Whole Site Water Plan

This is my response to the project for Unit 3: to prepare a water plan for my site. By way of introduction I’ll say that although we have a long way to go until we reach our water goals, we are already in an interesting place in that all our water originates on-site and our waste water is also treated here. Furthermore, both of these situations are fully legal and entirely unregulated, so in essence my only design constraints are those originating in the permaculture principles (i.e. using small and slow solutions, not generating waste, not doing anything that adversely affects my neighbors, the community, or the area [not sure exactly what principle that is contained in, but I'm pretty sure it's a key factor in any permaculture design]).

The overall goals of the plan are to:

  • increase  and diversify sources of clean water
  • create water storage to build resilience
  • greatly improve grey- and blackwater systems to make them efficient and healthy while recycling all water and nutrients.

Current Situation

Water enters the site from overhead, as rainfall, and underground, from a 60-meter borehole with an automatic submersible pump that feeds the 1100 L tank on the roof. At present, the latter is the only water we actively use on the site. As I mentioned in an earlier post, this borehole replaced a prior (and much shallower) well with a simple electric pump which we could not drink from and which eventually became unuseable (it’s unclear whether it dried up or was just very shoddily dug). As there is no mains water in the area, all of our neighbors have set-ups that are similar to one of these.

Rooftop tank and ground level pump control house

Rooftop tank and ground level pump control house

The borehole water is from the Puelches aquifer and is very high quality (we drink it without filtering it). The use of borehole water is very controversial. I’m not going to explore the issue in any detail here, other than to say that it basically consists of tapping a non-renewable resource using often non-renewable energy to do so. There is additional controversy in this case as there are no restrictions on aquifer use in Argentina and there is really no awareness of water conservation in Buenos Aires province (there’s an overview of the Puelches aquifer situation here). Although I know that household use is minuscule in comparison to that of industry, I still would like to set up a rainwater harvesting system to at least be able to provide some of the water we use from a more sustainable source. It would also be an important step towards disaster mitigation, not just in the case of aquifer depletion, but also because the borehole pump requires electricity, and as a bad storm can knock the power out here for a few days, a back-up water source seems very high on the priorities list to improve both our short-term and long-term resilience.

Water from the bathroom (i.e. all household blackwater and probably half the greywater) goes through a digestion/sedimentation chamber (possibly more than one) of some sort and then to an underground septic tank where it is reabsorbed. Greywater from the kitchen sink and washing machine goes to another sedimentation chamber at the back of the house which is not connected to a septic tank. I describe the two systems in more detail in the waste water section below.

The following diagram shows the key water features currently on the site.

Water plan flows and features

Key flows and features for clean and waste water


As a family, I think we have pretty good water habits. This is partly due to conscious decisions: we do the the “if it’s yellow let it mellow” approach, soap all dishes before rinsing, take short showers, and were recently lucky when our old top-loading washing machine broke down for the third time, the old guys at the genius repair shop let me trade it in for a reconditioned front-loader for about a third of what a new one would have cost me). And it’s partly to do with situational constraints: we don’t have a pressure pump on our system so water pressure in both the kitchen and bathroom is low but still functional, we only have one inside bathroom, etc. (I often find myself observing that convenience is the enemy of austerity.) The following diagram shows all the areas where we consume water on the site.

Places where water is currently consumed on the site

Places where water is currently consumed on the site

To get a better idea of exactly how much water we are using, I did two kinds of water audits for this project. The first involved calculating approximate use for each individual item and the second involved shutting off the pump and seeing how long it took for the tank to run out. The two counts roughly coincided at somewhere between 350 L and 500 L per day, depending largely on how many times I use the washing machine (which uses at least 50 L per cycle, possibly more). If we count my two little kids as one person, then that averages between 116 L and 166 L per person per day, or 816 L and 1166 L per person per week. That still seemed like a lot to me, although it pales in comparison with that of the city of Buenos Aires, which has one of the highest per capita water consumption rates in the world: a massive 636 L per person per day.

There’s another bigger post in the pipeline (ahem) about waste water, but for now, suffice to say that using drinkable water to flush the toilet and wash clothes and dishes is really beginning to appall me. Seriously, we wouldn’t poo in a bottle of drinking water we just bought from a shop, but we don’t have a problem doing our business water of equal (actually, usually superior) quality that comes out of our tap.

Using water in toilets at all doesn’t make sense, but anyway, until I talk my husband into installing a loveable loo I’d at least like to be using rainwater rather than bore water to flush with, wash clothes with, water the garden with, etc. Or even rig some kind of cascade system (used shower water into toilet and then to filtration, for instance) but not entirely sure how without rebuilding totally. Anyway. So, on to…

Rainwater harvesting and water storage

I calculated our rainwater harvesting capacity based on those parts of our buildings that have corrugated metal roofing.

