it is best to make a single clean slice, (with a very sharp knife), so as to get a continuous smooth cut, rather than whittling the cut, (which never gives a smooth flat surface).  But all you really need is cambium-to-cambium contact over as large an area as possible, and the trees will forgive less-than-perfect technique.  If your grafts fail, it is simply that you did not get cambium in contact with cambium.
And, unless it was not apparent in my first posting, those delta grafting tools look good, but work only under ideal conditions, (when both stock and scion are almost perfectly matched in diameter).  They are damned expensive, and useful only in very restricted circumstances.  A good sharp knife is a lot cheaper, and works in all circumstances.  And if anybody wants to pursue grafting to its full depth, the "bible" is a book by R.J. Garner, called "The Grafter's Handbook"

This is pretty straightforward:  the scion wood for apple and pear has to be dormant, and the best success with field grafting will be when the rootstock has just broken dormancy.  You can cut your scion wood any time between now and spring before the scion wood breaks dormancy.  (I cut mine about March.  I am in zone 5b).

But you are going to have difficulty with your grafting tool if you are grafting onto established trees.  These gadgets work only when both scion and rootstock are virtually identical in diameter.  If you are trying to graft onto established trees you need completely different techniques - cleft grafts or what Stephen Hayes calls rind grafts, (on this side of the Pond, more usually called Bark grafts), (or, if you want to get fancier, things like oblique side grafts, inlay grafts, kerf grafts etc.).  I second the recommendation to check out Stephen's videos.  He is a bit wild, but one of the best on the web - YouTube, look for HayesUK, or just enter "grafting" and Hayes

Stone fruit indeed does best budded in June up to August, and here indeed you go from growing mother tree to rootstock tree.  But this, again, is a different technique from grafting, and is done with a single bud, cut with a sharp blade. (Stephen has excellent videos demonstrating the technique)

There are degrees of complexity in "Active".  How complex is perceived as appropriate for a "Permie" is probably variable, but I do agree that there is a bit of a bad rap in many people's minds for any form of active controls.  In practice, my system does have some technology.  The pumps to pump the water through the radiators (my "heat exchangers") are controlled by a differential thermostat, so that they run only when the temperature of the air, (in the peak), is higher than the temperature of the water.  It was apparent that the heat was coming out of the water in the tanks reasonably efficiently, as the temp in the tanks drops as much as 20 degrees C overnight.  But I have gilded this lily a little - I am pumping the water through black plastic piles buried  8" down in the growing beds, 24 hours a day, (at least when the cheap pumps are still working).    I did this on the principle that heating the plants' toes made more sense that trying to heat the entire volume of air in the greenhouse.  But it had an unexpected effect - it added the soil in the growing beds to the thermal mass.  The temperature in the tank with a functioning soil pump runs about 2 degrees lower than the other one, and the soil temperature in the the "circulated" beds, runs 2 degrees higher than the ones without circulating water.  In addition to the pumps, I have 4 large computer fans which draw the air from the peak through the radiators.  All these  pumps, fans and controllers run off a 90 Watt P-V panel with a small battery storage (actually the battery from my ride-on mower) which carries the  soil pumps through the night.
Obviously, since I am able to quote these temperatures, I also have a monitoring system capable of storing a log of data drawn from, (in my case), 4 separate channels..  (Mine currently polls the temp sensors every 10 minutes).  
And that is the extent of my "active" technology.
Does movable thermal curtain insulation count as "active"? (That is my next "refinement", after I get, and install, some new pumps, which hopefully will last a little longer.)