Sunday, July 31, 2011

A Large Scale Wick Irrigation Experiment

The hoop house described in the last post  is intended to ultimately house many containers. With 200 to 400 containers, watering is a potential issue, especially in a Texas summer when containers often dry out between even twice-a-day waterings. The shade house is 18 by 40 feet. I first considered 40-foot rows or pairs of containers, each pair served by a forty foot length of ½ inch PVC pipe laid on top of the containers, appropriately drilled to direct water to the two rows. However, this makes for a complicated watering system and elaborate timers. I decided to try wick watering on a large scale. Here’s the prototype now being tested.

The 18-foot wide shade house will accommodate five double rows of containers, for a theoretical maximum total of 400 in the forty-foot house. I determine to test first with thirteen herb and vegetable starts in late July. No rain likely. Almost certainly near 100 degrees F. every day.

The container wick is simple. A 1/2 inch hole was poked in the side of each container, about halfway up the side, and each was filled to that point with soil.

About two feet of 1/2 inch nylon rope was cut for each container. One foot was run into the hole and arranged in a loop on top of the soil.  Each plant was removed from its nursery container and placed on top of that level, and the rest of the soil was added. The remaining rope trails outside the container and will lie in the water supply. Each container was watered by hand to afford a good start.

While preparing these containers, I temporarily let the rope ends trail into a container of water. Within a few minutes, each rope actively wicking and was wet all the way into the container. 

In the shade house, the water reservoir for one row is forty feet of four-inch PVC sewer pipe. Sewer pipe is thinner and less expensive than heavier PVC designed to handle water under pressure. Each end of the forty-foot reservoir was fitted with a 90-degree joint. I wanted to be able to see the water level from both ends. These will eventually be provided with easily removable caps to keep debris out.

The pipe was laid along the ground, and containers  placed either side of the pipe. At each container position, a 1/2-inch hole was drilled near the top of the pipe, and the trailing end of container wick/rope was inserted so that it lay inside the pipe. 

The pipe was then filled with water. In my installation, one end of the pipe happens to be slightly lower than the other end, and it’s rather difficult to get the lengths of sewer pipe glued together so that they are perfectly straight. But it is close enough that the whole length is reasonably full of water.  The pipe has enough flex, especially under the weight of the water, that shims could be used to level it. 

Twenty-four hours later, at 3:00 p.m. on an extremely hot and dry day, some of the new tomato plants showed the slightest wilt, and parsley had completely drooped, so all the containers were given a drink by hand. I reasoned that it would be foolish to force a failure simply on account of the need for a little extra moisture immediately after transplanting. By sundown, all but parsley had perked up. The parsley was still limp, likely a bit shocky.

By the next day, the parsley had also perked up. Even better, it had done it while being on the wick all night. That afternoon, there was less wilt than on the first day, and I gave them another drink by hand. The soil in each contained was slightly damp at root level. By the third day, the containers were clearly taking up plenty of water, as evidenced by the drop in water level in the pipe reservoir and the amount of water needed to top it off.

By day four, all thirteen young plants went through a 104-degree cloudless day with no sign of suffering and no supplemental watering. They are now on their own, or at least entirely on the wick system. On the afternoon of day four, I checked them and was dismayed to see that there were signs of moisture, areas of darker soil on the surfaces. Dismayed, because I thought my wife must have watered them and had spoiled the experiment for the day, even though she had told me they were doing fine. I checked them with an uncalibrated soil moisture meter and found that at root depth, the meter showed “5” on a scale of 1 to 10, just where it should be for nearly all plants that don’t require especially dry or wet soil. But I found she had not watered them at all. They were being maintained in good condition on the wick system.  Every third day the sewer pipe reservoir is visually down enough to take more water, but it clearly could go a long weekend without filling. Obviously, this will not be the case with 80 or even 40 containers on a row and probably not even with a dozen fully mature plants.

The system promises to provide the best benefits of wick watering, lack of need to intervene frequently to water and a steady on-demand supply of moisture direct to the soil. And fertilizers can be added to the water supply and will be taken up by the soils. Other benefits include the fact that watering doesn’t wash nutrients out the bottom and that evaporation is minimized on account of the water going directly to the deep root level. And the system is low-tech and relatively inexpensive, considering that one forty-foot length can handle 80 containers. And we can go away for days without fear of returning to dead plants.

There will be some question remaining to be answered. One is how much rope is really needed inside the soil. The rope is cheap enough and long lasting enough that I don’t mind using a lot of it, hoping to transfer moisture efficiently. Another is whether the amount of water used will be sufficiently predictable to safely put the refill operation on a timer or whether some sort of sensor will be needed. 

It also seems that the respective levels of the wick and the water source effects the rate of of water movement. I put out a second round of containers, this time mostly with seeds. Because I wanted he seeds to begin closer to the wick, I began with less soil upon which to lay the wick loop, which meant it entered the pot farther down the side than the transplant pots. This put the wick level in the pot below the water level of the pipe reservoir. The next morning, there was water pooling from under these new pots, apparently siphoning through the wick. I blocked those pots up two inches off the floor to duplicate the situation from the first pots. I will need to experiment. It may well be that varying the elevation of a pot can tailor the soil moisture somewhat to meet the differing water needs of different plants. 

The calculated volume of 40 feet of 4-inch pipe is about 26 gallons (208 pints). Estimates of how much water a mature tomato plant needs vary wildly, probably due to differences in climate container size, fruiting or not fruiting, and soil. A mature tomato plant under 30% shade on an otherwise clear, sunny day is said to require anything from two pints to a gallon of water a day. But clearly, if the system is supporting 40 mature tomato plants in fruit, at least two refills per day will be required. I suspect it will always be a matter of keeping an eye on it and adjusting any automatic filling mechanism according to how quickly the water level drops.

I have some ideas for both a visual indicator of water level that would allow me to spot any low reservoir at a glance and for and automated refill system. This is an ongoing project, and I'll report future results. 

A few notes. Avoid cotton wick material. It will rot. Nylon and poly are fine and reusable. It doesn't have to be rope. Twisted strips of cast-off nylon or polyester clothing will also wick. There are many other ways to adapt wick watering techniques to other situations, and there are many resources on the Web.

Meanwhile, the inspectors approve. 


  1. Hi,

    I love the system, I have found chux works great as a wick and here in australia I can get a roll of 100 for $5. You can spread it out in the bottom of the pot to evenly distribute the water too. I really like the idea of the pvc pipe, it'll raise the WAF of my container garden immensely!

    (WAF = wife approval factor)


  2. Interesting. I saw your post with the link here in the Gardenweb forum and submitted this to stumble upon. Hope you don't mind the extra traffic.

  3. Great thanks for the tips. I may actually use the Chux solution as we're trying to find a watering solution for large/tall pots we have for indoors which will be somewhat neglected but have an opportunity for us to put a water reservoir inside the pot (at the bottom) and the plant on top.

  4. Can you share which ropes wicked the best? The rope I've seen readily available at Ace hardware, for example, seems awfully tight in terms of wicking. I have a loose rope that wicks very well, but I can't find more to buy.