At the invitation of the Green Tree Foundation, I visited the town of Talupula in the drought-stricken Anantapur District in Andhra Pradesh. Once a dry tropical region, biotic pressures have changed the region into an arid landscape. Because of this, the Green Tree Foundation had me come in to design and implement water harvesting systems suitable for their area in an effort to assist with their regreening activities.
Having selected a site for our project, I mapped the boundaries of the site with a GPS unit, and assessed the site’s features and vegetation. Being ferrosols, the soil structure was generally pretty uniform from the surface down to as much as 8 metres deep. And being gravelly, it was not appropriate for dam construction. This meant we would stick to swales for this site. The cost of digging the swales was well within the total allotted budget for the project, so we next looked into choosing the best machine for the job.
While digging out swales with a bulldozer with a tilting blade can be a convenient way of making them, the soils were so hard that they would have created a near impossible situation for the dozer to handle. Add to that the fact that the nearest dozer would be 6 to 8 hours away, requiring a transport fee, and that they would only come out for more than 100 hours work, bulldozers were not an option anyway. We were left with the choice of a backhoe or a small excavator. The excavator had speed going for it, but availability was a problem (we would have access to it only one day a week). It also required transport to the site (increasing its cost) and was 35% more expensive than the backhoe. The backhoe looked to be the best choice by far.
Designing the water-harvesting system, I did not want the site to be plagued by undersized swales. For one thing, I wanted them to be able to hold a lot of water before excess would go over their spillways. Also, if they did not have gradual enough walls, they would be more prone to erosion. Over time, swales gradually fill in, too. With larger swales, they would be around longer.
The designer enjoying a fresh mango under a tamarind tree that saved him from heat stroke many times. Photo by Gangi Setty of the Green Tree Foundation.
As I wanted the swales to be around 1 metre from the bottom of the trench to the top of the mound, I designed the trench to be about 4 metres across (a little smaller on two of the swales) and about 4 metres across on the mound. The site was nearly devoid of vegetation, so to be on the safe side, I assumed 55% runoff, meaning a coefficient of runoff of 0.55. To determine the spacing of the swales, I used the formula
Spacing = Holding capacity per m ÷ (Runoff coefficient X Maximum rainfall in one event)
[Please note that this formula is flawed. We have a swale spacing calculator on the site to make your calculations for you.]
The volume of the swales per metre was to be around 1.7 cubic metres. The maximum rain in one large event in the area is 10 cm. From this, I calculated the approximate spacing for the spaces at 30 metres. Using the GPS, I was able to find the level for the second swale, 30m down from the top, then the 3rd level, 30 m down from there.
We needed to map out the contour sites, so we tracked down some local engineers with a dumpy level. It turns out that the engineers owed a friend of the landowner a favour, so they came out to the site for free and mapped out the contour points for 4 swales on 3 different contours. While I planned on mapping the site myself, their proficiency had them finishing the mapping in half the time it would have taken me. They put us ahead of schedule by one day.
The map of the site showing the swales in red and the level-sill spillways in yellow.
The swales were excavated without too much incident with the total excavation taking about 3 and a half days to move 600 cubic metres of earth. We didn’t have the backhoe do everything. We simply had it dig trenches 3 metres wide and 50 cm deep, placing the excavated earth on the downhill side of the trench. The rest we were leaving to be groomed by hand. At one point, we hit rock that would have taken hours for the backhoe to chip through. In such cases, it is simplest to try to go around either uphill or downhill around the rock. We opted to go uphill. Apart from that one little snag, the rest of the excavation went as quickly as one could expect considering the soil was nearly as hard as concrete.
We hired a team of ten labourers to groom the site. Because the soils were so hard, however, we had the backhoe come back and chip off the uphill edge of the trenches it dug to make the process faster. The first day of grooming was gruelling work for the work team. They had to rely on picks to be able to chip through the earth to smooth out the edges of the swale. Grooming the mound was much easier as the soil there was already broken up. In the end, I had hoped to get the mounds groomed to a more gentle slope while the workers were there, but time ran out before we could get everything as perfect as the ideal I held in mind. Still, the edges are not steep and erosion should not be a problem.
