By Richard Hanson, 
www.ecovieenvironmental.com and www.georgiawatertanks.com

Water. Second only to air in terms of value to life. Yet, most construction projects treat this valuable resource as a problem to be managed. We proclaim a commitment to sustainability, yet we demand a payback period of months. We can’t handle drought, but we can’t handle floods either. What’s the savings for preventing a flooded basement, or a combined sewer overflow? Who can accurately predict water and sewage costs over the life of a building?

Several factors contribute to the increasing viability (and popularity) of rainwater harvesting.

Manufactured products of all kinds are becoming cheaper, better, and smarter. Complex controls that once required human intervention can now be automated, failsafe, and visible through the Internet.

Water rates are growing. Water departments have changed from charging less per gallon for high consumers, to more. Since water departments are commonly government entities, hiking water rates can be less damaging politically than raising taxes. As of today, water departments still don’t place a value on the cost of the water they sell. Just operating expenses. What happens when they do?

Governments are beginning to charge for stormwater management. The system of storm drains in your 50-year-old subdivision are crumbling, and the developer isn’t going to pay for it. You are. In some enlightened communities, they offer credits for reducing the load.

In other areas, government doesn’t give you the option. Stormwater management is required by law in many urban areas. Your choices include retention ponds, underground chambers, pervious pavers, and rainwater harvesting. Civil engineering firm AMEC found rainwater harvesting to be the most cost-effective method of stormwater management; at least five times less expensive than any other.

There are some reasons that rainwater harvesting isn’t more popular.

Municipal codes often conflict. Plumbing inspectors are wary of bringing what they think is nonpotable water into such close contact with drinking water. We commonly see rainwater classified as more hazardous than boiler water, dairy equipment, and laboratories equipment. While water departments appreciate the reduction of consumption during droughts, they do want the income. They have legitimate concern over the loss of income, particularly when they are still treating the sewage from toilet flushing.

Many rainwater harvesting systems, let’s just say it, don’t work that well.

Lots of moving parts.

Interconnected logic that neither connected nor logical.

Lack of a holistic view on the part of the design and construction teams.

Manufacturers want to make standard products, and they often try to force-fit them.

Those that do work are often more expensive to purchase, install, operate, and maintain than they should be. Did you include electrical costs for that UV that runs continuously? Did you include annual bulb replacement in your costs? What about the cost and hassle of replacing bag filters quarterly?

Designing a rainwater harvesting system that’s practical, reliable, and sustainable isn’t impossible. Here are some guidelines.

First, decide why you want this system. Doing everything, e.g. stormwater management AND reducing municipal water use AND teaching the public about the benefits of rainwater harvesting AND and utilizing every drop of rain, is a sure way to do nothing well, and bust the budget. For this exercise, let’s assume that we are trying to reduce our municipal water consumption by 50 percent.

Second, evaluate how much water we have available. If this isn’t your first article on rainwater harvesting, you know that the formula is .62 gallons of water for a 1 inch rainfall on 1 square foot of rooftop. The annual rainfall in an area can be estimated using data from www.currentresults.com/Weather/US/average-annual-precipitation-by-city.php, or other websites. Incidentally, a 25 percent reduction in rainfall is considered a drought. So, even during a drought, quite a bit of water might be available.

For example, Washington, D.C. gets about 40 inches on average. Like most East Coast cities, rainfall is rather evenly distributed throughout the year. Yes, there will be a dry month, but sometimes that’s August and sometimes it’s October. A rule of thumb that I like to remember is that a 40,000 square foot roof on the East Coast has the potential to collect 1,000,000 gallons of rainwater per year. Is that going to be runoff or resource? A 4,000 square foot restaurant: over 250 gallons/day for toilet flushing.

Next step is to determine how to use this water. In America today, commercial buildings are essentially limited to nonpotable uses for rainwater harvesting. That’s fine, because it’s enough of an opportunity. The vast majority of water used could be nonpotable, so let’s start with that. Following are the most practical uses for the hospitality industry.

Cooling tower makeup

Cooling towers are most often the largest consumer of water on the property. As a rule of thumb, cooling towers use about a gallon of water/month for every square foot of conditioned space during the summer. So, a hotel with seven floors or more and a cooling tower will use all available water for cooling tower makeup.

Turf irrigation

A good rule of thumb is 1 square foot of rooftop to 1 square foot of turf. It depends on the plants, and could easily require more water than that for fescue in the south.

Washing machines

No cooling tower and not enough irrigation? We estimate 5 to 10 gallons per room per night, depending on how often the sheets are washed. Lots of water, but more treatment required in this application.

Toilet flushing

This is a common use of water in my experience, but perhaps not as cost effective. Many toilet flushing treatment systems make the water clean enough to drink, and we joke about making it safe for people to drink out of the toilet. We all know that it isn’t necessary, but consumers expect toilet water to be clean and odor free at the start. Many west coast, particularly Pacific Northwest, engineers and owners have progressed past this, and only filter the sediment out. Another added expense for using rainwater to flush toilets, is the need to run both potable and nonpotable piping to every bathroom in the building.

