Plant material and growing conditions

In Expt. I, seed rhizomes of ‘Bubba Blue’ ginger (average weight of 60 g) and ‘Hawaiian Red’, ‘BKK’, ‘White Mango’, and ‘Black’ turmeric (average weight of 26 g) were obtained from a commercial supplier (Hawaii Clean Seed LLC, Pahoa, HI, USA). On 19 Apr 2017, seed rhizomes were initially planted individually in 0.7-gal (6.4 inches top diameter and 7.2 inches height) round containers (C300; Nursery Supplies Inc., Orange, CA, USA) filled with a substrate mix composed of (v/v) 75% sphagnum peatmoss and 25% perlite (Fafard®2P; Conrad Fafard, Inc., Agawam, MA, USA). Plants in containers were placed in a polycarbonate greenhouse at the University of Florida (UF) in Gainesville, FL, USA, and grown for 90 d with a daily average (± SD) air temperature and relative humidity (RH) of 26 ± 1.7 °C and 77.8% ± 7.2%, respectively. Air temperature and RH were measured with temperature/RH probes (S-THC-M002; Onset Computer Corp., Bourne, MA, USA) interfaced to a data logger.

All plants were subsequently repotted on 17 Jul 2017 into either small (1.5 gal; 9.0 inches top diameter and 8.5 inches height) (C600, Nursery Supplies Inc.) or large (13.3 gal; 18.4 inches top diameter and 15.1 inches height) (C6900; Nursery Supplies Inc.) round containers evaluated as treatments. Containers were filled with a substrate mix composed of (v/v) 50% coarse coconut husk chips and 50% fine coconut fiber (Envelor Inc., Old Bridge, NJ, USA). For each variety, eight replicate containers of each volume were randomly placed on four benches (15.0 × 5.9 ft) inside a polycarbonate greenhouse at UF in Gainesville, FL, USA, with automated heating and pad-and-fan evaporative cooling. Air temperature, RH, and solar daily light integral (DLI) were measured with temperature/RH probes and a quantum sensor (S-THC-M002 and S-LIA-M003, respectively; Onset Computer Corp.) interfaced to a data logger.

For a long-day effect, night interruption (NI) was delivered by incandescent lamps providing a total photon flux density (TPFD) between 400 and 800 nm of 3.2 µmol·m⁻²·s⁻¹ from 2200 to 0200 HR (Flores et al., 2021). Plants were hand-irrigated every 2 to 3 d as needed using a fertilizer solution providing 200 mg·L⁻¹ nitrogen (N) (17N–1.8P–14.1K; Greencare Fertilizers, Kankakee, MI, USA). Plants were destructively harvested between 22 and 29 Jan 2018. The daily average air temperature, RH, and solar DLI during Expt. I were 22 ± 1.9 °C, 74.1% ± 16.1%, and 16.7 ± 4.5 mol·m⁻²·d⁻¹, respectively.

In Expt. II, seed rhizomes (average weight of 75 g) of ‘Bubba Blue’ and ‘Madonna’ ginger (Hawaii Clean Seed LLC) and ‘Indira Yellow’ and ‘Hawaiian Red’ turmeric (Hawaii Clean Seed LLC) were sprouted in flat plastic trays (21 cm height × 27.8 cm width × 6.2 cm depth; T.O. Plastics, Inc. Clearwater, MN, USA) filled with a commercial substrate composed of (v/v) 79% to 87% peatmoss, 10% to 14% perlite, and 3% to 7% vermiculite (PRO-MIX BX Mycorrhizae; Premier Tech Horticulture, Rivière-du-Loup, QC, Canada). Rhizomes were kept in a growth room for 60 d under a constant ambient temperature, RH, and DLI of 21 °C, 80%, and 9 mol·m⁻²·d⁻¹ (173 ± 5 µmol·m⁻²·s⁻¹ with a 14-h·d⁻¹ photoperiod from 0600 to 2000 HR), respectively. Light was provided by broadband white light-emitting diode (LED) lamps (RAY66; Fluence Bioengineering, Austin, TX, USA). Before transplanting, sprouted rhizomes with 5- to 20-cm shoots were selected to be used in the experiment.

