Sunteți pe pagina 1din 6

Forensic Science International 212 (2011) 158163

Contents lists available at ScienceDirect

Forensic Science International


journal homepage: www.elsevier.com/locate/forsciint

Factors determining yield and quality of illicit indoor cannabis (Cannabis spp.) production
Wouter Vanhove a,1,*, Patrick Van Damme a, Natalie Meert b,2
a Laboratory of Tropical and Subtropical Agriculture and Ethnobotany, Department of Plant Production, Faculty of Bio-Science Engineering, Ghent University, Coupure Links 653, 9000 Gent, Belgium b National Institute for Criminalistics and Criminology, Vilvoordsesteenweg 100, 1120 Brussel 5, Belgium

A R T I C L E I N F O

A B S T R A C T

Article history: Received 25 February 2011 Received in revised form 23 May 2011 Accepted 2 June 2011 Available online 7 July 2011 Keywords: Cannabis Indoor plantation THC Cannabinoid Yield

Judiciary currently faces difculties in adequately estimating the yield of illicit indoor cannabis plantations. The latter data is required in penalization which is based on the prots gained. A full factorial experiment in which two overhead light intensities, two plant densities and four varieties were combined in the indoor cultivation of cannabis (Cannabis spp.) was used to reveal cannabis drug yield and quality under each of the factor combinations. Highest yield was found for the Super Skunk and Big Bud varieties which also exhibited the highest concentrations of D9-tetrahydrocannabinol (THC). Results show that plant density and light intensity are additive factors whereas the variety factor signicantly interacts with both plant density and light intensity factors. Adequate estimations of yield of illicit, indoor cannabis plantations can only be made if upon seizure all factors considered in this study are accounted for. 2011 Elsevier Ireland Ltd. All rights reserved.

1. Introduction Over the last few years, it was found that the number of (illegal) indoor cannabis (Cannabis sativa L.) plantations in Belgium has risen considerably. Besides an increase in small-scale plantations, Belgian judicial authorities also observed a rise in number of large-scale plantations [1]. It is therefore of increasing importance that judicial response to these activities is consolidated in seizure (and subsequent conscation) of the capital gained by the perpetrators. Due to the current lack of hard facts, Belgian judiciary is forced to make rough estimations of the prots gained, based on estimated crop yield of seized cannabis harvests and wholesaler prices used in amongst others Dutch coffeeshops. Today, the Belgian judiciary uses a crop yield estimation made by the University of Wageningen, The Netherlands, [2] which is set at 28.1 g of female ower buds per plant (lower bound of the one-sided 95% condence interval). The latter study nevertheless has some important shortcomings. It is based on discovery and conscation of 77 indoor plantations of which crop yield was estimated at the spot. The study correctly accounts for plant density and light intensity as important yield

* Corresponding author. Tel.: +32 (0)9 264 60 89, fax: +32 (0)9 264 62 41. E-mail address: Wouter.Vanhove@UGent.be (W. Vanhove). 1 http://www.tropicallab.ugent.be. 2 Tel.: +32 2 240 05 06; fax: +32 2 241 61 05; http://www.nicc.fgov.be. 0379-0738/$ see front matter 2011 Elsevier Ireland Ltd. All rights reserved. doi:10.1016/j.forsciint.2011.06.006

determining factors, but neglects variation in crop yield that undoubtedly occurs between different cannabis varieties. Recent seizures of Belgian indoor plantations revealed that cannabis cultivation has become increasingly sophisticated, using automated lighting, ventilation and irrigation systems, and fertilization packages that follow technical grow schemes obtained from Dutch grow shops [3]. State-of-the-art knowledge on indoor cannabis production is mainly obtained from so-called grey resources [46]. Peerreviewed research on Cannabis spp. focuses on medicinal (e.g. [7]) or pharmacological properties [8,9] or on issues relating to cannabis use policy and/or criminology (e.g. [1]). Agronomic cannabis research has mainly focused on eld production and yield of bre hemp [1013]. So far, Toonen et al. [2] and Knight et al. [14] are the only researchers who reported on some agronomic features of indoor cannabis production in The Netherlands and New Zealand respectively. Concentrations of the main psychoactive substance, the cannabinoid D9-tetrahydrocannabinol (THC), have risen considerably until 2004 in so-called nederwiet cannabis varieties (i.e. those that are mainly used in Belgium and the Netherlands). Ever since, THC concentrations in nederwiet have stabilized between 15% and 20% [15]. Other cannabinoids are of minor importance. Cannabinol (CBN), the primary product of THC degradation, is only mildly psychoactive. Cannabidiol (CBD) is not psychoactive but attenuates the effect of THC [8,9].

