Air: VPD, stomata, leaf temperature and CO₂

Sooner or later every home grower stares at a little digital hygrometer and asks: is this humidity right? The answer from forums and grow apps is usually not a humidity at all but three letters, VPD, and a coloured chart of numbers in kilopascals. The tutorial in this app does the same. It asks for a vapour pressure deficit of 0.4–0.8 kPa for seedlings, 0.8–1.2 kPa in vegetative growth, 1.0–1.4 kPa through the flip and stretch, and 1.2–1.6 kPa while the buds fill out, with relative humidity falling from 65–75% at the start to 40–50% in late flower.

Those numbers are practice. They come from growers' experience in small rooms and tents, and they are sensible as a starting point. But what does VPD actually measure, what does a leaf do in response to it, and how much of the colour chart rests on research? This article goes through the physics and the plant physiology, the handful of studies that speak to it, and then a second question that often follows: whether adding carbon dioxide (CO₂) is worth it in a 1×1 m tent. One point before we start: in the UK, growing cannabis, even with a prescription, requires a Home Office licence. This article is educational.

What VPD actually measures

Air can hold only so much water vapour before it saturates and water starts to condense. That ceiling, the saturation vapour pressure, depends on temperature alone, and it rises steeply, not in a straight line, as the air warms (Grossiord et al., 2020). The amount of vapour actually in the air is the actual vapour pressure. Vapour pressure deficit is simply the gap between the two: how much more water the air could take before it is full. In words, VPD equals the saturation vapour pressure at the air temperature minus the actual vapour pressure; or, since relative humidity is the actual vapour pressure as a percentage of saturation, VPD equals the saturation vapour pressure multiplied by one minus the relative humidity as a fraction (Grossiord et al., 2020).

Relative humidity on its own therefore says little about how hard the air is pulling water out of a leaf. The same 50% means a very different thing at 18 °C and at 28 °C. Using the standard formula for saturation vapour pressure that Inoue et al. (2021) used to set their treatments, the arithmetic looks like this:

The table is calculated, not measured, but it shows the point. Warming the air from 20 °C to 28 °C at a constant 50% raises the deficit by more than half. Grossiord et al. (2020) make the same point on a larger scale: in their worked example, a heat wave that lifts the air from 25 °C to 35 °C at a constant 20% relative humidity pushes VPD from 2.5 to 4.5 kPa.

The same arithmetic explains why lights-off matters. Suppose a tent sits at 25 °C and 50% with the lamp on, a VPD of about 1.6 kPa. If the lamp goes off, the air cools to 18 °C and no moisture is removed, the relative humidity climbs to roughly 77% and the deficit falls to under 0.5 kPa. Nothing was added; the air simply got colder. In a real tent the extractor and the room around it soften this, but the direction is always the same: cooler nights mean wetter air unless something takes the water out.

Air VPD and leaf VPD

There is a subtlety that matters for anyone reading a VPD chart. What the leaf feels is not the deficit of the bulk air but the difference between the vapour pressure inside the leaf, usually taken as saturated at leaf temperature, and the vapour pressure of the air around it. Physiologists call this the leaf-to-air VPD, and Grossiord et al. (2020) describe it as the more accurate value for judging a leaf's water balance, because a leaf can be warmer or cooler than the air around it.

A degree or two makes a difference. In a calculated example, air at 26 °C and 50% has a VPD of about 1.7 kPa; a leaf sitting at 24 °C in that air experiences about 1.3 kPa, and one at 28 °C about 2.1 kPa. This is one reason the tutorial's humidity and VPD columns do not line up exactly. Its seedling stage asks for 24–26 °C, 65–75% RH and 0.4–0.8 kPa, but by air temperature alone, 25 °C and 70% works out at about 0.95 kPa. The two columns agree only if the leaves are assumed to be a couple of degrees cooler than the air. Neither column is wrong; they are approximate, and it helps to know which temperature a meter or chart is using.

