How Cannabis Is Fed: Nutrient Systems, Not Labels

R06 Verified Synthesis — Wave 2, systems comparison

A practitioner-record passage is withheld from the public copy.

The question, and why the labels dissolve

R06 asked what the evidence says about feeding cannabis “organically” versus with “synthetic” or mineral fertilizer. The honest answer is that the literature does not study those categories. It studies nutrient systems — chemical form, concentration, timing and release dynamics, substrate, root-zone chemistry, irrigation water, delivery architecture, genotype and the co-inputs that may accompany a nutrient program — and outcomes attach to those variables, not to the label on the product [R06-C001]. No located study compares “organic” with “synthetic” as such. The one controlled comparison that exists compares two specific commercial programs, and it is best read as a demonstration of how the categories fail rather than as a verdict between them [R06-C006].

The categories fail chemically before they fail agronomically. Certified-organic liquid fertilizers deliver soluble ions and are formulated with chelating agents or added salts; mineral programs can bundle organic chelators, kelp, sugars and microbes. In the one controlled trial, the “organic” stock included Epsom salt and the “mineral” stock was ammonium sulfate plus commercial soluble products [R06-C002]. Rock phosphate, permitted in organic systems, is a mined mineral. “Organic” does not mean non-mineral, and “synthetic” does not mean free of biology.

They fail legally too, in a way that matters for anyone who wants to put the word on a label. Cannabis cannot carry the USDA “organic” designation because it is ineligible under the National Organic Program; California’s OCal is a state “comparable-to-organic” certification with its own rules [R06-C004]. In Canada, cannabis cultivation is outside the Canada Organic Regime and cannot bear its logo; any “organic cannabis” there rests on a provincial or private scheme that must be named [R06-C005]. When a paper, a product, or a grower says “organic,” it can mean a certified input program, a manufacturer’s label, a biology-dependent feeding philosophy, or just the author’s category. R06 reconstructs the chemistry before it interprets the word.

The one comparison, and the three outcomes it separates

The keystone experiment is Massuela and colleagues’ 2023 indoor trial: one CBD-type genotype in a peat-and-perlite substrate, an organic and a mineral program each formulated to the same nominal nitrogen, phosphorus, potassium, calcium and magnesium concentrations at three dilutions, mycorrhizae applied to everything, no leaching allowed. The mineral program produced 20.9% more inflorescence dry matter (23.7 versus 19.6 g per plant). The organic program produced 12% higher CBD concentration (6.5% versus 5.8%). Total CBD per plant stayed higher under mineral feeding (1,319 versus 1,197 mg) [R06-C006].

Three outcomes, three directions. That is the interpretive rule for everything that follows in this report and much of what follows in the corpus: cannabinoid concentration, flower yield, and total cannabinoid yield are different numbers that routinely move against each other. A treatment that raises potency by shrinking the plant has not made more cannabinoid; a treatment that grows a bigger plant has not necessarily made stronger flower. Any claim about “quality” or “yield” that does not say which of the three it means is not yet a claim.

The equivalence basis matters as much as the numbers. The programs were matched on nominal solution concentration, and the plants still took up nitrogen differently: uptake efficiency was 0.55–0.58 under the organic program and 0.71–0.74 under mineral [R06-C003]. Equal total nitrogen, equal EC, equal label rate, and equal plant-available nitrogen are four different things, and matching on one does not deliver the others. The authors’ proposed mechanism, that mineralization-dependent supply fell behind sink demand once the flowers became the main sink, is consistent with the uptake data and is the cannabis instance of the nitrogen-timing asynchrony R01 found across organic agriculture [R06-C018; R01-C008, R01-C020]. It is a plausible mechanism from one experiment with one organic product, graded separately from the outcome, and not a law.

The same trial carries a quieter finding: the middle nitrogen rate delivered about 95% of the CBD yield of the highest rate with one-third less fertilizer [R06-C007]. That is a nutrient-use-efficiency result, not a footprint result, but it points the same direction as everything else in this report — more is rarely better past sufficiency.

Dose and form: where the evidence is actually strong

The literature is far stronger on nutrient chemistry than on categories, and most of that strength comes from one laboratory’s series of dose-response experiments in medical cannabis (see the same-lab note above).

Nitrogen rate follows the pattern the three-outcome rule predicts: vegetative development was optimal at 160 mg/L N and excessive above it; in flowering, yield rose with nitrogen to about that level while THCA, CBDA and terpenoid concentrations declined as nitrogen rose [R06-C009]. Organic liquid fertilizer has its own dose curve — a vegetative optimum near 389 mg/L N in coir, flowering optima that depended on the substrate, and cannabinoid dilution at higher rates [R06-C008] — which by itself retires the folk claim that you cannot overfeed with organics, at least in the weak sense that organic programs have optima.