On 28 Aug 2019, either one or three sprouted rhizomes were transplanted into large [13.3 gal (C6900, Nursery Supplies Inc.)] or medium [3.9 gal (11.9 inches top diameter and 11.0 inches height) (C2000, Nursery Supplies Inc.)] round containers evaluated as treatments. Containers were filled with the same substrate described previously mixed with (v/v) 50% composted pine bark. Powdered dolomitic lime (M.K. Minerals, Manhattan, KS, USA) was incorporated before planting at 3 g·L⁻¹. Immediately after transplanting, six replicate containers of each treatment per variety were randomly placed on four benches (15.0 × 5.9 ft) inside a polycarbonate greenhouse at UF in Gainesville, FL, USA. Air temperature, RH, and DLI were measured with temperature/RH probes (HMP60-L; Campbell Scientific, Logan, UT, USA) and quantum sensors (SQ512; Apogee Instruments Inc., Logan, UT, USA) interfaced to a data logger (CR1000; Campbell Scientific) and multiplexer (AM16/32B, Campbell Scientific). Each bench had one temperature/RH probe and one quantum sensor placed at above-canopy height, and measurements were made every 30 s and recorded at 60-min intervals.

All plants were grown under NI delivered by red + white + far-red LED lamps (Arize Greenhouse Pro; GE Lighting, Cleveland, OH, USA) providing a TPFD of 4.5 µmol·m⁻²·s⁻¹ from 2200 to 0200 HR. Throughout the experiment, plants were fertigated with drip irrigation every 2 to 3 d as needed, using a fertilizer solution providing 150 mg·L⁻¹ N (Peters Excel 15N–2.2P–12.5K Cal-Mag Special; ICL Specialty Fertilizers, Dublin, OH, USA). Plants were destructively harvested between 10 and 13 Mar 2020. The average daily temperature, RH, and solar DLI during Expt. II were 26 ± 4 °C, 76% ± 6%, and 20.2 mol·m⁻²·d⁻¹, respectively.

Data collected

Plants in both experiments were destructively harvested 28 weeks after transplanting to enable harvests of mature rhizomes while minimizing the risk of rhizome malformation due to potential restrictions imposed by smaller containers (Flores et al., 2021; Fig. 1). Before each harvest, shoot height was measured from the substrate base to the tip of the newest fully expanded leaf and the total number of new tillers (shoots) and leaves were counted for each ginger and turmeric plant, respectively. Shoot, root, and rhizome FW were measured following each destructive harvest. Except for rhizomes, DW was measured for all plant parts by placing bagged tissue in a forced-air drying oven at 70 °C for ≈10 d. For each treatment replication, a subsample of fresh rhizomes ranging from 100 to 150 g was oven-dried to estimate rhizome DW per plant. DW partitioning was then calculated as the ratio of the DW for each plant part divided by the whole-plant DW. In Expt. II, FW and DW of tuberous roots produced by each plant were also measured following each harvest. However, no distinction was made between roots and tuberous roots in Expt. I. Data are presented per container, rather than per plant, unless otherwise specified.

View Full Size

Signs of container deformation after turmeric plants were grown for 28 weeks in Expt. II using a medium-sized container.

Citation: HortTechnology 32, 5; 10.21273/HORTTECH05092-22

• Download Figure
• Download figure as PowerPoint slide

Experimental design and data analyses

In both experiments, containers were arranged in a randomized complete block design where each of the four benches were divided in three sections, each regarded as a block arranged by species. There were eight and six containers used as treatment replications per species and variety in Expts. I and II, respectively. In Expt. I, turmeric data were analyzed as a two-way factorial with four varieties and two container volumes, whereas ginger data were analyzed as a simple mean comparison between data collected from the two containers volume treatments. In Expt. II, data for each species were analyzed as a three-way factorial with two varieties, two planting densities, and two container volumes. In both experiments, blocks were considered as random effects and variety, container volume, density, and their interactions were considered as fixed effects. All data were subjected to analysis of variance using R version 3.6.1 (R Core Team, 2020) and analyzed with the Agricolae package in R (de Mendiburu and de Mendiburu, 2019). Except for ginger data in Expt. I, least-square treatment means were compared using Tukey’s honestly significant difference test (P = 0.05). Ginger data in Expt. I were analyzed using a Student’s t test (P = 0.05).

