Research site.

This 3-year trial was conducted during the Fall season of 2018, 2019, and 2020 on certified organic land at the University of Florida Plant Science Research and Education Unit in Citra, FL. A north–south oriented three-bay (25.6 m × 9.1 m each bay) gutter-connected high tunnel was used in this study (Atlas Greenhouse Inc., Alapaha, GA, USA). The high tunnel has a Quonset-shaped roof consisting of double layers of polyethylene film (152.4 µm) with air inflation and an automated E-Z roll-up curtain system operated by a control panel (based on high tunnel temperature settings) was installed to open and close the sidewalls and endwalls (3.05 m high). The soil at the research site has been classified as Candler sand (Hyperthermic, uncoated Lamellic Quartzipsamments). Before initiating the 3-year experiment, high tunnel soil analysis indicated a soil pH of 7.1 and composition of 95.2% sand, 3.1% silt, and 1.7% clay (Waters Agricultural Laboratory, Inc., Camilla, GA, USA).

Experimental setup.

A split-split-split plot design with three replications was used, with cover cropping as the whole plot factor, organic fertilizer as the subplot factor, and compost application as the sub-subplot factor. Each bay of the gutter-connected high tunnel served as a block (replication). Each sub-subplot bed (north–south orientation) was 0.9 m wide and 5.2 m long, with a bed spacing of 1.8 m. The whole plots were comprised of a weedy fallow control (Fallow) and a cowpea cover crop (Cowpea), arranged in randomized complete blocks. The ‘Iron & Clay’ cowpea cover crop was seeded by broadcasting at a rate of 112 kg/ha on 15 Aug 2018, 12 Aug 2019, and 10 Jul 2020. Before seeding, the untreated cowpea seeds (Tucker Seed Company, McRae, GA, USA) were inoculated with OMRI-listed Guard-N^® (Verdesian Life Sciences, Cary, NC, USA) at a rate of 7.5 g inoculant per 1 kg seed. The control plots without cowpea were maintained as a weedy fallow during the cowpea growing season each year. On 5 Oct 2018, 2 Oct 2019, and 1 Sep 2020, cowpea was terminated, and the aboveground biomass was chopped into ≈5-cm-long pieces using a flail mower (Frontier model GM1072; Deere & Company, Moline, IL, USA). Cover crop residues were then incorporated into the soil using a rotary tiller (model# TG-0G 48-JP; King Kutter^® Inc., Winfield, AL, USA) to a depth of 10 cm. A second tillage to a soil depth of 10 cm was conducted 5 to 7 d after cowpea termination. Weedy fallow control plots received the same tillage management as the cowpea plots following the same timeline.

The subplots consisted of preplant applications of granular organic fertilizer (Granular) and in-season fertigation with liquid organic fertilizer (Liquid). In each whole plot, four beds were formed, and two beds were randomly assigned to accommodate each fertilization treatment. The granular organic fertilizer 10N–0.9P–6.6K (Nature Safe, Irving, TX, USA) was applied preplant on 11 Oct 2018, 9 Oct 2019, and 9 Sep 2020. For the fertigation treatment, liquid fish fertilizers Aqua Power^TM 5N–0.4P–0.8K (JH Biotech, Inc., Ventura, CA, USA) and Big-K™ 0N–0P–41.5K sulfate of potash (JH Biotech, Inc.) were applied weekly through drip irrigation lines for 5 weeks during the pac choi production season. The first fertigation event started right after transplanting pac choi seedlings, with the last application conducted ≈4 to 7 d before harvest. The total application rates of N, phosphorus (P), and potassium (K) (i.e., 112.0 kg N/ha, 9.8 kg P/ha, and 74.4 kg K/ha) were kept the same for the two organic fertilizer treatments.

Each subplot was further divided into four equal sections for random assignment of the compost treatments: yard waste–based compost (Yard; Watson C&D, Archer, FL, USA), cow manure–based compost (Cow; Black Gold Compost Company, Oxford, FL, USA), vermicompost (Verm; Black Star Organic Products LLC, Archer, FL, USA), and no compost control (NC) (Table 1). Yard, Cow, and Verm were applied at 22.4, 22.4, and 5.6 Mt/ha, respectively, and mixed with granular organic fertilizer and incorporated into the soil to a 10-cm depth using a garden rake. A reduced application rate of vermicompost was used given its higher cost and N content (on average) relative to the other two types of composts. The treatment plots remained in the same location throughout this 3-year study, which helped determine the long-term impact of nutrient management treatments on crop performance.

Table 1.

Basic properties of yard waste compost (Yard), cow manure compost (Cow), and vermicompost (Verm) applied before transplanting pac choi in high tunnels (2018–20) in Citra, FL, USA.

View Table

Leafy green planting.