How a leaf answers: stomata

Leaves breathe through stomata, microscopic pores on the leaf surface flanked by two guard cells. Through the same pore, carbon dioxide diffuses in for photosynthesis and water vapour diffuses out. A plant cannot take in CO₂ without losing water, and VPD sets the price.

Grossiord et al. (2020), a long review in New Phytologist that draws on studies of many species and ecosystems, none of them cannabis, summarise what happens as VPD rises. Stomatal conductance, the ease with which gas passes through the pores, falls as stomata partly close. Transpiration still rises with VPD in most species, at least up to a threshold, because the air is pulling harder; beyond that point it levels off or even drops. Photosynthesis falls too, but less steeply than conductance, so the plant gives up a little carbon to save a lot of water. The authors report one telling detail about timing: when VPD jumps suddenly, stomata in seed plants briefly "pop open" before closing, within two to 25 minutes, and then settle to a new, narrower opening.

How much photosynthesis is lost varies widely. In the five species the review cites, photosynthesis declined by between 9.4% (maize) and 36.6% (castor bean) under daily swings in leaf-to-air VPD (Grossiord et al., 2020). Stomatal sensitivity differs between species, within species, and even between leaves on the same plant. Nothing in the review gives a figure for cannabis.

High VPD is not the only problem. The review also describes the only field experiment the authors know of that manipulates VPD on its own, by misting forest plots in Estonia to make the air more humid: the very humid air reduced transpiration, growth and photosynthesis. The authors are careful to add that some of this may have been caused by water condensing on the leaves from the misting system, which can itself impair stomata (Grossiord et al., 2020). Either way, more humidity is not automatically kinder to a plant.

Steady air beats swinging air

Most growers control humidity with devices that switch on and off: a humidifier with a set point, a dehumidifier, an extractor fan on a thermostat. The average reading can look perfect while the air see-saws around it. Does that matter?

Inoue et al. (2021) tested exactly this. They grew romaine lettuce in rockwool in a growth chamber at a light intensity of 200 µmol/m²/s, a 16-hour day, 24 °C and ordinary air at 400 ppm CO₂. Both groups of plants had the same average VPD, about 1.0 kPa, and the same average humidity and temperature. What differed was the swing. In the "moderate" treatment, VPD cycled between 1.32 kPa (55% RH) for seven minutes and 0.86 kPa (72% RH) for three. In the "drastic" treatment it cycled between 1.63 kPa (42% RH) for six minutes and 0.63 kPa (80% RH) for three.

When the researchers watched individual leaves in a gas-exchange chamber, the drastic swings caused stomatal conductance and CO₂ uptake to decline gradually after the first hour or two, while the moderate swings left them steady. After three weeks of growth, plants under the drastic regime had 15% less shoot dry weight and 29% less leaf area than plants under the moderate regime, even though they had the same number of leaves; their leaves were simply smaller. Before the third week there was no clear difference. The authors conclude that VPD should be controlled "not by intermittent regulation but by continuous regulation" (Inoue et al., 2021).

Some caution is in order. This is lettuce, not cannabis, at modest light, with cycles lasting minutes, and the growth comparison rests on a small number of plants (six biological replicates across two runs). Three of the authors work for chemical companies. Still, the physiology is general and the message travels well: a steady environment around a sensible average is better than a perfect average built out of big swings. Inoue and colleagues also note, citing earlier work, that most plants grow well between about 0.5 and 0.8 kPa, and that transpiration falls in most crops above about 2 kPa (Inoue et al., 2021). Those figures come from crops other than cannabis.

