R05 — Indoor vs. Greenhouse/Light-Dep vs. Sun-Grown Cannabis
ChatGPT Deep Research Pass · Brief v1.1
Date: August 25, 2026
Role: Independent primary research output for Claude/Fable editorial review
Status: Pilot first-pass evidence review
Executive summary
The strongest conclusion from the available evidence is not that one cultivation system is categorically “best,” but that the three broad environments solve different problems and externalize different costs.
Indoor cultivation offers the greatest environmental control, year-round scheduling, biosecurity potential, production consistency, and ability to manipulate light, temperature, humidity, COâ‚‚, irrigation, and crop timing. Those advantages come with a large energy burden. Modern LEDs and better HVAC/dehumidification can materially improve indoor efficiency, but recent field and life-cycle work does not support the claim that technological improvements have erased the basic energy/carbon gap between fully indoor cultivation and outdoor production. The magnitude of that gap is highly climate- and grid-dependent.
Outdoor sun-grown cultivation generally has the lowest operational energy demand and the lowest climate burden per kg of dried flower in the cannabis-specific LCAs reviewed. However, “sun-grown = sustainable” is too simplistic. Outdoor systems can impose meaningful water, nutrient-loss, pesticide, wildlife, land-use, erosion, and ecosystem costs, especially when cultivation occurs in water-stressed or ecologically sensitive regions. Illicit cultivation adds additional risks because environmental rules and pesticide controls may be absent.
Greenhouse and mixed-light cultivation span a continuum rather than a single category. A passive hoop house, a blackout/light-deprivation structure, a greenhouse with supplemental LEDs, and a sealed high-tech greenhouse with mechanical HVAC are not environmentally or economically equivalent. Greenhouse systems can reduce dependence on sole-source lighting while preserving partial environmental control, but their footprint depends strongly on heating, cooling, dehumidification, lighting intensity, climate, and structure.
On product quality, the evidence is much thinner than marketing claims imply. A notable 2023 direct cannabis comparison using two genetically identical commercial cultivars found substantial indoor/outdoor differences in cannabinoid and terpene profiles, including more diverse and abundant sesquiterpenes in outdoor samples and more oxidized/degraded cannabinoids in indoor samples. That study is important but small and confounded by cultivation media and incomplete environmental-control data. It does not establish a universal rule that outdoor cannabis is chemically superior. Likewise, the broad industry belief that “indoor is stronger” is not established by robust head-to-head evidence independent of genetics, crop management, harvest timing, curing, and testing-lab variation.
On yield and economics, controlled environments can generate more crop cycles per year and can deliver high production per unit floor area, but scientific yield estimates vary enormously and are often incomparable because of different plant densities, cultivars, cycle lengths, training systems, lighting levels, and functional units. Economics are even more jurisdiction-specific. Recent greenhouse work from New York illustrates that profitability can flip with cultivar type, yield, labor, planting material, and wholesale price. Older industry surveys consistently show lower per-pound production costs outdoors and in greenhouses than indoors, but those figures should be treated as dated market observations, not biological constants.
The most defensible synthesis is therefore:
Indoor has the highest controllability and usually the highest operational energy intensity.
Outdoor generally has the lowest climate burden, but not automatically the lowest total ecological burden.
Greenhouse/mixed-light is a family of systems whose sustainability depends on how much artificial environmental control is added.
Quality claims are substantially less settled than environmental claims.
Genetics and crop management frequently confound claims attributed to cultivation environment.
The correct comparison unit matters: per plant, per m², per crop, per year, per kg saleable flower, and per gram cannabinoid can produce different rankings.
Market and profitability comparisons must be date- and jurisdiction-stamped.
The public debate is more confident than the underlying comparative literature warrants.
Methodology
Evidence hierarchy used
This review followed the hierarchy specified in R05 v1.1:
Cannabis-direct controlled or comparative studies
Cannabis-specific observational studies, LCAs, government data, and field measurements
Adjacent controlled-environment agriculture evidence
General plant science
Industry/practitioner evidence, explicitly labeled
Modeled LCAs, direct field measurements, controlled experiments, regulatory datasets, industry surveys, and practitioner statements are treated as different types of evidence.
Functional units
Environmental results are reported using the source’s functional unit. The preferred unit is per kg of dried flower, but conversions were not made when doing so would require assumptions about moisture, trimming losses, marketable fraction, extraction yield, or crop cycles.
Yield data are particularly difficult to normalize because studies report combinations of:
g/plant
kg/plant
g/m²
kg/m²
g/W
production per crop
annual production
total biomass vs. trimmed saleable flower
Those values should not be pooled without a common production definition.
Lifecycle boundaries
The major cannabis LCAs reviewed use different boundaries. Some are cultivation-only or cradle-to-gate; newer work may include transportation, retail, and waste. Figures should therefore not be averaged into a single “cannabis carbon footprint.”
A. Definitions and cultivation landscape
A1. System definitions
Indoor / sealed warehouse
California defines indoor cultivation as cannabis grown in a permanent structure using exclusively artificial light, or any structure using artificial light above 25 W/ft². In practical cultivation language, indoor facilities range from retrofitted warehouses to purpose-built controlled-environment plant factories.
Typical environmental-control systems can include:
sole-source LED or HPS lighting
cooling and heating
dehumidification
ventilation or sealed-room air handling
supplemental COâ‚‚
fertigation and recirculating irrigation
filtration and sanitation controls
The defining characteristic for this report is that photosynthetically useful light is supplied primarily or entirely by electric lighting and the environment is mechanically controlled.
Source: California Department of Cannabis Control, cultivation license definitions.
https://www.cannabis.ca.gov/applicants/license-types/cultivation-license-types/
Outdoor sun-grown
California defines outdoor cultivation as mature cannabis production without artificial lighting in the canopy area at any point. In practice, outdoor production includes open-field and container/raised-bed systems exposed directly to seasonal conditions.
Outdoor production relies primarily on:
sunlight
ambient temperature and humidity
natural seasonal photoperiod
irrigation/rainfall
field soil, amended soil, containers, or raised beds
Outdoor is not synonymous with organic, regenerative, or low-input. Outdoor operations may use synthetic fertilizers, plastic mulch, irrigation pumps, pesticides, imported media, generators, and substantial postharvest infrastructure.