Pac choi (‘Mei Qing Choi’; Johnny’s Selected Seed, Winslow, ME, USA) transplants were grown in a research greenhouse at the University of Florida campus in Gainesville, FL, USA, with average day and night air temperatures of 24 and 18 °C, respectively. Untreated seeds were sown in 72-cell Styrofoam trays (Speedling Inc., Sun City, FL, USA) filled with peat-based potting soil (Natural & Organic 10; Fafard, Agawam, MA, USA) on 21 Sep 2018 (Year 1), 22 Sep 2019 (Year 2), and 18 Aug 2020 (Year 3). The pac choi seedlings were fertilized with 2N–1.3P–0.8K Fish/Seaweed Blend organic liquid fertilizer (Neptune’s Harvest, Gloucester, MA, USA) at a concentration of 100 to 200 mg of N per L every other day after the first true leaf emergence until the day before transplanting. Following the application of the granular organic fertilizer and compost treatments, the pac choi seedlings were transplanted in two rows in a staggered arrangement within each bed on 15 Oct 2018 (Year 1), 11 Oct 2019 (Year 2), and 11 Sep 2020 (Year 3), with both between-row spacing and within-row plant spacing at 0.3 m, resulting in 34 plants per sub-subplot. Double drip tapes (15-cm emitter spacing, 3.7 L/h/m flow rate; Jain Irrigation, Inc., Haines City, FL, USA) were placed on the soil surface in each bed. Based on soil moisture monitoring and environmental conditions, irrigation was run three to four times daily, with each event lasting for 15 min.

After harvesting the pac choi on 16 Nov 2018, 12 Nov 2019, and 16 Oct 2020, baby spinach or baby lettuce was grown as a catch crop without the application of any organic fertilizer or compost. On 19 Nov 2018 (Year 1), spinach (‘Corvair’, certified organic seeds; Johnny’s Selected Seed) was direct seeded in three rows within each bed (sub-subplot) using a push seeder (Jang JP-1; Mechanical Transplanter Company, Holland, MI, USA), with between-row and in-row spacings at 0.3 m and 1.3 cm, respectively. In the following two seasons, that is, on 13 Nov 2019 (Year 2) and 20 Oct 2020 (Year 3), ruffled red romaine lettuce (‘Outredgeous’, certified organic seeds; Johnny’s Selected Seed) was direct seeded in 12 rows within each bed (sub-subplot) using a six-row seeder (Johnny’s Selected Seed), with between-row and in-row spacings at 7.5 and 2.5 cm, respectively. Irrigation was supplied through overhead SpinNet^TM microsprinklers (Netafim Irrigation, Inc., Fresno, CA, USA) with a flow rate of 70 L/h during the spinach and lettuce seasons. The irrigation ran five times per day, with each event lasting for 2 min. During all crop seasons, the space between planting beds was covered with black ground fabric (Lumite, Inc., Alto, GA, USA) to control weeds. Tomato plants (data reported elsewhere) were grown after the leafy greens trials before planting the cowpea cover crop for the following year, and each sub-subplot remained in the same location for 3 years.

High tunnel management.

During the pac choi production trials in Years 1 and 2, the side curtains were automatically adjusted based on the air temperature inside the tunnel: fully closed at <18 °C and fully open at ≥25 °C, gradually opened as temperatures rose from 18 to 25 °C. In the following spinach or lettuce season, the automated side curtains were set to remain fully closed at <10 °C and fully open at ≥17 °C. In Year 3, high tunnel curtains were kept fully open throughout the growing seasons.

White light–diffusing shadecloth (∼40% shading level; HARMONY 3915 O E; Ludvig Svensson, Inc., Charlotte, NC, USA) was deployed to cover the high tunnel roof during the pac choi production season and was removed right before the subsequent spinach or lettuce season in Years 1 and 2. In Year 3, shadecloth was applied the day before transplanting pac choi seedlings and removed 2 weeks after seeding lettuce.

Environmental condition monitoring.

where $x_i$ = PAR measurement at $i^{th}$ observation (μmol·m⁻²·s⁻¹) (recorded every 30 min in Years 1 and 2, and every 5 min in Year 3); n = number of observations; and d = number of days.

Disease and pest management.

Pest and disease scouting was conducted each week, and organic pesticides were applied as needed during each pac choi season and the last lettuce trial for pest management. In the pac choi trials, the major pest problems were whiteflies (Aleyrodidae spp.), thrips (Thysanoptera spp.), and aphids (Aphidoidea spp.). Southern armyworm (Spodoptera eridania) was detected on lettuce in Year 3. Disease issues were generally absent in each trial. Sweet alyssum (Lobularia maritima) as a refuge crop was planted at the front and back ends of each leafy green bed and corners of each high tunnel bay to help attract beneficial insects throughout all seasons.

Assessment of crop yield and leaf mineral nutrient contents.

Pac choi was harvested on 16 Nov 2018, 12 Nov 2019, and 16 Oct 2020. Fifteen representative plants were harvested from each sub-subplot for marketable yield measurement. In Year 1, spinach was harvested twice by cutting 1 cm above the soil line; the first harvest was on 21 Dec 2018, and the regrowth resulted in a second harvest on 18 Jan 2019. Leaf fresh weight from each harvest was recorded and the total marketable yield was determined. In Year 2, baby leaf lettuce was harvested once on 16 Dec 2019, and in Year 3, lettuce was harvested twice on 24 Nov and 2 Dec 2020, using the “cut and regrow” method. After the determination of fresh weight to calculate marketable crop yield per hectare, selected plant samples (i.e., three harvested plants of pac choi and representative samples of spinach and lettuce leaves from each experimental unit) were dried for 5 to 7 d at 65 °C until a constant weight was reached. After weighing the dried samples for calculating total dry weight per hectare, well-mixed representative dried leaf samples were sent to Waters Agricultural Laboratories to measure the contents of macronutrients and micronutrients, including N, P, K, magnesium (Mg), calcium (Ca), sulfur (S), boron (B), zinc (Zn), manganese (Mn), iron (Fe), and copper (Cu).

