Desert Food Forest and Organic Commercial Production in Three Years – Update on Wadi Rum Consultancy (Jordan)

Outside the fence

On August 6, 2010 Craig Mackintosh posted on my initial consultancy for the site I’m reporting on today. Three years and three months later, I can give an update on what has transpired since.


It is with great pleasure and excitement — which I feel for this project, the people involved and the people I know who are interested in its evolution — that I bring you all up to date on Rum Farm Organics, as it is now called.

Carrots, eggplant and chilli, melons and corn behind

First I would like to show my appreciation and thanks to Sirin Al Masri for having the desire to see this project established on the ground — the vision, insistence, and belief in my ability and the permaculture design system. This all began for me with another consultancy job for Sirin and we have since then come a long way. This may only be a small part of the larger Rum Farm, and a tiny spot on the deserts of the world, but it can do an enormous amount of good and give many people an enormous amount of hope when they realise what is possible.

Tomato and corn

The first thing the Rum Farm technical team did was to direct the machinery department to simplify this beautiful, subtle, wind-blown desert landscape and flatten the whole 5ha (10ac) site. When you are dealing with the forces of nature and natural energies there is no action without a reaction. This is typical of agriculture and anyone who wants to compromise with modern industrial agriculture too much. How much is too much could be the question you ask, and the question you should ask. Bill Mollison states in Permaculture — A Designers’ Manual, the definitive book of permaculture design: “We need to work with permitted function and not forced function”. The wind-blown patterns of the hyper-arid lands can be the destroyer of our systems if not well observed and designed to be beneficial as positive deposition systems of wind energy and nutrient capture.


In the first year I heard very little in relation to the project, as to how it was going in the initial stages of implementation. It was just after twelve months that I got a call while I was in Jordan to say the project was not working and was a failure. I was asked to explain myself and was told that I owed the company a free consultancy visit to the site.

More lettuce

On my first re-visit to the site, one year into establishment, I must admit it did not look good. The agriculture expert installed by the company said that they had not produced one single thing. I was also told it would be impossible to get these soils into production using organic techniques. It was obvious to me that many elements of my design had been ignored, left out and considered not necessary. It was also obvious that the agriculture expert, although well qualified in conventional industrial agriculture systems, had no real understanding of my emphasis on the specific inclusions and their specific positions, and why I felt so strongly that they needed to pattern the site.

Onions and carrots

My advice was to include the specific elements that had been left out — especially the density of legume tree support species, and succulent ground covers. Also, that they should keep flooding the swales on a regular basis until there was good vegetative cover and reasonable tree growth and ground covers shading the soil in the system. And, to employ one of my trained permaculture students with organic farming experience, and then and only then would the system start to work.

Eggplant shipment

They agreed but were very sceptical that it would really make any difference. Like most industrial agriculturists, they do not normally use any support species. Instead they have very academically qualified staff as advisors, big machinery, big irrigation, agricultural chemical inputs and minimum wage immigrant workers en masse.

More organic products (Abdullah at right)

My student, Abdullah Jabali, was my first choice to take the position — a farmer by heritage. Since taking his PDC in the year 2000 he has been farming organically in the Jordan Valley, and is a full organic promoter and permaculture supporter. We had also brought him to the PRI Zaytuna Farm in Australia on an intern scholarship. Luckily Abdullah was keen to take the job, despite it being a 4-hour drive each way once a week; but we negotiated a company car for the drive. Abdullah insisted on wanting to only be following my instructions if there were problems, and it was agreed that I would be an advisor at a distance.

Eggplant, carrot, chilli, cucumber, tomato, okra

It did not take long and photos started to arrive, so that I could make recommendations on how to keep the project moving forward. I could see quite clearly that there were still two major element placements which were very deficient. One was the number of legume trees as support species, and the other was the lack of dense succulent ground covers as support species. This is often the case when consulting to conventional agricultural situations — elements that do not have any obvious direct yields are often left out. This is usually because modern industrial agriculture is fixated on production per area, in a simplified form, for ease of marketing, without any secondary design approaches towards stability increase over time and reduced inputs through positive interactivity of elements linked to beneficial diversity.