Greenhouse / mixed light
“Greenhouse” should not be treated as one exposure class.
For R05, the useful ladder is:
1. Passive greenhouse / hoop house
Sunlight is the primary light source, with little or no mechanical climate control.
2. Light-deprivation greenhouse
A greenhouse or hoop structure in which blackout fabric or other methods deliberately eliminate natural light to manipulate photoperiod and induce reproductive growth earlier than the natural season.
California regulation defines light deprivation as “the use of any technique to eliminate natural light in order to induce flowering.”
3. Supplemental-light greenhouse
Sunlight remains important, but electric lighting is added to increase daily light integral, extend day length during vegetative growth, improve uniformity, or maintain target production during low-light periods.
4. Highly controlled / sealed greenhouse
Sunlight is transmitted through the envelope but the facility may also use substantial electric lighting, HVAC, dehumidification, COâ‚‚ supplementation, screening, and computerized environmental controls.
These categories can have radically different energy footprints.
Regulatory source:
https://www.law.cornell.edu/regulations/california/4-CCR-15000
What light deprivation does photobiologically
Most high-cannabinoid photoperiod-sensitive Cannabis sativa cultivars initiate reproductive development in response to sufficiently long uninterrupted nights. Critical photoperiod varies genetically; it is not accurate to state that every cultivar has one universal threshold.
Light deprivation therefore works by artificially extending the dark period, typically using blackout curtains or covers, so that plants experience a flowering-inductive photoperiod while outdoor day length is still naturally long.
Mechanically this may involve:
manual tarps
automated blackout curtains
retractable greenhouse screens
covered hoop houses
The system permits growers to initiate flowering before natural autumn day shortening and, with appropriate vegetative scheduling, can support multiple seasonal flowering cycles.
Supporting sources: Cornell Hemp flowering-time overview and California regulatory definition.
https://www.law.cornell.edu/regulations/california/4-CCR-15000
Evidence classification: cannabis-direct/general photobiology.
A2. What share of production occurs in each system?
A defensible single percentage for “North American cannabis production by cultivation environment” was not located in the first pass.
This is an important data gap.
California
California tracks cultivation licenses and harvest data by indoor, outdoor, and mixed-light categories, but license counts are not equivalent to actual production volume because canopy size, utilization, yield, and number of cycles differ.
The 2024 California Cannabis Market Outlook reported 874 active indoor cultivation licenses in October 2024, after 926 in October 2023; outdoor and mixed-light licenses had declined more sharply. This tells us about license structure, not direct flower tonnage.
California’s licensing categories themselves are useful because they distinguish outdoor, indoor, Mixed-Light Tier 1 (0–6 W/ft²), and Mixed-Light Tier 2 (>6–25 W/ft²).
Source: California Department of Cannabis Control, California Cannabis Market Outlook 2024 / licensing data.
Canada
Health Canada reported, as of the latest dataset available in this search:
Indoor growing area: 1,412,077 m²
Outdoor growing area: 689 hectares = 6,890,900 m²
These figures describe licensed area, not production output. Outdoor area cannot be interpreted as five times indoor production because indoor facilities can run multiple annual crop cycles and have much higher production intensity per unit floor area.
Source: Health Canada, Cannabis data overview, updated July 9, 2026.
https://www.canada.ca/en/health-canada/services/drugs-medication/cannabis/research-data/market.html
Unlicensed / legacy production
The unlicensed market is intrinsically difficult to quantify. Environmental studies document illegal outdoor and indoor cultivation, but robust production shares by environment are not available. Estimates should not be mixed with regulated-market licensing statistics.
Conclusion: production shares remain a major unresolved evidence gap.
B. Environmental footprint
B3. Energy and greenhouse-gas emissions
Indoor cannabis: 2021 U.S. LCA
Summers, Sproul, and Quinn (2021) modeled commercial indoor production across U.S. climates and electrical grids using a cradle-to-gate LCA.
They reported:
2,283–5,184 kg COâ‚‚e per kg dried flower, depending on location.
Major drivers included:
electricity generation
natural-gas consumption
indoor environmental control
high-intensity lighting
supplied COâ‚‚
The geographic range is critical. A single national indoor carbon number is misleading because electricity mix and climate strongly influence results.
Source: Summers HM, Sproul E, Quinn JC. 2021. *The greenhouse gas emissions of indoor cannabis production in the United States*. Nature Sustainability 4:644–650.
https://www.nature.com/articles/s41893-021-00691-w
DOI: 10.1038/s41893-021-00691-w
Evidence type: cannabis-specific modeled LCA.
Boundary: cradle-to-gate.
Outdoor: 2024 cannabis-specific LCA
Desaulniers Brousseau et al. (2024) incorporated primary outdoor cannabis production data into an LCA. The study concluded that the tested outdoor system could be roughly 50 times less carbon-emitting than indoor production.
The paper also makes an important qualification: low direct energy does not mean zero environmental burden. Outdoor nutrient losses, fertilizer requirements, runoff, and mineralization can become material impact categories.
Source: Desaulniers Brousseau V, Goldstein BP, Sedlock C, Lefsrud M. 2024. *Environmental Impact of Outdoor Cannabis Production*. ACS Agricultural Science & Technology 4(7):690–699.
https://pubs.acs.org/doi/10.1021/acsagscitech.4c00054
DOI: 10.1021/acsagscitech.4c00054
Evidence type: cannabis-direct field trial + modeled LCA.
Canada: 2024 comparative environmental model
Desaulniers Brousseau et al. (2024) modeled Canadian cannabis production with primary HVAC energy modeling. The study found indoor cultivation substantially more carbon-intensive than outdoor cultivation and estimated that outdoor production could reduce carbon impacts on the order of 90% in the modeled Canadian context.
Cold climates can increase heating demand, while warmer climates increase cooling demand. The paper highlights natural-gas heating as a major issue in Canadian controlled-environment production.
