Growing nutrient-dense food starts two inches down.

Every value below comes from Logan Labs, Mehlich III extraction, same 0–6" depth, same point in the seasonal cycle: April 2022 and April 2023, same site (R2, Comptche, Mendocino County).

Total amendment cost: roughly seven cents per square foot — about $3,000/acre at 2022 pricing. Nine certified-organic materials, verified by lab testing at every step, not guesswork. Target values are the native-soil line from Logan Labs' Mehlich III "Ideal Soil" reference chart, not a one-off guess for this plot.

A lab report confirms what the ground has already been telling you. Before any test comes back, six things are visible in the field:

Six field signals

• Compaction — the spade test, how far your boot sinks, where water pools instead of soaking in.
• Color — chocolate-brown soil means biological activity; pale soil means leached and biologically thin. Black or very dark soil is a warning sign, not a good one — it often comes from thermophilic compost. A hard color line between topsoil and subsoil marks a compaction floor cutting the lower soil off from air and roots.
• Texture and structure — hand-feel aggregate size, the ribbon test for clay content, a soil-jar settling test for the sand/silt/clay split.
• Lipid feel — moist soil that feels slippery or soapy is holding stable microbial residue; soil that just feels gritty and mineral is biologically poor.
• Nutrient cyclers — springtails, mites, and earthworms visible near the surface mean a healthy community is working below it.
• What's already growing — what comes up on its own is a clue worth noting, not a diagnosis. Plants often persist on ground that doesn't match what they're said to indicate, so confirm with a soil test.

On this site, before any correction: surface cracking, winter puddling that gave way to brick-hard summer ground, shallow roots despite normal rainfall, almost no visible earthworm activity, and clover growing even though calcium was low — a good example of why weeds are clues, not a lab report. The one thing that didn't fit: a ribbon test said sandy, but the soil behaved like heavy clay all winter. That contradiction was the tell — this wasn't a texture problem. It was a magnesium problem.

Walking this ground in spring 2022, before any correction: a crust you'd feel through your boots, standing water after ordinary rain, soil that went from sticky-wet to brick-hard-dry with almost no workable window between, no earthworm castings anywhere, cover crop germinating unevenly and yellowing, and grasses crowding out the clover and brassicas — the visible signature of a soil that can't supply calcium.

Why each headline number mattered

• pH 5.3 — at this acidity, some aluminum can dissolve into a root-limiting form (the risk zone is roughly pH 5.0–5.5), phosphorus locks into unavailable compounds, and the bacteria that finish the nitrogen cycle shut down almost entirely.
• 786 ppm Mehlich-3 aluminum — normal to have, and not a toxicity reading on its own. Aluminum is one of the most common elements in any soil; at pH 5.3 some of it can dissolve into a root-limiting form, and raising pH above 5.5 locks it back up.
• 36% exchangeable hydrogen — more than a third of the soil's charged binding sites are holding acid ions instead of calcium, magnesium, or potassium. This is what "acid soil" actually means at the mineral level, independent of the pH number alone.
• Ca:Mg ratio 2.33:1 — squarely in what the report calls the magnesic danger zone, below 4:1, where soil structure starts to collapse and biology stalls out. This one number explains most of the others.

Bare, an ion of magnesium is physically smaller than an ion of calcium. In water — which is where soil chemistry actually happens — that flips. Magnesium's charge is packed into a smaller volume, so it grips water molecules hard and holds a tight, rigid shell of six of them. Calcium's charge is more spread out; it holds its water loosely, one or two molecules at most, and sheds it easily.

The result: a hydrated calcium ion is about 4.1 angstroms across. A hydrated magnesium ion, water shell and all, is 4.3–4.8 angstroms — bigger, not smaller. That reversal is the whole mechanism.

The finest soil particles — the little clay this soil has, plus its organic colloids — carry a negative charge. Calcium closes that gap and forms a real bridge between neighboring particles — that bridging is flocculation, and it's what makes soil structure that drains and breathes. Magnesium, trapped behind its own water shell, can only hold particles apart. That's dispersion: fine particles slide past each other, pores seal shut, and the surface crusts. It's why a soil with very little clay can act like heavy clay — the problem isn't the clay content, it's the magnesium. Every drainage problem, every crust, every dense and gummy handful of soil on this site traced back to this one piece of chemistry.