This landscape of shifting wind-blown sand with a hot, extreme arid climate has its own specific challenges; production areas are usually levelled and heavily irrigated to plant large areas of monoculture. Centre pivot irrigators cover up to 200 hectares per circle using up to 18,000 litres of water per second. Our system, using swale soaks and drip irrigation, is hard to establish initially because of sand drift — without new fast growing nitrogen fixing trees. These essential support species not only provide wind buffering as well as shade and mulch while reducing evaporation and the rapid drying of soils, but also reduce the negative effects of shifting sands.

Early swale photo

Very hardy succulent ground covers also greatly reduce evaporation while having very little water demand themselves — their gel content reduces soil surface temperatures and also reducing the negative effects of shifting sands.

Early swale photo

Once these two elements are well established they also help capture fine colloidal dust in the air that drops out as quality nutrient — a crucial component of soil building in the arid zones of the world that need re-vegetating.

Swales developing

In this situation the flat bed formed swales kept collapsing after being flooded and I was being told the system I designed would not work because of this. This was a golden opportunity to point out that my design had not been followed closely enough, and to plant a fast growing legume tree in between each fruit tree, on both sides of the swales, and succulent ground covers at the base of every fruit tree, legume tree and grape vine.

Swales developing — flooding

Swales developing — full flood

This was an agreed strategy to do with my insistence and Abdullah’s dogged persistence to only want to follow my advice. Unfortunately the farm management only supplied leaucaena as a fast-growing legume tree, which is fine, but I would have liked a little more diversity. Succulent ground covers were also introduced, but only Aptenia cordifolia (common names sun jewel, sun rose or baby sun rose), which is okay, but smaller and finer as a succulent ground cover, and I had specifically asked for Carpobrotus edulis (common names ice plant, highway ice plant, pigface, hottentot fig and sour fig) on account of its edible fruit. Carpobrotus edulis is ideal for the larger functions required in dryland food forests and Aptenia cordifolia is finer and more suited to dryland garden applications. I was still pleased to get some functioning elements well positioned and as the photos continued to arrive I could see the system beginning to stabilise and kept encouraging Abdullah to keep persisting.


The crop production photos then started to arrive — squash, tomatoes, beans, hot peppers, eggplant, okra, capsicum, cucumber, carrots, potato, lettuce, broad beans, onions and parsley. Rum Farm Organics was proven — thanks to Abdullah we proved the soils would produce organic crops.



The main fertilizer was aged animal manure of chicken, sheep, goat and cow, with a mulch of alfalfa and shredded weeds. Putting extra time into composting was very difficult for the farm management to accept because of the lack of understanding of the extended benefit over time for the soil life and soil eco-system. There has always been an interest in compost and the larger farm has massive resources of compostable material that could go to be compost-extended through various processes, but it has not yet happened. We hope with the interest in the organic results that quality composting, worm farming; compost tea, bio-fertilizer, and poultry interactions will all be trialled. Even without these we have had a reasonable success which does show the resilience of our design.

Fruiting cactus fence replacement

Apricot with ground cover

As soon as we got a productive result, the farm management wanted more and faster — thinking only about production out to organic fertilizer in, and not wanting to wait for long-term productive stability. Polytunnel organic crop systems were proposed to intensify and boost production. The polytunnel systems used in dry lands work because they are sheltered from the hot drying winds which increase evaporation — a large proportion of the humidity created inside by irrigation condensates back to the soil and the light is defused and softened, even shaded in Summer when shade cloth is often added to the top. The big difference is that there are no extra bonuses in a polytunnel, compared to outside growing where you have positive outside interaction helping to build increasing stability and soil fertility. Under plastic, “what you give is what you get”.

