Optimizing Carbon Dioxide Enrichment to Improve Fruit Quality and Expand Local Opportunities for Controlled Environment Strawberry Production

Plant Production

Two cultivars of everbearing strawberry (Fragaria ×ananassa ‘Albion’ and ‘Ozark Beauty’) were evaluated.  Bareroot strawberries (experimental runs 1 and 2) or aeroponically propagated vegetative runners (experimental runs 3 and 4) were planted in 1-gallon containers containing a soilless substrate comprised of perlite (50%) and a peat-based substrate (50%; Promix HP + Biofungicide/Mycorrhizae High Porosity).  Plants were grown in a walk-in growth chamber (GR64; Conviron, Temecula, CA) set to maintain 22/18 °C day/night air temperature, 55/65% day/night relative humidity (RH), and 450 µmol∙mol^–1 CO₂ for approximately eight weeks.  An 18-h photoperiod (0500 to 2100 HR) with a target photosynthetic photon flux density (PPFD) of 348 µmol∙m^–2∙s^–1 [daily light integral (DLI) ≈ 21 mol∙m^–2∙d^–1] was provided with light-emitting diode (LED) fixtures (Ray 44 Physiospec Indoor; Fluence, Austin, TX).  Finally, drip irrigation delivered a water-soluble fertilizer providing 100 mg∙L^–1 N (Jack’s 8-10-26 K Strawberry Part A and Jack’s 15-0-0 Calcium Nitrate Part B; J. R. Peter’s Inc., Allentown, PA) once per day.  Nutrient solution electrical conductivity (EC) was maintained no higher than 1.0 dS∙m^–1 with a pH target of 5.5-6.5.  Flowers were removed during this time to ensure vegetative growth was prioritized. 

After eight weeks, flower removal ceased and ten plants per cultivar were randomly assigned to four walk-in growth chambers (GR64), with two chambers set to an ambient CO₂ concentration (450 µmol∙mol^–1) and the others set to an enriched CO₂ concentration (900 µmol∙mol^–1).  Other environmental parameters matched those described above, except temperature, which was adjusted to 20/12 °C (day/night) in both chambers. Additionally, as the plants developed, a second and third daily irrigation event was added. Plants were hand-pollinated using a soft-bristled brush to ensure complete pollination, and new runners were removed to ensure resources were diverted to fruit development.  Fifteen minute averages of air temperature, RH, and CO₂ concentration for each chamber were logged using an air temperature and RH probe (EE08-SS; Apogee Instruments, Inc., Logan, UT) and CO₂ probe (GMP252; Vaisala, Woburn, MA) and were collected with a datalogger (CR1000X; Campbell Scientific, Inc., Logan, UT).

Data Collection 

Subobjective 1

By comparing photosynthetic responses of two everbearing strawberry cultivars, we can elucidate early signs of acclimation to elevated CO₂ conditions within the first week of production, allowing for a clear understanding of when photosynthesis may begin to decline under these conditions and, ultimately, what timing and duration of CO₂ enrichment may be most beneficial. 

Survey gas exchange measurements were collected every day for seven days on two plants per cultivar, per CO₂ treatment, per experimental run (n=8).  The same leaf on each plant was measured using a portable photosynthesis meter (LI-6800; LI-COR Inc., Lincoln, NE) fitted with a 2 cm² leaf cuvette.  Cuvette conditions matched chamber conditions (20 °C, 55% RH, and 348 µmol∙m^–2∙s^–1 PPFD), with reference CO₂ concentration based on treatment.  On day 1, three sets of survey measurements were taken on one recently fully expanded leaf: 1) all plants in both treatments were measured at 400 µmol∙mol^–1 (ambient) CO₂ to establish baseline photosynthesis, 2) plants assigned to the enriched (900 µmol∙mol^–1) treatment were surveyed again at a CO₂ concentration of 900 µmol∙mol^–1 to ascertain the initial response to CO₂ enrichment, and 3) plants assigned to the enriched treatment were surveyed at enriched CO₂ and saturating light intensity (1500 µmol∙m^–2∙s^–1) to determine maximum photosynthetic potential when light limitations are removed.  This procedure was repeated on subsequent days on the same leaf tagged as “old”, except plants assigned to the enriched treatment were not surveyed again at an ambient CO₂ concentration.

On days 7 and 8, rate of photosynthesis (A) in response to increasing internal leaf CO₂ concentration (A-C_i curves) was collected on all of the “old” leaves.  Cuvette conditions matched chamber conditions (20 °C and 55% RH), but with a saturating light intensity (1500 µmol∙m^–2∙s^–1 PPFD).  For plants grown under ambient CO₂ concentration, measurements began at 400 µmol∙mol^–1 CO₂, decreased to 50 µmol∙mol^–1, returned to 400 µmol∙mol^–1, and finally increased to a maximum of 1000 µmol∙mol^–1 in 100 µmol·mol^-1 increments to prevent feedback inhibition. For plants grown under the enriched CO₂ concentration, measurements began at 1000 µmol∙mol^–1 CO₂ and decreased to 50 µmol∙mol^–1 in increments of 100 µmol·mol^-1.

On day 14, survey measurements were taken on the next most recently fully expanded leaf that developed under treatment conditions, which was tagged as “new”.  Then, on day 15, A-C_i curves were collected on “new” leaves.  On days 28 and 29, A-C_i curves were again collected on “new” leaves, and finally on days 30 and 31, A-C_i curves were collected on the next most recently fully expanded leaf that developed under treatment conditions.