Magnesium dominance doesn't stay contained to soil chemistry — it cascades through every scale of biology working the ground.

Roots and larger organisms

Roots can't elongate into soil whose pores have collapsed under saturation; they stop, fork sideways, and stay shallow, cut off from subsoil moisture and minerals. Earthworms need air and moisture together — dispersed, waterlogged soil gives them moisture without oxygen, and populations collapse. Springtails, mites, beneficial nematodes, and the fungal networks that extend a root's reach all need the same air-filled pore space; in its absence they're replaced by the anaerobic pathogens that cause damping-off and root rot.

The microbial nitrogen and phosphorus cycle

Turning organic nitrogen into a form a plant can use is an aerobic relay: nitrogen-fixing bacteria seed the pool, protozoa graze bacteria and excrete ammonium, and nitrifying bacteria — the slowest-growing, most stress-sensitive link in that chain — convert ammonium to nitrate. Cut the oxygen and that last step stops first. Ammonium piles up un-converted, nitrogen-hungry crops stall, and the acid-tolerant weeds that match this chemistry move in. The same oxygen dependency holds back the fungi and bacteria that unlock phosphorus from mineral form.

Magnesium also competes directly with potassium, calcium, and ammonium for the same uptake channels in a root membrane — and it occupies those channels slower but holds on, so in a magnesium-rich soil solution the channels stay blocked. The plant can show a potassium or calcium deficiency in its tissue even when the soil test says there's plenty in the ground.

Trophobiosis — why pests and disease track the mineral chemistry

A stressed, mineral-imbalanced plant accumulates free amino acids and simple sugars instead of building them into proteins and complex carbohydrates — and free amino acids and simple sugars are exactly what aphids, slugs, and fungal pathogens are looking for. This is the framework French agronomist Francis Chaboussou called trophobiosis, later extended by researchers including Larry Phelan, Olivier Husson, and John Kempf: pest and disease pressure is very often a nutrition problem wearing an insect's face. An aphid outbreak or a run of powdery mildew is frequently the soil talking.

Recovery came from an ordered sequence, not any single input. Skip a step or do them out of order and the later steps stop working.

1. Remove the compaction zone. The compaction layer has to be broken all the way through, and a normal tillage pass rarely reaches it. Most ground needs deep rippers; this plot was opened with a backhoe, to an average depth of 24 inches. The goal is open pathways for air, water, and roots through the full depth of the sealed layer. After that, any future disturbance — tillage included — has to be done carefully and in a way that helps the soil, or it can undo the progress.
2. Rest, eight weeks (test only). Because this was a test, the plot was left alone for eight weeks after the compaction was removed, so the effect of opening the soil could be seen on its own, before any amendments went in. In practice, amending right away is recommended.
3. Balance the minerals, calcium-led. Modest, targeted amendments broadcast on the surface — never tilled in — calibrated in parts-per-million to this soil's own exchange capacity: calcium moved toward a 2000 ppm target, potassium toward 250, phosphorus toward 100, plus sulfur, boron, zinc, and cobalt. Every mineral on the Logan Labs report was taken into account, both in its own amount and in its ratio to the others. The ones named here were addressed first because their numbers and ratios made them the priority; the rest are managed the same way as the soil moves toward balance. Magnesium got nothing. Leaving it alone while everything else came up was the point — it's how the ratio corrects.
4. One biological inoculation. A single application of a liquid, biologically complete compost extract — not bulk compost, not manure — carrying bacteria, fungi, protozoa, and beneficial nematodes into soil that, thanks to steps one through three, finally had open pore space and stable structure for that biology to actually establish in.

Removing the compaction without chemistry fails. Chemistry without biology fails. Biology poured onto soil with no structure to hold it fails too — it has nowhere to live. The sequence is the intervention, not any one ingredient in it.

