Writing a Gel Spec an Inspector Can Check: Municipal Planting Contracts

A parks department in the English Midlands tendered 6,000 street trees with a landscape specification that ran to 40 pages. The polymer clause read: supply and incorporate a suitable water retaining gel at the manufacturer’s recommended rate. Eleven contractors priced it, six were shortlisted, and the trees that were planted under the winning bid contained somewhere between 0 and 90 g each. Nobody could prove anything either way, because the clause had no number, no method and no test.

Public planting contracts are where vague gel specifications do the most damage, because the buyer is spending other people’s money and the inspector needs something to check. This article sets out the three numbers an inspector can actually verify, how to write a sampling clause, and what gets rejected in practice.

Three Numbers an Inspector Can Verify on Site

Only three properties of a polymer delivery can be verified by a clerk of works with a clipboard and a set of kitchen scales: the mass of gel per tree or per square metre, the particle size distribution estimated by sieve, and the visible presence of swollen beads in the backfill at the time of planting. Everything else, including absorption capacity and crosslink density, requires a laboratory and belongs in the pre-delivery approval, not the site inspection.

So write the specification around those three. State grams per tree with a tolerance, for example 50 g plus or minus 5 g for a 45 litre root ball. State the screen cut as a range, 0.5 to 2.0 mm, with less than 5 percent passing 0.3 mm. And require that the polymer be pre-swelled in water for a minimum of ten minutes before mixing, which is visible on site because pre-swollen gel is translucent and heavy while dry granules are white and light.

Writing the Sampling Clause

A sampling clause should say who samples, when, how much, and what happens to the sample. A workable version reads: the engineer may take a 500 g composite sample from any delivered pallet, comprising five 100 g increments from five separate bags, and may retain it for 28 days. The sample shall be tested for moisture content, fines below 0.3 mm, and absorption in a 0.9 percent saline solution, and results outside the stated limits shall be grounds for rejection of the pallet.

That clause does three things. It makes the supplier responsible for consistency across a delivery rather than for one submitted sample. It gives the inspector a defined action. And it costs the contractor nothing if the product is what was ordered, which is the test of a fair clause.

Penalties, Retests and the 30-Day Window

Rejection is the strongest tool and the one least often used, because by the time a lab result arrives the trees are in the ground. Structure the payment so that 10 percent of the planting value is retained for 30 days, releasable on submission of the test certificate for the delivered batch. That single mechanism changes supplier behaviour more than any penalty clause, because it puts the money behind the paperwork.

For retests, name the laboratory and the method in advance, and state that a failed retest is at the contractor’s cost while a passed retest is at the client’s. Without that split, every dispute becomes a negotiation about who pays for the second test, and in practice nobody tests at all. Keep the tolerance generous enough to be enforceable: plus or minus 10 percent on mass, 5 percent on fines, 15 percent on absorption. A specification that demands 2 percent tolerance cannot be met by any real production process and will simply be ignored.

What Inspectors Actually Reject

In practice, three things get a load rejected: visible fines and dust in the bag, moisture that makes the granules clump into fist-sized lumps, and a particle size that does not match the approved sample. All three are detectable on arrival with a sieve, a moisture meter and a comparison against the retained sample, and all three correlate with a supplier who has blended cheap material to hit a price.

Add one more requirement that costs nothing: a certificate of analysis per batch, with the batch number printed on each bag, so that a sample taken in March can be traced to a production run. Municipal clients increasingly ask for this as part of their sustainability reporting, and it is the difference between a supplier who can answer a question two years later and one who cannot. For product data structured to slot into a specification, the agricultural grades in the SAP powder catalogue are supplied with the test methods already named.