Effects of different rain shelter coverings on volatile organic compounds in mature fruit and postharvest quality of sweet cherry

ABSTRACT This study investigated the effects of different shelter coverings (arched-shelter and umbrella-shelter) on the volatile components of sweet cherry mature fruits in rainy areas, and the postharvest quality of these fruits during storage at room temperature. A total of 68 volatile compounds were identified and semiquantified in mature fruits. Aldehyde compounds were the most abundant, followed by alcohols. Benzyl alcohol, acetaldehyde, 2-methyl-propanal, hexanal, (E)-2-hexenal and benzaldehyde were the major volatiles, and the proportions of these compounds were greatly affected by different shelter coverings. With the extension of the storage period, the color parameters (L*, a* and b*), texture parameters (hardness, springiness, chewiness, resilience, skin strength and flesh firmness) and titratable acidity of the fruit rapidly decreased. The total soluble solid content and weight loss of the fruit gradually increased. Principal component analysis indicated that the sweet cherry fruits from the arched-shelter had the best quality and the longest shelf life.


Introduction
Sweet cherry (Prunus avium L.) is native to the European Black Sea coast and western Asia. Sweet cherry is popular among Chinese consumers due to its early maturity, beautiful color, good taste and rich nutrition (Bal, 2012). However, the introduced sweet cherries often crack and rot due to the wet and rainy conditions in southwestern China, and these effects severely reduce the quality and yield of the fruit . Therefore, rain shelter application became an effective means to avoid sweet cherry fruit cracking and rot in areas with frequent precipitation (Blanke & Balmer, 2008;Sotiropoulos et al., 2014). In addition, rain shelter cultivation may effectively protect trees from attack by infectious diseases (Børve & Stensvand, 2007), thereby improving the quality, yield, and commercial value of the fruit. According to previous research, covering shelters can also reduce fruit quality during ripening on tree when trees are exposed to an excessive heat before or during bloom leading to floral malformations. This excessive heat can be produced or increased by the covering shelters reducing fruit firmness and TSS content (Blanke & Balmer, 2008). However, the studies of the effects of rain shelter covering on the volatile organic compounds and storability of sweet cherry after harvest have been very limited thus far.
The aroma, color, texture, total soluble solids (TSS), titratable acidity (TA) and weight of the fruit are the most important factors that contribute to sweet cherry fruit quality according to consumers. Aroma is one of the fruit characteristics that attracts the greatest attention and is usually composed of volatile flavor compounds that determine consumer acceptance of the product (El Hadi et al., 2013). As is well known (Hayaloglu & Demir, 2016), the aroma of fruits is the result of a complex mixture of esters, alcohols, aldehydes, organic acids, ketones, and terpenoids, among other compounds. Volatiles can be detected and analyzed by headspace solid-phase microextraction (SPME) (Felipe et al., 2011;Negri et al., 2015). Compared with other technologies, SPME is a simple, efficient and more realistic method to determine the basic composition of volatiles. SPME-GC/MS has been widely used in studies of the aroma components of flowers and fruits and has achieved phased results in the determination of the aroma components of different fruits (Vavoura et al., 2015). Color is derived from the natural pigments in fruits and vegetables, many of which change as the plant matures and ages (Crisosto et al., 2003). Sweet cherry fruit skin colors vary widely due to differences in red and yellow pigment profiles. Sweet cherries contain substantial amounts of anthocyanins (red and blue) and polyphenols (Franceschinis et al., 2015), which are not uniformly distributed in fruit tissue. For intensely red cherry cultivars, such as Bing or Lapins, the anthocyanins and polyphenols are mainly concentrated in the skin, although they are present in both the skin and the flesh (Chaovanalikit & Wrolstad, 2004). In addition, the color of sweet cherry fruit is also related to the TSS content (Yommi et al., 2012). The TSS is an important factor influencing the quality of sweet cherry fruits because its sugar, organic acids and vitamins are important nutrients in the fruits (Vavoura et al., 2015). Generally, a higher content of soluble solids is more popular among consumers.
Texture is not only an important indicator used to evaluate the internal quality of sweet cherry fruits but also a key factor used to determine the quality of storage and transportation (Yang et al., 2007). During the preharvest ripening process and the postharvest storage stage, the fruit texture, including hardness, springiness and cohesiveness, changes, and this change is regulated by genes and environmental conditions (Hajnal et al., 2012;Sirisomboon & Lapchareonsuk, 2012). Texture analysis technology is increasingly becoming a research hotspot in the evaluation and analysis of fruit quality postharvest storage, processing and shelf life (Fabrizio et al., 2011;Tiago et al., 2016). The available data have demonstrated that different sweet cherry cultivars have different fruit development and softening mechanisms, resulting in significant differences in fruit taste, flesh quality and other characteristics pre-and postharvest; in addition, there are also significant differences in fruit texture quality (Lufu et al., 2020;Vavoura et al., 2015). However, the differences and changes in the texture of sweet cherry fruits during postharvest storage under different types of rain shelter covering have not been reported thus far.
The objectives of the present research were to identify the volatile organic compounds in mature sweet cherry fruit under different rain shelter coverings and to investigate the changes in fruit skin color, texture, TSS and weight loss during storage at room temperature to provide a reference for the reasonable use of rain shelter coverings and the improvement of fruit quality after harvest.

