For the last six years, California has suffered through an extended drought with significantly reduced seasonal precipitation. As consulting engineers with a strong presence in the state’s northern reaches, we have observed an increasing number of older pools that recently have undergone differential movement and cracking. This comes as a result of the dry conditions.
In most cases, they are in areas with thick clay soil and mature trees near the pool. Due to the extended drought, the trees are pulling more moisture than normal from deeper down in the clay soil. This unusual moisture draw causes excessive desiccation and shrinkage of the soil, which leads to pool shell movement and cracking.
We also have seen similar problems in other parts of the country when unseasonably long dry periods occur and the same two conditions exist — clay soil and mature trees. This is a problem exclusive to clay soils since sands, silts and other soil types don’t normally change volume when moisture leaves them.
Perfect storm — or lack thereof
Soil contains three components: mineral solids, water and air. The last two typically fill the voids between soil grains, with water entering and exiting the soil matrix without any noticeable volume change.
However, if the soil grains are really small — less than 0.002 millimeter — they become what engineers refer to as “clay” sized. But not all clay-sized particles are created equal. When they are this small, the mineralogy of the particle becomes important. Certain minerals form flat and platy particles (think of cornflakes) and thus become clay.
Due to the chemistry of clay particles, a majority of the surface carries a slight negative electrical charge. This attracts the positively charged side of water molecules. The closer the water molecule, the stronger the attraction. Consequently, clay soil attracts and holds water with various levels of force. Think of it as a dirt sponge. This force causes clay soil to undergo volume changes with variations in moisture content. When water is available, it is attracted to the particles and pushes them further apart, causing the soil to swell. The particles hold on to this water until something comes along that can pull it away, resulting in shrinkage. Plant roots, particularly trees, have the ability to pull away this “stored” water in times of need, as they feed through capillary action.
We have noted this by observing small rootlets present in desiccated (dry) clay soil samples obtained in subsurface test borings.
The upper portion of a clay soil deposit is referred to as the “active zone.” This is the area where the soil seasonally changes moisture content, and volume, because of the drying action of the sun and typical plant growth. In northern California, this zone averages 2- to 3 feet in depth.
The clay soil below the active zone typically maintains a fairly uniform moisture content and volume year ‘round. This often is referred to as its “natural” condition. When it comes to building foundations, we generally recommend that they extend below the active zone to protect them against seasonal movement. We traditionally do not have to worry about constructing pools in areas with substantial clay soil deposits, since the shells usually are founded below the active zone. (The exception, of course, is decking, but that is a whole other issue.)
But clay soil is the perfect medium to help things survive during a drought, due to its ability to hold onto large amounts of water. Clay generally can hold in reserve 50- to 200 gallons of water per cubic yard. As certain trees mature, their roots often extend below the typical active zone. In times of prolonged drought, they can extend even deeper, searching for water.
Consequently, as a drought goes on, the trees pull more and more water from the clay soil, resulting in desiccation or shrinkage of the ground. If the trees are near the pool, and the roots extend under the shell, this can cause a loss of ground support, tilting, and cracking of the pool.
This condition has accelerated during the past year as homeowners reduce their landscape irrigation to comply with the recommended 25- to 30-percent water reduction required in most communities.
Proof of movement
The pool tile line typically is installed with a laser or water level for obvious reasons. Thus measurements between the water surface and the edge of a tile line can serve as a record of differential shell movement.
The diagrams above show the results of such measurements taken around the edges of two pools where large mature trees were nearby. The dashed lines are referred to as contour lines and they represent areas of a specific, equal elevation. For instance, the lines that read “-¼ inch” outline an area that rests ¼-inch below the intended elevation. The contour lines show that the pools, which were originally constructed level, now dip progressively in the direction of the trees, causing several cracks in the shell. In both cases, there was at least 10 feet of clay soil present. Soil samples taken underneath the shell show an unusual decrease in soil moisture content under the portions of the pools that settled along with the presence of small rootlets.
Sometimes the problem even becomes cyclic: As the pool cracks, it leaks; water enters the clay, causing it to swell back up; then the process repeats. We had one such case, where we repaired the cracks and everything was fine. Then the trees went back to pulling moisture out, causing the pool to settle and crack again. The pool started leaking, the clay swelled back up, and so the pool shifted back up.
This phenomenon doesn’t usually spell the end of pool and spa structures.
The cracks that form usually are hairline-type — measuring 5/1000 to 15/1000 inch in width — because the steel reinforcement holds the concrete together. But any crack is problematic, with the possibility of leaking and water loss.
When performed properly, high-pressure epoxy injection has proven to be an effective way to both seal the cracks and bond the shotcrete back together. It may have to be done more than once, if the problem becomes cyclic.
In the worst cases, those that cycle and become a perennial problem, we advise clients to apply a fiberglass coating to the shell, since that material can endure a bit more movement.
The jury is still out on how much recovery or “rebounding” the clay soil and pool shell will see once wetter weather returns. Theoretically, the soil should recover 100 percent of its volume, but our suspicion is there will be some permanent loss due to the maturing of the trees. The roots have grown to obtain water, and they’ll likely continue. And when clays go through this cyclic process, they tend to consolidate and shrink. So I believe there will be some permanent loss of volume, with only a partial recovery. That may not be a negative: If too much rebounding occurs, additional cracks may form that would require future injection work.
Unfortunately, there isn’t a cost-effective way to prevent this from happening in clay environments that suffer drought conditions for long periods of time. Contractors could place pools on piers, but of course this will add $30,000 to $50,000 to the cost of construction. This doesn’t seem cost-effective, especially considering that the problem can be remedied relatively for closer to $3,000.
If this is a concern, contractors may want to avoid placing trees near their pools, and not build near existing trees if possible. However, this may be unavoidable in the case of neighbors’ trees.
Neil Anderson is a principal with Terracon, a national geotechnical engineering firm based in Olathe, Kan.