The Farquhar–von Caemmerer–Berry model was fit to A-Ci curves, as described by Dubois et al. (2007) to estimate maximum rate of rubisco carboxylation (V_cmax) and rate of electron transport (J_max).  V_cmax is correlated to photosynthetic efficiency (Rubisco), so if enriched plants show a decline in V_cmax, we can conclude that they are using CO₂ less efficiently and acclimation has occurred.  

On day 58, all plants that were used for gas exchange were destructively harvested for biomass data.  Leaf number, leaf area, relative chlorophyll content, crown number, and crown diameter were measured.  Plants were separated into unripe fruit, flower/bud, truss, leaf, stem/petiole, and crowns, then dry masses weighed.  Leaf mass area (LMA) was calculated by dividing total leaf dry mass by leaf area (g∙cm^–2).  This destructive data will complement the gas exchange data by providing insight on whether CO₂ acclimation may also be facilitated by changes in plant morphology.  In our previous research focused on vegetative growth, we found significantly thicker leaves (greater LMA) in enriched treatments for petunia and strawberry, which was indicative of an acclimation response.  However, no differences in total dry mass were observed between the ambient and enriched CO₂ concentrations for these species, thus limiting the benefits of CO₂ enrichment for biomass accumulation.

Subobjective 2

Yield data was assessed for all plants grown in the above experiment.  Flower and fruit number per truss was counted at treatment initiation.  Days to first fruit harvest was recorded for all plants.  All strawberries that were at least 95% red were harvested and counted from all plants on days 23, 30, 37, and 44.  Additionally, all harvested berries from each treatment were weighed and marketability assessed using the USDA fruit grade rating scale.  On day 44, firmness of all harvested berries were tested with a fruit compression tester (FirmTech; BioWorks, Inc, Cleveland, OH), and sugar and acid content assessed on a pooled sample using a digital refractometer (PAL-BX|ACID4; Atago, Bellevue, WA).  This information will provide growers with expected yield and quality changes as a result of CO₂ enrichment.

Subobjective 3

Attributes of fruit quality will elucidate if there are value-added benefits with CO₂ enrichment.  Empirical measures of sweetness and sourness will be correlated with sensory assessments of the same attributes for a more holistic picture of desirable strawberry characteristics.  Nutritional profiling of plant protective phytochemicals like polyphenols and antioxidants will reveal potential benefits of enrichment for human health.

Eighty four fruits from each cultivar and treatment, not used for the fruit quality testing in Subobjective 2, were washed and stored in a food-safe cooler prior to the sensory test.  Participants were recruited from the Denver Metro area through the Sensory Testing Lab distribution list.  Paperwork was submitted to the CSU Institutional Review Board (IRB) for review and approval before the start of the study, and panelists, who were at least 18 years old, gave written, informed consent prior to participating.  Panelists were not financially compensated for participation but were provided non-financial incentives from the Sensory Testing Lab with non-grant funds.  All sensory analysis took place in good standard sensory evaluation conditions (Lawless and Heymann 2010).  The sensory tests took place at the Sensory Testing Lab at Spur, under the supervision of Dr. Martha Calvert, the Sensory Lab Manager.  Participants were asked to answer demographic information after taking the test.  Each panelist was served a whole berry from each CO₂ concentration and cultivar (four samples total) in individual booths, with serving order randomized between participants and participants blinded to the treatment.  For each sample, participants were asked to rate the intensity of aroma, flavor, and texture attributes from a list using a 10-point structured scale anchored by “not present” and “very strong”.  Lastly, participants assessed overall liking using a 9-point hedonic scale anchored by “disliked extremely” and “liked extremely”.  Participants were provided with crackers and water for palette cleansers between samples.  Data will be analyzed using standard multivariate procedures including analysis of variance (ANOVA) and principal component analysis (PCA) to determine drivers of liking for strawberry sensory quality. 

A representative sample of fruits from each cultivar and treatment weighing a combined total of at least 70 grams (fresh weight), were collected, frozen in dry ice, then stored in –80 °C until extractions. Standardized high-throughput omics platforms that have been developed as part of the Periodic Table of Food Initiative (PTFI) will be utilized to characterize the nutrient composition.  Specifically, metabolite analysis will be performed using ultra-performance liquid chromatography mass spectrometry (UPLC-MS) followed by data analysis using an open-source cloud based PTFI pipeline.  Metabolite annotation will be based on queries against the PTFI reference standard library.  All results from this analysis will be shared in the publicly accessible PTFI food composition database, which will increase the impact of these results beyond the current study.  This objective was achievable with the assistance of Dr. Jessica Prenni, Professor in the Department of HLA at CSU and CSU Bioanalysis and Omics Core Facility. Any remaining samples were preserved for future nutritional analysis of starch and vitamin C content in consultation with Dr. Sarah Johnson, Registered Dietician Nutritionist, Associate Professor in the Department of Food Science, and Director of the Functional Foods and Human Health Laboratory at CSU.

Data Analysis

The experiment was run four times, with assigned CO₂ treatments in chambers switched between runs to account for chamber effects.  A randomized complete block design was used with the replication in time serving as the block.  Each plant was considered the experimental unit.  An ANOVA will be conducted using the PROC MIXED procedure in SAS OnDemand for Academics (SAS version 9.4), followed by pairwise comparisons for variables that have a significant F-test (P-value <0.05).  Fixed effects in the model will be the two levels of CO₂, while the random effect will be run.

Table 1. Timeline tasks and dates completed for objective 1.

Research Objective Tasks Remaining

Data analysis is still ongoing and should be completed by May 31st, 2025. An abstract regarding this research was recently accepted for the 2025 American Society for Horticultural Science (ASHS) Annual Conference scheduled for July 28th - August 1st, 2025 in New Orleans, LA. Additionally, we will be working on manuscript preparation for this research over the course of the next few months with plans to submit for publication by August 31, 2025.