Partial budget analysis

Results from Expts. I and II were summarized in terms of cost and revenue using a partial budget analysis that only included resources that differed with respect to the scenarios evaluated in this study (container volume and resulting substrate volume, plant species, number and weight of rhizomes, and rhizome yield). The partial budget did not consider resources that would be left unchanged based on decisions related to container volume, species, or number of rhizomes. Because there are many other factors, such as labor, overhead costs, and other direct input costs, the partial budget only evaluated increases or decreases in income rather than overall enterprise profitability. In Expt. I, the scenarios focused on differences between species and container volumes. Because there was no yield benefit of using a larger container volume in Expt. II, only results in the lower-cost, medium container were included and the scenarios focused on differences between planting density and species.

Input costs were obtained in May 2022 for shipped prices for plastic containers ($0.57 for small, $1.59 for medium, and $4.21 for large containers) and substrate ($0.216/L) from a wholesale horticulture distribution company in Florida, seed rhizomes from a wholesale source in Hawaii ($22.27/kg for both ginger and turmeric), and fresh rhizomes from retail online sales prices averaged from six vendors each for ginger and turmeric ($18.76 and $22.22 per kilogram, respectively). The containers were assumed to be washed and reused for 2, 3, and 4 years for small, medium, and large containers, respectively, based on our experience. The cost per container was therefore annualized by dividing purchase cost over years of usable life. Substrate volume equaled the complete container volume for small and medium containers. Large containers were assumed to be only half-filled because most plant roots and rhizomes were only present in the upper half of filled containers. Seed rhizome weight, and therefore cost, for each experiment was based on the actual measured volume. Total cost included the annualized container cost, substrate, and seed rhizome cost. Net revenue based on the yield in each experiment was multiplied by the average retail price, and was calculated per container, per kilogram of seed rhizome, and per square meter of greenhouse (given an experimental spacing of 0.41 m² per container in Expt. I and 0.35 m² per container in Expt. II). Labor costs were not included in our budget analysis.

Effects of container volume

In Expt. I, there was an increase in plant growth and yield with increasing container volume, as both crops produced more than double the shoot, root, and rhizome FW when grown in large compared with small containers (Tables 1 and 2). The number of ginger tillers per plant also increased from six to 16 in the larger container, but no differences were measured in the number of turmeric leaves, regardless of container volume or variety. Shoot height was unaffected by container volume in both experiments (Table 3).

Table 1.

Final growth parameters measured in Expt. I for ginger and turmeric plants grown in a greenhouse using small (1.5 gal) or large (13.3 gal) containers.ⁱ

View Table

Table 2.

Total plant dry weight partitioning measured in Expt. I for ginger and turmeric plants grown in a greenhouse using small (1.5 gal) or large (13.3 gal) containers.ⁱ

View Table

Table 3.

Final growth parameters per container measured in Expt. II for ginger and turmeric plants grown in a greenhouse using different planting densities (one vs. three seed rhizomes per container) and container volumes (medium vs. large).

View Table

Trends in yield differed in Expt. II, whereby rhizome yield of ginger was 44% higher in medium than large containers, and container volume did not affect yield of turmeric (Table 3). Plants of both species grown in large compared with medium containers had up to 66% higher shoot FW and up to 35% lower root FW. Total DW was higher in plants grown in the larger container volume in both species in Expt. I, and turmeric only in Expt. II (Tables 2 and 4). However, ginger in Expt. II had an 18% higher plant DW in the medium compared with the large container. These findings suggest that medium containers could be used instead of large containers to minimize material and space costs for ginger and turmeric production under the conditions evaluated in our study. Nonetheless, large containers may be necessary for longer production cycles than the 28 weeks used in this study. The small containers used in Expt. I were likely too restrictive for adequate plant growth and development.

Table 4.