W. Vanhove et al. / Forensic Science International 212 (2011) 158163

159

Although hydroponics i.e. growing on substrates other than soil (using rock wool, coco bre, perlite, vermiculite) are becoming increasingly popular among cannabis producers, the present study focuses on cultivation in peat soil, which is the mainstream indoor cannabis production system in Belgium. The aims of this study are to (i) determine cannabis yield in indoor cannabis plantations; (ii) assess which are the most important yield-determining factors; (iii) assess whether cannabinoid concentrations correspond with values previously reported in The Netherlands [15]; (iv) evaluate which of the yield determining factors also affect cannabinoid concentrations.
2. Material and methods 2.1. Grow experiment 2.1.1. Experimental design Toonen et al. [2] evaluated the frequency of plant densities that prevailed in the 77 conscated grow rooms. 80% of grow rooms contained densities between 10 and 30 plants m2. Average density was 15 plants m2. In the present study, two plant densities were applied: 16 and 20 plants m2, corresponding with 43% of plantations in the 916 plants m2 interval and 29% of plantations in the 1724 plants m2 interval, respectively, that were reported by Toonen et al. [2]. Overhead lights should provide at least 54,000 lumens (lm) per m2 [4,6]. Nowadays, these light intensities are amply achieved by high-pressure sodium (HPS) or metal halide (HPI) lamps of either 400 W or 600 W [2]. As a result, both light intensities were used as a treatment in this study. The experiment hence consisted of three yield determining factors that were combined in a full factorial design: (i) light intensity (400 W m2 and 600 W m2); plant density (16 and 20 plants m2); and (ii) variety (Super Skunk, Northern Light #5 Haze, White Widow and Big Bud). Light intensity and plant density blocks were combined in 4 blocks of 4 m2 each. Each block was further divided in 4 subplots, each containing a different cannabis variety. A total of sixteen square experimental blocks (1 m2) were thus obtained, each containing 16 or 20 plants m2 and receiving light from one 400 W or 600 W lamp. Subplots with cannabis varieties were positioned in such a way that the complete experimental design formed a Latin square (Fig. 1). The experiment was conducted in a secured square grow room with sides of 5.5 m and a height of 2.4 m.

