Effect of Low Tunnels and Planting Dates on the Fruit Quality of Organic Day-Neutral Strawberries in High Tunnels in North Carolina

ABSTRACT

Fresh strawberry is in high demand because of the fruit’s unique flavor, taste, and high nutritional value. Berry quality is vital for fresh markets. Day-neutral strawberry varieties such as Albion and San Andreas have proven to produce a higher yield and better quality fruit than some short-day varieties. High tunnels (HT) and low tunnels (LT) are considered as good micro-environmental management options for the season extension of strawberries. However, the effects of LT inside HT on berry quality of day-neutral strawberries have not been reported. This 2-year study aimed to determine if LT and planting dates would affect the fruit quality of day-neutral strawberries in organically managed HT. The weight, size, color, texture, total soluble solids (TSS), and total titratable acidity (TTA) of berries were quantified. Planting dates and LT did not affect fruit quality variables except that fruit weight was reduced in LT in March and April. ‘Albion’ had larger, firmer fruit, and had higher TSS and TAA values than ‘San Andreas,’ except in samples collected in February. ‘San Andreas’ had more reddish and brighter fruit than ‘Albion.’ Although ‘Albion’ may produce better quality fruit than ‘San Andreas,’ both varieties would produce quality fruit for fresh markets in early spring. Including LT in HT did not benefit the fruit quality of organic day-neutral strawberries.

KEYWORDS:

• Fragaria × ananassa
• soluble solids
• titratable acidity
• season extension

Introduction

Fresh strawberry is known for its unique taste, flavor, and nutrient value (Montero et al., 1996). It is an excellent source of Vitamin C and other antioxidants (anthocyanin and ellagitannins) and has anticancer compounds polyphenols (Giampieri et al., 2012; Seeram, 2008; Voca et al., 2009). Strawberry fruit of 160 g will provide 140% of daily recommended amount of Vitamin C (Giampieri et al., 2012; Seeram, 2008). Strawberry is one of the top-five fresh fruits (USDA-ERS, 2017) in the United States (apple, banana, grapes, watermelon, and strawberry) based on per capita consumption, which is about 7.9 pounds in 2013 (Lewers et al., 2017; Rowley et al., 2011). The demand for fresh strawberry is always high in the United States. Fresh strawberry has a higher price in direct markets (e.g., farmer’s market, roadside stand, and pick-your-own) compared to wholesale markets (Ballington et al., 2008; Kadir et al., 2006; Poling, 1993). Although the average retail price is relatively low ($3.46/lb) during the regular season (USDA-AMS, 2018), the off-season price is always higher, ranging from 4.4 USD to 5 USD/lb (Ballington et al., 2008; USDA-AMS, 2019).

Fruit quality is crucial for fresh markets. Strawberry fruit quality is evaluated based on variables of appearance (size, shape, and color), texture (firmness), flavor (i.e., balance between sweetness, sourness, and aroma), and the content of minerals, vitamins, and other organic compounds (ElMasry et al., 2007; Hemming et al., 2006). The combination of organic acids and sugars can be used to determine the flavor of the fruit. Citric acid is one of the major acids in strawberry. Similarly, glucose, sucrose, and fructose are the main soluble components which are important to the taste of berries (Gunduz and Ozdemir, 2014; Kafkas et al., 2007). Strawberry fruit with the cap (calyx) attached, ⩾1.9 cm (3/4 inch) in diameter, ⩾75% of the surface showing red or pink color, and free from disease/pest infection is considered U.S. No.1. Fruit of ⩾1.6 cm (5/8 inch) in diameter with at least 50% of surface turning pink or red and free from decay or serious damage is considered U.S. No.2 (USDA-AMS, 2006). In scientific studies, however, researchers have considered a fruit as marketable when its weight is above 10 g/fruit and blemish-free (Fernandez et al., 2001; Rowley et al., 2011).

The fruit size, weight, firmness, and chemical compositions such as total soluble solids (TSS), sugars, total titratable acidity (TTA), and colors are affected by various abiotic factors. For example, planting dates affect the TSS and ascorbic acid content of fruits because the difference in temperatures during fruiting affected the accumulation of TSS and ascorbic acids (Rahman et al., 2014). The light and temperature also affect the accumulation of TSS in fruits by controlling the biosynthesis of soluble solids and stomatal opening and closure (Henschel et al., 2017). Cultivars and growing systems affect fruit quality as they can change the harvest dynamics of fruits (Voca et al., 2009). Temperatures affect the metabolism and the transpiration. Solar radiation affects photosynthesis, soluble solids, pH and acidity, and color development of fruits (Gude et al., 2018; Hemming et al., 2006).