That line is the reason this plot never received a truckload of anything. Years of consulting on farms and gardens across the region turned up the same pattern over and over: well-meaning compost, applied in bulk, with nobody able to say exactly what was in it, quietly working against the soil it was meant to help — carrying salts, imbalanced minerals, or simply the wrong microbial mix for what that ground needed. Compost is a process, not a product, and the word alone tells you nothing about what's actually going into your soil. This plot shows the alternative: functional, productive soil built without bulk compost — one targeted extract, applied once, onto ground that had already been given the mineral balance and physical structure to make good use of it. If you use compost — at farm or garden scale — know its inputs, and ask for elemental and microbial analysis the same way you'd ask for a soil test. "Compost" is not a specification.

"Compost" tells you nothing until you've seen the lab numbers. Below are two real, current lab reports on locally available compost — Cold Creek Compost (Ukiah) and Pacific Organics Solutions (Ukiah). Both are legitimate, EPA 503–passing products. Neither is wrong to use. One caution before reading the numbers: compost labs report total concentrations — the whole sample dissolved in strong acid — while the soil on this page was tested with Mehlich-3, which measures only the plant-available portion. The two can't be compared number for number. What can be compared is the balance inside each compost, and that's where the story is.

These are total concentrations, so they aren't shown against this soil's Mehlich-3 targets. Both composts passed all ten EPA 503 heavy-metal limits with room to spare; "passes EPA 503" means it's legal for general land application — those limits weren't set with food gardens in mind — not that its mineral profile matches what any particular soil needs.

That's the number that matters most here. A soil starting at 2.33 : 1 needs calcium brought up without adding magnesium alongside it. Either compost would nudge the ratio in the right direction, but only partway — and in bulk it also brings phosphorus, potassium, iron, and trace metals in amounts no soil test called for. None of that shows up if the bag just says "compost." It only shows up on a lab report — which is exactly why the standing advice is to ask for one before spreading anything in bulk, at farm or garden scale.

These aren't the only numbers that matter

The numbers above were chosen to show one thing: how hard it is to keep your ratios in balance when an input arrives carrying large amounts of minerals you didn't ask for. But minerals are only part of the picture. A full compost report lists many things a farm or garden soil doesn't need at all — metals, salts, and other compounds — and every one of them has to be managed once it's in the ground. You don't need to read all of it to use this page, but it's here if you want it: the complete lab reports are in Full Lab Reports.

What worked

pH climbed a full 1.2 points, past the threshold where nitrifying bacteria come back online, aluminum stops being soluble, and phosphorus releases. Calcium nearly doubled. Exchangeable hydrogen fell from 36% to 7.5% — those binding sites are now holding minerals instead of acid. Phosphorus increased tenfold. Organic matter rose 56%, the visible signature of aerobic biology switching back on. Estimated nitrogen mineralization rose 22%, from 78 to 95 lb/acre.

What didn't finish yet

The Ca:Mg ratio reached 4.14:1 — real progress, but short of the 6.5–7:1 structural target. Potassium is still 122 ppm under target, held back by residual magnesium still competing for uptake. Bulk density actually ticked up slightly, 0.92 to 0.96 g/cm³ — structure rebuilds over years, not one season, and that lag is expected, not a failure. Sodium doubled, 11 to 22 ppm — still safe, but worth watching if it's coming in through irrigation water.

Year two is chemistry finishing its work. Year three is where biology should take over as the driver. The structural payoff — falling bulk density, stable aggregates, soil that finally feels different in your hand — is a year three and year four story.

Same parent soil, same season, three different management histories side by side — the cleanest way to see what each variable actually contributes.

Plot 1 — Native grazing land

165 years of continuous grazing on this exact parcel. About one inch of true topsoil over a compaction pan; a spade under normal body weight stops at roughly that depth. Bearing resistance over 350 psi at the surface.

Plot 3 — Garden test plot

The intervention sequence, one season on. The full spade blade goes in with ease and lifts clean. A flexible fiberglass probe pushes to 24 inches — the average depth the compaction layer was broken. The top inch or so stays slightly firm from calcium lost to irrigation water; below that it probes easily, and it still does years later. Uniform chocolate-brown color to three feet — no compaction band, no color break.