important to have only female plants in the experimental blocks. To this aim, only so-called feminized seeds were bought. Indeed, during the experiment no male plants were reported. Seeds were sown in peat soil on February 4th 2010 in a nursery at the greenhouse in the Faculty of Bioscience Engineering, Ghent University. Seed plants were propagated by means of stem cuttings taken on March 3rd and March 24th 2010. Rooting of cuttings was stimulated by dipping each cutting in Rhizopon ATM powder (indole-3-acetic acid), as described by Arteca [16]. Cuttings were placed in rock wool cubes and covered with plastic in the greenhouse nursery until they were sufciently rooted (i.e. after between 2 and 3 weeks). Cuttings with rock wool cubes were subsequently transplanted in pots (diameter: 0.2 m) lled with peat soil in order to grow parent plants (15 per variety). On April 30th 2010, 150 stem cuttings per variety were taken from these plants using the same methodology as for the cuttings described above. The experiment consequently consisted of a random mixture of clones from 10 original plants per variety. Rooted cuttings were transplanted in pots at the experimental room on May 20th. Plants were placed on the ground either in round pots (diameter 0.22 m) of 5 L (in blocks of 20 plants m2) or in square pots (sides 0.25 m) of 11 L (in blocks of 16 plants m2). Lamps were positioned above the centre of each subplot of 1 m2. Pots were lled with peat soil (pH 6.4, OM: 20%) that was mixed with 5% perlite. 2.1.3. Nutrition Cannabis plants were uniformly treated with fertilizers that were added to the irrigation water. Fertilization and light regime were performed according to the Canna Terra grow scheme (except for the application of root stimulator Growstar1 Gold Excelerator) that according to Belgian federal police is frequently applied in illicit cannabis cultivation (Table 1). The scheme was found at http://www.onlinegrowsupplies.com/CANNA_Terra_kweekschema.pdf (last visit: 19 May 2011). Fertilizers and other additives were obtained from illicit plantations that were conscated by the Belgian federal police. Terra VegaTM (NPK 3-1-4) and Terra FloresTM (NPK 2-2-4) fertilizers stimulate vegetative growth and owering, respectively. Growstar1 Gold ExceleratorTM is a root stimulator in which auxins (no further specication) are active components. CannazymTM is, according to the label, an enzyme mixture (not specied) which is claimed to enhance mineralization of organic soil material, such as dead roots. CannaboostTM is advertised on its label as an additive which improves photosynthesis. Although some websites claim that the active substance in CannaboostTM is the natural growth stimulant triacontanol (C30H61OH), no reliable information on the composition of CannaboostTM was found. PK13/14TM (NPK 0-13-14) is added during one week in the middle of the owering stage in order to support the intensive owering in that period. Irrigation water was applied every two days in amounts that were arbitrary determined on the basis of plant requirements. 2.1.4. Environmental control Total grow cycle was 11 weeks and consisted of a vegetative stage (rst 4 weeks) in which light was provided during 18 h per day and a owering stage (subsequent 7 weeks) initiated by and with 12 h of light and 12 h of darkness. Switching lights from an 18 h to a 12 h light regime creates an articial short-day period, inducing owering in cannabis. Lamp height in each of the subplots was continuously adjusted so that lamps remained at 0.6 m above canopy level. Lamps of 400 W were Philips1 Master HPI-T Plus E40 1SL metal-halide lamps. Lamps of 600 W were Philips1 Master SON-T PIA Plus E40 high-pressure sodium lamps. Since plants remained visually healthy during the whole growth period, no pesticides were used. Plants that produced heavy apical female owers were supported by means of bamboo sticks in order to avoid stem bending. Temperature inside the grow room was kept below 30 8C by continuously removing hot air using a RoScro1 turbine with a maximum ow rate of 6000 m3 h1. Turbine speed was controlled by a Torinsifan1 Regulation Intelligent Controller (R.I.C.), functioning as a thermostat. Removed air was continuously ltered using a cylindrical carbon lter (brand unknown, height: 1 m, diameter: 0.44 m) in order to prevent the intense smell during owering from leaving the room. Fresh air entered the room through a vent hole of 1.2 m 0.8 m in the wall opposite the carbon lter. Air circulation inside the grow room was further enhanced by a Honeywell NV-1800E ventilator (ow rate: 6100 m3 h1) that was placed in the middle of one of the side walls, square to the walls with carbon lter and vent hole. All aforementioned equipment was obtained from a stock of conscated cannabis production material, held by the Belgian federal police.

2.1.2. Preparation Cannabis varieties were purchased in packages of 10 seeds per variety from growshops in Amsterdam (Netherlands). Varieties Super Skunk, Northern Light #5 Haze and Big Bud were obtained from the Sensi Seed Bank, a renowned cannabis seed bank and cannabis breeder. The White Widow variety was bought in the FantaSeeds growshop. Since only female plants produce cannabis buds (containing highest concentrations of the psychoactive cannabinoids), it was

Fig. 1. Experimental design of the cannabis cultivation experiment (SS = Super Skunk, WW = White Widow, NL = Northern Light #5 Haze, BB = Big Bud).

2.1.5. Data collection After 11 weeks (July 30th 2010), when pistils started turning brown, plants were labelled and harvested. The main stem was cut at the bottom after which plants were hung upside down to dry in the dark during 10 days. Subsequently, female owers were separated from the plants using scissors and leaves between owers manually removed. Finally, owers harvested were weighed per plant using an OHAUS AdventurerTM balance (precision: 1 mg). Yield per plant was converted to yield per square meters by multiplying per plant yield by 16 and 20 in blocks with 16 and 20 plants m2, respectively. Both yield gures were analyzed using ANOVA in SPSS 17.