High tunnels (HT) are used to grow various small fruits and vegetable crops as they modify microclimate inside. They extend the growing and harvest seasons, increase yield and fruit quality, and protect crops from low-temperature damage during winter (Carey et al., 2009; Gu et al., 2017; Kadir et al., 2006; Rowley et al., 2010, 2011; Wien, 2009; Xiao et al., 2001; Zhao and Carey, 2009). Similarly, low tunnels (LT) are smaller structural frames of about 0.7 m tall covered with thinner plastic films that usually increase fruit quality as they protect fruit and produce from adverse climatic factors such as rain, frost, and winds (Acharya et al., 2020, 2019; Jenni et al., 2006; Lewers et al., 2017; Petran et al., 2017; Singh et al., 2012).

There have been reports of higher fruit quality of strawberry in the organic production system, compared to the conventional system, especially on the phytochemical contents such as antioxidants (total phenolic compounds, anthocyanins, and ascorbic acids), sweetness, and color (Crecente-Campo et al., 2012; Khalil and Hassan, 2015). Day-neutral (DN) cultivars including California cultivars Albion and San Andreas produced better quality fruit in protected culture, compared to ever-bearing and June-bearing cultivars in Korea (Ruan et al., 2013), Romania (Tudor et al., 2014), Canada (Van Sterthem et al., 2017), and the U.S. (Anderson et al., 2019; Ballington et al., 2008; Gu et al., 2017; Gude et al., 2018; Lewers et al., 2017; Petran et al., 2017; Rowley et al., 2011). In the Southeast United States, DN cultivars are in research trials focusing mostly on yield and growth patterns (Rana and Gu, 2020). No reports have been available on the effect of planting dates and LT on fruit quality of DN strawberries in the organic HT production system. In this study, we aimed to investigate the effect of LT, planting dates, and cultivars on the fruit quality of DN strawberries using the annual strawberry plasticulture system inside HT. This study will help strawberry growers choose suitable DN varieties, planting time, and frost protection methods to produce quality fruit during late fall and spring in North Carolina.

Materials and Methods

The study was conducted at the University Farm of North Carolina Agricultural and Technical State University in Greensboro, NC (36°04ʹ09.8”N 79°43ʹ49.6”W, 228 m), which is in the plant hardiness zone 7b (USDA-ARS, 2012). The experiment was conducted from August 2016 to May 2017 in the first year (2016/17) and from August 2017 to May 2018 in the second year (2017/18). Soil type in the high tunnel (HT) is Enon sandy loam. Summer cover crops sorghum-sudangrass and pearl millet were planted inside the HT before the first year of research, and buckwheat and soybean were planted before the second year of study. The research site was in the second (2016) and third year (2017) of organic transition.

High Tunnel and Low Tunnels

One Quonset-style HT of 29.3 m long, 9.1 m wide, and 3.7 m high at the center (Jaderloon®, The Greenhouse Company. Columbia, SC) was used for this study. Low tunnel kits of 30.5 m long, 0.7 m wide, and 0.8 m high and covered with 0.038 mm (1.5 mil) transparent plastic film perforated in both sides (Dubois Agrinovation, Quebec, Canada) were installed in late October each year. The HT air temperature was maintained around 21–24°C by opening and closing the retractable HT’s sidewalls. All LT remained closed after installation except the time of data collection and pest management. During winter months, all plots (with or without LT) inside the HT were covered with non-woven floating row covers of 42.5 g yard⁻² (Gro-Guard®, Atmore Industries, Inc., Atmore, AL) at night to provide extra frost protections to the plants.

Plug Production, Field Preparation, Transplanting, and Pest Management

Strawberry plugs for this study were not available commercially during the planting time, so we raised our own plugs in a greenhouse, which is located at the Horticulture Unit of the University Farm, according to the protocol in the Southeast Regional Strawberry Plasticulture Production Guide (Poling, 1993). In brief, conventional runner tips of ‘Albion’ and ‘San Andreas’ were collected in mid-July from Norton Creek Farms (Cashiers, NC) and stored at 4.4°C and 75–80% relative humidity (RH) for two weeks. Before inserting into growth media, excessive leaves and runner cords of tips were trimmed off leaving two leaves per tip. Prepared tips were inserted 6 cm deep in cells of 50-cell trays containing a growing media of 1:1 ratio mixture of organic compost (Brooks Contractor, Goldston, NC) and Sunshine organic soil mix (Sun Gro Horticulture Inc, Canada). The trays with tips were placed on mist benches inside the greenhouse for 28 d.