Leaf temperature under LED

A common belief among growers is that LED lamps keep leaves much cooler than old high-pressure sodium (HPS) lamps did, so the air can be run warmer. Nelson and Bugbee (2015) put numbers on this. They did not grow plants; they built an energy-balance model of a single top leaf and fed it with measured radiation from four sources: clear-sky sun in a field, sun in a glass greenhouse, a 1000 W double-ended HPS lamp and a 400 W passively cooled red-blue LED fixture, both indoors and both among the most efficient fixtures then available. Leaf light absorption was measured on tomato, pepper, basil and broccoli, and the model was run at an air temperature of 25 °C, with the different sources compared at equal light intensity.

Their headline result is that water matters more than the lamp. If plants were not water-stressed, leaves under all four sources were typically within 2 °C of air temperature. The ranking never changed (HPS warmest, then greenhouse sun, then LED, then open-field sun), but under typical indoor conditions LED leaves came out only about 1.3 °C cooler than HPS leaves. The difference arises because LED fixtures shed much of their heat upwards by convection from their heat sinks, while HPS lamps radiate more heat down towards the plants.

The picture changes when transpiration stops cooling the leaf. In a "near worst-case" run with water stress, high light and still air, leaves rose 6, 8, 10 and 12 °C above air temperature under field sun, LED, greenhouse sun and HPS respectively (Nelson and Bugbee, 2015). The authors add that thick, dense structures that transpire little, and they name cannabis flowers among them, behave more like the worst case than a thin leaf does, and that LED technology "has the potential to reduce heating" of such structures.

For a grower this means LED leaves are not reliably several degrees cooler than the air, as is often claimed. A well-watered, actively transpiring canopy under LED sits close to air temperature. A canopy that reads clearly warmer than the air on an infrared thermometer is a warning sign: transpiration has slowed, through dry roots, damaged roots, very stagnant air or stomata that have closed for some other reason.

Humidity, VPD and mould

The strongest practical reason for the tutorial's dry late-flower air is not photosynthesis but grey mould. Botrytis cinerea causes bud rot, one of the most damaging diseases of cannabis flowers. Mahmoud et al. (2023) review what is known, mostly from greenhouse cannabis and from other crops such as grapes, strawberries, tomatoes and cucumbers. They report that the fungus can destroy cannabis flowers rapidly when relative humidity is above 70% and temperatures are moderate, around 17–24 °C.

Three points from the review link directly to VPD. First, low VPD, below about 0.4 kPa, means the air is close to saturation and moisture stays on the leaf, and that encourages fungi; this threshold comes from work on tomato and cucumber, not cannabis. Second, liquid water on plant surfaces is a key factor: dew forms when a surface cools to the dew point, and the authors recommend avoiding sudden changes in temperature and humidity that can cause condensation on the flowers. Third, air movement helps: circulating fans reduce humidity around the plants and prevent condensation (Mahmoud et al., 2023).

The review also reports a cannabis trial with two genotypes at one or two plants per square metre. Bud rot was significantly lower at the lower density, but yield per square metre fell by 18–23%, so the authors doubt it would pay commercially (Mahmoud et al., 2023). In a 1×1 m tent, the choice between one and two plants is exactly that trade-off.

The tutorial's bands, read against the research

Put side by side, the tutorial's stage targets look like this:

The logic of the progression is easy to follow. Seedlings and fresh cuttings have few roots, so low VPD limits how fast they lose water. As the root system grows, the deficit is raised to drive transpiration and nutrient flow. In flower the air is dried further, and the main reason given in the tutorial is protection from grey mould, with fans running round the clock.

The research supports the shape of this curve but not its precise numbers. None of the sources in this library tests VPD bands on cannabis for yield, potency or flavour. The general range often cited for greenhouse crops, 0.5–0.8 kPa (Inoue et al., 2021), sits well below the tutorial's flower band. That is not a contradiction: the flower band is not chosen for maximum photosynthesis but as a safety margin against bud rot, which the cannabis literature links clearly to humid air and wet surfaces (Mahmoud et al., 2023). It is a reasonable trade between growth and disease, made by practitioners rather than measured in a trial. The upper end of 1.6 kPa also stays comfortably below the roughly 2 kPa above which transpiration commonly falls in other crops.