On the night before the final day of work, the heavens opened up and released a torrent on the site. Excited to see the swales in action, the landowner rushed out in the middle of the night to see them fill up with water that would otherwise have washed down the hill in an erosive flood. When I arrived on site the last day, the top swale and one of the lower swales were full of water due to the slower infiltration from their slightly higher clay contents. Already they were a home to some very happy frogs that, with the rains, had come out of hibernation.
The rains transformed the earth from a concrete-like surface to a soft and yielding one. This made the final grooming much, much easier. In order to help make the swales more durable, I had the work team put in compacted, level-sill spillways set at 90 cm from the bottom of the swales. With them in place, water can spill gently over the top in very heavy rains, greatly reducing the chance of erosion damaging the swales. The workers seemed to get a kick out of me inspecting the spillway with a site level and having them fix spots that were a few millimetres out. While it may seem fanatical, if the spillway is not dead-level, flowing water will concentrate in the lower spots. When it is concentrated, it moves faster, and when it moves faster, it has more erosive potential.
Workers compacting the soil to create a level-sill spillway to allow overflow without eroding the swale.
Since the monsoon season hit just as the project was completed, it started collecting water right away. Within three weeks of the completion of the swales, they had already captured and saved over half a million litres of water. The land owner was initially worried about the amount of land that the swales took up – land that would otherwise have been dedicated to the mango tree crop that is to go in later. But upon seeing the results of the swales in action, he knew they were the right thing to do.
I was very fortunate to have the agroforestry expertise of the Green Tree Foundation to assist in the selection of tree species from the site. The plan was to plant a windbreak crop and living fence consisting of Gliricidia sepium, Caesalpinia crista and Sapindus trifoliate. G. sepium is a fast-growing nitrogen fixer with medicinal properties. C. crista makes a good windbreak and has anti-malarial properties. Sapindus trifoliate, as the name suggests, is rich in saponins, meaning it makes a great soap. Its fruit, which resembles a date, is a valuable crop that fetches a good price on the local market. I have received word from the Green Tree Foundation that these windbreak trees have already been planted on site and are doing well. When the mango crops go in, the Green Tree Foundation will provide nitrogen-fixing support trees to assist in the growth of the mango trees.
Given the swales and the addition of the trees, I suspect that within 3-years time, springs will appear at the bottom of the hill below the site. With the site’s exposure next to the national highway and the growing notoriety of the farmer we worked with, it is hoped that our project will be replicated by others throughout the area. I have been invited back by my friends at the Green Tree Foundation to do more work in the area, and I look forward to the day when funding permits me to go there again and carry out more projects.
The work team and the designer celebrate the project’s completion.
This month, we’ll take a look at the white mulberry, Morus alba. In the Species of the Month series, we’ve looked at some truly amazing plants and fungi. I thought I would make it easy on myself by doing a “simple” tree. Well, I thought wrong. I knew some of the uses of this tree, but as it turns out Morus alba offers many benefits and carries out many different tasks.
First off, this fast-growing tree is useful for controlling erosion. It also provides shade and can act as a windbreak. The leaf litter improves the soil. White mulberry has been adapted to many climates from tropical USDA zone 11 to chilly zone 4. The tree is coppiceable and survives short-rotation coppicing very well – I have seen M. alba thrive on a 2-month coppicing cycle, which is an amazingly short rotation. Mind you, this was in the tropics. Mulberry would not last very long on such a short cycle in temperate climates.
Mulberry can be propagated through coppicing or through seeds. Coppice shoots can be cut and treated with rooting compound or the coppice stool can be covered with soil after the shoots are around 30 cm tall. Left under the soil, the buried part of the shoots will grow roots. The shoot can then be carefully dug out and cut for transplanting. When propagating from seeds, soak the seeds in cold water for 1 week before planting. Mulberries grow best in dry to well-drained soils.