The soil inside the HT was tilled twice before making the raised beds. Based on the soil test results, pre-plant fertilizers at 13.8 kg N, 2.4 kg K, 6 kg S, and 0.14 kg B per acre were spread evenly inside the HT. Eight beds (along the length of the HT) measured at 25.5 cm high and 76 cm wide and spaced at 1.5 m (bed center to center) were made inside the high tunnel. The beds were covered with 122 cm wide, 0.03 mm (1.25 mil) embossed black plastic mulch (Berry Hill Irrigation, Buffalo Junction, VA). One 0.2 mm (8 mil) drip tape with 30.5 cm emitter spacing was placed in the middle of each bed (Toro Micro-Irrigation, EI Cajon, CA), under the plastic mulch. The drip tape has the capacity to deliver water at the rate of 1.0 L/h.

Strawberry plugs, with flowers removed, were transplanted into the raised bed on 1 Sept. (D1) and 29 Sept. (D2) in 2016 (the first year) and on 9 Sept. (D1) and 10 Oct. (D2) in 2017 (the second year). The six of eight beds in the middle of the HT were used for this study, and the two side beds were used for guard plants. Plugs spaced out at 30.5 × 30.5 cm within and between the rows. Staggered two rows of plugs were planted in each bed.

During the growing seasons, fish emulsion (5-1-1, Schafer Fisheries, Inc. Thomson IL) was used for fertilization. The plants were fertigated weekly from mid-February to the end of harvest at the rate of 1.1 kg/ha N. About 25 mm (1 inch) of water was irrigated each week through drip tapes, which was split up to four times a week based on weather conditions.

To manage pests (disease, insect, and spider mite), plants were scouted weekly for pest incidence. Dead leaves were pruned off as a preventive means. Predatory mites (Phytoseiulus persimilis) were released at the rate of 1.4 adult/ft² (Spidex, Koppert Biological System Inc., Howell, MI) when one 2-spotted spider mite per leaf (on average) was observed. Organic Materials Review Institute (OMRI) listed pesticide M-Pede (potassium salts of fatty acids), Trilogy (neem oil), Dipel (Bacillus thuringiensis), PFR-97 (Isaria fumosorosea), or PyGanic (pyrethrins) were applied based on the pests scouted.

Experimental Design

The experimental design was a completely randomized block design conducted as a split-split plot, with three replications. The protection methods [LT and no LT (NLT)], planting dates (D1 and D2), and cultivars (Albion and San Andreas) were randomized on the main plots (27.4 m in length), split plots (13.7 m), and split-split plots (6.8 m), respectively. There was a total of 42 plants in each split-split plot, with which 18 plants were used for harvesting and fruit quality analysis, the rest were used for a biomass study or as guard plants.

Data Collection

Temperature Measurement

Two WatchDog 1650 micro-stations (Spectrum Technologies, Inc, Plainfield, IL) were installed inside the HT at the height of the plant canopy; one under LT and another outside of LT (in the middle two rows) to measure air temperature and relative humidity. A WatchDog 2000 micro-weather station (Spectrum Technologies, Inc, Aurora, IL) was also installed at 1 m height off the ground outside of the HT. Measurements were recorded automatically every 30 min and data were analyzed with the SpecWare9 Pro software (Spectrum Technologies, Inc., Plainfield, IL).

Fruit Quality Analysis in Field

Blemish-free fruits of ⩾10 g were considered marketable (Fernandez et al., 2001; Rowley et al., 2011; Ruan et al., 2011). Six marketable fruits without visual defects from each split-split plot were randomly selected to measure the size, single fruit weight (SFW), and total soluble solids (TSS) on the same day of harvest at biweekly intervals from March to May each season. The measurements were done at the Horticulture Unit workstation at the University Farm. The TSS (%) was measured with a digital °Brix refractometer (Spectrum Technologies, Inc, Plainfield, IL). The SFW (g/fruit) and single fruit size (length and width in mm) were measured with a digital weighing scale and a digital Vernier caliper. The SFW, TSS, and single fruit size were presented as the mean of six data points for each plot.

Fruit Quality Analysis in Lab

The fruit color, texture, and total titratable acidity (TTA) were analyzed in the Food Science Lab of North Carolina Agricultural & Technical State University in mid-February and mid-April in 2017, and in the first week of May in 2018. All fruit measurements except the TTA were measured on the same day of fruit harvest.

For the lab analyses, five fruits without visual defect from each plot were sampled, transferred to the lab within 2–3 h of harvest, cleaned, and gently wiped with chemwipe to absorb water on the fruit surface. Fruit color was measured in Commission Internationale de I’Eclairage (CIE) L*, a* and b* color scale with a CM-3500d Spectrophotometer (Konica-Minolta, Tokyo, Japan). For each fruit, two readings of L*, a*, b* value were taken at two opposite sites. The brightness is represented by the L* value (L* = 0 and L* = 100 means completely dark and white fruit, respectively). The a* and b* value represents the color channels; a* = 0 and b* = 0 represent true neutral gray fruit, negative a* and b*values means green and blue fruits, and positive a* and b* values means red and yellow fruits (Smith and Yu, 2015; Voca et al., 2009). Among the five fruits in each plot, the brightest and most even-colored one was used as a reference, for the color variable measurements.