Swale starts to stabilise

Swale getting cover

Covered swale

Shaded swale

I have seen polyculture designed polytunnel systems on permaculture projects in England, but this is not the kind of setup that was being proposed at Rum Farm Organics — rather, it was going to be monoculture, organic polytunnel production. The obvious issue with this type of system is that if monoculture is used then it is very likely to have problems with pest, diseases and fungus.

Over the canopy

What really needs to be understood from the ancient traditional production systems of pre-industrial agriculture that have remained fertile and stable for millennia, is the positive edge relationships of crop production areas. Consider the three major climates of the world as temperate with winter rain and summer dry, tropics with winter dry and summer rain and arid with yearly evaporation much higher than precipitation (which is rain plus condensation as every drop counts in the arid climates).
Traditionally, cropped areas that were sustainable were surrounded by productive forests which were beneficial for nutrient interactions towards the soils of the area being cropped, as well as creating shelter, micro climates, pest and predator balances. These crop fields changed in area, and hence edge effect, to climate and shaping to landscape profile hence erosion control and slope stability.

Crop field edged in dates and olives

Traditional gardens in the tropical climates average no more than two-acre crop clearings, 8000 square metres. Temperature climate crop areas could be a little larger, 3 acres, or 12,000 square metres, but not much more. The arid climates had the smallest crop field clearings, hence the highest edge:area ratio for beneficial effect, using partial shade in function to reduce evaporation, and were often no more than a quarter to half an acre in size, or 1000 to 2000 square metres. This means if we want to hold highly efficient productive fertility we can design and implement narrow bands of productive forest to crop fields on contour to gain the extra potential of passive water harvesting that can also recharge shallow aquifers during the very infrequent large singular rain events that occur in drylands.

Chipper and mulch

Leaucaena chop ‘n drop mulching

An interesting reference is Banded Vegetation Patterning in Arid and Semiarid Environments: Ecological Processes and Consequences for Management, Ecological Studies, edited by David J. Tongway, Christian Valentin and Josiane Seghieri. These naturally occurring bands occur through wind deposits, where wind and slope align and natural vegetation takes place in bands of trees and shrubs on contour. In the extreme hot dry lands this has often meant that a little bit more than half the total landscape needs to be covered in trees. This is a very beneficial interactive edge relationship that becomes more efficient, effective and stable over time, greatly reducing the needs for outside inputs and labour.

Mixed fruit tree rows

Crop field

Polytunnel preparation

The polytunnels went in regardless, and luckily no serious problems have occurred that cannot be addressed with natural organic methods. All the crop systems in the open areas between the swaled food forest belts keep steadily increasing in production with reduced inputs, as the soil steadily improves. Deep organic crop residue is now being added as mulch on the crops, instead of plastic mulch strips. Deep organic crop residue is also being added on top of the aged manure around the fruit trees, with a separation from the bark of the trunks as this can cause bacteria collar root especially in high temperatures. The drip lines are being moved closer to the canopy drip line of the fruit trees. The concept that tree roots constantly move through the soil away from the tree trunk is a surprise to people. I am trying to encourage denser and more diverse legume support tree plantings, Carpobrotus edulis to be planted as the main ground cover, compost en masse with compost tea and bio-fertilizer, and even the interaction of solar-powered electric net chicken tractors with self-feeding compost elements. We will see.


Chilli peppers

Hot peppers

There is also interest in extending the system, which can be done more easily now. Overall I think we have achieved a great result so far, and in an extremely difficult landscape.

We also have two years of production sales records that verify the system’s commercial viability.

Two years of production sales records

Red hot chilli peppers



Cucumber and chilli

Tomato polytunnel


Cucumber polytunnel


Tomato shipment

Mixed fruit tree rows

Crops between trees

Crop field spacing

Completely covered swale

Dates and olives in one row, mixed fruit trees in the next row

A happy farmer