Determination of ascorbic acid content, total antioxidant capacity, and total phenolics.

The compositional quality of leaf samples was evaluated for pac choi and lettuce in Years 2 and 3. At the time of harvest of pac choi in each experimental unit, one fully expanded inner leaf was collected from the whorl immediately adjacent to the outermost leaves on 12 plants, and that 12-leaf composite sample was used for analysis. For baby leaf lettuce, approximately 100 g of representative leaf samples were pooled from the harvested leaves from each experimental unit, and samples from the first harvest in Year 3 were used for quality assessment. All leaf samples were placed into sealed plastic bags and packed on ice in a cooler until transported to the laboratory refrigerator (4 °C). Within 24 h after harvesting, the leaf sample from each experimental unit was cut into ≤1-cm-long pieces using a kitchen knife and mixed manually to form a well homogenized mixture. Then, 2 g was taken from the homogenized sample and placed in a 50-mL centrifuge tube (Thermo Fisher Scientific, Waltham, MA, USA) containing 20 mL of acid mixture with 6% metaphosphoric acid (Sigma-Aldrich, Inc., St. Louis, MO, USA) in 2 N acetic acid (Thermo Fisher Scientific). The samples in the acid mixture were then stored at –30 °C until further extraction for ascorbic acid analysis. The remaining homogenized samples were stored at –30 °C before further analysis of total antioxidant capacity and total phenolics.

The ascorbic acid content of leafy greens was determined using the AOAC method 967.21 (AOAC International 1995) with minor modifications. Standard ascorbic acid solutions were prepared with concentrations at 0.5, 1.0, 2.0, 3.0, 4.0, and 5.0 μg ascorbic acid per mL of 6% metaphosphoric acid in 2 N acetic acid. The absorbance of the solution was read at 540 nm using a microplate reader (Synergy HTX; BioTek Instruments, Inc., Winooski, VT, USA). Ascorbic acid of each leafy green sample was determined and expressed as milligrams of ascorbic acid per gram of fresh weight.

Approximately 100 g of frozen leafy green samples were homogenized into powder using liquid nitrogen with a mortar and pestle. After mixing 10 g of the ground leaf sample with 10 mL deionized (DI) water, the homogenized samples processed by a high-shear mixer were then centrifuged at 17,600 g_n for 20 min and filtered through Watman #4 paper (Thermo Fisher Scientific). The supernatant was used for measuring total antioxidant capacity using the ferric reducing antioxidant power assay following Benzie and Strain (1996), with absorbance measured at 593 nm. Total antioxidant capacity was calculated as millimoles of Trolox equivalents per gram of fresh weight.

Sample extraction for total phenolics measurement was performed following a similar process as total antioxidant capacity quantification except that 4 mM CaCl₂ solution was used instead of DI water. The total phenolic content was determined using Folin–Ciocalteu reagent (Sigma-Aldrich, Inc.) according to the method described by Young et al. (2005) with minor modifications. Briefly, 0.5 mL of filtered supernatant was used, and the absorbance was measured at 765 nm. A standard curve was made using gallic acid at concentrations of 0.02, 0.04, 0.06, 0.08, and 0.10 mg/mL. The total phenolic content of the sample was expressed as milligrams of gallic acid equivalents per gram of fresh weight.

Statistical analyses.

Data from each crop trial were analyzed following a linear mixed model using the GLIMMIX procedure of SAS (version 9.4; SAS Institute, Cary, NC, USA). Logarithmic transformation was applied to some data to help better meet the model assumptions as needed, but the results were reported using the original data. Fisher’s least significant difference (LSD) test at P ≤ 0.05 was conducted for multiple comparisons of different parameters among treatments. For the 3-year pac choi trial, repeated measures was used in the model to determine the year effects on different measurements as well as the interactions between year and nutrient management treatments.

Environmental conditions.

For the high tunnel pac choi trial, the average daily air temperatures were 22.2 °C, 23.2 °C, and 24.7 °C during the production season of Years 1, 2, and 3, respectively, and the average soil temperatures were 22.9 °C, 25.1 °C, and 26.7 °C, respectively, for the three seasons (Table 2). In Year 1, the average air and soil temperatures in the high tunnel baby spinach trial were 18.0 °C and 21.7 °C, respectively. For the high tunnel baby leaf lettuce trial in Years 2 and 3, the average air temperatures were 16.9 °C and 20.4 °C, and the average soil temperatures were 20.0 °C and 22.6 °C, respectively. The observed daily maximum high tunnel air temperatures were similar to the open field (<1 °C difference) in all seasons, whereas the daily minimum air temperatures in the high tunnel were 1 to 5 °C greater than the open field (data not shown). The difference in minimum air temperature increased during the colder lettuce seasons, indicating the ability of high tunnels to retain heat during cold nights, especially in the multibay high tunnel system used in this study.