Following the color measurements, fruit texture was measured with a TA.XT2 texture analyzer (Texture Technology Corp, Scarsdale, NY). Berries were cut halfway through the fruit (cutting equatorially at their maximum diameter) with a 3-mm thick knife blade probe at a speed of 3 mm/s loaded with a 5 kg load cell. The maximum force required to cut the berry in half (through the maximum) diameter was recorded and used as an indicator of fruit texture in newton (N) for all five berries.

After texture measurements, all 5-fruit samples were blended into puree using a Conair™ Waring™ Laboratory Blender (Fisher Scientific, GA) and stored in 50 ml centrifuge tubes at −20°C. Within one to two weeks of storage, the sample was thawed at room temperature and centrifuged at 3000 g for 20 min at 22°C using a 5180 R Centrifuge (Eppendorf). The TSS was determined by using a couple of drops of juice in a digital Brix/RI-chek refractometer (Reichert Inc, Depew, NY) and expressed as °Brix. For each plot’s sample, the 10 ml supernatant of centrifuged berry juice (from blended five berries) was diluted with 90 ml of distilled water, which was divided into four equal parts (i.e., four 25-ml aliquots) in 100-ml beakers. Each aliquot was titrated with 0.1 M NaOH dropwise from 10-ml burette until the pH of the aliquot sample reached 7.0. The net volume of 0.1 M NaOH needed to titrate a 25-ml aliquot sample was recorded. The TTA was expressed as anhydrous citric acid on weight basis using the equation below (AOAC, 1999).

% Acid (w/w) = [(net ml of titrant x N of NaOH)/W of sample] x 6.4

where N = normality of NaOH, W = sample weight in gram (2.5 g for each sample)

The color and texture values were expressed as mean of five replicated data points, and the TTA and °Brix were the means of the three replicated data points (one aliquot out of the four was used as a spare).

Statistical Analysis

The analysis of variance (ANOVA) was performed using the Proc Mixed procedure of SAS V9.2C for Windows (SAS Institute, Cary, NC). The LSMEANS statement was used for comparisons of means of the different treatments. Significance was determined by the Fisher’s least significant difference at a level of P < .05.

Result and Discussion

Fruit Size (Length and Width)

In both years, there were no 2-way or 3-way interactions among protection method (P), planting date (D), and varieties (V) for fruit length and width (Tables 1 and 2). In the first year, LT and planting date had no effect on fruit size except for fruit samples collected on 15 May and on 13 April (Table 1). On 15 May, plants of the LT had a larger fruit size than that of the NLT. On 13 April, D2 plants had larger fruit compared to D1 plants. ‘Albion’ had bigger fruit than ‘San Andreas.’ In the second year, fruit size was not affected by LT, planting dates, and varieties except that the fruit width of D1 and NLT plants was significantly larger than that of D2 and LT, respectively, on 5 March 2018. There were no differences in fruit size between the two varieties (Table 2).

Table 1. Effect of low tunnels, planting dates, and varieties on the fruit size in 2016/2017

	Fruit length (mm)				Fruit width (mm)
Treatment	9-Mar	30-Mar	13-Apr	1-May	15-May	9-Mar	30-Mar	13-Apr	1-May	15-May
Protection method (P)
LT^a	41.4	42.9	44.0	46.1	41.3	36.6	35.3	32.5	34.4	34.8
NLT	41.1	43.3	45.5	44.2	37.0	36.1	35.9	33.2	31.9	30.8
Planting date (D)
D1	41.4	42.8	43.6	44.9	38. 7	36.6	35.0	31.9	32.0	32.5
D2	40.2	43.4	45.9	45.4	39. 7	36.1	36.2	33.8	34.3	33.2
Varieties (V)
Albion	46.1	44.3	46.8	47.2	41.6	39.1	35.7	33.1	34.1	34.6
San Andreas	36.5	41.9	42.7	43.1	36.8	33.6	35.5	32.5	32.2	31.1
Significance
P	ns	ns	ns	ns	0.005	ns	ns	ns	ns	0.0031
D	ns	ns	ns	ns	ns	ns	ns	0.0212	ns	ns
P*D	ns	ns	ns	ns	ns	ns	ns	ns	ns	ns
V	0.0002	ns	0.0052	0.0445	0.0021	0.0012	ns	ns	ns	0.0076
P*V	ns	ns	ns	ns	ns	ns	ns	ns	ns	ns
D*V	ns	ns	ns	ns	ns	ns	ns	ns	ns	ns
P*D*V	ns	ns	ns	ns	ns	ns	ns	ns	ns	ns

^aLT: low tunnels; NLT: no low tunnels; D1:  1September; D2:  29September; ns: nonsignificant (P ≥ 0.05).