160

W. Vanhove et al. / Forensic Science International 212 (2011) 158163

Table 1 Grow scheme applied in the cannabis cultivation experiment, based on the Canna1 Terra grow scheme*. Root stimulation was performed using a product from the brand Growstar1 **. Weeks are calendar weeks. Week 2021 2223 24 25 2628 2930 Light (h/day) 18 18 12 12 12 12 Terra Vega* (mL/10 L) 20 40 Terra Flores* (mL/10 L) 60 55 60 Gold Excelerator** (mL/10 L) 40 15 5 5 5 Cannazym* (mL/10 L) 25 25 25 25 25 Cannaboost* (mL/10 L) 30 30 30 30 PK 13/14* (mL/10 L) 15

2.2. Analysis of cannabinoid concentrations For each of the 16 factor combinations, a mixed sample of ower buds from 10 randomly chosen dried plants was taken. Reference materials D9-tetrahydrocannabinol (THC), cannabinol and cannibidiol were purchased from Lipomed. Ethanol was purchased from Biosolve and tribenzylamine from Acros Organics. Identication and analysis of cannabis and cannabis products was based on recommendations by the United Nations Ofce on Drugs and Crime (UNODC) [17]. Homogenized plant material (2030 mg) was extracted with 10 mL internal standard solution (0.5 mg mL1 tribenzylamine in ethanol) for 10 min in an ultrasonic bath. Subsequently samples were placed in a rotator for 30 min. Gas chromatography mass spectrometry (GCMS) analysis (HP6890N-5973N, Agilent Technologies) was performed for identication, which was based on comparison with reference materials and library spectra (PMW and NIST.08). Gas chromatography ame ionization detection (GCFID) analysis was subsequently done for quantication, using HP6890 (Agilent Technologies). Separation was achieved on a HP 1 column (25 m 0.32 mm I.D. 0.52 mm lm thickness; J & W Scientic, Agilent Technologies). Carrier gas, helium, had a constant ow of 1.8 mL/min. One mL sample was injected with a split ratio of 25:1 at 275 8C. Oven temperature was programmed at 250 8C (hold 9.50 min). Detector parameters were set at a temperature of 300 8C, a hydrogen ow of 30 mL/min and an air ow of 400 mL/min. Chemstation software (Agilent Technologies) was used for data retrieval and calculations. Calibration was performed with cannabinol as reference material.

3. Results 3.1. Yield At harvest, almost every plant bore one apical and in some cases several lateral female ower buds. They consisted of dark green calyxes containing brown or orange pistils, that formed a compact owering mass; sometimes up to 8 cm thick and covering several internodes. Three plants were found to be stunted and without ower buds (two plants of the Big Bud variety under 400 W lamps, one in a block of 16 plants m2, the other in a block of 20

plants m2; and one plant of the White Widow variety grown under a 600 W lamp in a block of 16 plants m2). The highest yielding plant in the plantation was a Super Skunk plant grown under a 600 W lamp in a block of 16 plants m2 (48.32 g), whereas the lowest yield (0.37 g) was recorded for a White Widow plant in the same block. Average yield per plant and per m2 for each of the 16 factor combinations are shown in Table 2. Highest average yield per plant (28.00 g) was observed in the subplot of Big Bud plants, grown under a 600 W lamp with a density of 16 plants m2. Highest yield per m2 (483.33 g m2) was also found in a subplot of Big Bud plants, grown under a 600 W lamp, but with a density of 20 plants m2. Lowest average yield per plant (6.24 g) as well as lowest yield per square meters (124.88 g) are found in the subplot with Northern Light #5 Haze plants, grown under a 400 W lamp with a plant density of 20 plants m2. Table 3 shows the results of the analysis of variance. Considering only yield per plant, it was found that main effects of all three factors are signicant (p < 0.05) with higher yields for plants grown under 600 W lamps in comparison with plants grown under 400 W lamps; with higher yields for plants grown in subplots with a density of 16 plants m2 in comparison with yields of plants grown in subplots with a density of 20 plants m2; and with higher yields for Super Skunk and Big Bud varieties in comparison with Northern Light #5 Haze and White Widow varieties (Table 4). The same observations can be made when considering yield per square meters, except that there are no signicant (p < 0.05) differences in yield (in g m2) between blocks of 16 plants m2 and blocks of 20 plants m2. For both yield measurements, plant density and light intensity can be considered to be additive factors (p > 0.05 for light density interaction in Table 3). The variable variety is signicantly interacting with both