Table 2.

Environmental data during organic pac choi and baby spinach/lettuce production trials in high tunnels (2018–20) in Citra, FL.

View Table

In general, soil VWC at 0–10 cm depth remained above 8.0% throughout each pac choi season, which varied with the soil temperature and plant growth stage (data not shown). With Cowpea and Granular applied, the average soil VWC was consistently higher in Yard than in NC for the pac choi seasons (11.5% vs. 8.8% in Year 1, 13.9% vs. 8.1% in Year 2, 10.7% vs. 8.3% in Year 3), suggesting that compost application enhanced water retention for sandy soils, which is consistent with some previous studies (Mylavarapu and Zinati 2009). In the plots with Cowpea and NC, the average soil VWC was higher in Granular than Liquid (11.1% vs. 10.2%) during the pac choi season in Year 1. However, the average soil VWC was lower in Granular than in Liquid in Year 2 (11.8% vs. 13.8%) and Year 3 (9.1% vs. 10.8%). The cause of this discrepancy was unclear, but it might be related to the longer-term effects of granular vs. liquid organic fertilizer application, which deserves more examination.

The average PAR levels were consistently lower in high tunnels (Table 2) compared with open field in all production seasons, regardless of the shadecloth application. In each pac choi season with shadecloth, the maximum PAR inside the high tunnels was reduced by 33.8%, 65.3%, and 61.4% relative to the open field in Years 1, 2, and 3, respectively (data not shown). In the subsequent spinach or lettuce season without the use of shadecloth, the maximum PAR reduction in the high tunnel was 51.6%, 47.5%, and 42.8% in Years 1, 2, and 3, respectively (Table 2). Previous studies have shown that compared with open field, high tunnels with single-layer of polyethylene could reduce PAR by 20% to 30%, and by 47% to 54% when a shadecloth is added (Cowan et al. 2014; Frey et al. 2020; Retamal-Salgado et al. 2015). A high tunnel with double layers of polyethylene resulted in almost 20% less PAR than a single-layered polyethylene in the first season (Uchanski et al. 2020). Despite the further reduction in light intensity, high tunnels with inflated double-layer polyethylene are expected to provide enhanced protection from extreme weather events, reduce heat loss during cold nights, and offer improved capacities of coping with low temperatures for growing high-value warm-season vegetables (Maynard and O’Donnell 2018; Uchanski et al. 2020). In the present study, the overall reduction pattern of DLI values over time (Table 2) likely suggest the deterioration of the plastic film, indicating a need for regular monitoring of the light levels to ensure sufficient light transmission and intensity for high tunnel grown crops (Frey et al. 2020). For the lettuce trial in Year 3, the average DLI before and after removing the shadecloth was 8.0 and 11.2 mol·m⁻²·d⁻¹, respectively. The average DLI values over 3 years ranged from 8.9 to 14.1 mol·m⁻²·d⁻¹ (Table 2), which were generally above the 8 mol·m⁻²·d⁻¹ threshold required for greenhouse leafy green production according to Glenn et al. (1984).

Marketable yield and dry weight.

The marketable yield of pac choi was significantly affected by year, fertilization, and compost as well as the interactions of year × cover crop and fertilization × compost (Table 3). Cowpea and Fallow resulted in similar yields of pac choi in Years 1 and Year 3; however, a 15.6% yield reduction by Cowpea relative to Fallow was observed in Year 2 (Fig. 1A). This discrepancy might be attributed to a higher C:N ratio of cowpea at termination in Year 2 than other seasons along with lower aboveground dry biomass and N accumulation (data not shown). Similarly, the year × cover crop interaction effects on pac choi dry weight showed that Cowpea only significantly reduced dry weight by 12.0% compared with Fallow in Year 2 (Fig. 1B). The pac choi yield significantly increased from Year 1 to Year 2 regardless of the cover crop treatment (by 23.8% on average); however, pac choi yield in Fallow was significantly reduced by 11.4% from Year 2 to Year 3, with no significant differences found between Year 2 and 3 in Cowpea plots (Fig. 1A). The pac choi yield increase from Year 1 to Year 2 could be attributed to the improved soil fertility over time. However, this increase was not sustained in Year 3, suggesting that the yield ceilings might have been reached as the average yield of 333 g/plant in Year 2 is consistent with the range found in a previous study on organic pac choi in high tunnels (Zhao et al. 2009). On the other hand, the generally higher air and soil temperatures and reduced light intensity in Year 3 might have restricted plant growth and nutrient uptake.

Table 3.

Analysis of variance of the effects of year (Y), cover crop (CC), fertilization (F), and compost (C) on marketable yield, dry weight, ascorbic acid content (AAC), total antioxidant capacity (TAC), total phenolics (TP), and mineral nutrient contents of organically grown pac choi in high tunnels during a 3-year study (2018–20) in Citra, FL.

