Wednesday, September 26, 2012

The Desert Sponge

Today I made a  mini swale near one of my newly planted Screwbean Mesquite trees that hasn't been doing so well. I've been observing it all summer long and suspect that being at the lowest point on the property means that it's in deep sand that just drains water away. I read about mini swales for newly planted trees in Rainwater Harvesting for Drylands and Beyond  by Brad Lancaster. Basically you make a big sponge on contour (level with the topography of the land) on the up hill side of and next to the tree well. Theoretically, this will catch and hold any water run off from the land. It's possible that once this sponge is thoroughly soaked, the tree will need no more watering for many months. Here's how I went about it.

Using an A frame with a level I determined a contour line on the uphill side of the tree (flags mark the contour). Normally you would center the swale with the tree well but in this case I kicked the swale over to the left so it could catch an existing water run off course during a storm. I then marked out the dimensions of the swale (about 3' X 2') and saturated the ground with my water pic.

I dug out the swale to a depth of about 2 feet, piling the dirt on the downslope side. I then graded the uphill side so that it slopes gently into the swale to reduce turbulence once water is running down the land. I then put in a layer of old horse stall bedding and compost and dug it into the bottom of the swale. I wet that thoroughly.

I've been saving all mulchable paper goods and had a 32 gallon barrel full of junk mail, newspapers, old phone books, paper grocery bags. I put in a layer of this. I wet it throughly while stomping on it. When I was done with the swale, I'd used all contents of the 32 gallon barrel.

Next I sprinkled in a layer of horse stall bedding. Wet and stomp. I then continued alternating layers of paper stuff and horse stall bedding until I was a little below grade. Wet and stomp each layer.

For the next to last layer I put in this old bamboo fencing material. Wet and stomp. My thinking was that, no matter what, sand will start to fill in the surface of the swale and I want to be sure that there is lots of fluffy area for it to squeeze between to keep it sponge like.

I finished it off with another layer of horse bedding and then a thin layer of sand. Wet and stomp. Then rocks. Rocks are one of the great moisture holders of the desert. Notice the overflow trough in the middle of the top of the swale. VERY IMPORTANT when you are capturing water - always have an overflow route.

And last, I rebuilt the water well around the mesquite, packed it with lots of straw, then rocks. Notice the white capped tube in the water well just to the left of the tree. This is a deep watering tube that goes down 3'. Very important to get desert plant roots to dive deep as quickly as possible. It will speed up the time it takes for the tree to reach groundwater and be self-sustaining.

The method I read about in the book only used the above mentioned paper goods to fill the swale. I added horse bedding 1. because I had it and 2. because I wanted to see if it would speed up the composting process. As the tree grows, you can build a new swale outside the new water well. I figure that might be every 3 to 5 years. 

Since this swale is now thoroughly saturated, I'm going to stop watering this particular tree and see what happens. If this works, I'll do this with every tree I plant. I'll report back.

2 months later - reporting back. This one tree was so stressed that it's going to take me a season to see how it responds but so far, it's hanging in. Amazingly, the mini swale next to it still registers as WET with a water probe.

I've planted another Mesquite using this same method with one addition - a small amount of polymer water holding crystals mixed into the backfill. I researched these crystals quite a bit and so far they have proved to be completely non-toxic and biodegradable. I'm astonished at the water holding capacity. This newly planted tree needs watering about every 10 days (we're still fairly warm here in early November). The other Mesquites are needing to be watered twice a week. I'm hoping that once these new tree roots reach the mini swales, the irrigation need will even less.


  1. Lots of work... I'm sure the tree will appreciate and using the simple materials you had on hand was brilliant~ Love it!

  2. Yes, lots of work. Permaculture is often about lots of work up front to save resources and energy in the future. I figure, if it works, it will save water and irrigation time.

  3. Actually permaculture is about less work. Mesquite's should fairly much establish themselves.

    1. This is a jump start for these trees until they get their roots down into the water table. We get 4" of rain a year and that will possibly grow less with climate change. I believe we have to do everything possible to hold water in the land. I've been talking with land restoration experts who actually find it hard to establish Mesquites without initial artificial irrigation. I'm experimenting to see how much I can reduce my dependence on taking water from our ancient underground aquifer for surface irrigation.

      I do think that the general statement that permaculture is about less work can be misleading. Once the system is place it should lead to less work than the conventional farm, it's true, but less work according to who? The average American hardly moves at all. Permaculture is a very active life style. I don't now a single permaculturalist who isn't extremely physically fit. They didn't get that way by sitting on porch while fruit drops into their laps.

  4. These are Tendu patta or Tuniki aakulu (in Telugu). Beedies are rolled with these leaves.Apart from that, It is known for it's black heartwood,popularly known as ebony which is a precious material for carvings.The dried flowers of Tendu have medicinal values for Urinary,Skin, and Blood diseases.Its fruits are very sweEET yard fencing