Table 2 Average yield and THC, CBN and CBD concentration (%) of indoor cannabis for each of 16 factor combinations (light intensity, plant density and variety) (NLX = Northern Light #5 Haze). Light Density Variety n Yield (g per plant) Mean 400 W 16 m2 Big Bud NLX Super Skunk White Widow Big Bud NLX Super Skunk White Widow Big Bud NLX Super Skunk White Widow Big Bud NLX Super Skunk White Widow 15 16 16 16 19 20 20 20 16 16 16 15 20 20 20 20 9.91 11.63 18.58 8.91 12.61 6.24 11.25 13.01 28.00 17.42 23.41 15.49 21.06 16.83 16.42 11.54 Standard Deviation 4.32 6.72 7.95 6.55 5.90 4.21 6.19 6.20 8.34 8.86 11.79 7.82 9.34 8.61 6.08 7.64 Yield (g m2) Mean 142.51 186.15 338.54 142.52 252.10 124.88 225.08 260.11 447.97 278.72 443.87 268.29 483.33 357.55 375.98 230.80 Standard Deviation 69.16 107.59 165.28 104.86 118.03 84.24 123.86 124.05 133.42 141.69 225.36 144.23 231.49 191.84 187.18 152.76 Concentrations (%) THC 15.3 10.9 14.3 9.7 12.5 13.3 12.8 11.7 15.9 12.6 15.2 10.4 14.8 8.3 16.3 8.5 CBN 0.2 0.2 0.3 0.2 0.2 0.2 0.2 0.2 0.2 0.2 0.2 0.2 0.2 0.2 0.2 0.1 CBD 0.3 0.2 0.3 0.2 0.3 0.2 0.3 0.3 0.3 0.3 0.4 0.2 0.4 0.4 0.3 0.2

20 m2

600 W

16 m2

20 m2

W. Vanhove et al. / Forensic Science International 212 (2011) 158163 Table 3 Results of the analysis of variance on indoor cannabis yield. Yield (g per plant) F-value Corrected model Intercept Light Density Variety Light Density Light Variety Density Variety Light Density Variety 10.87 1019.25 73.24 9.71 12.97 1.55 6.43 4.70 3.15 p 0.000 0.000 0.000 0.002 0.000 0.214 0.000 0.003 0.025 Yield (g m2) F-value 9.76 984.92 70.01 0.18 11.89 0.09 6.37 3.83 3.17 p

161

0.000 0.000 0.000 0.669 0.000 0.769 0.000 0.010 0.025

Table 4 Main effects of light intensity, plant density and variety on indoor cannabis yield and quality. For each of the 3 factors, different lower-case letters (a and b) indicate signicant differences (p < 0.05) between means. T-tests were used for comparing means between different light intensities and between different plant densities. For comparison of mean yield between varieties, Dunnett T3 method was used (unequal variances); comparison of mean THC concentrations between varieties was done using the LSD method. Yield (g) Per plant Factor Light intensity 400 W m2 600 W m2 16 plants m2 20 plants m2 Super Skunk Big Bud Northern Light #5 Haze White Widow N 142 143 126 159 72 72 70 71 Mean 11.65 20.08 17.62 14.49 19.21 18.43 13.28 12.46 a b a b a a b b Standard Deviation 7.49 11.12 11.33 9.34 11.54 11.13 8.97 7.68 Per m Mean 210.48 361.72 281.93 289.88 340.83 335.78 239.65 227.08 a b a a a a b b
2

THC Concentration (%)

