Expanding Northeast Strawberry Production in Controlled Environment Agriculture with Naturally-Derived Nutrient Source

Objective 1: Evaluate the effect of substrate and cultivar on plant growth and fruit production of greenhouse grown strawberries.

This experiment evaluated 3 strawberry cultivars and 2 substrates for a total of 6 treatments. Plants were grown in 16 Liter Bato WAVE meter buckets (A.M.A Horticulture, Ontario), with six plants in each bucket, staggered so that plants were approximately 8 inches apart.

Cultivars and substrates: Strawberry fragaria cv. Albion, Seascape, and San Andreas runners were propagated from mother plants, placed into 25 mm Terra Plug Oasis Cubes (Smithers-Oasis, OH) and rooted in a propagation house at the UNH Macfarlane Greenhouse for 14 days. After two weeks, strawberry plugs were transported to the Kingman Research Facility and transplanted into Bato WAVE meter buckets. Buckets were placed in 3.05-meter-long troughs placed on a table with a 25.4 mm slope to allow leachate to drain out of buckets and collect at the end of the table. Leachate was not captured for reuse and was drained into a large holding tank behind the greenhouse.

The two substrates tested in this experiment (Table 1) were chosen based on their composition and commercial use in strawberry CEA production in Ontario and Quebec. To improve uniformity, each substrate was further mixed prior to potting using a mortar mixer attached to a drill was used to break up all aggregates larger than 3 cm in diameter. Water was then added to the substrate and further prior to filling the pots.

Table 1. Composition of the Peat (P) and Peat amended with coconut coir (PC) substrates used in this study. BVB substrates were supplied by A.M.A plastics (Ontario, CA)

Plant production: Plants were irrigated via drip irrigation based on weight. The weight of each of each bucket at 55% saturation was calculated based on wet and dry weights. Soil moisture sensors (Decagon 10hs, Pullman, WA) were placed in four of the Bato Buckets and calibrated to irrigate for 4 minutes when soil moisture in the buckets fell to 55%. Netafim Drip emitters with 3 mm irrigation tubing (Griffin Greenhouse Supply, Tewksbury, MA) were used to supply nutrient solution to the plants at a rate of 2 Liters per hour. Plants were fertilized with Peters Professional 5-11-26 NPK Hydroponic Solution in Tank A, and Jack’s Professional Calcium Nitrate (JR Peters Inc., PA) in Tank B. The nutrients were mixed in stock solutions that were injected at a 1:100 rate using a D14MZ2 Dosatron (Dosatron USA, FL). At each mixing date, 500 ml of solution was collected from the sample port prior to measure pH and electrical conductivity (EC). If the samples were within an EC of 0.2 µS/cm of the ideal range (1.0-1.20), nothing was changed. If not, Dosatron flows were adjusted until the target pH/EC was achieved. Strawberries were grown in a triple polycarbonate greenhouse. The greenhouse control system was set to vent when temperatures reached 30ºC and heat when temperatures fell below 18ºC during the daytime.

Plants were manually pollinated daily using a leaf blower which is a substitute for bees and other natural pollinators (Kubota, 2015). Yellow sticky cards were placed throughout the crop for the purpose of scouting and pest removal. Fungus gnats and thrips were managed by applying the beneficial nematode Steinernema feltiae biweekly to the pots at approximately 33,000 nematodes per plant. Precautionary measures were taken to prevent thrips and whitefly using Swirskii predatory mite sachets (Amblyseius swirskii) and Encarsia cards. Slow-release Swirskii sachets were placed on each strawberry plant every 6-8 weeks, while Encarsia cards were placed on 2 plants per greenhouse every other week. A bran formulation of Crysopa carnea (green lacewing) was used to control aphids, spider mites, whiteflies, and thrips. Phytoseiulus persimilis (predatory mites) were sprinkled in a bran formulation biweekly to plant leaves to control spider mites.

Runners were removed every three weeks from all plants at the base of the plant using pruners. Dead and dying leaves on the bottom of plants were also removed every three weeks.

Data collection: The following data was collected to compare the effect of cultivar and substrate on greenhouse strawberry performance.

1. Agronomic data – Data was recorded on number of days until runner and flower production per treatment when 50% of the plants in each treatment had at least one or more runners or flowers. Flowers were counted when fully opened and petals were still present.
2. Fruit yield – Ripened fruit was harvested 3 days a week from July 23rd to September 5th, 2018 to determine yield in grams. Ripened fruit was defined as fruit with 80% or more red color development. Fruit were considered Marketable if they exhibited 80% red color development, weighed equal to or greater than 10g, and had no deformities. Percent of marketable fruit was calculated by dividing the number of marketable fruit by the total number of fruit from each treatment at the end of the experiment.

Data analysis: Data were analyzed using a split-plot analysis in JMP Pro (JMP Pro. 14.1 Statistical Software, Cary, NC) to evaluate differences in effects of the substrate, cultivar and the interaction between substrate and cultivar on total fruit yield and fruit yield each week. A Tukey’s HSD test was used to separate treatment means. Statistical significance was determined at α=0.05.