Why CO₂ enrichment rarely fits a home tent

Commercial growers often enrich the air with CO₂, and grow shops sell cylinders and fermenting bags for tents. The physiology behind it is real. Wang et al. (2022), reviewing CO₂ enrichment in greenhouse vegetable production, report that a moderate rise to 550–650 ppm improves the yield of various C3 crops (the large group that includes cannabis, tomatoes and most vegetables) by an average of 18%, and they describe 800–1000 ppm as optimal for crop growth, against an outdoor level of 350–450 ppm. They also note that in a closed greenhouse full of plants, daytime CO₂ can drop to only 100–250 ppm, well below the outside air.

The trouble lies in keeping the gas where the plants are. Wang et al. (2022) report that the CO₂-use efficiency of greenhouse enrichment is generally below 60%, and below 50% at a set point of 1000 ppm even in an unventilated greenhouse, because of leakage. A grow tent is far leakier than a greenhouse: the tutorial's own set-up runs an extractor that pulls air through a carbon filter and out, keeping the tent under slight negative pressure. Any CO₂ added is pulled straight out with the stale air. For enrichment to pay, the space has to hold the gas, and a tent with a constantly running extractor does not.

The cheaper home methods have their own problems. Wang et al. (2022) note that making CO₂ from baking soda and acid is cheap, but that in practice the operation is complicated, and gas generated "out of control" is wasted and can damage plants. Burners, used in greenhouses that need heat anyway, produce harmful gases such as nitrogen oxides, sulphur dioxide and carbon monoxide that must be cleaned out. Compressed gas is the cleanest source but expensive and bulky.

There is also a link back to the rest of this article. Elevated CO₂ generally makes stomata close somewhat. Xu et al. (2016), reviewing stomatal responses, report that doubling CO₂ cut stomatal conductance by about 40% in almost all enclosure experiments such as greenhouses and chambers, while free-air enrichment experiments averaged a 22% reduction across about 580 cases. They also report a 5% average drop in stomatal density in a meta-analysis, and stress that the response is "general rather than universal", varying with species and conditions. Narrower stomata mean less transpiration and warmer leaves: Nelson and Bugbee (2015) calculated that a 30% fall in stomatal conductance would raise leaf temperature by about 1 °C under any of their light sources. In a tent, that is one more thing pushing leaf temperature and VPD in a direction the grower then has to correct.

What a small tent needs is the opposite: enough fresh air that the plants never deplete it. Wang et al. (2022), citing earlier work, note that crop yield depends more heavily on CO₂ at low concentrations, below about 450 ppm, than at higher ones. Steady air exchange with the room keeps the tent out of that depleted range, which is where the plants would lose most.

What this means for a 1×1 tent

None of this replaces the tutorial's targets; it explains them and shows where to be flexible.

• Treat the bands as a corridor, not a bullseye. Aim for the tutorial's stage ranges, but think of the boundaries as roughly 0.4 kPa at the bottom and about 2 kPa at the top. Below 0.4 kPa, especially in flower, moisture lingers on leaves and buds; above about 2 kPa, many plants begin to close down.
• Keep it steady. Swings of about one kilopascal every few minutes cut lettuce dry weight by 15% in three weeks. A humidifier or dehumidifier set with a wide margin, or a fan controller that ramps speed rather than switching on and off, tends to give smoother air than tight on-off control.
• Watch the transition to lights-off. This is when relative humidity jumps, and in flower it is the riskiest moment for condensation. Avoid sudden large temperature drops in late bloom, and keep air moving through the canopy at night.
• Measure where the plants are. Hang the sensor at canopy height in the middle of the tent, not on the wall, as the tutorial says, and note whether your meter calculates VPD from air or from leaf temperature.
• Use leaf temperature as a health check. Under LED, a well-watered canopy should read close to air temperature on an infrared thermometer. Leaves several degrees warmer than the air suggest that transpiration has slowed and are worth investigating.
• Spacing is climate control too. Dense canopies and crowded plants trap moisture. Thinning interior growth, and in a small tent choosing one plant rather than two, lowers bud-rot risk at some cost in yield.
• Skip bottled CO₂. In a ventilated tent most of it will be extracted. Good air exchange with a well-ventilated room is the cheaper way to keep CO₂ from becoming limiting.

