Plant material and treatments.
The experiments were performed in July to September 2020, at the Zhuanghang (ZH) experimental vineyard, Shanghai Academy of Agricultural Sciences, Shanghai, China (30°51′N, 121°13′E) and at Shanghai Pingqi (PQ) Grape Planting Professional Cooperative (30°11′N, 121°25′E), using the table grape cultivar ‘Jumeigui’. The study site has a subtropical monsoon climate. Due to the high rainfall levels in this area, grapes are planted under plastic film coverings to reduce the incidence of plant disease and insect pests. Facility agriculture (设施农业 or shèshī nóngyè in Chinese) is a production method that uses clustered plastic-roofed greenhouses and irrigation to produce crops10. The own-rooted grapevines were planted in the spring of 2008 in a north–south orientation at a 4 m × 2.8 m spacing. The vines were grown in a rain shelter with a Y-shaped training system at ZH (8 m ⋅ 45 m) and a flat scaffolding system at PQ (6 m ⋅ 41 m). The vines were managed according to standard viticulture and disease control practices.
Blue, green, and black shade nets and aluminum foil gray shade nets were purchased from an agricultural market (Jiehang Agricultural Materials Factory, Yangzhou, Jiangsu, China). The colored shade nets (6 m ⋅ 41 m at PQ, 8 m ⋅ 45 m at ZH) were used to cover the single-shed facilities. The experiment adopted a random block design and was repeated three times. For the control, no shade net was used. The processing period was from July 24, 2020 (50% color change) to September 3, 2020. Fruit harvesting occurred on August 18, 2020, and leaf-related indicators were measured on September 3, 2020. The grapes were judged to be mature when the total soluble solid content (TSS) was higher than 16 Brix. The grape harvest period occurred when the grapes were mature, and the leaf photosynthetic performance measurement period occurred after the grape harvest. RC-4HC (Jiangsu Jingchuang Electric Co., Ltd., Nanjing, China), which placed near the berries, was used to record the temperature. All methods were performed in accordance with the relevant guidelines and regulations.
Spectral analysis.
A PMS-2000 UV-VIS-near IR spectrophotocolorimeter (Everfine Co., Ltd., Hangzhou, China) was used to analyze the physical properties of the transmitted light. Spectral data analysis of transmitted light was performed under different color shading nets, and the analysis wavelength range used was 300-800nm.
Total soluble solids (TSS) measurement.
For each treatment, 30 grape berries from the top, middle, and bottom of ten bunches were prepared, and the TSS were determined after pressing the juice separately. TSS was measured in degrees Brix using a PAL-1 digital refractometer (Atago, Tokyo, Japan).
Fast chlorophyll a fluorescence kinetic parameters.
A Pocket-PEA fluorimeter (Plant Efficiency Analyzer, Hansatech Instruments Ltd., King’s Lynn Norfolk, UK) was used to determine the fast chlorophyll a fluorescence kinetics of the grape leave11. The definitions of parameters are shown in Table S1.
Leaf gas-exchange parameters.
Photosynthetic measurements were performed by selecting five old lower leaves near the base and five adult upper leaves near the top. The gas-exchange parameters comprised the net photosynthetic rate (A), stomatal conductance (gs), transpiration rate (E), and intercellular CO2 concentration (Ci) of the leaves, which were measured using a portable photosynthetic CIRAS-3 system (PP Systems, Amesbury, MA, USA)12.
Sugar analysis.
For this analysis, ten berries were selected from the top, middle, and bottom of three bunches and mixed. The pulps were ground with liquid nitrogen prior to further use. Each treatment was replicated three times. The extraction of soluble sugars was performed and determined via high-performance liquid chromatography, using the Waters E2695 system (Waters, Milford, MA, USA), as described by Zha et al.11.
Anthocyanin content analysis.
For this analysis, ten berries were selected from the top, middle, and bottom of three bunches and mixed. The skins were ground with liquid nitrogen prior to further use. Each treatment was replicated three times. Anthocyanin content was determined using the pH-differential method13,14.
Classifying symptoms of abnormal softening.
We examined the abnormal softening of the grapes in the field, and developed a grading standard where we investigated 30 clusters per treatment and estimated the softening percentage using specific equations. Softening was graded according to the following scale: level 0, no abnormal softening symptoms (N1); level 1, abnormal softening up to 10% (N2); level 2, abnormal softening up to 30% (N3); level 3, abnormal softening up to 50% (N4); level 4, abnormal softening up to 70% (N5); and level 5, abnormal softening up to 100% (N6).
Softening index = (0 ⋅ N1 + 1 ⋅ N2 + 2 ⋅ N3 + 3 ⋅ N4 + 4⋅ N5 + 5⋅ N6) / [5⋅ (N1 + N2 + N3 + N4 + N5 + N6)]
0 to 5 refers to the plant replicate number.
Grape texture analysis.
For each treatment, 30 grape berries from the top, middle, and bottom of ten bunches were prepared, and the berry texture was analyzed using a TA.XT.Plus type physical property tester (Stable Micro System, Godalming, UK). Fruit firmness was expressed as the force (N) required to deform the berries, according to Lijavetzky et al.15 (2012).
Berry skin color.
For each treatment, 30 grape berries from the top, middle, and bottom of ten bunches were prepared, and berry color was analyzed using a hand-held C410 Chroma Meter (Konica Minolta, Chiyoda-ku, Tokyo, Japan) at the equatorial portion of each berry. The Color Index of Red Grapes (CIRG), which is based on CIELab data, was calculated using the formula CIRG = (180 − h) / (L* + C*)11. The index used for the evaluation of the appearance and color of the berry was: CIRG < 2 represents yellow-green, 2 < CIRG < 4 pink, 4 < CIRG < 5 red, 5 < CIRG < 6 deep red, and CIRG > 6 blue-black16 (Amiri et al., 2010).
Statistical analysis.
The TSS and CIRG data was tested to confirm that they were normally distributed. These data were compared through two dimensions, average and distribution. The frequency distributions were generated using the function “NORMDIST” in Microsoft Excel 2010 (Redmond, WA, USA). The differences among treatments were assessed using a one-way ANOVA followed by the Tukey-test in SPSS v22.0 (IBM Corp., Armonk, NY, USA).