The agricultural scientist knows that corn is a demanding crop, so after plowing the land and applying glyphosate herbicide on upstart weeds, he or she fertilizes the soil with a commercial synthetic fertiliser. After all, plants grow better when they have a good nutrient supply. Unfortunately for the scientist, plowing kills beneficial worms and fungi and damages soil structure, increasing erosion. Plowing also oxidizes some of the carbon in the soil, releasing it to the atmosphere as CO2. Apart from being a significant contributor to green house gas emissions, this also reduces the carbon content. Reducing the carbon content reduces the cation exchange capacity of the soil, which is the ability of the soil to transfer essential minerals to plants. In other words, it makes the soil less fertile.
The glyphosate applied might be a brand that has a surfactant which is highly toxic to amphibians. In any event, glyphosate increases the risk of fusarium, a toxic fungus. Glyphosate is also linked to acute health risks including but not limited to headaches, skin and eye irritation, nausea, numbness, increased blood pressure, and heart palpitations and longer term health risks including lesions in salivary glands, inflamed stomach linings, genetic damage in human blood cells, reduced sperm counts (in testing on rats) and abnormal sperm (in testing on rabbits), and cancer (non-Hodgkin’s lymphoma in humans and liver tumors and thyroid cancer in rats). So, let’s hope the scientist is careful and has good health coverage.
Hopefully, the fertiliser is not one that holds the nutrients in a cadmium salt as cadmium will further kill off fungi. On its own, the synthetic fertiliser will shift the soil in a bacteria-dominant direction and reduce fungal content with or without cadmium. The synthetic fertiliser also reduces the carbon content of the soil making it less hospitable to life and less fertile, reducing its capacity to retain water, and degrading soil structure thus increasing erosion.
Being clever, the scientist goes into the genetics lab and isolates the cry 1Ab gene from Bacillus thuringiensis ssp. kurstaki, the nptII gene, an intron (a non-coding section of a gene) from the heat shock protein hsp 70, the CaMV 35S promoter gene from the cauliflower mosaic virus and the NOS terminator sequence from Agrobacterium tumefaciens and sets them on the plasmid vector pV-ZMBK07. Plasmid vector pV-ZMGT10 carries the CP4 EPSPS gene from Agrobacterium tumefaciens and the gox gene from Achromobacter strain LBAA and the nptII gene. These are coated on microscopic BBs and fired into corn cells to transfer the DNA. Simple enough, right?
Climate permitting, the permaculturist might adopt the Central American technique called frijol tapado. This method involves allowing the land to fallow for 2 or 3 years until woody weeds are dominant. Grasses will be competitive with the crop, but the woodier weeds will not be. Beans and corn are scattered directly into the weeds. Then the weeds are chopped and dropped to create mulch for the crop. (This system works well enough to produce 60 to 70% of the beans grown in Costa Rica.)
The corn is planted, but to help meet nitrogen needs, it is intercropped with clover and beans to fix nitrogen. Beans have also been shown to decrease outbreaks of leafhoppers and fall armyworm when intercropped with corn. Clover and soybeans have been found to decrease losses from the European cornborer. Weeds want to be avoided, so to back up clover as a ground cover, squash is planted. The added benefit to this groundcover is the food it yields. Squash also reduces losses in corn to spider mites and aphids. Bee balm is thrown in as a beneficial attractor, encouraging predatory insects and attracting pollinators. Over-concentrating corn is avoided as this would be too attractive to pests. Dr. Jane Mt. Pleasant of Cornell University has run trials of 3-sisters plots (3 sisters being a corn, bean and squash mix) against conventionally grown monocultural corn plots and found the calories produced in the 3-sisters system were 17% higher per unit area.