Source: Desaulniers Brousseau V, Goldstein BP, Lachapelle M, Tazi I, Lefsrud M. 2024. *Greener green: The environmental impacts of the Canadian cannabis industry*. Resources, Conservation & Recycling 208:107737.
https://doi.org/10.1016/j.resconrec.2024.107737
Evidence type: cannabis-specific industry/LCA model.
2025 U.S. systems-level analysis
A 2025 One Earth analysis concluded that indoor cultivation drives most modeled U.S. cannabis industry emissions and that shifting production outdoors could reduce emissions by as much as 76% in the study’s scenario.
This study used a broader system boundary than many early cultivation LCAs, encompassing cultivation energy and other inputs plus transportation, retail, and waste disposal.
It estimated approximately:
44 Mt COâ‚‚e/year industry-wide
595 PJ/year energy consumption
Those national estimates require caution because total legal and illicit production itself must be modeled. They are best used as scenario estimates rather than direct census measurements.
Source: *Energy-intensive indoor cultivation drives the cannabis industry’s expanding carbon footprint*. One Earth 8(3), 2025, 101179.
https://www.sciencedirect.com/science/article/pii/S2590332225000053
DOI: 10.1016/j.oneear.2025.101179
Bottom line
The exact magnitude varies, but the direction of the carbon comparison is unusually consistent:
Fully indoor cultivation generally carries a substantially greater operational energy and climate burden per kg flower than open-field outdoor production.
That conclusion is much better supported than many potency or terpene claims.
B4. Do LEDs eliminate the indoor problem?
No.
They materially improve lighting efficiency, but lighting is only one part of a tightly coupled indoor energy system.
The 2022–2024 PG&E field study of two Sacramento indoor cannabis facilities found:
HVAC ≈ 45% of facility energy use
flower rooms ≈ 86% of total facility energy use
flower-room energy intensity ≈ 760 kBtu/ft²/year
about 70% of water use was transpired into the room and became HVAC condensate
dehumidifier performance in real cultivation conditions could be only 40–65% of rated efficiency
This illustrates an important rebound/coupling issue: efficient LEDs lower lighting load and sensible heat, but plants still transpire water, dehumidification still requires energy, and environmental-control requirements remain substantial.
Source: PG&E / Energy Transition Coordinating Council, *Controlled Environment Horticulture: Energy Consumption and Environmental Control Field Study*, 2022–2024.
A 2025 large review of indoor agriculture likewise found cannabis among the most energy-intensive controlled-environment crops in its dataset. It emphasized that end-use shares vary strongly with climate: dehumidification can dominate in hot/humid locations while heating may dominate in cold climates.
Source: *The emergence of indoor agriculture as a driver of global energy demand*. npj Sustainable Agriculture, 2025.
https://www.nature.com/articles/s44264-025-00091-z
Cannabis-specific lighting experiments also show that LED spectrum, photoperiod, interlighting, and placement can improve grams of cannabinoid or flower per unit electricity. These are efficiency gains, not evidence that indoor cultivation is environmentally equivalent to sunlight-based production.
B5. Water use
Water comparisons are more difficult than energy comparisons because the correct sustainability question is not simply “liters per plant.”
Relevant distinctions include:
source of water
local water scarcity
rainfall contribution
irrigation efficiency
runoff
recapture/reuse
transpiration condensate recovery
yield per unit water
Outdoor water withdrawals can be locally severe
Bauer et al. estimated water demand in four northwestern California watersheds using mapped cannabis plants and a working estimate of 22.7 L/plant/day over a 150-day season. Estimated daily cannabis demand ranged from approximately 523,000 to 724,000 L/day per watershed in the study.
The ecological significance was greatest during seasonal low flow, when diversions could represent a substantial fraction of streamflow.
Source: Bauer S et al. 2015. *Impacts of Surface Water Diversions for Marijuana Cultivation on Aquatic Habitat in Four Northwestern California Watersheds*. PLoS ONE.
https://pmc.ncbi.nlm.nih.gov/articles/PMC4364984/
This is a regional diversion-risk study, not a universal water-footprint estimate.
Greenhouse supplemental light and water-use efficiency
A 2024 controlled cannabis greenhouse experiment on ‘Suver Haze’ found that increasing sunlight + supplemental LED light increased total evapotranspiration but improved biomass water-use efficiency. Water requirement per gram biomass declined from about 0.37 to 0.24 L/g across the tested light treatments.
This is a useful reminder that absolute water use and water-use efficiency can move in opposite directions.
Source: Collado CE, Hwang SJ, Hernández R. 2024. *Supplemental greenhouse lighting increased the water use efficiency, crop growth, and cutting production in Cannabis sativa*. Frontiers in Plant Science.
https://www.frontiersin.org/journals/plant-science/articles/10.3389/fpls.2024.1371702/full
DOI: 10.3389/fpls.2024.1371702
Indoor recapture potential
In the PG&E field study, roughly 70% of cultivation water ultimately appeared as HVAC condensate, creating substantial reuse potential in sealed facilities.
That does not make indoor cultivation “water-free,” but it demonstrates why liters applied per plant can be misleading: controlled environments can capture water that outdoor crops release to the atmosphere.
Synthesis: water sustainability is highly site-specific. Outdoor can minimize energy but be problematic in drought-sensitive watersheds. Indoor and greenhouse can recirculate/recapture water but expend energy to do so.
B6. Land, ecosystems, pesticide drift, light pollution, waste
Outdoor land and wildlife effects
Outdoor cannabis production can affect:
habitat conversion/fragmentation
riparian zones
stream flow
erosion
wildlife movement
fencing
pesticide exposure
road construction and human disturbance
California research has documented cannabis cultivation intersecting with sensitive wildlife habitat, especially in legacy production regions.
Illicit outdoor cultivation
A 2025 study of inactive trespass grows on California federal land detected pesticide residues in soils, including bifenthrin, cyfluthrin, malathion, imidacloprid, fluopyram, myclobutanil, and triadimefon. The study also identified plasticizers and other contaminants associated with cultivation infrastructure.
Source: *Identifying organic contaminants at trespass cannabis grows on federal land in California, USA*. Science of the Total Environment, 2025.
https://www.sciencedirect.com/science/article/pii/S0048969725022168
Earlier watershed research detected pesticides including diazinon and carbofuran downstream of illegal public-land cultivation sites.