Plot 2, the middle case — a pile of excavated subsoil left alone with a self-planting cover crop — whatever wind-blown seed found it: thistle, forbs, grass, classic ecological succession with zero chemistry or inoculation. By late winter it was the most carbon-rich soil on site, chocolate-brown three feet down, ~30 worms per shovelful. But it couldn't hold that gain — by May the same soil had hardened back to roughly one inch of spade penetration and the worms had moved out. Biology built structure all winter with no calcium to bridge it in place, and lost almost all of it by early summer. Biology alone, without chemistry, doesn't last.

Early structure, first two months

Soil holding together around a clover root mass, with root channels running through it and no dust falling away. A photo can't show whether those larger clumps are held by fungi and roots or are still partly magnesium-tight — a slake test tells the difference.

The same ground, in production

Vegetable growth held up through the full season rather than fading with summer heat — the practical payoff of structure that persists past spring.

Across all three: only the garden plot combined winter structure with summer persistence. Native ground failed on every metric. The self-planted cover-crop pile won in winter and lost by summer. Chemistry-led, lightly inoculated soil held its gains year-round.

• Calcium, never magnesium. Any further liming stays calcitic — calcitic lime, gypsum, or oyster shell flour — no dolomite or other magnesium-bearing inputs. Target: nudge the ratio from 4.14:1 toward 5.5:1 this year; correcting faster risks overshooting into a magnesium deficiency in the plant tissue.
• Potassium as sulfate, not chloride. Potassium sulfate closes the remaining 122 ppm gap while adding sulfur for amino-acid synthesis, split across two applications rather than one heavy dose.
• Carbon, the right kind. Finished compost is fine; fresh manure isn't — it would spike nitrogen and destabilize the aerobic community that's still consolidating. Cover crop biomass gets cut and left as surface mulch, not tilled in.
• Watch the irrigation water. Sodium doubled, 11 to 22 ppm. Still low, but worth a water test for sodium, bicarbonate, and electrical conductivity before it becomes a slow, invisible drag on the recovery.
• Test the plant, not just the soil. A soil report says what's present. Sap or tissue testing mid-season says what the plant can actually get to — the real check on whether residual magnesium is still blocking potassium and boron at the root.

Every number from every test used on this page, exactly as the labs reported it. You don't need any of this to use the kit — it's here for anyone who wants to check the work or read further.

Two kinds of test, not comparable number for number. The soil reports are Mehlich-3 (Logan Labs): a mild extraction that measures the plant-available portion. The compost reports are totals (Soil Control Lab, McCampbell Analytical): the whole sample dissolved in strong acid (EPA method 3050B).

Soil — the demonstration plot (Logan Labs, Mehlich-3)

Compost — Soil Control Lab (totals)

Cold Creek Compost, Ukiah: sample AB21125-FP, reported Feb 27, 2026, and sample TopPad 530247, reported June 2024. Pacific Organics Solutions, Ukiah: sample "Pacific Organics – Medium," reported Jan 29, 2026. Dry weight unless noted. The last column is the US EPA 503 limit, where one exists.

Pacific Organics — second-lab metals check (McCampbell Analytical)

Same compost, tested independently for California Title 14 metals (extraction SW3050B, ICP-MS SW6020B, mg/kg dry weight, collected Jan 12, 2026). Soil Control Lab's figures are shown alongside. Most agree closely; nickel differs by about half (48.4 vs. 32).

All soil data: Logan Labs, LLC, Mehlich III extraction. Two samples, same site (R2, Comptche, Mendocino County, CA), same 0–6" depth, same point in the seasonal cycle — April 2022 baseline, April 2023 follow-up.

Published by the Organic Regenerative Certified Apprenticeship (ORCA), a California nonprofit founded 2025 (CA Division of Apprenticeship Standards #101310; U.S. DOL registered), in partnership with its affiliated consulting practice, Surprise Valley Agroecology, LLC. Released as open educational material for farmers, gardeners, apprentices, and community partners.