Experimental Garden

I find this very interesting. I have been doing no till, mulch beds or years. I have 3, 4 x 50' beds that are doing ok. I just put them in last fall. Have amended with composted wood chips, mushroom compost, and hay twice. As I know it will take a bit for it to really take off, I've done it before. I'm contemplating digging in and doing hugels. It's a lot of work, I have the material on the property. Would you say go for it per your experiment? Lots of brush and dead trees down that I need to do something with.  Sitting on the fence post.

@Karl R.: "Subtracting the weight of tomatoes of Bed C from bed A is one test for the hugel variable as well as subtracting the lbs of tomatoes of bed B from bed D.  The no-till variable will have two trials as well. By Subtracting Bed C from Bed D and bed A from bed B both will give data that should be due only to whether or not it is no-till(mulch covered) or tilled (bare dirt)."

Looks like a nice design, Karl.  I can see where this would provide some useful information for those wanting to test out scenarios for their own operation.  One concern would be with the treatment replication within the experiment.  For example, although I can't see the whole area of study to be sure, it looks as if there is some shade creeping in at the left side of the photo.  This would put bed A under shade earlier than the other plots....maybe?  Although it may be an ugly increase in work to do so, if you were to replicate the beds at random, increasing the experiment to 12 beds so that each bed-treatment was replicated 3 times, it may reveal (or reduce) variability between readings that would be attributable to "noise", bed location, or otherwise in the experiment.  In addition (no pun intended), as noted in the clipped statement above from your post, there often are variables that are not additive....cannot be separated out by simple addition and subtraction:

"A problem with data that we actually look at is that you do not know in advance whether the effects are additive or not. Because of random error, the interaction terms are seldom exactly zero. You may be involved in a situation that is either additive or non-additive, and the first task is to decide between them."   -- https://onlinecourses.science.psu.edu/stat503/node/31

So for example the variable "mulching" might provide a plant-yield benefit, the variable "watering" might provide another yield benefit, but under certain conditions, put the two together and you may get over-saturation of the soil and see a yield penalty due to increased root rot or anoxia.  Not the best example, but something to be aware of.  Nevertheless, for bulk effects that give a good starting point for other tests and trials, a nice layout here.