View Table

View Full Size

Effects of two-way interactions between year and cover crop, fertilization, or compost on yield, dry weight, and mineral nutrient contents of organically grown pac choi in high tunnels (2018–20). (A) Marketable yield, (B) dry weight, (C) K content, (D) S content, (E) B content, and (F) Mn content. Bars sharing the same uppercase letter within each year, and bars sharing the same lowercase letter within a cover crop, fertilization, or compost treatment are not significantly different at P ≤ 0.05 according to Fisher’s least significant difference test. Cowpea = cowpea cover crop; Fallow = weedy fallow control; Granular = preplant of granular organic fertilizer; Liquid = weekly fertigation of liquid organic fertilizer; Yard = yard waste compost; Cow = cow manure compost; Verm = vermicompost; NC = no compost control; DW = dry weight.

Citation: HortScience 58, 12; 10.21273/HORTSCI17327-23

• Download Figure
• Download figure as PowerPoint slide

The limited Cowpea effect on pac choi yield was in line with a 2-year open field study in Florida sandy soils by Allar and Maltais-Landry (2022), indicating a lack of impact from Summer cover crops (i.e., monocultures of sunn hemp/sorghum sudangrass, mixtures of two or three species, or a mixture of five species) on yields of Fall bell peppers (Capsicum annuum) and soil N. In contrast, Kruse and Nair (2016) reported improved soil N and lettuce growth and yield in the fall after the summer planting of cowpea (54–65 d) in an open field study under fine-loamy soil conditions in Iowa. Cover crop selection and timing of termination before planting leafy greens can impact vegetable growth and yields (Hutchinson and McGiffen 2000; Reberg-Horton et al. 2012). In our study, only one legume cover crop was evaluated with a single termination schedule and one termination method. Future studies on high tunnel systems may focus on more systematic examinations of different cover crop species and mixtures, as well as management practices to help determine the role of cover crops in high tunnel soil fertility management. According to Miguez and Bollero (2005), the positive effects of winter leguminous cover crops on the subsequent corn crop yield were greater with reduced N fertilization rates. Allar and Maltais-Landry (2022) also highlighted the importance of improved management of cover crop N to enhance its efficiency and reduce potential N losses, and they suggested that cover crop N may be more impactful in systems with closer row spacing and increased plant density. Hence, more studies are warranted to examine the effect of varying fertilization rates and optimize cover crop termination and management to maximize its benefits in N cycling especially in high tunnel growing systems where N leaching due to rainfall events is minimized.

The impact of compost on pac choi yield and dry weight was dependent on the organic fertilizer treatment (Table 3). With the preplant application of granular organic fertilizer, Yard and Cow led to significantly greater marketable yield than NC by 11.6% and 14.3%, respectively (Fig. 2A). Verm did not differ significantly from Yard, Cow, and NC. In contrast, with the weekly application of liquid organic fertilizer, significant yield improvement was only observed in Yard (by 11.7% on average) compared with all other treatments, which had similar yields. Furthermore, pac choi yield was significantly higher in liquid vs. granular organic fertilizer plots for Yard, Verm, and NC, whereas no differences between Granular and Liquid were found in Cow. The compost × fertilizer interaction on pac choi dry weight showed results similar to the yield response (Fig. 2B), indicating the improved pac choi yield from Liquid could result from increased dry matter accumulation and/or water content. The combined use of compost and organic fertilizer may synergistically increase soil microbial activity and soil fertility, and subsequently enhance nutrient availability for plant uptake (Chen 2006; Molina-Herrera and Romanya 2015). It is likely that the potential of compost to enhance seasonal nutrient availability to meet crop demand could be largely affected by the compost application and fertilization program.

View Full Size

Effects of two-way interactions between fertilization and compost or cover crop on yield, dry weight, and nutrient contents of pac choi and baby spinach or baby leaf lettuce. (A) Marketable yield of pac choi, (B) dry weight of pac choi, (C) total yield of baby leaf lettuce in Year 3, (D) total dry weight of baby leaf lettuce in Year 3, (E) Mg content in baby spinach (first harvest) in Year 1, (F) S content in baby spinach (first harvest) in Year 1, (G) Zn content in baby spinach (second harvest) in Year 1, and (H) Mg content in baby spinach (second harvest) in Year 1. Bars sharing the same uppercase letter within each fertilization treatment, and bars sharing the same lowercase letter within a cover crop or compost treatment are not significantly different at P ≤ 0.05 according to Fisher’s least significant difference test. Cowpea = cowpea cover crop; Fallow = weedy fallow control; Granular = preplant application of granular organic fertilizer; Liquid = weekly fertigation of liquid organic fertilizer; Yard = yard waste compost; Cow = cow manure compost; Verm = vermicompost; NC = no compost control; DW = dry weight.

Citation: HortScience 58, 12; 10.21273/HORTSCI17327-23

• Download Figure
• Download figure as PowerPoint slide

The pronounced benefit from Liquid over Granular on pac choi marketable yield (Fig. 2A) indicates that weekly fertigation during the growing season may have better matched the plant nutrient demand compared with preplant fertilization. The benefits from Liquid may also be attributed to its nutrient sources, as well as its finer particle size and greater reactive surface area, which could result in a faster rate of mineralization compared with Granular. In this study, the granular organic fertilizer was primarily derived from hydrolyzed feather, bone, and blood meal, whereas the liquid organic fertilizer was derived from enzymatically digested fish proteins. As pointed out by Li et al. (2021b), the slower release of N from the granular organic fertilizer may be associated with the presence of complex proteins and long-chain saturated fatty acids. Hartz and Johnstone (2006) compared the N availability of different types of organic fertilizers and found the N mineralization from the fish powder was significantly higher than that of blood meal, feather meal, and seabird guano within the first week at 10, 15, or 20 °C. However, after 2 weeks, the mineralization of all fertilizers slowed down, highlighting the need for split applications of organic fertilizers. Given the relatively high cost of liquid organic fertilizer products, it would be worthwhile to assess the cost-effectiveness of organic fertilization programs when comparing different types and application methods.