Objective 2: Compare growth and yield of day-neutral strawberries fertilized with a synthetically derived hydroponic solution and a naturally-derived aquaponic solution.

The goal of this study was to evaluate the effect of nutrient source on plant nutrition, plant growth, and yield of greenhouse grown strawberries. The experiment was designed to (1) to understand the relationship between nutrient availability and nutrient uptake of strawberry plants grown under a naturally-derived aquaponic nutrient source and a synthetically-derived nutrient source, (2) evaluate the effect of nutrient source on yield of greenhouse grown strawberry plants, and (3) evaluate the effect of phosphoric acid supplementation of a naturally-derived nutrient source on strawberry yield.

This study evaluated three nutrient sources (Table 2); two naturally derived and one synthetically derived. Each of the solutions were assigned to 24 replicate plants. For the synthetically derived treatment, a Japanese hydroponic strawberry recipe known as the Yamazaki solution (Yamazaki, 1982) was used (H). For the naturally derived treatments, aquaponic solutions were collected from the Aquaponic facility at the UNH Kingman farm facility (described below). This aquaponic solution was used alone (AQ) or amended with phosphoric acid (AQ+PA).

Table 2. Concentration of macronutrients of the three nutrient solutions used in this study (concentrations are reported in mg/L). The Aquaponic solutions were obtained from the UNH Kingman Farm Aquaponic facility.

Plant production: Strawberry plants were propagated as described in objective 1 on September 17, 2018. Runners were transplanted into Classic 300 20.3 cm pots filled with Pro-Mix HPCC substrate. Plants were grown on three 91 cm tall tables at the Kingman Farm Greenhouses and irrigated as described in objective 1. The greenhouse was set to 30ºC for cooling, 18ºC for daytime heating, and 15ºC for nighttime heating. For the Yamazaki solution, an A/B irrigation system was installed using two 19 L tanks each with a D14MZ2 Dosatron nutrient distribution system (Dosatron USA, Fl). The EC of the Yamazaki solution was maintained at 1.0 mS/cm and pH 5.5. The EC of the aquaponic effluent was maintained at 1.1 mS/cm and pH 7.0. A third Dosatron was installed to the AQ+PA table to supply phosphoric acid to the aquaponic effluent to reach a target pH of 5.5-6.0. A tank was filled with 20mL of phosphoric acid and 19L of RO water and injected into the Dosatron at a rate of 1-50. The EC of the supplemented aquaponic effluent was maintained at 1.1 mS/cm and the pH at 5.9. A pour-thru method was used to monitor the pH and EC of the leachate of 3 pots per treatment. If the pH and EC were not within 0.2 µS/cm of the target range (1.0-1.5), clear water was run through emitters until saturation occurred.

Four-hundred-Watt high pressure sodium lights were placed above each table for supplemental lighting (PL Light Systems, Beamsville, ON). Lights were programmed to ensure 24 mol·m^-2·d-1 was achieved at a maximum of a 16-hour photoperiod. Plants were pollinated, and pests managed as described in objective 1.

Data collection: The following data was collected to compare the effect of nutrient source on plant production.

1. A nutrient mass balance was established by quantifying nutrient input and output masses, and assimilated plant nutrient mass. The mass balance calculations were based on the mass of nutrients in the input (irrigation solution), output (leachate) and tissue samples (captured) at four time points throughout the study: at the time of transplant, at runner production, at fruiting, and at harvest.

1. 1. To determine nutrients in plant tissues, root, shoot, and fruit tissues of three plants from each nutrient solution treatment were collected. Fresh and dry weights of samples were recorded. Dry samples were ground into a powder and sent to JR Peters Inc. for nutrient analysis.
   2. To characterize nutrient inputs, irrigation solution was collected from four drip emitters per treatment at each irrigation event to quantify macro and micronutrient concentrations in the source water. Samples were sent to JR Peters Inc. for nutrient analysis.
   3. To characterize nutrient output, leachate was collected from the same 3 pre-selected plants per treatment. Ten minutes after am irrigation event ended, leachate volumes from each replicate pot were measured to determine the volume lost per pot. The volume lost was then subtracted from the input volume to estimate the captured volume (or volume retained in the pot). Leachate from each irrigation event was combined into one bucket per treatment as a composite sample.

2. Agronomic data – The number of days until runner production, number of days until flower production, and number of days until ripened fruit was collected for all treatments. Symptoms and signs of nutrient deficiencies were recorded weekly. Percentage of plants that showed signs of nutrient deficiency per treatment was also recorded.
3. Fruiting data- Ripened fruit was collected three times a week for 8 weeks. Marketable fruit was defined as fruit with 80% red color development, weighed equal to or greater than 10 g, and has no deformations (Cantliffe et al., 2007). Brix content, a measurement of soluble sugars, was collected on all harvested fruit using a Hanna H196801 Digital Refractometer (Hanna Instruments, Smithfield, RI).

