Thank you Fetters. Really good post. Quite timely for us as DH has suggested putting in a garden pond. My first question was, "What about the mosquitoes?" I was intrigued by the mention of Lemon Eucalyptus. E maculata sounded somewhat familiar so I assumed we call it something else down here. I googled it and this is what I found. E maculata is known as Spotted Gum or Honey Gum (good for bees). E citriodora is known as Lemon-Scented Gum. And there is a hybrid E maculata citriodora. Recently these particular gums have been reclassified from Eucalyptus to Corymbia (useful to know in case the nurseries change the labels). I debated whether to mention the next bit. In some areas they are getting a reputation as an invasive weed. It's hard to know when such a reputation is truly warranted. (I have some birch trees, which are most definately not indigenous to my area, and I do get seedlings coming up, but after all these years my birches haven't taken over the nearby bushland. Or is that because they have no chance against the Patterson's Curse, which was once an introduced garden flower.) I suppose all I can say is do some research if you're thinking of planting one. Perhaps conditions in the areas of the US where they are grown are such that they are happy enough to grow okay but not so happy that they get out of control. And the good news is if you change your mind after you've planted one, it apparently makes really good timber. Hello OzOrganic. Sorry to hear about your landlord issue. Don't get me started on that subject. Hope you find a solution soon.

I set up a spreadsheet much like the sunnyjohn calculator, and ran some numbers for my climate, Edmonton, Alberta. I came to the conclusion that in this climate I needed to double the depth of the layers. E.g. 6-8 foot deep. This starts getting awkward, as they don't rent ditchwitches that do a narrow trench that deep. Ballpark calcs: The worst case is a 'frozen fog' weather cycle. Cloudy days, so minimum heating, coupled with temps of -20C (-30F). Given an R2 greenhouse (typical for 2 layers of poly with an inflation fan, then every square foot of surface has a 90 degree temperature differential across it. Because of the sides of a hoop house, I am going to make the approximation that there are 2 square feet of outer envelope for each square foot of ground. This is pessimistic. It's somewhere between 1.57 (pi/2) for a very long green house and 2. So each square foot of greenhouse floor has to store heat for 2 square feet of R2. With a 90 degree F temperature differential, it takes 45 BTU/hr x 2 square feet = 90 BTU per hour. Again, being pessimistic 5 day operation = 120 hours. So 90* 120 = 10,800 BTU The heat capacity of dirt is about 1/3 that of water on a volume basis. Call it 20 BTU/cuft/F If we are willing to let the greenhouse temp drop 20 degrees, then each cubic foot can store 400 BTU. That would mean 25 cubic feet of dirt per square feet of floor. Ouch! If we are willing to go 2 days, it's easier. 2/5 as much heat = 10 feet. Saving grace: If we are are going for long term heat storage, the tubes can be further apart. Sunny John estimated on the basis of storing 1 days heat, so his pipes were a foot apart. If you want 5 days heat, you can put them further apart. It starts to make sense to drill vertical holes instead. E.g. Drill a 1 foot diameter hole 20 feet deep every 5 feet. Each one gets a loop of pipe going down and up. This does make for a less efficient heat exchanger, as the air going down will transfer some heat to the air coming. Worst case, the air out will be the average of the ground temperature and the air going in. E.g. With 50 degree ground, and 80 degree air, you will get 65 degree air coming out of the ground. This ignores the latent heat factor. You might be able to get better efficiency with concentric pipes, but this makes the plumbing complicated. A cheaper way to do this may be to dig a pit 3 feet deep over the entire interior space. Put barrels in the space, filled with water, with a bit of space between them. Put in a lightly insulated floor. (R4 should do.) Barrels are about 22 inches. Put in a 2 inch space, and you have 1 barrel for 4 square feet. A barrel has 480 lbs of water. A 20 degree temperature variation is 10,000 BTU, plus whatever heat goes into the ground below the barrels. But that's for 4 square feet, so it's 2500 BTU/sqft Would do for a day and a bit of our worst case. Lets look at a different solution: Natural gas. Winter gas prices are about $6/GJ and a GJ is about a million BTU. So our one day worst case is 2000 BTU/sqft. That .002 GJ will cost us about a cent a day. How much digging are you prepared to do to save 3 bucks a year? That's worst case too. It's not always that cold. Actual greenhouse operations are worse than this. When plants are actively doing their photosynthesis thing, they will use up the CO2 in a greenhouse in about 10 minutes. So you need an air change, or another source of CO2 every 10 minutes. Air has about the same specific heat as water on a weight basis. A cubic yard of air is about 2 lbs. Your square foot of ground with a 15 foot ceiling has about a pound of air to heat. Heating from -30 to +60 takes another 90 BTU. Every 10 minutes. So the air heating load is 6 times the building heating load. However, no sunlight, no photosynthesis. And when there is sun, it's providing even in winter about 600 W/m2 which is 200 BTU/sqft/hr One greenhouse operator near here heats water in the day, pipes the CO2 into the green house to help the plants, does an air exchange every few minutes on top of that, and uses the stored hot water at night. They claim energy costs of $3.50/square foot. Conclusion: It's a cool idea. If you are off grid, it's a great idea, especially if you have sunny winters. Is it practical as a 'best solution' at current energy prices for a cold climate? I don't think so.

Autumn, we investigated in advance of building the walls for our closet, so we knew that 4 of the large units would fit (since you're building, you could do the same). As far as the exact design, we did it at the store with an associate's help. We have 5 short hanging sections, one section with all shelves above a short hanging sections, two sets of 5 drawers each, and some extra shelves where we could fit them. We actually measured what we needed in advance, although I will say I forgot to account for the [very few] long hanging items I have (a few dresses, even though I hardly wear dresses). I'm not going to modify the closet, I'll just hang those elsewhere. One thing you should know is that there is one color that you have to order (we couldn't wait for it - called something like "whitewashed oak" or "driftwood" or "weathered" I can't remember). We just got birch. We paid to have someone else put them together, and the whole thing including delivery and assembly was less than $1500.

We don't have heat strips and haven't had any trouble with the system in the 7 or 8 years we've had it. I think Mitsubishi's HyperHeat is able to extract heat from air down to -13, but how that translates into keeping a house warm probably depends on square feet, insulation, etc. I was reading somewhere else this week about units that work in even colder climates (someone on Reddit asking about Chicago, I think) so it's certainly possible for a heat pump alone to keep a house warm here. I'll admit that my anxiety goes up when the temperature goes down, and I keep an eye on frost on the outdoor unit and on the indoor temp, but we haven't been uncomfortable at all. I was just googling, and it looks like heat strips are pretty easy to install; if that's the case, you could maybe add them later if this sales person turns out to be right?