View of the back gallery corrugated metal roof, and above it the roof over the kitchen (left) and living area (right)

View of the back gallery corrugated metal roof, and above it the roof over the kitchen (left) and living area (right)

In the main house, that means the kitchen (13.5 sq m), living room (18 sq m), back gallery (11.2 sq m), and side gallery (13.5 sq m), which combine to give 56.2 sq m, plus the storeroom (15 sq m), and carport (22.5 sq m). I multiplied these by the monthly average rainfall for our area to get the average potentially harvestable water, shown below.

Harvestable rainfall graph

Harvestable rainfall graph

Assuming we could store it, this is a reasonable yield, but the bottom line is an average harvestable 97,447 L per year, when our current use, although quite meager, is up to double that, which is a scary thought if the aquifer dries up or the power goes down for good… or even just for a few weeks. If we were able to add harvestable roof material to the front half of the house, the yield would increase considerably.

At this stage, I am envisioning a super-simple guttering system on all these roofs, leading via a first flush system to three separate tanks (partly out of necessity because of location, but partly because more tanks means more resilience—if one fails you’ve got more.

Locations of roofs from which rainwater can be harvested, and potential guttering and tank locations.

Locations of roofs from which rainwater can be harvested, and potential guttering and tank locations.

Where my parents live in Australia, everyone harvests rainwater: in fact, rain is the only freshwater source available to them, and they have a huge underground concrete tank under their carport. Freestanding rainwater tanks abound and are easy to buy. In Buenos Aires province, the concept is practically unheard of, and there are no systems or tanks on the market. So for the storeroom and carport I’d probably have to rig a system using old plastic barrels of some sort, or buy a tank like we have on the roof, but they don’t come any bigger than mine (1100L). For the house, we’d need a bigger storage capacity.

I haven’t been able to find figures on the longest ever dry spell in this area, but this winter was particularly dry: around 60 days without rain, which I have certainly never experienced in the 13 years I’ve lived in Buenos Aires. And 20-day dry spells are commonplace in winter. So, if we ever had to depend entirely on our rainwater system, at maximum usage we’d need 10,000 L to ride out 20 days with no rain, and 30,000 L for 60 days, or maybe two-thirds that if we were careful with it (well, I suppose probably much less than that if we were very careful). At this point, a 20,000 L tank is probably not feasible, but 10,000 L might be if we made one ourselves from curved corrugated metal. I really like this idea of combining the tank with cold storage beneath it, while also raising it to increase gravity.

Abdhallah House raised tank and cellar. Source: abdallahhouse.blogspot.com.au/2009/11/cellar-tank-stand-cool-cupboard.html (definitely worth reading the entire post)

Abdhallah House raised tank and cellar. Source: abdallahhouse.blogspot.com.au/2009/11/cellar-tank-stand-cool-cupboard.html (definitely worth reading the entire post)

I feel I don’t yet know enough about water systems to have a clear idea on what system we would use to access this tank water, but at this stage, I think the two small tanks can just have a tap and hose at the bottom, whereas the large tank should include a pump system so that we can pump it up into the roof tank in an emergency. The pump would need to be non-electric: I really like bicycle-driven ideas like this one from Guatemalan organization Maya Pedal

Bicimáquina Source:   http://mayapedal.org/gallery.en

Bicimáquina Source: http://mayapedal.org/gallery.en

I hope to install at least the guttering and tanks (even if I start using old plastic olive barrels until we can afford a bigger tank) within the next six months. One question: traditionally gutters here are made of zincalum, although PVC is now available. Any thoughts from anyone on what the better option is?

Waste water

This part of the plan is more observations than decisions, as I feel I need to understand grey- and black-water systems much better before I actually design ours. I am about to order Art Ludwig‘s book Create an Oasis with Grey Water, and I’m still looking around for reliable information on the sewage side of things.

As I said above, all our waste water is treated onsite, but that’s about as far as our knowledge goes. We don’t know when this system was built, how many stages it contains, what the stages involve, what the tanks are made from, whether they follow even basic sanitation norms…. In short, nothing. Given the state of the rest of the house infrastructure, the chances are that the system is old, very randomly designed, and not very effective. Evidence supporting this includes the fact that, by coincidence as I was in the middle of writing this, the toilet tank started leaking such that there was a small but constant dribble of water into the toilet bowl, the kind of thing that someone living in the city might leave for months before bothering to fix. Within a couple of days, the whole system had become full (the toilet was acting like it was blocked and wouldn’t flush, and one of the chambers in the patio started leaking water (not sewage, but probably not very nice water) at EXACTLY the same rate the water was dribbling out from the tank at. My husband wanted to have a septic truck empty the system, but I persuaded him to get the leak fixed first and see what happened. The plumber came, and I also flushed a packet of septic tank enzymes, and within a day everything was working fine. So the water reabsorption capacity is pretty limited, and it can’t deal with much more water than we currently put down. All of which perhaps goes to show that just about any system I come up with will probably function better than the current one.