Nitrogen form is the sharpest result in the corpus for the “an ion is an ion” argument. At a fixed 200 mg/L of total nitrogen, raising the ammonium fraction to 50% cut inflorescence yield by 46% relative to nitrate-only nutrition and produced foliar toxicity; 100% ammonium killed the plants; 10–30% carried smaller penalties [R06-C010]. Same total nitrogen, different species, half the crop. Plants do acquire nutrients as defined chemical species rather than as marketing categories, but nitrate and ammonium are different species handled by different transport and assimilation machinery, and form, ratio, availability, root-zone chemistry and delivery materially change the result.

Phosphorus is the best-replicated finding in the report and the one with the clearest practical edge. Thirty mg/L P gave about 80% of maximum yield and phosphorus above 5 mg/L reduced THCA and CBDA by up to 25%; raising P from 25 to 75 mg/L produced no yield or CBD gain while leachate phosphorus rose twelvefold and use efficiency fell from 80% to 31%; raising P from 15 to 90 mg/L and doubling solution EC in closed hydroponics produced no gain either; a response-surface model placed the optimum at 59 mg/L P [R06-C011]. Three laboratories, four experiments, one direction, and one preserved disagreement about where exactly the ceiling sits. Above sufficiency, added phosphorus did not improve flower or cannabinoid output in the located studies, and in one experiment the extra input sharply increased phosphorus loss to leachate. That is the evidence the “bloom booster” claim has to answer to: the products were not tested as products, but the nutrient they add was, and it did not do what the label promises.

Potassium and magnesium round out the picture. Potassium at 15 mg/L was deficient, 60–175 supported yield, 240 produced oversupply symptoms in one of two cultivars, and most cannabinoid and terpenoid concentrations declined as potassium rose [R06-C012]. Magnesium was optimal at 35–70 mg/L in vegetative growth, with a small physiological cost and no biomass penalty at 140, while at the reproductive stage high magnesium reduced reproductive biomass by about 12% [R06-C013]. “More magnesium is better” is not supported; calcium was not studied, so the commercial “Cal-Mag” bundle is only half addressed. Single-element deficiency, meanwhile, cut floral yield by 33–72% for every nutrient tested except iron and manganese — and only iron and manganese were tested among the micronutrients, so the finding stops there [R06-C014].

Two systems-level results belong here. Augmenting phosphorus, potassium and iron above a standard regime did nothing significant to inflorescence mass or cannabinoid concentration, while the fertigation architecture did matter: recirculating delivery significantly raised total THC yield relative to drain-to-waste, even though inflorescence mass did not conclusively differ [R06-C015]. And solution EC, the number growers watch most, is an operational control and not an equivalence measure: doubling it at matched phosphorus bought nothing, and it says nothing about nitrogen form or availability schedule [R06-C021].

Flushing: a disagreement the corpus can finally state properly

Pre-harvest flushing is the practice the archive taught most confidently and the evidence treats least kindly — but the evidence is now more interesting than “it doesn’t work.”

For short, water-only nutrient withdrawal of 7–14 days, no study has shown a change in flower mineral content, yield, potency, terpenes, taste, smoothness or ash color. The one industry trial found no significant differences across 0, 7, 10 and 14-day flushes, with flower iron and zinc actually rising; a blind expert panel trended toward preferring the unflushed flower (36% rated it “smooth”), not significantly; and a Guelph master’s thesis reports flushing “ineffective in removing any significant amount of nutrient from the bud” [R06-C024; R08-C026]. Neither source is peer-reviewed.

A peer-reviewed 2025 greenhouse study reports something different: under prolonged leaching — up to four weeks of flushing to a target leachate EC of 0.2 mS/cm — inflorescence nitrogen, phosphorus and potassium concentrations declined with duration, inflorescence mass declined, THC was unchanged, and CBD concentration rose in one of two high-CBD cultivars [R06-C025]. R06 has that paper only as an abstract; the magnitudes stay out of this synthesis until the full text is read.

The two findings do not contradict each other because they are not the same treatment. One is a short water-only withdrawal; the other is a prolonged leaching regime to a low-EC target, and duration was not isolated experimentally. The defensible statement is regime-dependent: the located evidence on short 7–14-day flushes has not demonstrated meaningful nutrient removal from flower, and one longer leaching regime did move tissue chemistry at a yield cost. “Flushing pulls nutrients out of the flower” is neither confirmed nor refuted; it is disputed by regime, and it now has a peer-reviewed anchor on the side the archive never had.

A practitioner-record passage is withheld from the public copy.

Safety: no halo in either direction

Canada’s regulator names fertilizers and supplements as pathways for heavy metals, pathogens, pesticide residues, dioxins and furans into cannabis, regardless of whether the input is organic or conventional [R06-C029]. The Health Canada-led market comparison found more arsenic, cadmium, lead and mercury in illegal dried cannabis and about threefold more chromium in legal product; the authors propose fertilizer or micronutrient additives as a possible chromium source but did not test it [R06-C030; R03-C025]. In one general survey of fertilizers and amendments, trace-element loads generally decreased from rock phosphate to sewage sludge to manufactured phosphorus fertilizer to organic amendments, and commercial-grade phosphorus fertilizer raised seed cadmium in flax — but no cannabis study traces a known input metal load through the substrate into compliant or failing flower, so input contamination and flower contamination remain separate claims [R06-C031]. Microbial compliance depends on the assay, and the pathology literature attributes threshold failures to conventional and organic methods alike [R06-C032; R08-C022]. “Organic is safer” and “mineral is cleaner because it’s controlled” are both unsupported as category claims. Compliance and contaminant risk are properties of specific inputs, batches, matrices and tests — not of an organic or mineral label.