Upon arriving in Talupula, I found the lateritic soils to be slightly more yielding than asphalt. The soils are ferralsols, which are an iron-rich lateritic soil that becomes hard after the land is stripped then subjected to repeated wetting and drying – just the conditions that occur in the dry tropics. When wet, they are very workable and may even bog down machinery working on it. When dry, they are are hard as pavement. In these soils, calcium, magnesium, potassium and sodium are weathered out, and there is next to no humus content in the soil. As a result, the cation exchange capacity is low, meaning that plant health suffers. While these soils do tend to lose potassium easily (another argument against the common practice there of burning pasture land as a management strategy), they do hold onto phosphorus well. They also respond well to amendments of lime and gypsum, though this was beyond the scope of our project.
The Indian government is in the process of building a large irrigation channel to divert the flow of the nearest major river to the drier regions of the south. While I am personally skeptical of the ecological viability of this project, visiting the excavation site for the channel did give me the opportunity to examine the typical soil strata of the area. Lateritic soils are deep – sometimes running down to 20 metres in depth. I could see from the channel excavation that the gravelly soil continued down at least 8 metres to the bottom of the channel. All that gravel meant that I would not be designing and constructing an earthen dam as the gravelly conditions require considerable engineering for dam construction. The focus then became on swales and possibly gabions.
Before I left Canada, a site in Talupula on public land on a small mountain outside of town was suggested. It is said that the deity Obula Swami lives at the summit of that hill. Our potential site there was the highest practical site to work on. On firsthand inspection, however, I found that the access to the site was difficult and there was very limited space to work in. There also were a number of rock walls already build on the site to combat erosion. And as it was public land, would it be subject to neglect, or destruction? A further problem was that shepherds regularly burn the land there, so establishing trees would be difficult at best.
For a few days, we were stuck without a site to work on. Then we got the approval of an organic farmer outside of town to do whatever we liked on a 7-acre hillside patch of his land. The farmer, Gangahadr, had already greatly benefited from the agroforestry advice of the Green Tree Foundation and was eager to see what we could do. The land was not too steep, and the nick point on the land (the point at which the hillside goes from convex to concave) was high enough that we could get up near the top of the hill and put in a series of swales. Being private, the land would be well cared for and access to it controlled. Gangahadr had a growing reputation in the area for excellent results, and the site was visible from the national highway, giving the project more exposure. As an added bonus, the site was adjacent to and would thus compliment a rock check dam built in 2005 by the Rural Development Trust. The effect of that dam has been to change the land downstream from desert-like conditions to a rich oasis. This was the site. I met with the Ganghadr and got his permission to build swales on his land.
Azadirachta indica, neem, the village dispensary. This amazing tree has so many uses that it’s hard to imagine a tropical garden being complete without it.
The dried leaves are used as a moth repellent to protect clothes, in grain and dried fruit stores to protect from insects, and as a general insect repellent. Fresh leaves are sometimes eaten to rid the body of parasites. Twigs from the tree are chewed on one end, then used as a tooth brush.
During the month of May, I was in the semi-arid region of Andhra Pradesh, India at the invitation of the 
Some forest is lost by people cutting trees for fuel wood but, as many species in the tropics coppice vigorously and firewood cutting tends to coppice the trees cut, this is not the major cause of deforestation there. The major causes of deforestation are land clearing for agriculture and land destroyed by grazing animals or by shepherds setting fire to the land as a management strategy. Also, grazing is communal by default. If land can be accessed by shepherds, both private and public land will be grazed.
The quickest means of repairing the land might arguably be by the controlled grazing techniques of Holistic Management®. This method would involve using one of the current degraders of the environment – grazing animals – to regenerate the land. I gave the practicalities of this approach some thought. To do just one District of the province would involve a large scale educational program to teach Holistic Management® to thousands and thousands of shepherds. It would also require each shepherd to have portable electric fencing to contain their herds in a controlled area for a controlled amount of time. Additionally, a sophisticated database, easily accessible by cell phone (the only communication tool at a shepherd’s disposal) would be needed to track grazed land and grazing schedules of different plots of land. The alternative, if Holistic Management® were to be employed, would be to have land owners fence their land into small plots and control animal access themselves. One problem here would be to convince the land owner that there would be benefit to him were he to shoulder the expense of fencing and managing the grazing on the land. The alternative is to only allow grazing on one’s own land. The problem here being that this approach would ruin the livelihoods of thousands of families, throwing them into poverty.