With respect to the baby spinach and baby leaf lettuce grown as catch crops following the pac choi, Cowpea did not show positive impacts in their yield and dry weight performance (Table 4), except that Cowpea significantly increased the spinach yield and dry weight in Year 1 for the second harvest (by 142.8% and 131.4%, respectively) and total harvests (by 121.5% and 104.9%, respectively) compared with Fallow (data not shown). The catch crop results supported the cover crop × year interaction observed in pac choi production, suggesting the complexity of using rotational cover crops for improving soil nutrient pools and crop productivity. According to Muchanga et al. (2019), ending hairy vetch 2 to 3 d before transplanting tomato in high tunnels with clay loamy soils caused an increase in soil inorganic N concentration up to 16 weeks after transplanting compared with the bare soil control, indicating the contribution from the cover crop may have had a long-lasting effect. In the catch crop trials of the present study, fertilization treatments also showed little influence on crop yield and dry weight except that in Year 3 cover crop × fertilization interaction significantly impacted lettuce yield and dry weight for the first and total harvests, with the three-way interaction of cover crop, fertilization, and compost influencing the first harvest of lettuce (Table 4). The two-way interaction showed higher total lettuce yield in Cowpea than Fallow for Liquid, but it was the opposite for Granular (Fig. 2C). On the other hand, compared with Liquid, Granular enhanced yield with Fallow, but reduced yield with Cowpea. The cover crop × fertilization interaction had a similar impact on lettuce dry weight (Fig. 2D). These results were possibly associated with the residual nutrients from the previous crop season and reflected the long-term or legacy impact of cover cropping and organic fertilizer and compost application because no nutrient sources were applied during the catch crop season. Given that the composition of different organic fertilizers (Li et al. 2021b) and cover crops (Woodruff et al. 2018) may determine their shorter- and longer-term nutrient availability, it is likely that different organic fertilizer and cover crop combinations could show differential impacts on nutrient supply to vegetable crops, together with the site-specific environmental effects. Interestingly, the cover crop × fertilization × compost interaction on the first harvest of lettuce in Year 3 also indicated that Yard and Cow further enhanced crop yield of the Cowpea + Liquid combination compared with NC, but the compost effect was lacking in other cover crop and fertilization combinations (Table 5).

Table 4.

Analysis of variance of the effects of cover crop (CC), fertilization (F), and compost (C) on marketable yield and dry weight of organically grown baby spinach (Year 1) and baby leaf lettuce (Years 2 and 3) as catch crops in high tunnels during a 3-year study (2018–20) in Citra, FL.

View Table

Table 5.

Effects of three-way interactions among cover crop, fertilization, and compost on marketable yield from the first harvest and total antioxidant capacity of organically grown lettuce as catch crops in high tunnels in Year 3 during a 3-year study (2018–20) in Citra, FL.

View Table

Yard significantly increased spinach yield compared with Verm (by 59.0% on average) and NC (by 74.4% on average) for the second and total harvests of Year 1 and resulted in higher total lettuce yield by 52.3% compared with NC in Year 3 (Table 4, Fig. 3A–C). The advantages of compost application on increasing catch crop productivity, particularly using yard waste compost in Year 3, may be related to the gradual release of certain amounts of nutrients from the compost, the carryover effect of continued compost application and the overall improvement of soil fertility and health (Reeve and Drost 2012).

View Full Size

Effects of fertilization or compost on yield and mineral nutrient contents of pac choi and baby spinach or baby leaf lettuce. (A) Yield of baby spinach (second harvest) in Year 1, (B) total yield of baby spinach in Year 1, (C) total yield of baby leaf lettuce in Year 3, (D) N content of pac choi, (E) P content of pac choi, (F) B content of baby spinach (first harvest) in Year 1, (G) B content of baby spinach (second harvest) in Year 1, (H) Fe content of baby spinach (second harvest) in Year 1, (I) K content of baby leaf lettuce (first harvest) in Year 3, and (J) ascorbic acid content (AAC) of pac choi. Bars sharing the same letter are not significantly different at P ≤ 0.05 according to Fisher’s least significant difference test. Yard = yard waste compost; Cow = cow manure compost; Verm = vermicompost; NC = no compost control; Granular = preplant application of granular organic fertilizer; Liquid = weekly fertigation of liquid organic fertilizer; DW = dry weight; FW = fresh weight.

Citation: HortScience 58, 12; 10.21273/HORTSCI17327-23

• Download Figure
• Download figure as PowerPoint slide

Leaf mineral nutrient contents.