Potable

Full potable rainwater harvesting might actually be less expensive than for toilet flushing alone, since we don’t have to run separate piping. However, negotiating the compliance regulations is beyond the scope of this article.

Back to our example in D.C., if this is a 15-story luxury hotel, there is a good chance that they will use a water-cooled chiller and cooling tower. All the rainwater from the roof can be utilized to fill the cooling tower.

Sizing the tank is equal parts science, art, and 20-year weather prediction. We rainwater specialists have great debates about which method is best: monthly average, daily history, etc. Two truths to consider: we can’t use more than we have, and yet usually collecting everything available isn’t cost-effective. Go back to our purpose on this project, because the most cost effective solution means that we will probably waste some water during floods and run out during a drought. Tanks are the most expensive component, so consensus with the owner is important.

Type of tank, underground or above, and materials, is an important consideration. In general, above-ground tanks are $1/gallon less expensive than below. Plastic tanks are by far the most cost effective in sizes 10,000 gallons or less. Concrete is durable and cost-effective, especially when integrated into the foundation. Concrete has the added benefit of reducing rain’s natural acidity. Steel tanks are most common for above-ground commercial installations, giving visibility to the conservation efforts while minimizing cost. Freeze protection is not as complicated as some think, with aeration and recirculation being sufficient in all but the coldest climates.

Treatment is another large expense. It’s a multi-step process in some cases. Most important is the prefilter that keeps leaves and larger debris out of the collection tank. These filters have coarse screens, and are installed in the downspouts or other conveyance piping prior to the tank. The first of these products use a vortex action, but other manufacturers use a cascading or “jump” action. While we suppliers have great debates about our technology versus others, especially efficiency and maintenance requirements, the important message is that all the water that reaches the tank is quite clean. Left alone, the water actually gets cleaner as it sits in the tank. A biofilm will grow on the tank surface, and those organisms clean the water naturally.

We sometimes see designs with continuous treatment of collection tanks. The need for this is rare, in our experience. Design the inlet and overflow to eliminate stagnation, standard prefiltration, and the biofilm will keep the water clean naturally.

For our irrigation and cooling tower makeup applications, the water in the tank is clean and safe enough to use without further treatment. It’s almost certainly cleaner than the water in a cooling tower, which will likely have a treatment system already. You might want to add a sediment filter for irrigation systems to protect any drip emitters, or that might already be a part of the irrigation design. Coordinate to eliminate redundancy.

For toilet flushing applications, sediment filtration is definitely needed to keep flush valves working correctly. To make the water clear, use either a carbon filter for 20 gpm or smaller jobs, or UV for higher. When using UV, coordinate sediment filter screen size with UV requirements. One sizing tip for UV is to use a day tank. Toilets have a large diversity factor, as we know. If we size the day tank for an expected daily use, we can fill the tank once, then turn off the UV the rest of the day. The filters and transfer pump can be substantially downsized from peak demand, with no loss in service.

Sometimes we see continuous UV treatment here, also. Again, in most applications, treated water will remain so for weeks. Let’s save the energy and bulb life, if possible. If there’s a concern about water quality after a seasonal shutdown, drain the day tank prior to use.

Sizing sediment filtration is another situation where experience is valuable. Rainwater is not always perfectly clear, but it’s certainly much clearer than creek or river water. The small screen perforations (commonly 25 or even 5 micron) can lead to filter manufacturers recommending larger filters than necessary. Find an application expert that you can trust. This is good advice for many products that you specify.

The pumping requirements are similar to other systems you commonly specify. We need a pump to serve the fixtures, which we call a Building Pump. These are not that different from the domestic water booster systems common in many commercial buildings. Be wary of suction lift/NPSH requirements and capabilities. Some specify submersible pumps, and install them either in the collection tank or in a separate vault. Some pump manufacturers are offering special pumps designed for cistern use, but so far they are more residential size and durability. ASPE members in Puerto Rico are well familiar with lifting water, as this is common in commercial buildings there.

One final component to discuss is standby connection. Here in America, we will want to air condition the building, flush the toilets, and maybe irrigate the lawn even during drought. So, we will usually have a standby connection to municipal water. For cooling tower and irrigation applications, an air gap connection into the collection tank is reliable and inexpensive. Toilet flushing applications can benefit from more sophisticated methods that are beyond the scope of this article.

The most important part of your specification isn’t a component, it’s execution. Rainwater harvesting requires much more site coordination and installation than, say, a domestic water booster system. Collection tanks go outside. Pumps and filters and piping inside. About a dozen communication points. As much as we would like to have a manufactured unit, it’s just not possible on a commercial project. Do everyone a favor, and require a single entity, a sole source, to coordinate all the components of your rainwater harvesting system.