What the archive taught, and what R06 says back

A practitioner-record passage is withheld from the public copy.

The ratios the archive taught (3-1-2 in veg, 1-4-5 in bloom) were never concentrations, so the phosphorus ceiling cannot be checked against what was actually fed — only against the doctrine, which it contradicts. And a 2017 course outline promised two sessions, “Comparing Food/Fertilizer Systems” and “Passing Lab Tests, Reading Labels, and Products to Avoid,” that were never written. R06 is those sessions, nine years late, with a ledger.

The public claims, calibrated

Organic tastes better: unsupported — no blinded sensory trial in any class; chemistry differences exist [R06-C022, C023]. Synthetic residues make harsh smoke and white ash proves a clean flush: unsupported in cannabis [C026, C028]. Organic yields less: unsupported as a category claim, observed in one program comparison [C006, C003]. An ion is an ion: misleading as agronomy [C010]. You can’t overfeed with organics: unsupported — organic programs have optima, and no organic-program toxicity trial exists [C008]. Flushing pulls nutrients out of the flower: regime-dependent, not refuted [C024, C025]. Organic is safer, or mineral is cleaner: no halo either way [C029–C031]. Synthetics kill the microbiome: too broad; no cannabis threshold located [C019, C021]. High-P bloom boosters raise yield and resin: not supported above sufficiency [C011]. Organic or living soil has a lower footprint: unverified; no matched LCA, and per-area versus per-product rankings can reverse [C034; R04-C029].

What we do not know

No cannabis mineralization curves exist for meals, manures, guanos or composts [R06-C019]. No tracer study shows cannabis taking up nutritionally meaningful intact amino acids from the products sold for that purpose [C020]. No trial isolates the microbes, humics, kelp and sugars from the nutrient source they are bundled with [C036]. No peer-reviewed short-flush trial with tissue chemistry and blinded sensory exists, and the one peer-reviewed leaching paper is unread beyond its abstract [C024, C025]. No fertilizer-to-flower metal transfer study exists [C031]. No economics dataset compares programs on cost per kilogram or per gram of cannabinoid [C035]. No matched life-cycle assessment compares fertility categories [C034]. None of the located trials reports irrigation-water chemistry in a form the ledger could reconstruct, so every dose-response row carries “water-undisclosed.” And copper residue from organic-approved fungicides on a combusted crop was not searched by either researcher — an open question, not an empty one.

Implications for practice, hedged to the grades

Target evidence-supported sufficiency ranges rather than assuming additional nutrient supply improves performance; phosphorus provides the clearest cannabis example, while nitrogen, potassium and magnesium also show non-monotonic dose responses [C009–C013]. Where nitrogen form is disclosed, consider it alongside total N: in one controlled cannabis experiment, high ammonium fractions at the same total nitrogen sharply reduced yield [C010]. Decide which outcome you are growing for, because potency, flower weight and total cannabinoid do not move together [C006]. Do not assume a short pre-harvest flush provides the claimed quality or nutrient-removal benefits; the located 7–14-day evidence did not demonstrate them, and a separate prolonged-leaching study, known so far only from its peer-reviewed abstract, reports tissue changes alongside yield loss [C024, C025]. Assume no input is safe by category, and test batches, not philosophies [C029–C031]. And when a claim says “organic” or “synthetic,” ask what was actually in the bottle, at what rate, in what water, in which plant — the questions the literature answers and the labels do not [C001–C003].

What would change these conclusions

A replicated, multi-genotype organic-versus-mineral trial with analytically characterized inputs and equal plant-available nitrogen; cannabis-specific mineralization curves for the common organic inputs; a co-input isolation trial; a peer-reviewed short-flush study with tissue chemistry, combustion products and a blinded panel; the full text of the 2025 leaching paper; a fertilizer-to-flower metal transfer study with known input loads; and any economics on a per-product unit. Until then, the corpus’s line is that cannabis nutrition is well measured as chemistry and unmeasured as ideology.

Changelog

• v1.0 (2026-09-07) — Danny sign-off after ChatGPT’s editorial pass (approved with minor calibrations; no structural or grading changes). Five wording edits: “every bottle” → “may accompany a nutrient program” and “mineral programs bundle” → “can bundle”; the phosphorus conclusion narrowed to dose evidence, with the bloom-booster inference stated separately; the short-flush conclusion keeps the bounded-search wording (“the located evidence on short 7–14-day flushes has not demonstrated…”); practice implications hedged to evidence-supported ranges, single-experiment nitrogen-form evidence, and Alden & Faust as abstract-level; the safety aphorism tightened to inputs, batches, matrices and tests.
• v0.1 (2026-09-07) — initial draft from ledger v1.0.