Source: *Passive monitoring of soluble pesticides linked to cannabis cultivation: a multi-scale analysis*. Water Quality Research Journal 57(4), 2022.
https://doi.org/10.2166/wqrj.2022.101
These findings should not be generalized to licensed outdoor cannabis. They specifically demonstrate the potential ecological externalities of poorly regulated or illicit production.
Greenhouse light pollution
Greenhouse lighting can create nighttime skyglow and ecological disturbance when supplemental light escapes through the structure. Cannabis-specific quantitative ecosystem studies remain limited. California requires lights used for indoor and mixed-light cultivation to be shielded from sunset to sunrise to reduce nighttime glare.
Source: California DCC cultivation regulations.
Waste streams
Cannabis cultivation can generate:
disposable growing media
plastic pots and bags
irrigation tubing
trellis
fertilizer containers
HVAC/lighting equipment replacement
crop residues
packaging and postharvest waste
Comparative peer-reviewed life-cycle quantification of these waste streams by cultivation environment remains weak relative to the energy literature.
Data gap: no robust normalized indoor-vs-greenhouse-vs-outdoor waste dataset was found in this pass.
C. Product quality
C7–C8. Cannabinoids, terpenes, secondary metabolites
Best direct indoor-vs-outdoor comparative study identified
Zandkarimi et al. (2023) compared two commercial cultivars grown from genetically identical clone stock in indoor and outdoor environments.
Important findings:
primary cannabinoid differences were not uniformly large
indoor samples contained more oxidized/degraded cannabinoid products
outdoor samples contained higher levels of several unusual cannabinoids
outdoor samples contained greater abundance/diversity of several terpenes, especially sesquiterpenes including β-caryophyllene, α-humulene, α-bergamotene, α-guaiene, and germacrene B
Source: Zandkarimi F et al. 2023. *Comparison of the Cannabinoid and Terpene Profiles in Commercial Cannabis from Natural and Artificial Cultivation*. Molecules 28(2):833.
https://www.mdpi.com/1420-3049/28/2/833
DOI: 10.3390/molecules28020833
Critical limitations
This paper should not be read as a universal outdoor superiority study.
Limitations include:
only two cultivars
small sample counts
proprietary growing conditions
indoor and outdoor systems differed in more than light exposure
outdoor plants used living soil; indoor plants used a proprietary artificial medium
postharvest details could contribute to oxidation/degradation differences
commercial environments were not experimentally standardized
Thus, the study supports:
Environment and cultivation system can materially alter cannabis secondary-metabolite profiles.
It does not securely support:
Sun-grown cannabis always has more terpenes or is always chemically “better.”
Does indoor cannabis have more THC?
The direct evidence does not justify a universal rule.
The 2023 study found some cultivar-specific cannabinoid differences but did not establish consistent indoor superiority in primary cannabinoids.
A 2024–2025 Australian home-grow observational analysis found that greenhouse cultivation was associated with higher total THC than several outdoor categories in that particular sample, and living soil was associated with higher total THC than hydroponic cultivation. However, this was observational and susceptible to cultivar, grower skill, and practice confounding.
Source: *An analysis of the cultivation, consumption and composition of home-grown cannabis following decriminalisation in the Australian Capital Territory*.
https://pmc.ncbi.nlm.nih.gov/articles/PMC11794705/
Conclusion: “Indoor is stronger” remains insufficiently established as an environmental effect independent of genetics and production practice.
UV and environmental-stress hypotheses
Sunlight exposes outdoor plants to a broader spectrum, including UV wavelengths absent or reduced in many indoor lighting regimes. General plant science and cannabis lighting studies show that light spectrum and intensity can influence cannabinoid and terpene metabolism.
However, moving from:
light spectrum affects secondary metabolism
to:
sunlight therefore produces universally superior terpene or cannabinoid profiles
is an unsupported leap.
Modern indoor spectra can be manipulated, and genotype × environment interactions are strong.
Status: biologically plausible mechanism; insufficient basis for a universal cultivation-system ranking.
Physical/commercial quality
Scientific comparisons of:
flower density
visual appeal
trim quality
color
resin appearance
aroma retention
market-grade acceptance
across otherwise matched indoor, greenhouse, and outdoor cannabis are sparse.
This is a major place where commercial perception outruns research.
Indoor cultivation may allow growers to avoid wind, rain, dust, temperature excursions, and certain pest/weather damage while tightly managing density and finish. Outdoor cultivators may achieve excellent morphology under favorable conditions. The literature does not presently justify treating “bag appeal” as an intrinsic property of indoor production.
Data gap: standardized blinded grading across matched genetics and postharvest conditions.
Consumer perception
Market pricing historically reveals a premium for indoor flower in many jurisdictions. That premium may reflect:
visual conventions
THC selection
brand positioning
historic scarcity
consumer expectations
perceived cleanliness/control
actual production consistency
A price premium is evidence of market perception, not proof of biochemical superiority.
C9. Contamination risk
Microbial contamination
Outdoor cannabis is naturally exposed to diverse airborne bacteria and fungi. NIOSH-linked research at an outdoor farm documented occupational exposure to microbial bioaerosols during harvesting and processing.
Source: Green BJ et al. *Microbial hazards during harvesting and processing at an outdoor United States cannabis farm*.
https://pmc.ncbi.nlm.nih.gov/articles/PMC6314012/
Indoor cultivation is not inherently sterile. High humidity, dense canopy, poor sanitation, recirculated air, or inadequate dehumidification can produce serious Botrytis, powdery mildew, and other microbial problems.
Recent cross-state research on total yeast and mold testing also emphasizes that regulatory thresholds and testing practices differ substantially, complicating comparisons among markets.
Source: *Regional (in)Consistency and Comparability of Cannabis Laboratory Testing: A Statistical Exploration of Total Yeast and Mold*. Clinical Therapeutics, 2026.
DOI: 10.1016/j.clinthera.2026.03.019
Data gap: a robust multi-state dataset linking contamination failure rate to declared cultivation environment while controlling for testing rules.
