Why Paver Bases Fail in Central Florida Sand, and How to Tell Which Failure You Have

Almost everything written about installing pavers was written for a climate with a frost line. Dig below frost depth, worry about heave, use a base that drains so trapped water cannot freeze. None of that applies in Sumter County. There is no frost line, no freeze-thaw cycle, and no clay layer to bridge across. What there is instead is sand - the USDA maps this county as dominated by Astatula, Candler and Lake fine sands, all classed excessively drained, with the Candler series described as sand exceeding eighty inches in thickness with very rapid to rapid permeability. That single fact changes which failures are likely and which ones never happen.

In practice, paver work in Wildwood and the surrounding towns fails in three ways, and they are easy to confuse from a standing position. Bedding sand washout, where water has carried the thin sand course out from under the pavers. Edge restraint failure, where the perimeter has let go and the whole field is creeping outward. And root heave, where a mature live oak has run a lateral root under the pavement and lifted it. All three look like uneven pavers. All three get treated with a bag of joint sand by somebody hoping the problem is cosmetic. None of them is, and each one needs a different repair.

What is actually under a paver job in Sumter County?

Sand, and then more sand. The soil survey for this county is dominated by Astatula fine sand, Candler sand and Lake fine sand, all classed excessively drained, with Tavares and Apopka at the moderately-well and well-drained end. The official series description for Candler puts sand thickness at more than eighty inches and permeability at very rapid to rapid, formed in thick beds of eolian or sandy marine deposits. There is no clay horizon at any useful depth, and the material has essentially no cohesion. Squeeze a handful and it falls apart.

That has two consequences worth understanding before you buy anything. First, the subgrade contributes very little bearing capacity of its own, so the compacted base carries nearly all the structural work - which is why base thickness is the number that matters most on a quote and the first one a cut-price bid shortens. Second, water moves through this ground fast and does not pond for long, which sounds like an advantage and mostly is. The catch is that water moving quickly through loose material is also water capable of carrying that material along with it. In Sumter County the enemy is not standing water. It is moving water.

• No frost line means excavation depth is set by load and base design, not by a freeze depth.
• Excessively drained sand offers minimal subgrade support, so base thickness and compaction do the work.
• Very rapid permeability means water leaves fast and takes fine material with it when it goes.
• Nothing about a Sumter County paver installation is governed by freeze-thaw, so northern spec sheets mislead.

Where does the bedding sand actually go?

Between the compacted base and the pavers sits a screeded course of bedding sand - the published spec is an uncompacted nominal one inch. Its job is to give each unit a consistent seat and to allow a small amount of interlock to develop when the field is vibrated. It is not structural and it is not stabilised. It is loose sand, one inch thick, sitting where water can reach it.

It leaves by two routes. Downward, when the base underneath is too open-graded or has no geotextile separation, so the fines migrate into the voids and the whole field settles a fraction of an inch at a time. Sideways, when there is an unrestrained edge or an open joint at a slab, step or coping, and rainfall running across the field finds it. Central Florida delivers water in exactly the pattern that maximises this: the normal precipitation record at the nearest long-record stations shows June through September carrying roughly 29 of Orlando's 51.45 annual inches, arriving as concentrated afternoon storms rather than as a steady drizzle.

The tell is external. Sand that has come out from under a paver field has to end up somewhere, so look for a fan of clean sand on the grass, the mulch bed or the driveway below the lowest edge after a heavy storm. That fan is bedding material, not joint sand, and it is the clearest evidence available that the field is losing its seat rather than just its joints.

• Look for a fan of clean sand on grass or mulch below the lowest edge after heavy rain.
• Check where downspouts and roof valleys discharge - concentrated flow onto a field is the most common single cause.
• A repeat low spot in one place, refilled every year, means the water route was never addressed.
• Joint sand disappearing faster than it can be replaced usually means the bedding course beneath is also moving.

Why does the edge give way before anything else?

Interlocking concrete pavement is not held together by the pavers. It is held together by the perimeter. Every unit in the field pushes outward slightly under traffic, and the only thing resisting that is the edge restraint plus the compacted base it is anchored into. The published standard calls for a minimum of one inch of vertical restraining surface in contact with the side of the paver. Less than that and the units ride up over the restraint instead of pushing against it.

The Sumter County failure is almost always in what the restraint is anchored to. A plastic or aluminium restraint spiked into loose sand outside the compacted base has nothing holding it, and it walks outward under two or three seasons of normal use. The correct build extends the compacted base six to twelve inches beyond the paver edge so the spikes go into compacted material. It is invisible once the sod goes back down, which is exactly why it is the first thing skipped on a low bid.

Once the perimeter releases, everything else follows in order: the outside course drifts, the joints open across the field, joint sand escapes, rainwater reaches the bedding course, and units start to rock. Homeowners usually notice at step four and treat it as a sand problem. It is not. Refilling the joints on a field with a failed edge is refilling a bucket with a hole in it.

What do live oak roots do to a paver field?

Live oak is the signature tree of this part of Florida, and it is the third failure mode. UF and IFAS state directly that live oak can form large surface roots and that in time those roots lift sidewalks, curbs and driveways, and they put mature spread at 60 to 120 feet with a height of 60 to 80 feet. A tree thirty feet from a driveway is well inside the zone where its lateral roots run under the pavement.

Root heave looks different from the other two failures. It produces a hump rather than a dish, the units on top of it tilt or crack rather than sink, and the joints on the crown of the hump open while the joints around it stay tight. It also does not respond to weather - a settlement problem gets worse after a heavy wet season, while a root keeps going in any month. If a hump is growing steadily and the nearest large tree is a live oak or a laurel oak, that is almost certainly the answer.

The decision it forces is genuinely uncomfortable. Cutting a large structural root close to a mature live oak buys a flat driveway at the cost of tree stability, which on a lot exposed to tropical systems is a poor trade. The usual middle path is to lift the pavers, prune only small-diameter roots at a distance the tree can tolerate, install a vertical root barrier between the tree and the pavement, rebuild base and bedding, and reset. Sometimes the honest answer is a reset that will need repeating, planned for and priced accordingly.

• A hump with tilted or cracked units points to a root; a dish with tight joints points to washout.
• Root damage worsens in any month; washout worsens after heavy wet-season rainfall.
• Cutting large structural roots close to a mature oak trades pavement for tree stability.
• A vertical root barrier plus a reset is usually the least-bad outcome on a mature tree.

How do you tell them apart before paying anyone?

Three checks, none of which needs a tool you do not own. Walk the perimeter and push the outside course of pavers sideways with a foot. Movement there means edge restraint failure, and it is the first thing to fix because every other symptom is downstream of it. Then lay a straightedge or a long level across the worst area and look at the shape: a dish means material has left from underneath, a hump means something is pushing up from below. Finally, stand outside during or immediately after a storm and watch where water arrives from and where it leaves.

Those three answers determine the repair and the price. An edge failure needs the perimeter rebuilt into compacted base, which usually means lifting a course or two around the whole field. A washout needs the water route dealt with first - a downspout extended, a grade corrected, a slab joint sealed - and then the affected footprint lifted, base and bedding rebuilt, and the original units reset. Root heave needs a decision about the tree before anyone touches the pavers.

What none of them needs is more joint sand. Joint sand is the last step of a repair, not the repair, and it goes in after the cause has been dealt with rather than instead of dealing with it. If someone looks at a moving field and quotes sand and sealer, they have priced a cosmetic job on a structural problem, and the field will be exactly where it is now in eighteen months with your money spent. Ask for the cause to be named before you ask for a number.