Permaculture Research: The Reality of Food Forests

A newly established swale food forest at PRI Australia,
backdropped by an 8 year old food forest. Photo: Fraser Bliss

We have heard about the wonders of permaculture food forests, whereby nature does all the work and we can simply walk around harvesting more food than we can possibly eat. Bill Mollison, the founder of the permaculture movement, is known for saying that the world is "in grave danger of falling food". This is an incredibly appealing idea that certainly has its roots deeply embedded in the human psyche that craves for a paradise lost, a Garden of Eden and the freedom from the toils of work. But is this achievable? What data supports these claims?

These questions I had in mind when I went to the Permaculture Research Institute of Australia to take Geoff Lawton’s Permaculture Design Certificate (PDC) course and 10-week internship program. At his farm, I walked through food forests that had been established a decade ago and observed their diversity and stability with my own eyes. It was apparent that these forests of food didn’t plant themselves. They were a direct result of thoughtful design and obviously someone had planted them and guided their growth over the years.

The challenge working with natural systems is that there is never one single variable that affects outcomes. In nature, there are more factors that we can identify and measure. This holds especially true when one works together with nature, not against it with force. That is why I took it upon myself, together with fellow intern Theron Beaudreau, to conduct research and measure the amount of work required to maintain and establish food forests.

We performed chop and drop maintenance on food forests of all ages and sizes from those newly planted six months ago to very stable nine year old systems. We also designed and established a new food forest measuring 550 m2, containing a great diversity of sub-tropical trees and plants.

All labour was recorded over a three month period of time. Everything was done manually with hand tools and no fossil fuels were used (except a short visit by car to a local tree nursery).


  • All food forests were located on or below swales.
  • Research occurred during spring-summer in the subtropics of northern New South Wales in 2011.
  • The experiment was for educational and research purposes. The amount of labour input could be potentially reduced and thus results are on the conservative side.
  • Some food forests had been relatively well maintained and others had hardly or never been touched since they were planted.
  • The food forests were located on different zones and parts of the property, and some had their own unique variations of microclimate.

From the analysis of the research data several conclusions can be drawn:

As a food forest ages and stabilizes, the maintenance required to maximize productive output (mostly chop and drop) decreases up to 93% over 10 years. Another way to look at this is one can cover up to 14 times more area per hour of work on food forests that are a decade old and well maintained. If this pattern is extrapolated, it can be assumed that food forests, like normal forests, will eventually reach a stage when little to no maintenance is required.

If chop and drop maintenance of a food forest is not regularly performed, this leads to a four-fold increase in the amount of future work to clean it up and optimize the productive output of food. It is therefore recommended to do a small amount of maintenance preemptively to avoid unnecessary work later and to keep growing conditional optimized. The research data aligns with the general recommendation that food forests should be chopped and dropped ideally every six months (in the subtropics, less in cooler climates).

In the first three years of a food forest’s life, the maintenance effort drops 70% and then plateaus for at least five years if not adequately maintained. Thus without regular maintenance, one is not getting maximum benefit from their food forests.

It is only slightly more effort to plant and establish a new food forest than to try to recover a young one (six months old) that has been overgrown due to a lack of maintenance and perennial cover cropping. That reinforces conclusions drawn above that it is a better use of one’s time to properly maintain existing food forests and have time to plant new ones, rather than spend all one’s time trying to recover forgotten ones. This also illustrates that an essential part of food forest establishment is to fill in the niches between trees with plant species beneficial to the system and your desired outcome. For example a sweet potato cover crop that blocks the overgrowth of grasses and weeds.

The results of this research speak for themselves, however in order to fully answer my original question of the achievability of permaculture food forests that require no work, more time and research is obviously required. It was beyond the scope of my project to measure and evaluate the productive output of food relative to labour inputs. However, my personal observations were that well maintained food forests did seem more productive than those that weren’t, which makes sense, but doesn’t exactly qualify as research data.

Does that mean that the claims of permaculturists are exaggerated? Well, it depends how you look at it. The data does suggest that stable food forest systems at their successive climax no longer require the input of humans to produce decent amounts of food. But of course, one cannot get away without strategic design, planning and a few hours at the end of a good pair of loppers in a regular session of chop and drop to tweak the system to your advantage. Yes, nature does the vast majority of the heavy lifting, but you still have to walk around and harvest it. Food won’t magically fall out of the sky into your mouth.


Fraser Bliss is a permaculture consultant, designer and teacher with a background in international business systems consulting.