• Post Reply Bookmark Topic Watch Topic
  • New Topic

Permaculture: A Designers' Manual - Chapter 10 THE HUMID TROPICS  RSS feed

Burra Maluca
Mother Tree
Posts: 10012
Location: Portugal
bee bike books duck forest garden greening the desert solar trees wofati
  • Mark post as helpful
  • send pies
  • Quote
  • Report post to moderator


10.1 Introduction
10.2 Climatic types
10.3 Tropical soils
10.4 Earthshaping in the tropics
10.5 House design
10.6 The tropical home garden
10.7 Integrated land management
10.8 Elements of a village complex in the humid tropics
10.9 Evolving a polyculture
10.10 Themes on a coconut- or palm-dominant polyculture
10.11 Pioneering
10.12 "Animal tractor" systems
10.13 Grassland and range management
10.14 Humid tropical coast stabilisation and shelterbelts
10.15 Low island and coral cay strategies
10.16 Designers' checklist
Erica Wisner
Posts: 1183
Location: Okanogan Highlands, Washington
books cat dog food preservation hugelkultur
  • Likes 2
  • Mark post as helpful
  • send pies
  • Quote
  • Report post to moderator
OK, I have been waiting patiently for this chapter to come up. A presentation about aquaculture ponds for the Amazon basin was my first introduction to permaculture, after all, in a college class that was about 15 years ago now.

I still know next to nothing about the tropics; I've spent a lifetime total of maybe 2 weeks in the Andes, and a little bit in the parts of California where palm trees don't look utterly miserable.

So I offer a spatial critique of the lovely Gangamma's Mandala garden diagram.
This is billed as a 'least path,' zero-runoff approach, maximizing accessible growing surfaces and 'edge', and stacking functions like the shower water runoff.

I can't argue with the function stacking - again, I am not familiar with tropical plants, but on paper I can see how this diagram fits with his palm-circle space-saving, mulching idea. Crowded plantings do save space, even if they make it a bit tricky to get into the crowded plantings themselves.

My question about this was, if this is such a lovely and space-saving way to garden, why isn't it more common in my experience?

The most popular bed shape I've seen is long linear beds, followed by various types of broken-contour beds, or beds arranged in artful blobs around patches of lawn, stepping stone, or big feature plants.
Less-organized layouts may have big courtyard or grove areas, sometimes rows of cash crops, and big tangled edible stuff that you can clamber around and get to when it's in season.

In gardens where I've installed keyhole beds, I am very often tempted to step over them to get to the next bed. The dead-end space only works well for weeding, and often I see the next patch of weeds from a place where I can't get to it without going out and around to the next keyhole. So I step over things - which I've also done in row crops, sure, but much lower portions of the time.
These keyhole structures are also difficult to scale for line-of-sight progressions like harvesting salad from multiple beds, sprinkling each bed with a little booster like mycelial starter spores, or any tasks where the size of the wheelbarrow-load and the size of the keyhole unit don't match. I've heard they're hard to irrigate - could see that, if you want to irrigate uniformly without specialized equipment.

The places I've found them useful were as a compromise to get into a deeper area, when the existing bed was for some reason the 'wrong' size - too big a space for one reachable garden bed, too narrow for an extra path. I'm experimenting with incorporating some buried wood for water storage, in the spurs between keyholes.

The other question I had was, how does this mandala actually compare with traditional garden layouts for the qualities it's supposed to maximize? Least path, most edge, reachable space?

It is not a least-path diagram.

It has approximately
2.5% shower
17% path
29% easily-accessible (arms-reach) garden beds
52% "beyonds" - hedges or one-touch perrennials out of arm's reach

It becomes even less so if you need to tesselate it - additional path must be added, off more than one of the keyholes, to connect with similar shapes. So you add maybe 2 paths per wheel, and your path ratio is more like 18%. Not bad, but not the minimum.