The approach that my associates at the Green Tree Foundation are taking is a tree-based approach. The problems are well defined: poverty from a degraded environment; the environment degraded by biotic pressures – mostly from grazing animals. The solution they are using is to try to replace the forests that once stood and to shift agriculture in the region to an agroforestry-based system.
They are surveying local villages, finding the material needs of the villagers, selecting appropriate tree species based on local conditions and local needs, and supplying trees at low or no cost to the farmers and citizens of the village. To address the problem of grazing animals specifically, they often seek out productive tree species that are non-browsable or use non-browsable species as a living fence to exclude grazing animals from a site. They are planting agave as a firebreak to protect their planted areas from the fires set by shepherds. They are also encouraging the use of fodder trees for animal feed. Penned animals cause no damage to surrounding lands and feed can be cut and carried from fodder trees to the animals.
Being a mushroom nut, it has taken all the discipline I can muster not to have a fungal species as Species of the Month yet. I can wait no longer. This month’s species is Trametes versicolor, the turkey tail or yun zhi mushroom. This saprophytic, polypore mushroom is a white rot mushroom, meaning that it breaks down lignin (the organic polymer that gives trees their strength). This mushroom is found in boreal, temperate, sub-tropical and tropical regions.
It has many uses for bioremediation. It can be grown on woodchips in burlap bags. The bags can then be stacked in runoff channels below animal paddocks to filter E. coli, Listeria, Candida and Aspergillus, protecting watersheds from contamination.
First, create a map of the site. It needn’t be anything too fancy, just something you can use for planning purposes. Freehand maps will do fine.
After laying out the paths, you will know what floor space is available to you. Now you need to consider sun exposure and wind direction. You’ll generally be placing taller plants so that they do not obstruct the sunlight of other plants. You may need to break this guideline if you need either the taller plants such as bamboo or a trellis to act as a windbreak to prevent wind damage to other plants. In arid conditions, the additional shade from taller plants or trellises can be beneficial by helping to retain soil moisture.
The installation of trellises will also help us to grow plants up walls and across ceilings. A trellis with a mesh pattern is good for plants that climb using tendrils such as grapes and bitter melon (Momordica charantia). Poles or lengths of string can be used for plants like beans or hops (Humulus lupulus) that twist around objects as they grow. In addition to natural climbers, plants such as squash and kiwis (which have a variety of species suited to climates from sub-tropical to cold temperate) can be tied to trellises to grow where you want them to. An added benefit of climbing plants is that they shade buildings in the summer, helping them to stay cooler.
First, put in a base layer of soil – either sand or potting soil – about 15 cm (6 inches) thick or more, depending on how deep the planter is. Place some of the natural soil you collected outside in the planter. Water the soil. The deeper the planter, the more you can put in. Next, add a dusting of compost on top of the soil followed by 10 cm (4 inches) of mulch and water it. Next add about 5 cm (2 inches) of compost and water it. Finally, add on about 5 cm of mulch and water it.
The mulch serves several purposes: it suppresses the growth of weeds, it helps retain moisture in the soil, it breaks down over time to feed the soil, and it creates a niche for spiders, which will help control any unwanted pests that show up. This mulch is a key element, and without it your garden will be less likely to be successful and will surely take more effort from you to maintain.
It will be helpful if you sow white clover seeds (Trifolium repens) in the mulch. Just sprinkling it on top will do. As the clover grows, it will create a weed-suppressing groundcover and it will fix atmospheric nitrogen into the soil, helping to fertilise your garden naturally. Groundcover also reduces soil moisture loss due to evaporation.
Keith: I’m Keith Johnson. I teach permaculture design with Peter Bane. I help him on