In the pac choi trial, the year was included in the data analysis model to help examine the carryover effects from soil fertility management. On a dry weight basis, the leaf N, P, Mg, Ca, Zn, and Fe contents varied with year (Table 3), with a higher level of N in Years 2 and 3 than in Year 1 but a lower level of P in Year 2 than in Years 1 and 3 (Table 6). The contents of Ca and Fe decreased from Year 1 to 3, whereas Mg and Zn reached the highest level in Year 3. The cover crop did not significantly impact the leaf mineral nutrient content, except for the year × cover crop × fertilization interaction on Mn (Table 3). Under Cowpea, Granular significantly lowered Mn content in Year 1 but showed an increase in Year 3 compared with Liquid. In contrast, under Fallow, the Mn content did not differ between the two fertilization treatments (Table 7). In terms of the change across years, Year 2 showed the lowest level of leaf Mn in all cover crop and fertilization combinations with a greater reduction in the Cowpea plus Liquid treatment.

Table 6.

Effects of year on ascorbic acid content (AAC), total phenolics (TP), and nutrient contents of organically grown pac choi in high tunnels during a 3-year study (2018–20) in Citra, FL.

View Table

Table 7.

Effects of three-way interactions among year, cover crop, and fertilization on Mn content and total antioxidant capacity of organically grown pac choi in high tunnels during a 3-year study (2018–20) in Citra, FL.

View Table

The significant main effect of fertilization on N and P (Table 3) indicated increases (2.9% and 11.2%, respectively) in Liquid vs. Granular (Fig. 3D and E). For K and S, the fertilization effect varied with year (Table 3). Compared with Granular, Liquid only increased K and S contents in Year 1, no differences were found between Liquid and Granular in Years 2 and 3. The highest contents of K and S were found in Year 3, and the yearly increase was more pronounced in Granular vs. Liquid (Fig. 1C and D), which may be associated with the higher soil and air temperatures in Year 3 as well as the carryover effects of fertilization treatments. On the other hand, despite the enhanced K and S contents in Year 3, there was no further improvement in yield compared with Year 2, indicating potential luxury consumption. The positive effects of Liquid on pac choi leaf N and P contents were congruent with the increased yield benefit from Liquid, likely suggesting the enhanced plant uptake with Liquid vs. Granular. Similarly, Woldetsadik et al. (2018) observed positive correlations between lettuce yield and certain nutrients, such as P, K, and Mg, when grown with biochar application in a pot experiment. However, according to Laur et al. (2021), while arugula (Eruca sativa) yield was linked to higher levels of P, S, and Cu, a similar correlation was not found in lettuce or basil (Ocimum basilicum) grown in high tunnels. Further investigation is warranted to explore the impact of mineral nutrient uptake on the yield of different leafy greens grown in various production systems.

The significant effect of compost on pac choi leaf B and Mn contents was year-specific (Table 3). Both B and Mn contents reached the lowest level in Year 2. For B, the level was higher in Yard than NC in Years 2 and 3, with no treatment difference in Year 1 (Fig. 1E). For Mn, the compost treatments did not differ from the NC in Years 1 and 2, whereas Yard and Cow caused a reduction compared with Verm and NC in Year 3 (Fig. 1F). This discrepancy was likely due to the variable nutrient compositions of composts in each year (Table 1), and the soil nutrient pool dynamics over 3 years from repeated applications of composts (Rudisill et al. 2015).

The mineral nutrient contents of baby spinach and baby leaf lettuce were analyzed separately for each catch crop season. During the spinach trial in Year 1, significant cover crop × fertilization interaction effects were found on leaf Mg and S contents for the first harvest and on Zn contents for the second harvest, in that Granular and Liquid led to similar nutrient levels under Fallow, but higher contents were observed in Granular under Cowpea (Fig. 2E–G). In contrast, higher Mg contents were found in Liquid vs. Granular under Fallow for the second harvest (Fig. 2H). For the lettuce trial in Year 2, Liquid significantly increased leaf P and Ca contents but reduced Zn content in comparison with Granular (Table 8). A similar impact of Liquid on increasing P contents was also found for the first lettuce harvest in Year 3 (Table 8).

Table 8.

Effects of cover crop (CC), fertilization (F), and compost (C) on mineral nutrient contents, ascorbic acid content (AAC), total antioxidant capacity (TAC), and total phenolics (TP) of organically grown baby leaf lettuce as catch crops in high tunnels in Citra, FL.

View Table

Some legacy effects of compost application on improving mineral contents in spinach and lettuce following pac choi production were also observed. For example, in Year 1, Yard increased B content in both spinach harvests compared with all other treatments and Fe content was higher in Yard relative to Cow and Verm in the second harvest (Fig. 3H). However, this impact was not consistent in the following two seasons of baby lettuce, which may be due to different nutrient uptake in spinach vs. lettuce production. In contrast to other compost treatments, Cow significantly increased Cu content of baby lettuce in Year 2 and K content in the first harvest of Year 3, yet reduced Ca content in Year 2 (Table 8, Fig. 3I). Additionally, Yard and Cow reduced Mn contents relative to Verm and NC in lettuce of Year 2 (Table 8). When comparing Verm to Yard and Cow applied before transplanting pac choi seedlings over 3 years, the latter two contained higher levels of P, Ca, Mg, and Mn (Table 1). Furthermore, because the application rate of Verm was only a quarter of that used for Yard and Cow, the latter two led to greater inputs of almost all nutrients and organic matter, which might have contributed to the increased uptake of K, Ca, Fe, Cu, and B in baby spinach or baby leaf lettuce.