It may maximize edge space between beds and paths, but in my experience as a gardener this is some of the least-useful edge space: plants that creep into paths get trampled.
There are lots of ways to create edges within planting beds that can host more abundance. Edges defined by something solid, on which the gardener can rest and which deflects wheels or tools, help preserve the distinction between path and growth. Edges defined by stepping-stones within the garden encourage gentler access, eliminating the invitation to trundle in with heavy tread. Edges around actual differences in growing conditions, like light and shade, or water accessibility, are interesting.
The mandala seems to allocate the wettest water locations to the paths, allowing salad greens to sip equally from the raised beds. Seems like you'd be walking on salad greens soon.

This is an identical-size rectangle purposely designed for least-path, using the same features.

At the same path width, it has
2.5% shower (and more private, too!)
14% path
36% easily-accessible (arms-reach) garden beds
48% "beyonds" - hedges or one-touch perrennials out of arm's reach.

It is equally elegant, if a little boring until you put actual plants in it. The water is equally central, though perhaps not as evenly distributed; it is a zero-runoff layout with wetter and drier zones. The advantage is both perennial and fully-accessible growing beds right up against the water source, where you could concentrate shade-loving water-hog type plants under a couple of central trees. Across the paths would be moist (sides) and less wet (corner) zones. A buried pot in each corner could be added if more even watering is needed - or if water isn't an issue, and privacy is, the shower could go in a corner. The shower is already more private located in the central trees in this garden, than facing the entry of the mandala.


This is a more typical garden layout such as I might use, combining keyholes at the edges with standard rows and connecting paths toward the middle. It is designed not to minimize path, but to maximize accessible bed space, and easy access (straight paths) to various points.
The drawing doesn't show the same path width as the others, but in a previous pencil sketch I ran the numbers at:
Single unit: (solid lines)
21% path
45% accessible (arm's reach) beds
33% beyonds (border) beds

Tesselation: (dashed lines)
34% path
55% accessible (arm's reach) beds
11% beyond (border) beds

If a shower was wanted, it could be located at the truncated end of any of the beds for privacy (the highest corner of the garden would help distribute the water), or in the center of one of the main beds for easier water-harvesting. Side and central units can be any height appropriate for climate - sunken hugels, raised beds, raised hugels, etc.
The advantage of redundant paths is it does give you the option to walk around each other with a wheelbarrow, or to abandon a path when it is overgrown by a particularly lovely rampant squash.
It also gives you a lot more access to beds. If your garden is for the purpose of zone 1 harvesting, accessible beds trumps.
It is not optimized for water distribution from a central source - but much easier to irrigate with linear hoses or tapes, with a minimal number of hose-stakes to prevent mashing of corner plants.

(This is the first time I've used Facebook as a place to plunk up images for a post, so please let me know if the links break and I'll re-post from a more controlled location....
Edit: the links broke, I have re-posted the images by hosting them from our website.)

Any comments on favorite garden layouts?

Any other thoughts about humid tropics?
What are lateritic and loess soils? (Answer: Googleable... edit 2015: loess is wind-deposited soils, there is some AMAZING video documentation now online of Chinese reclamation of the Loess plateau and highlands using permaculture methods (terracing, slowing erosion, etc).)


[Edit: summary:
I thought this post was about the mandala garden not being the true least-path approach... "challenging sacred cows"
but on revisiting it to replace the diagrams, I realize it's always been more about the accessible bed spaces in the garden. The conventional row method emphasizes access, and larger rows allow access for larger equipment (hoes, scuffles, and wheelbarrows all the way up to automated combines)
The access to beds depends on paths - therefore if you have a garden where it's useful to be able to see, tend, and pick all of your plants on a regular basis, where plants are vulnerable to being stepped on or disturbed and you can't climb into beds to handle the occasional wallflower tending, etc, then more paths usually equals more access and more practical management. If you have a garden area where things don't need tending much (like dense old rhododendron patches or forest or whatever), then it doesn't need paths.
Only where there is no tending or harvesting needed are paths truly a waste of space.
If your paths are mulch or dirt, not gravel or pavement, then it makes sense to include an adequate number of paths in any garden section where you expect to use unskilled helpers, tools and wheelbarrows, or frequent tending visits. With more than one person or heavy harvest equipment, more space and paths that circle back via intersections help allow harvesters to cooperate without damaging the crops.