Standard Deviation 131.31 197.94 181.20 186.85 191.05 205.63 165.32 139.08

N 8 8 8 8 4 4 4 4

Mean 12.6 12.8 13.0 12.3 14.7 14.6 11.3 10.1 a a a a a a b b

Standard Deviation 1.8 3.3 2.5 2.8 1.5 1.5 2.2 1.3

Plant Density

Variety

plant density and light intensity (p < 0.05 for light variety and density variety interactions in Table 3). Fig. 2 shows that yield differences (in g m2) in blocks with 600 W lamps in comparison with yield per square meters in blocks with 400 W lamps is relatively much higher for varieties Big Bud and Northern Light #5 Haze than it is for varieties Super Skunk and White Widow. Within the blocks with 400 W lamps, yield (in g m2) of plants of the Super Skunk variety is signicantly (p < 0.05) higher than yield (in g m2) of plants of the other varieties, whereas in the blocks with 600 W lamps, yield (in g m2) of Super Skunk plants signicantly (p < 0.05) equals yield of Big Bud plants but yield (in g m2) of both varieties differs from yields of White Widow and Northern Light #5 Haze plants.

3.2. Cannabinoid concentrations Highest THC concentration (16.3%) was found for a mixed sample of Super Skunk plants grown at a density of 20 plants m2 under 600 W lamps. Lowest THC concentration (8.3%) was detected with a mixed sample of Northern Light #5 Haze plants grown under the same conditions. Mean values of CBN and CBD concentrations were 0.2% and 0.3%, respectively, with very little variation (standard deviations of 0.04 and 0.07, respectively) (Table 2). Since only 16 THC measurements were performed, interaction effects of factors on THC concentrations could not be examined. Considering main effects of three grow factors on THC concentrations, only the variety factor proved to be signicant (p < 0.05, LSD). Highest average THC concentrations were found with the Super Skunk and Big Bud variety (14.7% and 14.6% respectively), which also had the highest average yield (see previous section) (Table 4). Factor effects on CBN and CBD content were not evaluated given the little variation in both concentrations. 4. Discussion and conclusion Yield of indoor cannabis depends on a number of factors, three of which proved to have signicant effects in this study: light intensity above the canopy, plant density and variety. Cannabis grown under 600 W lamps has a higher yield than cannabis under 400 W lamps. Yield per plant is higher when plant density is 16 plants m2 in comparison with yield of plants grown at a density of 20 plants m2. However, there was no signicant difference (p < 0.05) between yields from both plant densities when yield was expressed per m2. The latter indicates that light interception is the limiting factor in indoor cannabis yield. Plants at lower densities intercept more light, have a higher photosynthesis and consequently a higher production [11]. In illicit cannabis grow

Fig. 2. Mean yield (g m2) of four different cannabis varieties grown indoors under 400 W and 600 W lamps. Error bars show standard deviation. Different lower-case letters (a) and (b) indicate signicant (p < 0.05) differences (LSD) between varieties within the 400 W series, different capitals (A, B and C) show signicant (p < 0.05) differences (LSD) between varieties within the 600 W series.