All those caveats aside, this is one way we could go:

Rough plan for onsite waste water treatment

Rough plan for onsite waste water treatment

As the grey water from the kitchen and washing machine are already routed to a surface-level sedimentation chamber, the simplest solution at the back of the house seems to be to eventually replace the vegetable bed along the back wall with some sort of reed bed or similar, which would then feed into a small pond, and any overflow could go straight into the vegetable beds.

However, apparently there are lots of minuses to using reed beds, mainly involving have to de-gunk them every five years. If this turns out to be true, then I really like the ultra-simple solution used by Milkwood

The only problem I see with this is that the tank needs to be lower than grey water source, which would probably mean doing the whole thing underground, which complicates things. I could debate this all day (indeed, I have been)… There may be a way of making it work above ground if we cancel out the current piping and bash a hole in the kitchen wall and route the grey water out that way instead of under the house as it currently goes. To be continued… (ADVICE WELCOME!)

The black water is trickier. First of all the piping out of the building needs to be completely dug up, inspected, and replaced if necessary. The system that appeals to me the most at present is a worm-farm-based one like the one described in this PRI article. Basically the black water goes into a big worm-filled sedimentation chamber which you can also chuck other waste into, and then the clean water is pumped out (or flows downhill, depending on your site) for use in orchards etc.

If our current reabsorption tank turns out to be structurally safe (and that’s a big if: our next-door neighbors’ tank just collapsed in on itself after a recent storm. Very dangerous… especially with two little children running around the garden), we could retrofit it, taking advantage of its location in the middle of the planned food forest. Another similar system combining flushing with composting is described at Solviva. Conclusion: I need to investigate this a lot more and then will choose the simplest method for our site to include in my final design.

Rainwater run-off

The final part of my water plan is dealing with non-harvestable rainwater, i.e. the water that falls on the garden and side patio and causes (mild) flooding. Just to be clear, I’m talking big puddles that take several days to reabsorb, not life- or property-threatening floods.

Here’s the plan:

Flooding mitigation plan

Flooding mitigation plan

Hopefully the guttering system will sort out most of the patio flooding, and eventually this area will either be built over (with a roof with guttering) or will be reconverted to garden, and the soil will do lots of absorbing. Part of that would involve a rain garden, probably using Anne Spafford’s Rain Gardening in the South (thanks James for pointing us to this video with her which explains the concept). Once I have a clearer idea of where the new structures on the site will go, I can define this.

The larger floodable area basically coincides with where most of our fruit trees now are, and where the eventual food forest will be. Here I plan to use dual purpose swales/paths, basically following the idea that Milkwood used in their food forest, which basically entails narrow on-contour swales filled with wood chip to serve as main paths, connected by smaller “spur” paths, which run on diagonals between the main swales so that water doesn’t just flow straight down them. Here is Kirsten’s description of how they function:

The main contour paths were surveyed and then dug out [...] as small swales, about 60cm across. These trenches were then filled with woodchip. This feature will allow the main paths to:

  • Catch and store rainfall, releasing it slowly and gently downslope, through the topsoil.
  • Allow the woodchips to decompose over time, creating fungally dominated compost which can then be dug out onto the forest garden

We’re finding that, unlike all the other larger swales here at Milkwood, these small ones are working brilliantly to catch, store and release moisture through the garden. I think it’s because they’re the right size and in context with the system surrounding them.

Here is their design:

Milkwood paths as swales/swales as paths. Source: http://plantingmilkwood.files.wordpress.com/2011/11/mw-ff-patchwork.jpg

This would work well with my site because the main paths would run the length of the property, roughly on contour, though that’s probably academic as the site is basically flat. This is my first design, which is 100% on paper (i.e. I haven’t paced it out yet), just to get an idea.

Paths to work as swales through the floodable part of the garde

Paths to work as swales through the floodable part of the garden

As soon as I drew this out, it made total sense. In fact, for the first time, I can start to envision the forest garden, with the paths as the structure around which different patches and fruit-tree guilds are built, defined probably by the mini-microclimates within this area. Stoked. The main challenge for this part is going to be the digging, of course, and sourcing wood chips or other suitable material in large enough quantities while still being affordable.

Even though I haven’t ironed out all the details in this project, as a result of doing it I have a much clearer idea of how water on our site works and what action I can take to improve our use of it, and which areas I need to research more.

Talk to me...

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s