The mandala garden is a great way to playfully, creatively, and beautifully experiment with different patches growing different things, micro-zones based on solar aspect, or individual personal plots for students or interns. The row planting is a great way to make it easier for inexperienced helpers to easily recognize the crops, however, and to use automated or relatively indiscriminate equipment (including clumsy helpers) to speed chores like weeding and harvesting.]

Ann Torrence
Posts: 1191
Location: Torrey, UT; 6,840'/2085m; 7.5" precip; 125 frost-free days
bee books chicken duck goat trees
  • Mark post as helpful
  • send pies
  • Quote
  • Report post to moderator
I still need to write up my thoughts, but I want to say that Erica, I love how you challenge the sacred cows with data. I have long thought the mandala as a one-size-fits-all solution was wrong-headed simplification, but I never did the math. Thank you. Can't wait to hear what you have on the next chapter.
Ann Torrence
Posts: 1191
Location: Torrey, UT; 6,840'/2085m; 7.5" precip; 125 frost-free days
bee books chicken duck goat trees
  • Mark post as helpful
  • send pies
  • Quote
  • Report post to moderator
Just when I think an entire chapter applies only to a climate I left with great relief (6 long years in Houston, 6 species of household pests, running the A/C to keep the drywall from molding, hurricanes and floodwaters, never again), I find a critical nugget on p280. In fact, this is so critical, I wonder why it is buried in the tenth chapter, because I can't imagine it applies only to the humid climates. Good design reduces guild complexity as we move outward from zone 1.

"While a complex polyculture of many hundreds of species delights both the naturalist and (in food plants) the householder or villager, and the benefits to settlements are numerous, it becomes difficult to control and extensive rich polyculture and collect its {Mollison's emphasis and apparently a missing word as the next sentence begins}
Our very complex polycultures work best at small scale and with close attention from people."

Continuing on p282, he gives some typical species counts. I wonder how these numbers compare to drylands systems.
"While the fiscal return peaks at about 6-8 species in a system,the nutritional-total yield system peaks at 50-100 species, well-distributed over all seasons. These two factors (extensive-intensive, fiscal-nutritional) must be defined by ourselves for our needs, and will have a profound effect on design. What we may arrive at is a sensible zonation of species richness close in, and a concentration of less species of high value as we extend the system."

Zone 1 and 2 then are our test-beds and horticultural playgrounds:
"It is in the garden, however, that we may learn the value of such successful extension without sacrificing large amounts of energy and capital. Thus, our gardens are trial areas for the outer zones."

In planning our new and distant Zone 4 system, I have been worrying a great deal about how to manage to up the species count. the answer is that the standard is not the one I hope to achieve in our Zone 1/2 system, which is 3,000 square feet and not 3 acres. Do the best I can. It's already going to be (# of new species * 100%) better than the monocropped alfalfa that was there for 40, 50, 60 years.

I'd like to hear from some other members who live in hurricane zones about how they are safely incorporating trees into their systems, shading their houses but not getting crushed by them. I lived in a house with three 80 year old oaks that were terrifying in storms.

Were I planning systems for the tropics, more attention need be given to pests, especially food storage. I had geckos in my kitchen, they didn't help one bit when the roaches outweighed them. Gross.
And I hate coconut. Nope, the tropics are not for me. I'm still glad I read the chapter.
Janet Dowell
Posts: 32
Location: Kennewick, WA
  • Mark post as helpful
  • send pies
  • Quote
  • Report post to moderator
Ann Torrence wrote: I had geckos in my kitchen, they didn't help one bit when the roaches outweighed them. Gross. ....
Nope, the tropics are not for me. I'm still glad I read the chapter.

You made me laugh. I lived in San Diego for awhile, 20 years ago. The roaches there were.....quite healthy as well. Moved back north after 14 months & never looked back.

Tropics are not for me, either. And I'm still glad I read the chapter, too.
Consider Paul's rocket stove mass heater.
  • Post Reply Bookmark Topic Watch Topic
  • New Topic
Boost this thread!