How to move a large Japanese Maple tree

So the story is that we're remodelling our house, and there was a 20 year old Japanese Maple that was thriving next to where a deck used to be,  in the same location where a new foundation and addition needed to be, so we thought it would be great if only we could move the tree over about 15 feet.  We first called several tree moving companies, which quoted in the range of $6,000-$15,000, but none of them could say whether the tree would survive or not.  After we regained our breath, Pam posted a notice to give away the tree in the Master Gardener community, but the respondents, like us, lost interest when they found out how much it would cost to move and transplant the tree.  As a last resort, Pam posted a notice making the cut wood available, and several woodworkers responded with great interest in using the wood to make furniture, but it was killing us to chop down such a beautiful tree.  So we decided we should make an effort to move it ourselves.  I'm closely related to your local village idiot - merely believing a thing can be done is enough to convince me that I should do it, and then once the idea is set, there is no alternative but to keep trying and scheming until something works (or the tree dies - the jury's still out on that one ;-).  Here is our saga:

After consulting with a few arborists, we were told to:
  1) only move the tree during winter dormancy
  2) preserve as wide a diameter of the root ball as possible
  3) make clean cuts of the roots
  4) Water regularly after moving the tree to compensate for root injury
  5) Let the tree decide which limbs to kill off in spring, and prune those after it is clear which ones the tree intends to lose

The theory we started with was that we could dig a perimeter trench around the tree, then dig a slot in the direction where we wanted to move the tree, part the dirt under the tree with a cable, and drag the tree to its new location.  Here is Pam and a worker with the tree to be moved, standing in the freshly dug slot.

The next idea was to part the rootball from the ground below.  This seemed to be a good method at the time, as the adobe clay is like modeling clay when wet  The scheme half worked - the cable attached to the rented winch cut through the 12' diameter clay below the root ball, like a wire through cheese, until it hung up on a rock or root deep below, and became permanently embedded in the muck.  I cut the cable to free the (rental) winch, and installed a new hook and we pressed on.. 

Once the root ball was (mostly) parted from the ground, a winch to the upper branches finished the parting.  In a grand bit of wishful thinking, we had soaked the ground around the tree in the hopes that this would make it easier to drag the root ball.  This mostly created a water obstacle to jump through or over.  The 750lb pull-rated  rental winch didn't have enough power to move the tree, so it was time to return it to the rental yard.  The first lesson - get your own winch instead of a $50 per day rental, since you're going to be spending quite a bit of time either freeing yourself from the last scheme, or coming up with the next one ;-)

Here is a view of the back of the root ball separated from the dirt below.

The new winch setup.  This is a 12V Warn M6000 with a snatch block (doubler pulley) to generate 12,000 lbs of pulling force.  We bought this one used for $150 on Craigslist, and they are usually bolted to a bracket on 4WD vehicles, and sold as self-rescue winches.  We bolted this one to an angle iron with brackets leading to a clovis pin for attachment at the rear.  The tree is to the right, and the winch is connected to the base of the small tree trunk, which is in turn tied to a series of stakes (see below).  Note that the pulling capacity of a winch is rated using  the innermost layer of the capstan, so the winch was periodically unwound and rewound to wrap the cable on the smallest diameter layer for maximum pulling strength.

The winch is at the top, with the anchoring tree stump in the middle, attached to the base of 3 stakes (white ropes), the tops of which are tied to the bases of 3 stakes below (blue ropes).  There was a 6" deflection of the tree anchor point during the pull, but everything stayed together during the move, amazing considering the 12,000 lb winch stalled several times, indicating it was at maximum pulling force (DC motors generate maximum torque at 0 RPM).

This post was stabilized with an A-frame and centered below the line of action of the cable thereby creating an upward pulling component on the tree.

View of the top of the post with the cable cutting through.  A crosswise 4x6 was placed on top so we wouldn't split the post.

The tree is being dragged through the slot using the 12K lb winch setup above.  The rate of movement was about 6-12" per minute

12 feet of dragging later, the tree is in its final resting place (so to speak).

Then pulled (mostly) upright

The final resting place (with rear deck roof framed).  If the tree dies, it's actual final resting place will be with the woodworkers who wanted it in the first place, after we amortize our effort over 2 years of Christmas decorations.

And viewed from the rear.  For reference, the tree had been located directly in front of the rear door, where the deck roof and scaffolding are now.

A few months later in mid-April - it survived!

And here is the Japanese Maple 9 years later in 2015 (in the foreground above the beanpole posts, the pointy darker tree in the background and extending above the Japanese Maple is a redwood from the other side of the house).  Curiously, some previously shaded areas of the Japanese Maple bark appeared to have gotten "sunburned" over the years in areas where the formerly shaded trunk regions now get sun after the tree rotated ~45 degrees from the move.  But overall, the tree is doing well, and the move was a success.  This is mostly a testament to the good advice we wannabe tree-movers received!