Pesticides
There is no basis to assume that indoor automatically means pesticide-free. A Belgian study of illicit indoor plantations found pesticides in 64.3% of cannabis plant samples.
Source: Cuypers E et al. 2017. *The use of pesticides in Belgian illicit indoor cannabis plantations*.
https://pubmed.ncbi.nlm.nih.gov/28609661/
Conversely, outdoor illicit/trespass production has well-documented pesticide risks in some regions.
The key variable is likely regulatory compliance and pest-management practice, not simply roof/no roof.
Heavy metals
Cannabis is capable of accumulating metals from substrates and irrigation sources. A 2026 outdoor reclaimed-water study found most tested metals low or nondetectable but reported lead in dried flower slightly above the cited California inhaled-product limit despite low water/soil signals.
Source: *Feasibility of medicinal cannabis cultivation in outdoor conditions using municipal reclaimed water*. Journal of Cannabis Research, 2026.
https://link.springer.com/article/10.1186/s42238-026-00433-9
Heavy-metal risk therefore depends heavily on soil/media, fertilizer, water, and cultivar—not only cultivation environment.
D. Yield, economics, and risk
D10. Comparative yield
The yield literature remains heterogeneous.
A 2019 cannabis yield meta-analysis found yield influenced by:
cultivar
plant density
light intensity
fertilization
container size
light type
flowering duration
The authors found enormous reported/projection ranges across commercial facilities and warned that facility area alone is a poor predictor of actual yield.
Source: Backer R et al. 2019. *Closing the Yield Gap for Cannabis: A Meta-Analysis of Factors Determining Cannabis Yield*. Frontiers in Plant Science 10:495.
https://www.frontiersin.org/journals/plant-science/articles/10.3389/fpls.2019.00495/full
DOI: 10.3389/fpls.2019.00495
This is why per-plant comparisons are often misleading. Outdoor plants may become very large; indoor systems may use much higher plant densities and several annual cycles.
Indoor
Advantages:
year-round production
crop scheduling independent of season
high plant density
high light levels
multi-tier possibilities in some systems
Limitations:
electricity/HVAC constraints
disease can spread rapidly in dense monoculture
equipment failure can quickly threaten crops
Outdoor
Advantages:
sunlight at no direct energy cost
potentially very large individual plants
low infrastructure cost per unit land
Limitations:
generally one main natural-season harvest for photoperiod cultivars
climate constrains geography and scheduling
storms, wildfire smoke, frost, rain, heat, and drought can reduce saleable yield
Light deprivation / mixed light
The major agronomic advantage is control over flowering timing. By manipulating photoperiod, growers can initiate flowering earlier than the natural fall season and potentially schedule multiple annual harvests in suitable climates.
However, the legacy claim that light deprivation can increase cycles “up to 5x” should not be accepted as a general scientific conclusion without defining:
climate
greenhouse heating
propagation schedule
cultivar
crop duration
whether five cycles includes autoflower or short-cycle strategies
whether the count is theoretical or commercially realized
A 2026 New York greenhouse study modeled five two-month autoflower cycles versus four three-month photoperiod cycles, showing that very high annual cycle counts are technically conceivable in protected cultivation but are not automatically profitable.
D11. Production cost and economics
Older industry benchmark
Cannabis Business Times reported median 2021 production costs of approximately:
Outdoor: $214/lb
Greenhouse: $256/lb
Indoor: $472/lb
These are industry survey figures, not controlled scientific measurements.
Source: Cannabis Business Times, 2021 State of the Cannabis Cultivation Industry Report.
A 2018 Marijuana Business Factbook reported approximately:
indoor: $635/lb
greenhouse: $490/lb
outdoor: $305/lb
Again, these are dated industry economics, but both datasets show the same qualitative ordering: indoor production tended to cost more per pound.
Why those numbers cannot be universalized
Cost is dominated by:
electricity and natural gas prices
climate
labor rates
building costs
licensing fees
local taxes
required testing
security
insurance
financing
wholesale price
crop failure
cultivar yield
facility utilization
California itself charges substantially different annual cultivation license fees by production class, adding regulatory cost differences unrelated to plant biology.
2026 New York greenhouse evidence
Vergara and Ruterbories (2026) evaluated greenhouse production economics in New York.
Under their baseline assumptions:
Autoflower system
assumed yield: 0.04 lb/plant (0.018 kg)
assumed price: $260/lb
estimated annual return above total costs: –$1.48/ft²
Photoperiod system
assumed yield: 0.18 lb/plant (0.082 kg)
same assumed price
estimated annual return above total costs: +$7.18/ft²
Labor represented a major cost component; profitability changed sharply across yield/price scenarios.
Source: Vergara D, Ruterbories KA. 2026. *High-THC Cannabis sativa in a New York greenhouse: yield and economic factors*. Journal of Cannabis Research 8:85.
https://link.springer.com/article/10.1186/s42238-026-00429-5
DOI: 10.1186/s42238-026-00429-5
This study is particularly useful because it demonstrates that “more crop cycles” does not necessarily mean “more profit.”
D12. Cultivation risk profile
Indoor
Advantages
weather exclusion
tight environmental control
predictable photoperiod
potential pest-exclusion/sanitation programs
repeatable scheduling
Risks
catastrophic HVAC/power failure
high fixed and operating cost
dense canopy disease spread
humidity/dehumidification failure
pathogen persistence in recirculating systems
economic sensitivity to energy and wholesale prices
Greenhouse/light dep
Advantages
protection from rainfall and some weather
sunlight reduces electric-light dependence
photoperiod scheduling
supplemental light can stabilize seasonal DLI
Risks
heat buildup
humidity spikes/condensation
blackout-system failure
supplemental lighting cost
pest entry
storm damage
heating requirements in cold climates
light pollution if poorly shielded
Outdoor
Advantages
low energy demand
low structure cost
natural solar spectrum
potential large plant size
Risks
weather and seasonal limits
drought
wildfire/smoke
rain near harvest
frost/heat
pest exposure
wildlife
pesticide drift from neighboring agriculture
theft/security
ecosystem and water constraints
climate-change sensitivity
E. Synthesis
E13. Where evidence is strong, weak, or absent
Strongest evidence
Indoor cannabis is energy-intensive.