Table 2. Effect of low tunnels, planting dates, and varieties on the fruit size in 2017/2018

	Fruit length (mm)			Fruit width (mm)
Treatment	5-Mar	23-Mar	6-Apr	5-Mar	23-Mar	6-Apr
Protection Method (P)
LT^a	43.9	44.1	38.0	30.9	31.9	36.1
NLT	46.5	42.6	41.2	33.7	31.9	38.2
Planting Date (D)
D1	46.8	44.0	40.7	33.1	32.6	35.7
D2	43.6	42.7	38.5	31.5	31.1	38.5
Varieties (V)
Albion	47.0	43.7	38.8	33.1	31.5	35.5
San Andreas	43.4	43.0	40.4	31.5	32.2	38.8
Significance
P	ns	ns	ns	0.0305	ns	ns
D	ns	ns	ns	0.0344	ns	ns
P*D	ns	ns	ns	ns	ns	ns
V	ns	ns	ns	ns	ns	ns
P*V	ns	ns	ns	ns	ns	ns
D*V	ns	ns	ns	ns	ns	ns
P*D*V	ns	ns	ns	ns	ns	ns

^aLT: low tunnels; NLT: no low tunnels; D1 = 9 September; D2: 10 October; ns: nonsignificant (P ≥ 0.05).

Fruit Weight

For the single fruit weight (SFW), there were no 2-way or 3-way interactions in the first year among P, D, and V (Table 3). LT did not affect fruit weight except on 15 May, when LT plants had higher SFW than NLT plants. Planting dates did not affect SFW except for samples in April when D2 plants had higher SFW than D1. ‘Albion’ always had heavier fruit than ‘San Andreas.’

Table 3. Effect of low tunnels, planting dates, and varieties on single fruit weight (g/fruit) in the first (2016/2017) and second year (2017/2018)

	2016/17					2017/18
Treatment	9-Mar	30-Mar	13-Apr	1-May	15-May	5-Mar	23-Mar	6-Apr
Protection Method (P)
LT^a	24.2	22.3	20.1	22.4	20.8	19.7	19.1	18.0
NLT	22.9	22.9	21.5	19.2	15.9	22.8	18.2	22.0
Planting Date (D)
D1	23.7	21.4	19.4	19.3	17.8	23.2	19.5	19.4
D2	23.3	23.8	22.2	22.3	18.9	19.3	17.8	20.6
Varieties (V)
Albion	28.9	23.6	22.0	23.2	20.8	23.3	18.6	18.3
San Andreas	18.2	21.6	19.6	18.4	15.9	19.2	18.7	21.8
Significance
P	ns	ns	ns	ns	0.0033	0.0368	ns	0.0055
D	ns	ns	0.0192	ns	ns	0.0108	ns	ns
P*D	ns	ns	ns	ns	ns	ns	ns	ns
V	0.0002	ns	0.038	0.0219	0.0031	0.0078	ns	0.0126
P*V	ns	ns	ns	ns	ns	ns	0.0337	ns
D*V	ns	ns	ns	ns	ns	0.0241	ns	ns
P*D*V	ns	ns	ns	ns	ns	ns	ns	ns

^aLT: low tunnels; NLT: no low tunnels; D1: 1 September 2016 or 9 September 2017. D2: 29 September 2016 or 10 October 2017. ns: non-significant (P ≥ 0.05).

In the second year, there were also no three-way (PxDxV) interactions. However, 2-way interactions existed between planting date and varieties (DxV) on 5 March and between protection method and varieties (PxV) on 23 March (Table 3). On 5 March and 6 April, LT plants had smaller SFW than NLT plants. Looking at the D x V interaction, ‘Albion’ of D1 had higher SFW than of D2, but SFW of ‘San Andreas’ was not affected by planting dates. For D1, ‘Albion’ had higher SFW than ‘San Andreas’; for D2, no difference occurred for SFW between the two cultivars (Table 4). As for the P x V interaction, the two cultivars had no difference in SFW in both LT and NLT. LT increased the SFW of ‘San Andreas’ compared to NLT, but LT had no effect on SFW of ‘Albion’ (Table 5).

Table 4. Simple effect of planting dates (D) and varieties (V) on single fruit weight (g/fruit) on 5 March in 2017/2018

	Single fruit weight (g/fruit)
Treatment	5-Mar
Albion
D1^a	26.9
D2	19.6
p-Value	0.0016
San Andreas
D1	19.4
D2	18.9
p-Value	Ns
D1
Albion	26.9
San Andreas	19.4
p-Value	0.0012
D2
Albion	19.6
San Andreas	18.9
p-Value	ns
D*V	0.0241

^aD1 = 9 September, D2 = 10 October. ns = non-significant (p ≥ 0.05).

