Soil testing: pH, water repellency, depth and nutrition

You have a hole, a colour, and a depth to something hard. That is the Perth soils lesson done. This lesson turns those observations into numbers you can cost from.

Four tests, all cheap, none needing a laboratory — although there is one situation where a laboratory is worth the money, and this lesson says when. When you get to Phase 7 and price amendment by the cubic metre, these numbers are what stop you guessing.

What a professional does at this stage

A landscaper does not send soil to a laboratory on every residential job, because field tests and laboratory tests answer different questions and only one set changes the design.

Field tests — repellency, depth, drainage — change what you build: whether the irrigation cycles, whether a tree pit needs breaking out, whether a bed needs raising. Laboratory tests mostly change what you feed the garden afterwards, a maintenance decision you can revise cheaply later. So field-test everything, and laboratory-test only where something expensive or edible is at stake: a large lawn, a vegetable garden, a $4,000 specimen tree, or a spot where the last three plantings have died.

The second habit is test by zone, not by block. A renovation site is never uniform. The strip against the house is builder’s fill; the bed under the old jacaranda has thirty years of leaf litter in it; the side where someone parked a trailer is compacted to 300 mm. Use the same zone names you used when you measured up.

Test 1 — The water-drop test

The most important test in the lesson, and it takes ninety seconds.

Find bare soil. Scrape off the mulch and the top 5 mm — the waxes that cause repellency concentrate right at the surface and you want the layer just under them. The soil must be genuinely dry, which in Perth means January to April; testing damp soil in August tells you nothing. Put ten drops of water on the surface from an eyedropper and time how long they take to disappear.

Penetration time	Severity	What you do about it
10 seconds or less	Non-repellent	Nothing. Note it and move on.
10–60 seconds	Low	Wetting agent annually, good mulch cover.
Over 60 seconds	Moderate	Wetting agent twice a year, and irrigation designed to cycle and soak.
Minutes to hours	Severe to very severe	Clay and organic matter incorporated, not just a surface treatment. This is a real line item.

Very repellent soil can take longer than three hours to take the droplet in. Do not stand there — mark the spot, set a timer, come back.

The failure mode this predicts is everywhere in Perth: the sprinklers run, the surface looks wet, water runs sideways or channels down a few preferential paths, and the plant dies of drought in what appears to be a watered garden. Repellency severity feeds the wetting agent programme in turning Perth sand into soil and the run times in irrigation scheduling.

Test 2 — pH, and why your number may be misleading

Take five or six subsamples per zone from about 100 mm depth, mix them in a bucket, and test the mix. Do not sample within a few weeks of fertilising or liming, and stay a metre away from limestone walls, concrete paths and anywhere mortar was cut — those spots read a full unit higher than the bed two metres away.

Two pH scales are in circulation and confusing them is the standard mistake. WA laboratories report pH in a weak calcium chloride solution; garden kits give a water-based reading, roughly 0.7 units higher. Whichever you use, write down which one it was. Most plants sit happily between 5.5 and 7.5 on the CaCl2 scale. On the coastal plain, grey Bassendean sand is often mildly acidic and rarely the limiting factor, while anything over limestone is alkaline and coastal calcareous sand strongly so.

The honest professional position on alkaline sand over limestone is that you do not fix it. Carbonate buffers the soil — you can drop the pH inside a raised bed of imported mix, but not across a whole garden at a price worth paying. The consequence is iron and manganese lock-out, showing up eighteen months later as yellow leaves with green veins on gardenias, citrus, camellias and magnolias. So this number goes into your planting plan rather than your soil budget.

Test 3 — Depth to a hard layer

You dug one hole in the last lesson. Now cover the block. A 1.2 m length of 12 mm steel bar driven with a hammer gives you refusal depth in seconds; probe on a rough 3 m grid wherever you intend to plant or excavate, and record both the depth and what stopped the bar.

• Limestone rings and does not give. A crowbar bounces.
• Coffee rock is dark and cemented but crumbles when you hit it.
• Compaction stops the bar, then lets it through with more effort. On a renovation this is the most common finding and the most misdiagnosed — old driveways, scaffold runs and brickie’s sand leave a plate at 150–300 mm that people call clay. It is not clay. It is a machine.

Under about 400 mm to rock, medium and large trees need the pit broken out mechanically; 400–800 mm, you can plant trees but should price rock breaking per pit. Compaction is usually rippable — a labour cost, not a machinery cost. The free DPIRD NRInfo soil-landscape mapping is a cross-check, but trust the steel bar over the map: regional mapping is not drawn at the scale of a 600 m² lot.

Test 4 — Drainage rate

Dig a hole roughly 300 mm across and 300 mm deep in each major zone. Fill it, let it drain away completely, then fill it again — the first fill wets the surrounding soil so the second measures the soil rather than its thirst. Measure the drop over ten minutes and convert to millimetres per hour.

Deep Perth sand normally empties absurdly fast. If a zone drains slowly you have found something worth money: a compacted plate, coffee rock perching water, a clay lens, or a high winter watertable.

Two Perth warnings. A summer test is optimistic — water tables on parts of the coastal plain rise substantially through winter, and a hole that empties in two minutes in February can hold water in July. And this number decides whether soakwells will work on your block, so take it through to water on site and the drainage plan.

The nutrition question, and when to pay for a laboratory

Perth sand holds very little of anything. The WA-specific measurement worth asking for is the Phosphorus Retention Index, because deep sandy soils are the exception to the rule that phosphorus stays where you put it — here it leaches, which is why SERCUL’s Fertilise Wise programme tells Perth turf managers to measure PRI before planning a fertiliser regime. More fertiliser is not the fix; more retention is, via clay and organic matter in Phase 8. And many WA proteaceous natives, banksias in particular, are phosphorus-sensitive, so a standard high-P lawn food broadcast near them is a slow execution.

If you send a sample away, ask for pH in CaCl2, organic carbon, Colwell phosphorus, PRI and electrical conductivity. On an older block where you plan to grow food, screen for metals as well — old painted walls shed lead, and parts of Perth carry a market-garden history. VegeSafe does that for a $20 donation, and DWER’s free contaminated sites map is a five-minute check before you dig.

What this feeds

Repellency sets the wetting agent programme and the irrigation run times. pH sets the plant palette. Depth to hard layer sets excavation method and tree selection. Drainage rate sets the drainage design.

The next lesson leaves the ground and looks up — sun, shade and aspect — and the two together are what let you draw hydrozones later instead of scattering plants and hoping.