For the area of yellow rectangles in the image below, the slope goes from shallow (on the left) to steep (on the right). The slope faces east.
I've redesigned those beds 5 times.
Design Permutations
The first was a zigzag ditch (suggested by a permaculture designer) that was supposed to be for drainage, with little stopgaps and catchment pools on the way down. When I realized what he was designing and digging, I said, "Wait, it's for water? We don't get enough drainage here to fill it, even in the rainy season. I thought you were doing a hugel ditch."
So, we turned it into a hugel ditch and prepared beds above and perennial slopes below, but what a pain. You can't barrow-in organic material easily when you have a zigzag of raised ground.
So, I redesigned them as rows that went straight across (north to south), but with a curve, like a bow. I decided that would help catch water a little more and let it permeate deep into the soil. I gave up on the ditches and decided to berm using scrap wood. Massive pieces of scrap wood. I'd backfill uphill of the wood to make flat beds. In essence, terracing the slope using short wood retaining walls.
Then I realized the curve was pointless without swales (yup, the swales that the permaculture guy was essentially creating). Instead of focusing on water catchment (since I failed so spectacularly), I decided to focus on creating critter habitat, especially for snakes, so I thought laying the wood like a sine wave would create more "edge effect". What a pain to mow, though. And I tripped a lot.
Then I realized I should space the rows differently and have paths through the 100 feet of wood and backfill. This meant that all the wood had to move again. I also made the rows straight instead of wavy. :( What design permutation are we on? Yeah, 5.
Problem
I grew crops in the rows after that 5th design. Sadly, the super-wide beds were painful to plant in. My body is old. I only want to have to reach 1.5 feet or less.
Retaining wood higher than the soil level because I ran out of time to backfill.
Also, I hadn't had time to backfill to the point where the soil was level with the tops of the wood for the rows that were lower down on the slope (the deeper rows). And guess what I learned? The greater the difference in height between the wood and the soil, the colder the bed was in spring. In fact, there was a 2+ week difference between when the soil in the top-most row (closest to the house) reached and maintained 60 degrees F and when the soil in the lowest row reached 60. Why?
Every morning, chilly, humid air flowed down from the top and got caught and held by the exposed wood. Even if the uphill side of the bed had been raised, it probably still might've been a problem. We also have cold morning fog that moves up from the valley. I live in zone 8b, but the top most beds were more like zone 8a and the bottom was more like zone 7a.
What to do? I mean, coldframes, sure, but that's quite an expense.
New Idea
So now I'm toying with the idea of square raised beds, where the soil area is the size of a standard scrap square window or 2 rectangular windows (in case I do want to use cold frames). However, I want the squares to be rotated a quarter turn so that any cold air moving up or down the slope hits a point and flows around the raised bed. It has the added benefit of splitting and redirecting the super cold winter winds that blow north and south.
I did some calculations to compare new straight across rows of raised beds (designed 3 feet wide in order to fit scrap windows or plexiglass from Home Depot and be comfortable to reach across), with quarter-turned square beds (4 feet by 4 feet, otherwise I'd have to make a gazillion beds).
I mistakenly started with a space 100 feet by 88 feet, but I only have 100 feet by 44 feet.
Rows:
- 6 rows, 100 feet long, minus about 54 feet where paths are put through, times 3 feet wide = 1600 sqft of growing surface.
- Requires 546 feet of wood retaining walls, plus ~72 feet of wood to flank the paths = 618 feet of wood retaining walls.
- Ratio of growing surface area per foot of assembled wood walls = 2.6.
Diamond beds:
- 32 beds, 4 feet by 4 feet = 512 sqft of growing surface.
- Requires, say, 5.5 feet of wood on each of the 4 sides of the beds, times 32 beds = 704 feet of wood retaining walls.
- Ratio of growing surface area per foot of assembled wood walls = 0.72. Yikes.
I suppose I could do a chevron design in order to make better use of the space and more efficient use of the wood, but then water would drain too fast, and it would be too hard to get organic material in. Maybe. Maybe I'm wrong. Actually, the more I think about this chevron design, the more I like it, but now I realize that because of the slope, I wouldn't be able to have flat bed unless the bottoms of the chevron rows were 4 feet tall. :)
Are there other benefits to the square beds and less growing surface area?
OK, so the row idea is more efficient and will take less time to construct, but are there other reasons that make the diamond beds a better choice? Like...
- Higher percentage of woody material, and thus might attract more fungi.
- Easier traversal with a wheel barrow to dump organic matter.
- Can comfortably reach into the center from all sides.
- Plants, as they grow, often spread outwards, covering soil anyway, and I can let that happen into the paths instead of letting it take up valuable growing surface area.
- Would it be better for snakes or worse? As long as there's foliage or wood chips in the paths to hide them from birds as they cross from one bed to the next, it should be ok.
Anyhoo, I'm on the fence. I should do a combination to test things out, but I'm feeling too rushed. Wahh!!!
Update 2/5/2021:
Cons for the square/diamond beds
- Because of the 18" of wood framing the beds, I'd have to make the beds about 6' x 6', which means a reach of 3 feet to the center. I don't even want to reach 2 feet. This is a big enough con for me to throw away the square/diamond bed idea.
I'm back to the chevrons (broken into lengths that allow for making the soil level), and regular rows. Need more think time. ...
Update 7/17/2023:
Straight lines for the win
I can't even. I'm on design permutation, what 7? I have learned A LOT. But my body's broken.
I'm going to make terraces on contour such that there are 4 levels. The surface of each level will be... level. That way, I can use cloches (cold frames) if I need to, not only for weather problems, but also to keep slugs off of some things. Each level will have 2 or 3 planting rows, where the paths in between will be swales backfilled with rocks and wood chips.
I've already been trying to level out the places where a terrace on contour will create a wavy line. I just need the logs. For now, I'm not letting it stop me plant. Just harvested a potato dinner for 5 today--gorgeous, big, worm-free potatoes! Something at 2 from the same plant, but left the 4 that I grabbed untouched. Hurray for weekly dumps of beer mash, hops paste, and wood chips!
My instructors at Raven's Roots Naturalist Farm prefer to make their planting rows and paths each 2 shovel-widths wide. I'm going to start with that and adjust as needed.