Total plant dry weight per container and dry weight partitioning measured in Expt. II for ginger and turmeric plants grown in a greenhouse using different planting densities (one vs. three seed rhizomes per container) and container volumes (medium vs. large).

View Table

Yield was expected to increase in larger containers that enable the growth of plants with multiple, fleshy, and thick rhizomes without physical root restriction (Landis et al., 2014). A higher shoot FW and total plant DW in larger containers was also expected, as plant growth of various crops is known to increase with container volume (NeSmith and Duval, 1998). In a meta-analysis of studies evaluating the effects of rooting volume on growth of several plant species, Poorter et al. (2012) found that doubling the container volume increased biomass production by 43%. In that study, reductions in growth and yield using small containers were mainly attributed to decreases in photosynthesis per unit leaf area, rather than to changes in leaf morphology or biomass allocation in smaller plants. Accordingly, Sakamoto and Suzuki (2018) showed that tuber yields of sweetpotato (Ipomea batatas) were higher in plants grown under medium (0.8 gal) or large (1.2 gal) containers compared with small (0.4 gal) containers. In contrast, Massa et al. (2005) showed that yield of sweetpotato can be increased by restricting root volume using smaller containers, which corresponds with our findings for Expt. II. Considering that ginger and turmeric rhizomes tend to grow horizontally as they develop in the soil or substrate surface (Ibrahim, 2018), it is plausible that these crops may benefit more from being grown in containers with a large diameter, rather than a large depth, provided that the overall volume is not too restrictive for adequate rhizome growth and development.

Rhizome DW partitioning was higher in ginger than in turmeric plants, especially in Expt. II (Tables 2 and 4). Shoot dry mass was greater than the rhizome DW for both species in Expt. I, and for turmeric in Expt. II. In contrast, ginger rhizome DW was greater than shoot DW in Expt. II. Roots had the lowest DW partitioning in both experiments, but values were generally higher in the smaller container volume. The effect of container volume on root DW corresponds with the findings of Massa et al. (2005) who showed that root density of sweetpotato was higher in small compared with large containers. Turmeric had considerable DW partitioning to tuberous roots, especially for medium containers with ‘Indira Yellow’. Flores et al. (2021) found that partitioning to tuberous roots in turmeric was correlated with less DW partitioning to rhizomes. Therefore, it is likely that the tuberous roots in turmeric in Expt. II competed for resources with rhizomes.

Effects of planting density in Expt. 2

Using a high planting density generally increased yield and biomass production per container compared with a low density, regardless of variety and container volume (Table 3). Increasing density from one to three rhizomes more than doubled the number of tillers in ginger and leaves in turmeric. Similarly, using a high planting density resulted in a FW increase of shoots (66% and 46%), roots (85% and 86%), tuberous roots (66% and 165%), and rhizomes (42% and 71%) for ginger and turmeric, respectively, compared with a low density. Total DW was also higher in plants grown under high density (Table 4). However, ginger and turmeric plants under low density had higher rhizome and shoot DW partitioning, respectively, than those under high density.

The increase in yield at high planting density corresponds with the findings of Kratky and Bernabe (2009), who compared four planting densities for ginger grown in 2.5-gal containers. The authors proposed that at higher densities, container volume should be increased to prevent rhizome growth restriction, which can lead to rhizome deformation and cause busting of containers (Kratky and Bernabe, 2009). A study evaluating containerized production of sweetpotato found that increasing density reduced the amount of light intercepted by plant leaves, thereby reducing tuber productivity per container (Massa and Mitchell, 2012). Based on our observations, plants grown in medium containers using three seed rhizomes were affected by some level of rhizome growth restriction, as indicated by a visual deformation of the containers (Fig. 1). However, no overall growth or yield reductions were measured with a higher density, most likely attributed to the relatively short production cycle used in our experiment. It isplausible that larger containers are necessary to prevent rhizome growth restrictions when increasing planting density, especially for longer production cycles than the 28 weeks used in our study.