Table 5. Simple effect of protection methods (P) and varieties (V) on single fruit weight (g/fruit) on 23 March in the 2017/2018

	Single fruit weight (g/fruit)
Treatment	23-Mar
LT^a
Albion	17.8
San Andreas	20.5
p-Value	ns
NLT
Albion	19.4
San Andreas	17.0
p-Value	ns
Albion
LT	17.8
NLT	19.4
p-Value	Ns
San Andreas
LT	20.5
NLT	17.0
p-Value	0.0396
P*V	0.0337

^aLT: low tunnels; NLT: no low tunnels; ns = non-significant (p ≥ 0.05).

‘Albion’ had higher SFW and fruit size than ‘San Andreas’ in the first year, but no differences were found between the two cultivars in the second year. LT plants had larger fruit size and SFW in mid-May than NLT plants in the first year, but the LT plants had reduced the SFW on 5 March and 6 April in the second year. Our first-year results agreed with previous studies that ‘Albion’ had bigger fruit size and weight than other day-neutral cultivars including ‘San Andreas’ (Rowley et al., 2011; Ruan et al., 2011). Berry weight was found to vary in different studies. In HT systems, SFW was 16.4–20.4 g/fruit for ‘Albion’ and 15.9–18.7 g/fruit for ‘San Andreas’ (Martin, 2013; Rowley et al., 2011; Ruan et al., 2011), although ‘Albion’ and ‘San Andreas’ were reported to have light berry (around 8–11 g/fruit) in an HT planted in spring (Gude et al., 2018). In an organic LT system, Petran et al. (2017) reported the weight of ‘Albion’ and ‘San Andreas’ of up to 18.2 and 18.1 g/fruit. Compared to these findings, our results showed higher fruit weight and size for the two cultivars. The difference in production system, season and duration of harvest, yield, and microclimate between previous studies and our study might explain the difference in the fruit size and SFW.

Berry size and SFW increased from March to May, which was opposite to our expectation, as we anticipated reduced fruit size and weight but increased yield as the number of fruits (fruit load) increases after winter.

Field Total Soluble Solids (TSS)

In both years, there were no 2-way or 3-way interactions among P, D, and V for the TSS (Table 6). LT did not affect TSS in both years except for the samples in April 2018, when the fruit of LT had a higher TSS value (8.7) than that of NLT (7.8). Planting date generally did not affect the fruit TSS except on 16 Feb. of the first year, when D2 fruit had higher ºBrix (9.3) than D1 fruit (7.8). ‘Albion’ had higher TSS value than ‘San Andreas’ on 13 April and 1 May in the first year, but there was no difference between the cultivars in the second year (Table 6). The TSS value of ‘Albion’ and its superiority over the ‘San Andreas’ were similar to the previous studies in organically managed HT production (Martin, 2013) and in the field plasticulture system (Petran et al., 2017).

Table 6. Effect of low tunnel, planting dates, and varieties on the total soluble solid (TSS, ºBrix) in 2016/2017 and 2017/2018

	2016–17						2017–18
Treatment	16-Feb	9-Mar	30-Mar	13-Apr	1-May	15-May	5-Mar	23-Mar	6-Apr	5-May
Protection Method (P)
LT^a	8.4	6.9	6.7	7.4	7.1	9.1	6.1	7.5	8.7	9.8
NLT	8.7	8.5	6.6	7.0	7.7	9.0	6.3	7.8	7.8	9.2
Planting Date (D)
D1	7.8	7.3	6.9	7.3	7.4	9.2	6.0	7.4	8.0	n/a^b
D2	9.3	8.1	6.5	7.1	7.3	8.9	6.4	7.9	8.5	n/a
Varieties (V)
Albion	8.6	6.6	6.6	7.5	7.8	9.3	6.3	7.9	8.7	10.5
San Andreas	8.6	8.9	6.8	6.9	6.9	8.8	6.2	7.4	7.9	8.5
Significance
P	Ns	ns	ns	ns	ns	ns	ns	ns	0.0301	ns
D	0.0041	ns	ns	ns	ns	ns	ns	ns	ns	n/a
P*D	ns	ns	ns	ns	ns	ns	ns	ns	ns	n/a
V	ns	ns	ns	0.0235	0.004	ns	ns	ns	ns	ns
P*V	ns	ns	ns	ns	ns	ns	ns	ns	ns	ns
D*V	ns	ns	ns	ns	ns	ns	ns	ns	ns	n/a
P*D*V	ns	ns	ns	ns	ns	ns	ns	ns	ns	n/a

^aLT: low tunnels; NLT: no low tunnels. D1 = 1 September 2016 or 9 September 2017. D2 = 29 September 2016 or 10 October 2017. ns :non-significant (P ≥ 0.05).