Data analysis: Analysis of nutrient mass balance, agronomic and fruiting data were conducted as described below.

1. Nutrient mass balance
   1. To quantify nutrient uptake of each strawberry plant, a series of calculations were carried out. The leachate nutrient composition subtracted from the nutrient irrigated to each pot to calculate the volume and composition of nutrient solution that remained in the pot. We considered nutrient solution captured in the pot as solution that was taken up by the plant.
   2. Based on nutrients supplied to the plant, the composition of nutrients within the plant tissue and fruit was then analyzed to determine the nutrient uptake efficiency of each plant.
   3. An ANOVA was used to compare mean percent assimilation of Nitrogen, Phosphorous, and Potassium in sample plants grown in each nutrient source at two stages of growth: 1st at the fruiting stage and then at final harvest. In all analyses, statistical significance was assessed at α=0.05.
2. The effect of nutrient source on plant fresh weight, number of fruit, and brix was analyzed using an ANOVA. Tukeys HSD post hoc tests were conducted to separate treatment means.
3. A mixed model ANOVA with a repeated measure was conducted using SAS 9.4 (SAS Institute, Cary, NC) to evaluate the cumulative yield and cumulative marketable yield over the 8 harvest dates per plant (g) in each treatment. Orthogonal contrasts were then used to compare treatment means across the three nutrient sources. A mixed model ANOVA was also conducted to compare total yield and total marketable yield per plant of aquaponically grown plants compared to hydroponically grown plants.

In our initial project proposal, we planned to repeat this study at the farm scale. Since additional funds were needed for this work than originally budgeted for, this replication was not completed but will be included in follow-up projects. 

Objective 3: Evaluate daily light integrals that optimize plant growth and yield of greenhouse production of strawberry

This experiment evaluated three daily light integral (DLI) regimes (14, 20, and 26 mol/m²/day) with three replicate blocks of each lighting regime. Each block consisted of three chambers (one for each lighting treatment) with 9 plants for a total of 27 plants per treatment. Treatments were randomly assigned to a chamber on each bench.

Plant material and plant production: Strawberry cv. Albion was propagated from mother plants as described in objective 1 on December 31, 2018. Plugs were then transplanted into Classic300 20 cm pots filled with ProMix HPCC substrate 27 days post rooting. The greenhouse temperature was set to a 16°C daytime temperatures and 13°C nighttime temperature. Pests and diseases were managed, and plants were pollinated as described in objective 1.

Lighting control: ARC 600 full spectrum LED lights with integrated dimmers (HortLED, Ithaca, NY) were used for the study. All plants were exposed to an 18-hour photoperiod. Lights were set to one of three target DLI treatments: 14, 20, or 26 mol•m^-2•d-1 with a light intensity of 216, 308, or 401 umol•m-2•s-1 respectively. Plants were grown inside chambers covered with UV stabilized black ground cover (VMInnovations, Lincoln, Nebraska) to block natural sunlight and ensure plants were exposed to 100% supplemental light. Photosynthetically active radiation (PAR) was measured underneath the ground cover using a line quantum meter with ten integrated PAR sensors (MQ-301, Apogee, UT) to ensure equal light intensity was provided to each plant.

Irrigation: Plants were irrigated with the Yamazaki solution as described in objective 1.

Data collection: The following data was collected to compare the effect of light exposure on plant productivity and fruit quality.

1. Fruit yield. Ripened fruit was harvested two times a week. The number of fruit, fruit weight, and marketability of fruit was recorded for each plant. These data were also averaged to determine yield per replicate chamber.
2. Sugar, measured as total soluble solids, was analyzed in fruit using the Hanna digital refractometer (model HI96801) (Hanna Instruments, Smithfield, RI).
3. Chlorophyll index. A Soil-Plant Analyses Development (SPAD) meter (Minolta Camera Co., Japan) was used to analyze leaf chlorophyll index for each treatment. SPAD measurements were taken on the youngest fully expanded leaflets of 3 plants within each chamber. Chlorophyll index was measured at 4, 6 and 11 weeks from transplant.
4. Plant biomass. To determine the effect of lighting treatment on plant growth, above ground plant biomass was measured. Shoots from each plant were cut at the crown, flush with the substrate. All ripened and unripened fruit on the plant on harvest day were removed before weighing. Plants were dried for 48 hours at 65°C, removed from the drying oven, and weighed for a dry weight.

Data analysis: Effects of DLI treatment on total yield and marketable yield per plant (g) over the 10-week harvest period were analyzed using an ANOVA in SAS 9.4 (SAS Institute, Cary, NC). Differences between treatment means were evaluated using a Tukey’s honestly significant difference (HSD) test at the α=0.05. Analyses were also conducted to evaluate the effect of DLI on fruit production per chamber, fruit sugar content (brix), plant biomass, and chlorophyll index using a one-way ANOVA in JMP Pro version 14.1 Statistical Software (SAS Institute, Cary, NC). A Tukey-Kramer HSD test was used to compare differences between treatments.