Short answer. CO₂ is the raw material of photosynthesis. In cannabis leaves, raising CO₂ from ambient to 750 ppm increased net photosynthesis by about 50 % and water-use efficiency by about 111 % [1]. It only turns into yield when light is the strong side of the equation (roughly 900–1,500 µmol/m²/s in flower) and the room is sealed so the gas stays put. Typical target: 800–1,200 ppm during lights-on only. Safety: 5,000 ppm is the 8-hour workplace limit; 40,000 ppm is immediately dangerous to life [4][5].
What's happening
At ambient CO₂ (~420 ppm today) and high light, the enzyme Rubisco runs short of CO₂ and wastes energy on photorespiration. More CO₂ fixes more carbon per photon and lets stomata close slightly (cannabis at elevated CO₂ showed lower stomatal conductance and transpiration and higher water-use efficiency [1][2]). The plant can then use more light and tolerate slightly higher temperatures.
Cannabis yield kept rising linearly with light up to 1,800 µmol/m²/s even at ambient CO₂ in one well-run indoor trial [3]. CO₂ is therefore a multiplier on light, not a substitute for it.
Numbers that matter
| Situation | CO₂ target (lights-on) | Notes |
|---|---|---|
| Light below ~600 µmol/m²/s | Ambient, just ventilate | Little benefit; spend on light first |
| Flower 700–900 µmol/m²/s | Ambient to 800 ppm | Marginal; only if sealed |
| Flower 900–1,500 µmol/m²/s | 900–1,200 ppm | Raise day temperature 2–3 °C (4–5 °F) |
| Above 1,500 µmol/m²/s | 1,200–1,500 ppm | Commercial practicebeing reviewed |
| Lights-off | None | Plants respire, they don't fix carbon at night |
| Last 1–2 weeks of flower | Often tapered to ambient | Grower practice, no published evidence |
| Level | Meaning |
|---|---|
| 5,000 ppm | 8-hour time-weighted workplace limit (OSHA PEL, NIOSH REL) |
| 30,000 ppm | NIOSH 15-minute short-term limit |
| 40,000 ppm | IDLH (immediately dangerous to life or health) |
How to do it
- Seal first. Exhaust fans must be off (or on a controller that locks out CO₂ when they run). Otherwise you are heating the street with gas.
- Measure with an NDIR sensor at canopy height, not near the emitter. Calibrate yearly (outdoor fresh-air reference ≈ 420 ppm).
- Choose a source: compressed cylinder with regulator and solenoid (clean, no heat); combustion generators add heat and water vapour and need a CO monitor; fermentation/bags give small, uncontrollable amounts. Combustion indoors in a small space is not recommended.
- Control: dose on a controller with a high cut-off; release from above the canopy (CO₂ is denser than air) with circulation fans running.
- Safety: a separate CO₂ alarm at head height outside the canopy; ventilate before entering a sealed room; never enter if the alarm is above 5,000 ppm; cylinders strapped upright.
How to confirm it worked
Readings hold within ±100 ppm of target during lights-on; plants tolerate the higher light without bleaching; dry yield per m² and per kWh improve over a matched run.
Common mistakes
- Running CO₂ with the exhaust on.
- Adding CO₂ to a 400 µmol/m²/s tent.
- Dosing at night.
- Placing the sensor next to the emitter.
- Forgetting a people-safety alarm.
Evidence
Leaf-level photosynthetic gains in cannabis are Supported [1][2] (single group, a few cultivars, gas exchange not yield). The yield benefit under high light is Supported by horticultural consensus and the light-response work [3]; a full cannabis CO₂ × light yield trial is still missing. Exposure limits are Established regulation [4][5].