Cannabis-specific LCAs, modern field measurements, and broader CEA evidence all support this.
Outdoor production can dramatically reduce climate impacts.
Recent cannabis-specific LCAs support large reductions relative to indoor production.
Climate and grid mix matter.
Carbon intensity varies strongly by location.
Greenhouse is not one technology.
Supplemental lighting and HVAC can move a greenhouse much closer to indoor energy profiles.
Moderate evidence
Cultivation environment can alter cannabis secondary-metabolite profiles.
Direct research supports this, but generalization is limited.
Controlled environments can improve production consistency and enable more annual cycles.
Outdoor systems can create substantial local ecological harm when poorly sited or managed.
Weak / insufficient evidence
Indoor flower is universally more potent.
Sun-grown flower universally has richer terpene profiles.
Light deprivation is universally the sustainability “sweet spot.”
One cultivation environment has inherently lower contamination risk across all contaminant categories.
A universal per-pound cost ranking can be transported across jurisdictions and years.
There is one defensible current North American production-share breakdown by environment.
E14. Repeated public claims and evidence check
Claim 1: “Indoor weed is stronger.”
Verdict: Not established as a universal environmental effect.
Indoor producers often select high-THC cultivars and optimize controlled environments, but direct matched-genetic studies do not establish that the presence of a roof and artificial light inherently produces higher THC.
Claim 2: “Outdoor cannabis is automatically sustainable.”
Verdict: False as stated.
Outdoor generally has a major energy/carbon advantage but can still impose significant water, land, nutrient, pesticide, erosion, and wildlife impacts.
Claim 3: “Sun-grown has more terpenes.”
Verdict: Plausible and supported by at least one notable direct study, but not established universally.
The 2023 matched-genetic study found greater terpene abundance/diversity in outdoor samples, especially sesquiterpenes. Sample size and environmental confounding prevent universalization.
Claim 4: “LEDs fixed indoor cannabis’s environmental footprint.”
Verdict: False.
LEDs improve lighting efficiency. HVAC, cooling, dehumidification, COâ‚‚, and the remaining lighting load still create high energy demand.
Claim 5: “Greenhouse is always the sustainability middle ground.”
Verdict: Too simplistic.
A passive hoop house can be very low-energy. A sealed greenhouse with supplemental LEDs, cooling, heating, and dehumidification can be much more energy intensive.
Claim 6: “Light dep gives growers several harvests per year.”
Verdict: Technically well founded, exact cycle count context-dependent.
Artificially manipulating darkness can induce flowering before natural seasonal photoperiod would do so. Annual crop count depends on climate, cultivar, propagation, vegetative duration, flowering duration, and environmental control.
Claim 7: “Indoor is cleaner.”
Verdict: Not established generally.
Indoor excludes some outdoor contamination pathways but can develop severe microbial problems and can still involve pesticide residues. Heavy-metal risk is more closely tied to inputs and media.
Claim 8: “Outdoor is cheaper.”
Verdict: Often true in reported industry datasets, not a timeless rule.
Outdoor avoids most lighting/HVAC costs, but labor, land, crop loss, regulation, water, postharvest infrastructure, and wholesale pricing matter.
Claim 9: “Indoor yields more.”
Verdict: Depends on denominator.
Indoor often produces more annual yield per unit floor area due to multiple cycles and high-density/high-light production. Outdoor may produce much larger plants. Yield per plant, per crop area, and per year are not interchangeable.
Claim 10: “The best cannabis is indoor.”
Verdict: Scientific evidence does not support a universal ranking.
“Best” mixes chemistry, potency, appearance, aroma, contamination, consistency, environmental footprint, economics, and consumer preference. Different systems optimize different dimensions.
Legacy-claim re-test notes for editor
“Greenhouse + light deprivation can increase cycles up to 5×”
Status after first pass: Needs qualification, not immediate rejection.
Evidence strongly supports the mechanism by which light deprivation enables off-season flowering and multiple cycles. A 2026 greenhouse economic paper explicitly models five annual autoflower cycles and four photoperiod cycles, showing that multiple protected-culture cycles are feasible in some systems.
What remains unsupported is a general statement that light deprivation itself increases crop cycles fivefold across cannabis cultivation. A canonical revision should specify climate/system assumptions.
Suggested rewrite candidate:
Light-deprivation and protected greenhouse systems can enable multiple harvests per year by decoupling flowering initiation from natural seasonal day length; the achievable number of cycles depends on cultivar, propagation strategy, crop duration, climate, and facility controls.
“Flushing ensures you won’t be smoking trace chemicals.”
Status for R05: Not resolved here; defer to nutrition/input report.
This claim concerns preharvest nutrient management and contaminant chemistry rather than cultivation environment. It should receive a dedicated evidence review rather than being inferred from R05.
Data gaps
No robust current North American production share by indoor / greenhouse subtype / outdoor was identified.
Greenhouse energy LCAs are insufficiently stratified by passive, light-dep, supplemental-light, and sealed greenhouse type.
Direct matched-genetic indoor-vs-greenhouse-vs-outdoor quality trials are rare.
Consumer blind-testing studies comparing matched genetics and postharvest treatment are scarce.
Physical/commercial quality metrics such as density, visual grade, and aroma retention lack standardized comparative research.
Contamination data by cultivation environment are fragmented and confounded by different state testing rules.
Waste and plastics are undercharacterized compared with energy.
Water footprint per kg saleable flower is not standardized across environments.
Climate-change resilience comparisons need more explicit modeling.
Commercial yield datasets often mix projected and actual production and inconsistent functional units.
Postharvest energy can materially affect environmental comparisons but is inconsistently included.
Carbon accounting boundaries differ enough that published carbon values should not simply be pooled.
Genotype × environment interactions remain a major unresolved issue for potency/terpene comparisons.
Illicit production volume and environmental mix are inherently uncertain.
True life-cycle economics including financing, depreciation, crop loss, taxes, compliance, and wholesale quality premiums are rarely reported consistently.