162

W. Vanhove et al. / Forensic Science International 212 (2011) 158163

rooms where oor space is often scarce, loss of yield per plant at higher plant densities is compensated by a higher number of plants per m2. Further research is required to evaluate whether there are signicant differences in cannabis production per m2 between plant densities lower than 16 plants m2 and higher than 20 plants m2. According to the production model of Toonen et al. [2], optimum plant density is 32 plants m2 which would yield 608 g m2, or 19 g per plant. However, the latter model does not include differences that might arise across cannabis varieties. It was shown that the factor variety interacts with the factor plant density for both yield per plant and yield per m2. Considering only the main effect of cannabis varieties, it was found that varieties with the highest yield also have the highest THC concentration. It was shown that differences in THC concentrations are predominantly linked with genotypic variation, rather than with cultivation factors. THC concentrations are slightly below the range found by Rigter and Niesinck [15] in nederwiet, but well above THC concentrations of imported cannabis (varieties not specied) that range between 5% and 10%. A substantial amount of THC in the plant material is present as the precursor D9-tetrahydrocannabinolic acid (THC-A) which is converted into the psychoactive THC when heated. It occurs by decarboxylation when cannabis products are smoked [18]. According to Dussy et al. [18], only around 70% of THC-A is converted to THC during an optimized GC analysis. In order to check the conversion factor in the analyses of this study, some additional experiments were carried out. In one experiment decarboxylation was performed prior to analysis, as proposed by UNODC [17]. A second analysis was done following the methods described in the material and methods section. Both analyses, with and without decarboxylation, resulted in similar THC concentrations. Based on these results it can be assumed that the conversion from THC-A to THC was complete in our original experiments. The lower conversion reported by Dussy et al. [18] might be attributed to differences in injector geometry, as suggested also by UNODC [17]. In jurisdiction in the Netherlands, until 2008 yield gures were based on an investigation by Huizer & Poortmanvan der Meer [19] who claimed that a condent estimation of indoor cannabis yield was 22 g per plant. However, this research was based on immature, conscated plants entailing inaccurate yield estimations. Since 2008, Dutch jurisdiction is based on the research from Toonen et al. [2] who stated that when plant density is unknown, the median grow room consists of 15 plants m2, uses a light intensity of 510 W m2 for which the lower-bound of the one-sided 95% condence interval for yield is 28.1 g per plant. The latter authors accounted for 10 different developmental stages of cannabis plants: stage 1, being the onset of owering up to stage 10, being the harvest stage with fully developed owers. Yield of plants that are at the moment of conscation in a developmental stage lower than 10, is extrapolated to obtain an estimate of yield at harvest. The present study differs considerably from earlier approaches because cannabis was produced and harvested by researchers and because it is the rst study assessing differences across cannabis varieties. Recently Knight et al. [14] also produced cannabis in three grow cycles of six plants each, using the so-called screen of green method in which branches are supported by a wire mesh, allowing the formation of a very even canopy. Varieties used are unknown, but genetic ngerprinting and subsequent principle component analysis revealed 3 distinguished groups. Yield gures obtained in this study are considerable: between 350 g and 1340 g per plant. Unfortunately, the work of Knight et al. [14] is of little value as a benchmark to the present study due to the limited number of plants, the lack of information on the varieties used and particularly because of the non-disclosure of cultivation methods (light intensity, plant density, fertilizers used, etc.). The latter

information is restricted to the New Zealand law-enforcement community in order to address police operational concerns [14]. The company Bedrocan BV obtains an average yield of 315 g per m2 for the sativa-type Bedrocan1 variety and an average yield of 251 g per m2 for the indica-type Bedica1 variety (hydroponics system with a plant density of 2.33 plants per m2 and a overhead light intensity of 423 W per m2), which concurs with mean yields found in plots with 400 W lamps in this study (information obtained from Bedrocan BV Nederland, ofcial grower of cannabis for medicinal purposes under the authority of the Ministry of Health, Welfare and Sport, Bureau of Medicinal Cannabis, The Hague, The Netherlands). It was shown that light intensity and plant density are additive factors, whereas the factor variety is interacting with both light intensity and plant density factors. Apart from the Big Bud variety, produced at a density of 16 plants m2 and under a light intensity of 600 W m2, all subplots in the research had average yields lower than 28.1 g per plant, the gure that was proposed by Toonen et al. [2]. However, neglecting differences between varieties is an important drawback of the studies of Huizer and Poortmanvan der Meer [19] and Toonen et al. [2]. Our results are far below the yield gures found by Knight et al. [14], but as stated earlier cultivation practices used by the latter authors are not known. Since Knight et al. [14] used the screen of green method, it is likely that one overhead lamp was used per plant. The present study shows that yield of illicit cannabis plantations cannot be straightforwardly estimated as a number of grams per plant. It depends on growth factors such as plant density, light intensity and variety used, all of which interact in their effect on cannabis yield. According to the Belgian Police, the yield gures presented in this study are below the average yield found in common illicit cannabis indoor plantations. In future grow experiments cultivation practices need to be optimized in order to achieve yield gures that more accurately approach real indoor cannabis yield gures. Further research will also have to include otherpossibly higher-yielding varieties, or varieties that are frequently conscated by police (but of which the variety name is usually unknown). Acknowledgements We would like to thank The Belgian Federal Science Policy Ofce for nancing our research. Special thanks to Mr. Benny Van Camp, head of the Cannabis Section within the Central Bureau of Drugs, Directorate of Offence Against the Person, Federal Police, Brussels, Belgium, for sharing with us experiences in the eld. References
[1] T. Decorte, The case for small-scale domestic cannabis cultivation, Int. J. Drug Pol. 21 (2010) 271275. [2] M. Toonen, S. Ribot, J. Thissen, Yield of illicit indoor cannabis cultivation in The Netherlands, J. Forensic Sci. 51 (5) (2006) 10501054. [3] B. Van Camp, Berekening van het wederrechtelijk vermogen (geschatte winstopbrengst) Indoor-Cannabisplantages, Belgian Federal Police, Directorate of Offence Against the Person Central Bureau of Drugs, Brussels, Belgium, 2009. [4] P. Adams, Marijuana Indoor Growing, second ed., Positive Publishers b.v.b.a, Amsterdam, The Netherlands, 2007. [5] J. Cervantes, Marijuana Horticulture: The Indoor/Outdoor Medical Growers Bible, Van Patten Publishing, Vancouver, WA, 2006. [6] G. Green, The Cannabis Grow Bible: The Denitive Guide to Growing Marijuana for Recreational and Medicinal use, fourth ed., Green Candy Press, San Francisco, CA, 2001. [7] F.K. Engels, F.A. De Jong, R.H.J. Mathijssen, J.A. Erkens, R.M. Herings, J. Verweij, Medicinal cannabis in oncology, Eur. J. Cancer 43 (2007) 26382644. [8] T.P. Freeman, C.J.A. Morgan, G.L. Schafer, H.V. Curran, In smoked cannabis, cannabidiol attenuates the acute memory impairing effects of delta D9-tetrahydrocannabinol (THC), J. Psychopharmacol. 24 (suppl. 3) (2010) 61A. [9] C. Henquet, R. Kuepper, Does cannabidiol protect against the negative effects of THC? Br. J. Psychiatry 197 (4) (2010) 259260.