^bn/a: not applicable. Fruit samples were collected only from D2 plants on 5 May 2018.

‘Albion’ had lower total yield (261.1 g/plant) than ‘San Andreas’ (411.4 g/plant) during April 2018. This difference in yield may have affected the TSS of fruits as crop load affects fruit properties (Ruan et al., 2013). The TSS value, in general, was higher in February and started to decrease in March and April, and again increased in mid-May. This could be due to the low crop load in February, the increasing crop load in March and April (harvest peaked in April), and again decreased crop load in May.

TSS is closely related to temperatures. Mackenzie et al. (2011) and Wang and Camp (2000) reported that the TSS value decreased as temperatures increased. High temperatures would increase the metabolic activities and reduce the starch content of plant parts as plants use more soluble carbohydrates (fructose, glucose, and sucrose), which account for >65% of TSS in strawberry (Henschel et al., 2017; Wang and Camp, 2000). In the first year, the 2-week temperature averages (high, low, and average) before the date of the TSS test was increasing (Figure 1c,e,g), and the TSS value was decreasing (Figure 1a). In the second year, the 2-week temperature averages before the date of TSS test was decreasing (Figure 1d,e,h), and the TSS value was increasing (Figure 1b). This agrees with the findings of Mackenzie et al. (2011) and Wang and Camp (2000). However, for LT, TSS seems to increase as temperature increased (in the later phase, from 23 Mar. to 6 Apr.) as found in a study on ‘Chandler’ and ‘Sweet Charlie’ by Kadir et al. (2006).

Figure 1. The total soluble solids (ºBrix) and average of maximum, mean, and minimum air temperatures (ºC) in 2017 and 2018

Each temperature data set represents a 2-week temperature average prior to the date specified (when a fruit quality test was conducted). LT: low tunnels inside a high tunnel; NLT: without low tunnels (high tunnels only)

Lab Analysis of Fruit Color, Texture, and Total Titratable Acidity (TTA)

In the first year, there were no 2-way or 3-way interactions among P, D, and V for fruit color, texture, TTA, and TSS/TAA in the February samples (Table 7). LT and planting date did not significantly affect fruit color (L*, a*, and b* values), texture, and TTA, but varieties did. ‘San Andreas’ had higher a* and b* values than ‘Albion,’ which means that ‘San Andreas’ had more reddish-yellow fruit than ‘Albion.’ Because planting dates did not affect any quality parameters in samples from February, we only sampled fruits from D2 later on. For fruit samples collected in April 2017 (D2 plants only), there were no P x V interactions for fruit color (L*, a*, and b*), texture, and TTA (Table 8). LT did not affect color (L*, a*, and b* values) and TTA but produced softer fruit compared to NLT (8.4 N vs 9.9 N). The fruit firmness (texture) was not different between ‘Albion’ and San Andrea’ in 2017 (8.9 vs 9.6 N). The TTA of ‘Albion’ (0.8) was significantly higher than that of ‘San Andreas’ (0.6).

Table 7. Effect of low tunnel, planting dates, and varieties on color, texture, and total titratable acidity (TTA) in February 2017

	Color			Texture	TTA
Treatment	L*	a*	b*	(N)	(%, w/w)	TSS/TTA
Protection Method (P)
LT^a	43.1	14.7	6.2	11.2	1.1	9.4
NLT	43.0	14.0	5.6	12.0	0.9	11.3
Planting Date (D)
D1	43.2	14.4	6.0	12.2	1.0	9.5
D2	42.9	14.3	5.8	11.0	1.0	11.2
Varieties (V)
Albion	42.8	13.0	5.3	11.5	1.0	10.8
San Andreas	43.4	15.7	6.6	11.7	1.0	9.9
Significance
P	ns	ns	ns	ns	ns	ns
D	ns	ns	ns	ns	ns	ns
P*D	ns	ns	ns	ns	ns	ns
V	ns	<0.0001	0.0007	ns	ns	ns
P*V	ns	ns	ns	ns	ns	ns
D*V	ns	ns	ns	ns	ns	ns
P*D*V	ns	ns	ns	ns	ns	ns

^aLT = Low tunnels, NLT = No Low tunnels, D1 = 9 September, D2 = 10 October, ns = Non-significant (P ≥ 0.05).

