These before and after shots of my water harvesting basins taken only two days apart illustrate neatly why I created them, and the purpose they serve. The first pair of photos shows one flood retention basin completely full on Monday Sept 8, 2014 with about 4.5 feet of water (1.4 meters) and then virtually empty only two days later on Wednesday Sept 10, 2014. All that water has soaked via gravity into the ground – about 25,000 gallons (nearly 100,000 liters). My well shaft is located between these basins and this is water that I will be able to draw upon in dry times.

Only a thin ridge of sand now separates the basins. The larger basin in the background fills from a different source than the smaller foreground one, and frankly I have never seen both full to the same level before now. What I didn’t know is that a breach in the sand ridge between them had occurred, allowing water to flow between them. The breach is underneath the branches of the catclaw acacia (Acacia greggii) to the right edge of the photo – I only discovered this two days later after the water had receded.

In the second picture below, you can see to the right where there is a cone of sand that eroded out into the basin when the thin separating ridge between them failed, allowing water to flow from the other basin into this one. While that sounds bad, actually it was good, because I believe that at least half of the water came from the larger basin and was retained in the smaller, where it quickly percolated into the soil and was not lost to runoff downstream. It had never occurred to me to actually connect the two basins before (they get runoff from different wash channels), but now that it has happened anyway, I think I’ll leave it that way so that future overflow from the larger basin, which fills much more often than the smaller one, will be diverted over there and not be lost.

Now that the water has receded, you can better see the breach between the basins – the cone of sand under the catclaw to the right is where it occurred. I am going to leave it there so that overflow can occur between the two basins in the future. I will lose less water that way and get to keep more for aquifer recharge. About 25,000 gallons of water has percolated into the soil in only two days.

Closer-in view of the sediment cone that occurred when water from the larger basin broke the narrow sand strip separating them and flowed into the smaller one. Eventually both of the pools equalized in depth.

The second pair of images below depicts the accidental blowout of the dam of the larger basin, again two days apart. Too much water came through too quickly and overtopped the lowest spot in the dam, which was a few inches higher than the spillway I created on the other edge (not visible here) but clearly not high enough. Or course it’s a tiny earth fill dam made mostly of sandy, loose soil, so erosion was quick and about a cubic yard or two of soil was washed downstream.

An unintentional spillway and temporary stream flows past the solar panels that power my well pump, providing me with the critical water resources I need to live here.

Fortunately, the blowout occurred near some catclaw acacia trees and other shrubs, whose roots definitely helped stop further erosion and kept damage minimal. Once the water in the basin recedes I will patch the dam back up and make it wider and higher to prevent this from happening again. I knew it was a potential issue, but had not had any problems in the past 6 years so this is the first major test of its structural integrity. But at least now I know what to do for sure. I will also have to fill in the new arroyo that was downcut about 6 inches into the soil, since now the once-buried cable that runs between my solar panels and the well pump has been exposed, and it’s also harder to walk there now. I have plenty of sediment on the other side of the dam for patching purposes since every single flood brings in a couple of tons of sand and mud….

Two days later, damp sand and an eroded channel are all that remain to indicate the water that was flowing here. I’ll have to fill that back in since now it’s harder to walk to my solar panels to turn them on and off as needed, and the electrical cable is exposed.

The electrical cable that runs from my solar panels to the well pump 300+ feet down the shaft lies exposed via erosion after the breach of the dam.

Here’s another shot of the dam breach on Sept 8 causing the flooding and erosion seen above.

The unintended spillway. Before this storm, in the prior 6 years I had not seen enough water flow into both basins to entirely fill them. I knew the dam was likely to be a bit low at the breach site but hadn’t taken the time and effort to raise it by a few inches. This is what happens when you procrastinate.

Floodwater draining past my solar panels. Well at least I wasn’t using them that day….

This effort at water harvesting via flood catchment, retention, and percolation is a good example of sustainability – doing something that can actually be replicated indefinitely without depletion or excessive damage to the ecosystem. This is runoff that would not have been put into the aquifer without me capturing it, concentrating it, and allowing it to soak in via a dedicated spot.