Mineral nutrients are important vegetable quality parameters that can be affected by fertilization programs and environmental conditions (Laur et al. 2021; Zhao et al. 2007). The average macronutrient and micronutrient levels of pac choi found in the current study were adequate to high, suggesting nutrients were not a limiting factor for growth (Johnson et al. 2012; Zhao et al. 2007). Combining organic fertilizer with soil amendments has been suggested to increase plant nutrient availability in organic production systems, especially during the crop’s early growth stage (Barker et al. 2017). When examining the mineral content on a fresh weight basis, cover cropping did not demonstrate any significant effects, and compost treatments showed limited and inconsistent impacts; however, fertilization treatments exhibited consistently positive effects on N, Mg, and Zn contents in pac choi (data not shown). Granular significantly increased N (by 7.2%), Mg (by 10.0%), and Zn (by 9.7%) compared with Liquid, and showed a tendency to enhance Ca (P = 0.057, by 7.9%), B (P = 0.071, by 6.1%), Mn (P = 0.085, by 9.8%), and Fe (P = 0.078, by 11.0%) in pac choi. Mineral contents on a fresh weight basis may better represent the nutritive value as most leafy greens are consumed fresh; however, the observed differences may not be able to impact human nutrition meaningfully (Marles 2017).

Ascorbic acid content, total antioxidant capacity, and total phenolics.

The phytochemical contents of pac choi and lettuce were only evaluated in Years 2 and 3. In Year 3, only the lettuce samples from the first harvest were examined. Year was continued to be included in the data analysis model for pac choi but not for lettuce.

Cowpea and composts did not show any significant main impact on ascorbic acid content, total antioxidant capacity, or total phenolic content of pac choi, although a three-way interaction of year × cover crop × fertilization significantly affected total antioxidant capacity (Table 3), showing a greater level in Granular vs. Liquid for Cowpea in Year 3 but not for Fallow or Year 2 (Table 7). Compared with Liquid, Granular also significantly increased the ascorbic acid content of pac choi by 14.0% on average across both years (Fig. 3J). In some previous studies the production of antioxidant compounds in plants may be altered in response to different fertilizer sources, possibly due to the changes in available nutrients and microbial activity around the root zone (Lombardi-Boccia et al. 2004; Nicoletto et al. 2014; Zhao et al. 2009). However, there is a lack of definitive evidence on the relationship between yield and phytochemical content in leafy greens. Considering the higher crop productivity in Liquid vs. Granular in the present study, it would be interesting to further explore the relationship between crop yield and antioxidant phytochemical levels as affected by fertilization programs in high tunnel organic leafy green production. The seasonal variation showed that the total phenolic content of pac choi was lower in Year 2 than in Year 3 (Table 6), possibly linked to higher air and soil temperatures and PAR levels in Year 3. Greater levels of total antioxidant capacity were also observed in Year 3 except for the Cowpea + Liquid combination (Table 7). However, ascorbic acid content was found to be higher in Year 2 than in Year 3 (Table 6). While previous studies reported the pronounced impact of environmental conditions on antioxidant compounds and lower levels in high tunnel grown leafy greens in contrast to open field production (Oh et al. 2011; Zhao et al. 2009), our study pointed out the need for understanding the changes in antioxidant properties of leafy vegetables produced in high tunnels as related to environmental management.

For baby lettuce as the catch crop, ascorbic acid content and total antioxidant capacity differed significantly between the compost treatments and NC in Year 2. Compared with NC, all compost treatments significantly reduced ascorbic acid content, whereas Yard and Cow also decreased total antioxidant capacity (Table 8). The increased phytochemical contents in NC may be associated with the enhanced plant stress relative to the compost treatments. Higher levels of abiotic stress, such as nutrient deficiency or suboptimal supply of nutrients, or water stress, may affect not only crop yield but also phytochemical production. According to Zhao et al. (2009), increased phenolic contents in pac choi plants might be associated with less N availability when comparing organic and conventional fertilizer treatments in both high tunnel and open field systems. Oh et al. (2010) reported that mild water stress applied to lettuce in open field before harvest could improve its crop quality in terms of phytochemical contents without causing notable negative impacts on lettuce growth or yield. In Year 3, the total antioxidant capacity of baby lettuce was affected by the cover crop × fertilization × compost interaction (Table 5). It is noteworthy that compost did not impact total antioxidant capacity in the Cowpea + Liquid combination in which the highest lettuce yield was produced, whereas certain compost treatments reduced the total antioxidant capacity of lettuce for other Cowpea and organic fertilization combinations.

Overall, the levels of ascorbic acid, total antioxidant capacity, and total phenolics of pac choi and baby lettuce in this study were comparable to those reported in previous studies (Galieni et al. 2015; Unal et al. 2014; Vargas-Arcila et al. 2017; Wang et al. 2010; Zhao et al. 2009). As research continues to develop high tunnel soil and nutrient management programs with the goal of improving organic vegetable production, high tunnel vegetable quality attributes also deserve more in-depth studies.