Most studies evaluating planting density for ginger and turmeric have been conducted for open field production in soil. Similar to our findings, significant increases in plant growth and yield with higher densities have been reported in ginger (Ghosh and Hore, 2011; Tiwari et al., 2019) and turmeric (Preetham et al., 2018; Temteme et al., 2017) (Tables 3 and 4). However, some studies have indicated that although overall yield per hectare increases, the rhizome number and weight per plant can be reduced by increasing planting density (Kiran et al., 2013; Kumar and Gill, 2010; Modupeola et al., 2013). Further, Sharma et al. (2012) reported a lower ginger yield using higher densities per hectare, but their findings are likely attributed to an increase in disease incidence with closer plant spacing. It appears that environmental conditions and other factors such as soil quality or harvest time can directly impact vegetative biomass production and yield of ginger and turmeric plants grown in the field (Berbeć and Matyka, 2020; Peter et al., 2005). In addition, some environmental conditions such as excess rainfall could increase humidity and disease incidence with high planting densities, particularly under limited airflow caused by excessive vegetative growth (Pilkington et al., 2010; Sharma et al., 2012).

Varietal differences

In Expt. I, ‘BKK’ turmeric produced the tallest plants, whereas shoot FW was highest in ‘Hawaiian Red’ turmeric (Table 1). ‘Hawaiian Red’ and ‘White Mango’ turmeric produced up to 110% more rhizome yield than the two other varieties. ‘Hawaiian Red’ and ‘Black’ turmeric also produced the largest total DW, but both ‘Hawaiian Red’ and ‘BKK’ partitioned more DW to shoots compared with the other varieties (Table 2). ‘Black’ turmeric had the highest rhizome DW partitioning, and no differences were measured in root DW partitioning among varieties.

Overall, there were minor varietal differences in Expt. II (Tables 3 and 4). ‘Madonna’ ginger produced 59% more tuberous roots than ‘Bubba Blue’, and ‘Bubba Blue’ had a slightly greater rhizome DW partitioning. ‘Indira Yellow’ turmeric had a 57% greater tuberous root DW partitioning than ‘Hawaiian Red’ turmeric, but no yield differences were measured between the two varieties.

Economic analysis

Sales revenue exceeded direct costs of container, substrate, and seed rhizomes in all scenarios, resulting in a positive net revenue (Table 5). Net revenue per container was higher in Expt. II than Expt. I due to the higher rhizome yield (Tables 1 and 3). In Expt. I, the higher yield of ginger compared with turmeric increased sales revenue of this species, despite a lower sales price per kilogram. In contrast, the higher yield of turmeric in Expt. II resulted in higher sales revenue and net revenue per container compared with ginger. In Expt. I, turmeric seed rhizomes were particularly small (28 g) compared with those of ginger (60 g), and all rhizomes in Expt. II (75 g). The higher yield in Expt. II is likely attributed to the use of larger seed rhizomes (Girma and Kindie, 2008; Hossain et al., 2005; Whiley, 1990), which can more than offset differences in seed rhizome cost.

Table 5.

Partial budget analysis based on inputs and yield from Expts. I and II using different container volumes (small = 1.5 gal, medium = 3.9 gal, large = 13.3 gal).ⁱ

View Table

The yield multiplier (kilogram of rhizome yield/kilogram of seed rhizome) may be useful for growers who consider producing their own seed rhizomes for use in a following year, whereby growing plants one rhizome per container resulted in the highest multiplier (Table 5). This may be an efficient practice because the cost of seed rhizome and retail sales price of produced rhizomes were similar. Reusing large containers and filling the containers at no greater depth than necessary could be recommended for commercial production, considering the high cost of these inputs ($4.21 for a large container, and $5.45 to fill the container) compared with the cost of seed rhizomes ($0.59 to $1.32) in Expt. I. However, a local source of low-cost substrate would be desirable, as substrate reuse over years may increase economic risks associated with potential transmission of root-borne pathogens.

Increasing container volume and increasing the rhizome number per container both increased return per container, and per square meter (Table 5). In our experiments, the different container volumes were given the same spacing on the bench. However, growers could grow plants in smaller containers placed closer together to increase return per square meter. Space efficiency is an important consideration because production area in a greenhouse or high tunnel is likely to be more limited and expensive than in the field (Janke et al., 2017).