Strongest sources — editorial shortlist
1. Summers, Sproul & Quinn (2021)
The greenhouse gas emissions of indoor cannabis production in the United States. Nature Sustainability 4:644–650.
DOI: 10.1038/s41893-021-00691-w
https://www.nature.com/articles/s41893-021-00691-w
Reliability note: Peer-reviewed cannabis-specific cradle-to-gate LCA with geographic climate/grid sensitivity. Modeled rather than facility census data.
2. Desaulniers Brousseau et al. (2024)
Environmental Impact of Outdoor Cannabis Production. ACS Agricultural Science & Technology 4(7):690–699.
DOI: 10.1021/acsagscitech.4c00054
https://pubs.acs.org/doi/10.1021/acsagscitech.4c00054
Reliability note: Cannabis-specific outdoor production trial integrated into LCA; especially important because outdoor evidence is sparse.
3. Desaulniers Brousseau et al. (2024)
Greener green: The environmental impacts of the Canadian cannabis industry. Resources, Conservation & Recycling 208:107737.
DOI: 10.1016/j.resconrec.2024.107737
https://doi.org/10.1016/j.resconrec.2024.107737
Reliability note: Current Canada-focused LCA with HVAC modeling and regional climate sensitivity.
4. One Earth (2025)
Energy-intensive indoor cultivation drives the cannabis industry’s expanding carbon footprint. One Earth 8(3):101179.
DOI: 10.1016/j.oneear.2025.101179
https://www.sciencedirect.com/science/article/pii/S2590332225000053
Reliability note: Broad industry-scale LCA/scenario analysis; useful for system-wide emissions but total production includes modeled assumptions.
5. PG&E / ETCC (2022–2024)
Controlled Environment Horticulture: Energy Consumption and Environmental Control Field Study.
Reliability note: Direct facility field measurements from two California indoor cannabis sites; small sample but strong operational detail.
6. Zandkarimi et al. (2023)
Comparison of the Cannabinoid and Terpene Profiles in Commercial Cannabis from Natural and Artificial Cultivation. Molecules 28(2):833.
DOI: 10.3390/molecules28020833
https://www.mdpi.com/1420-3049/28/2/833
Reliability note: Rare matched-genetic indoor/outdoor commercial comparison. Important but small and environmentally confounded.
7. Collado, Hwang & Hernández (2024)
Supplemental greenhouse lighting increased the water use efficiency, crop growth, and cutting production in Cannabis sativa. Frontiers in Plant Science.
DOI: 10.3389/fpls.2024.1371702
https://www.frontiersin.org/journals/plant-science/articles/10.3389/fpls.2024.1371702/full
Reliability note: Controlled greenhouse cannabis experiment; directly relevant to supplemental light, growth, and water-use efficiency.
8. Collado & Hernández (2025)
Vegetative and reproductive stage lighting interactions on flower yield, water use efficiency, terpenes, and cannabinoids of Cannabis sativa. Scientific Reports 15:43641.
https://www.nature.com/articles/s41598-025-27437-4
Reliability note: Controlled greenhouse flower-production research connecting light, water, yield, cannabinoids, and terpenes.
9. Vergara & Ruterbories (2026)
High-THC Cannabis sativa in a New York greenhouse: yield and economic factors. Journal of Cannabis Research 8:85.
DOI: 10.1186/s42238-026-00429-5
https://link.springer.com/article/10.1186/s42238-026-00429-5
Reliability note: Rare recent peer-reviewed greenhouse economics/yield study. Highly jurisdiction- and assumption-sensitive.
10. Backer et al. (2019)
Closing the Yield Gap for Cannabis: A Meta-Analysis of Factors Determining Cannabis Yield. Frontiers in Plant Science 10:495.
DOI: 10.3389/fpls.2019.00495
https://www.frontiersin.org/journals/plant-science/articles/10.3389/fpls.2019.00495/full
Reliability note: Valuable synthesis of yield drivers; predates much of the modern legal-market literature.
11. Health Canada (2026)
Cannabis data overview: Inventory, sales and licensed area.
https://www.canada.ca/en/health-canada/services/drugs-medication/cannabis/research-data/market.html
Reliability note: Official regulated-market dataset; area is not equivalent to production.
12. California Department of Cannabis Control
Cultivation license types / market data / regulations.
https://www.cannabis.ca.gov/applicants/license-types/cultivation-license-types/
Reliability note: Authoritative regulatory definitions and licensing structure; license counts are not yield estimates.
13. Bauer et al. (2015)
Impacts of Surface Water Diversions for Marijuana Cultivation on Aquatic Habitat in Four Northwestern California Watersheds.
https://pmc.ncbi.nlm.nih.gov/articles/PMC4364984/
Reliability note: Important regional ecological study; water-use assumptions should not be generalized nationally.
14. Green et al.
Microbial hazards during harvesting and processing at an outdoor United States cannabis farm.
https://pmc.ncbi.nlm.nih.gov/articles/PMC6314012/
Reliability note: Direct occupational/microbial field evidence; not a comparative contamination-rate study.
15. California/illicit cultivation environmental studies
Passive monitoring of soluble pesticides linked to cannabis cultivation and 2025 trespass-grow contaminant work.
https://doi.org/10.2166/wqrj.2022.101
https://www.sciencedirect.com/science/article/pii/S0048969725022168
Reliability note: Strong evidence of localized illicit-cultivation externalities; should not be transferred to regulated outdoor production.
Candidate claims for R05 Claim Ledger review
These are candidate statements only. Final evidence status and confidence belong to Claude/Fable.
R05 candidate 01
Indoor cannabis cultivation generally has substantially higher operational energy demand and greenhouse-gas emissions per kg dried flower than outdoor cultivation, although the magnitude varies strongly with climate, grid mix, facility design, and lifecycle boundary.
R05 candidate 02
Modern LED lighting improves indoor cannabis energy efficiency but does not eliminate the major HVAC, dehumidification, environmental-control, and remaining lighting loads of fully indoor production.
R05 candidate 03
Climate can shift the dominant energy end use in controlled cannabis facilities: cooling/dehumidification may dominate in warm humid regions, while heating can dominate in cold climates.