This one flood event alone captured about 80,000 or more gallons of water between the two basins, and I definitely do not use more than that in an entire year. Plus the rain arrives free of carbon (it’s rain! no canals and pumps and pipes to deliver it to my property!) and it is also pumped back out of the soil when needed without carbon emissions (because the electricity used is solar) so there’s little carbon dioxide pollution there as well. It’s one of my best efforts at conservation of a critical resource and it’s very low impact. Moreover, since embarking upon the journey of water harvesting in 2008, not only has the quantity of water in my well increased, but the quality has also. These several-times-yearly influxes of fresh water dilute the dissolved mineral salts present in native groundwater here in Arizona, so it is less “hard”. I love the taste of it. So do my plants.

This new erosion channel formed after the dam breach following Hurricane Norbert’s two inches of rainfall in two days on Sun and Mon, Sept 7 and 8, 2014.

This smaller basin did empty/soak in a great deal more quickly than the other basin does. That’s because the larger basin has much more fine silt and mud in it, because it gets used a lot more thanks to the runoff from Alamo Road, and that greatly retards percolation rates. When both basins were newly-built in 2008, the perc rates were similarly fast, dropping 2 feet a day. Now, the larger basin, with more than a foot of mud and fine sediment in it, drops only 2 to 3 inches per day, meaning that several feet of water in it stays there for several weeks rather than only 2-3 days.

Of course as the smaller basin fills more regularly since they are connected now, and more mud enters the system, the perc rate will slow there too. It seems that about every 4 or 5 years I will need to hire a bulldozer to scoop out the sediments and open the basins back up, because over time they will totally fill up again. Basin maintenance…. who would have thought I’d be doing geoengineering?

The smaller basin fills up with water primarily off of my property, comprised of native coarse desert sand and rocky hillsides. It takes a pretty heavy or sustained rainfall to create enough runoff for the feeder wash channel to flow and fill it. Therefore, it fills less often and with less water (usually) and less fine mud and silt is deposited. This makes for a fast percolation rate, and the 4.5 feet of water (about 25,000 gallons) of water in this basin after the storm has mostly soaked away within 48 hours on the evening of Sept 10, 2014 when this image was taken. Less than 6 inches of water remains in the very bottom, and that was gone by the third evening of Sept 11.

A view of the smaller basin filled up on the late morning of Monday Sept 8, 2014, two days before, for comparison. Note how dramatically the water level dropped in only two days from the photo that preceded this one. The creosote bush several feet tall  in the center right of the photo is nearly submerged.

I actually would like quick percolation, but that mud really slows things down after only a flood or two. The silt suspended in the floodwaters is so fine and dense that even after weeks of sitting, it still looks like weak coffee with milk in it. The water pretty much never goes clear, indicating that the particle sizes are so tiny that they won’t settle on their own via gravity. I believe that those tiniest particles are what slow the percolation so much. There is a reason why in artificial ponds that aren’t using rubber liners that they put down a layer of clay on the bottom – it actually is effective at dramatically retarding water loss.

A view of the larger basin two days after Hurricane Norbert’s two inches of rainfall produced tens of thousands of gallons of runoff.

The larger basin has a watershed that includes open desert as well, but critically, it also includes about 3/4 mile of Alamo Road, which is the county road I live on. While Alamo Road is not paved, due to traffic and road maintenance it is hard-packed and therefore sheds water much more effectively than open desert does. I have seen usable amounts of runoff enter the larger basin from as little as a quarter inch of rainfall, and a good storm of a half inch that falls fast enough (usually in less than 20 minutes) can yield 15,000 to 25,000 gallons of water.

The larger basin covers a surface area that is about 2.5 times larger than the smaller one and is about the same depth of 4.5 feet at most. It can hold as much as 70,000 gallons of water before it overflows.

Ripples in the sand and mud indicate the entry point of the upper basin, where the inflow comes in.

The well shaft sits between the two basins. At full pool, it sits in a few inches of water. This usually soaks away into the soil within an hour or two, although the deeper regions of the lower basin may contain water for several weeks. The dam breach is semi-visible in the shadows between the top of the well shaft and the catclaw acacia behind it.

The garden under a rising full moon the night of Sept 10, 2014. The last clouds of the monsoonal rains enhanced by Hurricane Norbert have cleared the area, leaving the air washed clean and scented of juniper and creosote. My plants and the desert are grateful for the rains after a long, hot summer. And so am I.