W. Vanhove et al. / Forensic Science International 212 (2011) 158163 [10] S. Amaducci, A. Zatta, F. Pelatti, G. Venturi, Inuence of agronomic factors on yield and quality of hemp (Cannabis sativa L.) bre and implication for an innovative production system, Field Crops Res. 107 (2008) 161169. [11] H.M.G. Van Der Werf, The effect of plant density on light interception in hemp (Cannabis sativa L.), J. Int. Hemp Ass. 4 (1) (1997) 813. [12] P.C. Struik, S. Amaducci, M.J. Bullard, N.C. Stutterheim, G. Venturi, H.T.H. Cromack, Agronomy of bre hemp (Cannabis sativa L.) in Europe, Ind. Crop. Prod. 11 (2000) 107118. [13] H.M.G. Van Der Werf, M. Wijlhuizen, J.A.A. De Schutter, Plant density and selfthinning affect yield and quality of bre hemp (Cannabis sativa L.), Field Crops Res. 40 (1995) 153164. [14] G. Knight, S. Hansen, M. Connor, H. Poulsen, C. McGovern, J. Stacey, The results of an experimental indoor hydroponic Cannabis growing study, using the Screen of Green (ScrOG) methodYield, tetrahydrocannabinol (THC) and DNA analysis, Forensic Sci. Int. 202 (2010) 3644.

163

[15] S. Rigter, R. Niesinck, THC-concentraties in wiet. nederwiet en hasj in Nederlandse coffeeshops, Netherlands Institute of Mental Health and Addiction (Trimbos Instituut), Utrecht, The Netherlands, 2010. [16] R.N. Arteca, Plant Growth Substances: Principles and Applications, Chapman & Hall, New York, 1996. [17] UNODC, Recommended methods for the identication and analysis of cannabis and cannabis products, Manual for use by national drug analysis laboratories, Laboratory and Scientic Section United Nations Ofce on Drugs and Crime, Vienna, Austria, 2009. [18] F.E. Dussy, C. Hamberg, M. Luginbuhl, T. Schwerzmann, T.A. Briellmann, Isolation of D9-THCA-A from hemp and analytical aspects concerning the determination of D9-THC in cannabis products, Forensic Sci. Int. 149 (2005) 310. [19] H. Huizer, A.I. Poortmanvan der Meer, Report on the yield of Indoor Cannabis Plants, Forensic Laboratory of the Department of Justice, Rijswijk, The Netherlands, 1995.

S-ar putea să vă placă și