Table 8. Effect of low tunnel and varieties on color, texture, and total titratable acidity (TTA) in April 2017 and May 2018

	April 2016						May 2017
		Color		Texture	TTA	TSS/TAA		Color		Texture	TTA	TSS/TAA
Treatment	L*	a*	b*	(N)	(%, w/w)		L*	a*	b*	(N)	(%, w/w)
Protection Method (P)
LT^a	43.6	13.9	5.1	8.4	0.7	9.0	42.7	12.1	3.7	8.6	0.9	11.3
NLT	43.9	13.8	5.6	9.9	0.7	8.7	42.5	12.1	3.6	8.9	0.8	10.8
P-value	ns	ns	ns	0.0165	ns	ns	ns	ns	ns	ns	ns	ns
Varieties (V)
Albion	43.3	11.1	4.1	8.7	0.8	9.2	42.5	10.6	3.1	9.9	0.9	11.7
San Andreas	44.1	16.6	6.6	9.6	0.6	8.5	42.8	13.6	4.2	7.5	0.8	10.3
P-value	0.0097	<0.0001	0.0002	ns	0.0149	ns	ns	0.001	0.0016	0.002	0.0331	0.0101
significance
P*V	ns	ns	ns	ns	ns	ns	ns	ns	ns	ns	ns	ns

^aLT = Low tunnels, NLT = No Low tunnels. ns = Non-significant (P ≥ 0.05).

In the second year, there was also no P x V interactions for fruit color, texture, TTA, and TSS/TTA. LT did not affect fruit color, as the L*, a*, and b* values were similar between fruit from LT and NLT. ‘San Andreas’ had brighter, more reddish-yellow fruit than ‘Albion,’ but ‘Albion’ had higher TTA than ‘San Andreas’ in both years (Table 8). There was a significant difference in fruit firmness (texture) between ‘Albion’ (9.9 N) and ‘San Andreas’ (7.5 N), which is opposite to the findings by Gude et al. (2018).

The TSS/TTA ratio is a very important parameter in evaluating fruit quality because it determines fruit flavor harmony (Voca et al., 2008). Lower TSS/TTA ratio usually corresponded to stronger sour taste (Ikegaya et al., 2019). Table 7 shows that the strawberry planted under NLT or at D2 had a numerically higher TSS/TTA than those under LT or at D1, although not statistically significant. Table 8 shows that the protection method did not affect the TSS/TTA ratio. Regardless of the protection method and planting date, ‘Albion’ berry had a higher TSS/TTA ratio (Tables 7 and 8) and there is a significant difference in year 2 (Table 8). These results indicate that the variety, not planting date and protection method, determines the sensory quality, especially, flavor of strawberry. The effect of planting date and protection method on fruit flavor is limited.

This superiority of ‘San Andreas’ in color and of ‘Albion’ in TTA was similar to the findings of previous studies (Martin, 2013; Ruan et al., 2011; Tudor et al., 2014). However, the L* value was higher, and the a* and b* values were lower in our study compared to the values in a previous study by Gude (2016). The textural difference between cultivars was not consistent in our study, which agrees with the results of the study by Ruan et al. (2013). Changes in light intensity, plant growth condition, and crop load over the growing period (February to April) may have affected the fruit quality (Gunduz and Ozdemir, 2014; Josuttis et al., 2011; Martin, 2013).

Conclusion

In our study, planting dates and LT did not significantly affect the fruit quality (size, weight, TSS, TTA, color, and firmness) of the two DN strawberry cultivars inside HT. Plant genotype (cultivars) contributed more to fruit quality. ‘Albion’ had higher TSS, TTA, firmness, and larger and heavier fruit than that of ‘San Andreas.’ This organoleptic property is important as higher TSS combined with higher TSS/TTA ratio results in better taste and flavor that attract consumers. On the other hand, ‘San Andreas’ had a brighter and more reddish color than ‘Albion.’ From the consumers’ perspective, the first impression of fruit is color and size, so ‘San Andreas’ could be consumers’ first pick in a market. However, from the growers’ perspective, taste along with fruit size and yield play a decisive role when choosing a variety. In our study, the yield difference between the two cultivars was not consistent (Rana and Gu, 2020). We suggest growers plant both varieties for diversified product and better quality for fresh markets. This will attract new consumers by presenting good-looking strawberry fruit from ‘San Andreas’ and at the same time draw consumers back by providing better taste and flavored fruit from ‘Albion.’ Future research should include more DN varieties and examine the content of fruit health-promoting chemicals such as the levels of antioxidants, with more planting dates and different color films for LT.

Acknowledgments

We thank John E. Beck, John E. Kimes, Ivy Smith, and other staff from the Extension Horticulture Unit for their help with crop management and fruit quality analysis.

Additional information

Funding

This project was supported by the USDA NIFA Evans-Allen project [NC.X-301-5-16-130-1] and OREI project [2015-51300-24134].