What we know, and what we don't

The physics is solid. Vapour pressure deficit is defined the same way in a forest, a greenhouse and a tent, and the steep rise of saturation vapour pressure with temperature is textbook science. The direction of the stomatal response is also well established across many species: as VPD rises, stomata partly close, transpiration rises up to a point, and photosynthesis falls more slowly than conductance (Grossiord et al., 2020). Leaf temperature under LED is, on the best available modelling, close to air temperature when plants are well watered, and only about 1.3 °C cooler than under HPS (Nelson and Bugbee, 2015). Elevated CO₂ usually reduces stomatal opening (Xu et al., 2016), and enrichment loses much of its gas through leakage even in greenhouses (Wang et al., 2022).

Less certain is everything that carries numbers into cannabis. The case for steady over fluctuating VPD rests on one chamber study on lettuce (Inoue et al., 2021); the 0.4 kPa mould threshold comes from tomato and cucumber, and most bud-rot data from greenhouses (Mahmoud et al., 2023); the leaf-temperature figures come from a model built on vegetable leaves (Nelson and Bugbee, 2015); the CO₂ yield gains are greenhouse vegetable averages (Wang et al., 2022).

What we do not know, from this library, is which VPD bands give the best yield, potency or aroma in cannabis, whether those bands differ between cultivars, how cannabis stomata compare with those of other crops, or how much CO₂ enrichment adds to cannabis grown at the modest light levels typical of a small tent. Until such trials exist, the tutorial's bands remain what they are: a well-reasoned grower's practice that fits the physiology, with a deliberate tilt towards dry air in flower to keep the buds clean.

Sources

Grossiord, C., Buckley, T.N., Cernusak, L.A., Novick, K.A., Poulter, B., Siegwolf, R.T.W., Sperry, J.S. and McDowell, N.G. (2020) 'Plant responses to rising vapor pressure deficit', New Phytologist, 226(6), pp. 1550–1566. doi:10.1111/nph.16485.

Inoue, T., Sunaga, M., Ito, M., Yuchen, Q., Matsushima, Y., Sakoda, K. and Yamori, W. (2021) 'Minimizing VPD fluctuations maintains higher stomatal conductance and photosynthesis, resulting in improvement of plant growth in lettuce', Frontiers in Plant Science, 12, 646144. doi:10.3389/fpls.2021.646144.

Mahmoud, M., BenRejeb, I., Punja, Z.K., Buirs, L. and Jabaji, S. (2023) 'Understanding bud rot development, caused by Botrytis cinerea, on cannabis (Cannabis sativa L.) plants grown under greenhouse conditions', Botany (accepted manuscript). doi:10.1139/cjb-2022-0139.

Nelson, J.A. and Bugbee, B. (2015) 'Analysis of environmental effects on leaf temperature under sunlight, high pressure sodium and light emitting diodes', PLoS ONE, 10(10), e0138930. doi:10.1371/journal.pone.0138930.

Wang, A., Lv, J., Wang, J. and Shi, K. (2022) 'CO2 enrichment in greenhouse production: towards a sustainable approach', Frontiers in Plant Science, 13, 1029901. doi:10.3389/fpls.2022.1029901.

Xu, Z., Jiang, Y., Jia, B. and Zhou, G. (2016) 'Elevated-CO2 response of stomata and its dependence on environmental factors', Frontiers in Plant Science, 7, 657. doi:10.3389/fpls.2016.00657.