R05 candidate 04
Outdoor cannabis cultivation generally has a much lower cultivation-stage carbon footprint than fully indoor production in the cannabis-specific LCAs currently available.
R05 candidate 05
A low carbon footprint does not make outdoor cannabis automatically sustainable because site-specific water diversion, nutrient loss, erosion, pesticide use, habitat disturbance, and land-use impacts can be substantial.
R05 candidate 06
“Greenhouse” is not a single environmental category; passive, light-deprivation, supplemental-light, and highly controlled greenhouse systems can have materially different energy and resource profiles.
R05 candidate 07
Light deprivation induces flowering in photoperiod-sensitive cannabis by extending the uninterrupted dark period while natural day length remains longer than the cultivar’s flowering-inductive threshold.
R05 candidate 08
Light-deprivation and protected greenhouse systems can enable multiple annual cannabis harvests, but achievable crop count depends on cultivar, climate, propagation schedule, crop duration, and facility controls.
R05 candidate 09
Comparative cannabis sustainability results are sensitive to the functional unit used; per plant, per m², per crop, per year, and per kg saleable flower can produce different apparent rankings.
R05 candidate 10
Cultivation environment can alter cannabis cannabinoid and terpene profiles even among genetically identical clone stock.
R05 candidate 11
Current direct evidence is insufficient to conclude that indoor cannabis is inherently or universally higher in THC than outdoor cannabis independent of genetics and crop management.
R05 candidate 12
A small direct matched-genetic commercial study found greater abundance and diversity of several terpenes, particularly sesquiterpenes, in outdoor-grown samples than indoor-grown samples, but the result is not sufficient to establish a universal sun-grown terpene advantage.
R05 candidate 13
Claims that sun-grown cannabis is chemically superior or indoor cannabis is chemically superior are both stronger than the present direct comparative evidence base.
R05 candidate 14
Indoor cultivation can reduce exposure to weather and some outdoor contamination pathways but is not inherently free of microbial, pesticide, or heavy-metal contamination risks.
R05 candidate 15
Heavy-metal contamination risk in cannabis depends substantially on growing media/soil, fertilizers, irrigation water, and cultivar uptake characteristics rather than cultivation environment alone.
R05 candidate 16
Outdoor and illicit cannabis cultivation can create locally significant pesticide and watershed impacts when production occurs without effective environmental controls.
R05 candidate 17
Indoor cultivation can support year-round production and multiple annual crop cycles, while conventional outdoor photoperiod cultivation is more strongly season-limited.
R05 candidate 18
Yield comparisons among cultivation environments are frequently misleading when studies use different plant densities, cultivars, crop durations, lighting levels, and yield units.
R05 candidate 19
Historical industry surveys generally report higher production cost per pound for indoor cannabis than greenhouse or outdoor cannabis, but these values are not transferable across jurisdictions or market years without adjustment.
R05 candidate 20
Cannabis profitability can change sign with wholesale price, yield, labor, planting material, energy, and crop-cycle assumptions; cultivation environment alone does not determine economic viability.
R05 candidate 21
The environmental advantage of greenhouse cultivation over indoor production is conditional rather than guaranteed because heating, cooling, dehumidification, and supplemental lighting can materially increase greenhouse energy use.
R05 candidate 22
Direct matched-system evidence comparing contamination rates across indoor, greenhouse, and outdoor cannabis is currently insufficient for a universal ranking.
R05 candidate 23
Controlled environments provide greater ability to standardize temperature, humidity, light, irrigation, and crop timing, but this control is purchased with infrastructure, energy use, and equipment-failure risk.
R05 candidate 24
Outdoor cultivation transfers a larger portion of production risk from mechanical systems to weather, seasonal climate, pests, fire/smoke, water availability, and land/ecosystem conditions.
R05 candidate 25
Greenhouse supplemental lighting can increase cannabis growth and improve water-use efficiency per unit biomass even while increasing total crop evapotranspiration.
R05 candidate 26
Water-use sustainability cannot be ranked reliably by liters applied per plant alone; source scarcity, rainfall, recapture, runoff, and yield-normalized water use all matter.
R05 candidate 27
A current authoritative North America-wide breakdown of cannabis production volume by indoor, greenhouse subtype, and outdoor system is not available from the sources identified in this review.
R05 candidate 28
Market price premiums for indoor flower demonstrate consumer and market valuation, but they do not by themselves demonstrate biochemical or toxicological superiority.
R05 candidate 29
Postharvest conditions are a significant confound when comparing flower chemistry across cultivation environments because oxidation, drying, curing, storage, and handling can alter measured cannabinoids and terpenes.
R05 candidate 30
The strongest current evidence differentiating cannabis cultivation systems concerns energy and carbon impacts; evidence for universal product-quality rankings is substantially weaker.
Overall limitations
This is a first research pass for a pilot workflow, not the final canonical R05 synthesis.
Major limitations include:
incomplete access to some full-text papers
inconsistent system definitions across jurisdictions
limited direct three-way cultivation comparisons
heavy reliance on modeling for life-cycle comparisons
underdeveloped greenhouse subtype literature
rapidly changing legal-market economics
poor visibility into illicit production
cultivar and postharvest confounding in quality studies
regulatory differences in testing thresholds
mismatch among production functional units
The evidence is nevertheless sufficient to support a strong provisional conclusion: energy/carbon differences among cultivation environments are real and large; product-quality hierarchies are not nearly as settled as cannabis marketing culture implies.
Search stopping rationale
The first pass stopped after the major evidence lanes produced diminishing returns:
multiple current cannabis-specific indoor/outdoor LCAs converged on the same broad energy/carbon direction;
modern indoor field data established the continuing HVAC/dehumidification burden;
direct quality searches repeatedly converged on a small number of matched-genetic studies rather than a large comparative literature;
market-share searches produced authoritative licensing/area data but no defensible North America-wide production split;
economics searches confirmed high jurisdiction sensitivity and limited peer-reviewed comparative evidence.
The remaining gaps are best handled by targeted follow-up during Claude’s cross-examination, rather than by adding more loosely comparable sources to the raw research pass.
R05 ChatGPT Deep Research Pass | August 25, 2026