Fruit sample preparation
Sweet cherry fruits were sampled from 6-year-old trees (cv. 'Summit' grafted on 'Gisela 6ʹ root stock) grown under two types of shelters as well as from control trees (shelter-free) in the Orchard Basement attached to the Key Laboratory of Plant Resource Conservation and Germplasm Innovation in the Mountainous Region (Ministry of Education), Wudang District, Guiyang City, P. R. China (27°03′3.89″N and 106°25′ 47.23″E). The two types of shelters were a permanently mounted 4-span arched-shelter and a single umbrellashelter, and the covering was a colorless polyvinyl chloride (PVC) anti-aging plastic film. The length, width and height of the arched-shelter were 30 m, 8 m, and 5.3 m (above ground), respectively. The umbrella-shelters had a radius and height of 1.9 m and 3 m (above the ground), respectively, and covered only one tree. In 22 May 2020, fruits with 80-90% maturity were sampled during the commercial harvest period based on subjective evaluation of fruit color, from 9 different trees in each treatment as three replicates (three trees per replicate), after which they were immediately placed in thermal bags, cooled and transported to the laboratory. The fresh sweet cherry fruits immediately after harvest were used to determine the volatiles. Six kilograms of fruit per replicate, with uniform size and bright color that were free of pests, diseases and mechanical damage, were selected and then incubated at a room temperature of 25 ± 1°C and 70 ± 5% relative humidity to observe and record changes in the fruit skin color, texture characteristics, TSS contents, TA content, and weight loss (%) at 0, 4, 8, 12, and 16 days of storage, respectively.

Determination of volatiles
The research process for determining fruit aroma compounds mainly includes five aspects: sample collection and preparation, sample preprocessing, data collection, data processing and analysis, and biological interpretation. Two grams of sample was added into a head space bottle (20 mL) and sealed. Volatiles were isolated by solid-phase microextraction (SPME) (50°C), which required oscillation for 15 min and extraction for 30 min (250 rpm) using the CTC triad automatic sampler (extractor head: 50/30 livm DVB/ CAR on PDMS and the length of fiber is 1 cm). An Agilent 7890B series gas chromatograph (GC) equipped with an Agilent 5977B mass spectrometer (MS) was used for the analysis of volatiles adsorbed onto the SPME fiber. The extractor head is desorb at the injection port for 2 min and the desorption temperature is 250°C. GC was performed on a DB-wax (30 m × 0.25 mm × 0.25 m) to separate the derivatives with helium at a constant flow of 1 mL·min −1 . The injection temperature was 260°C. The temperatures of the column and ion source were 40°C and 230°C, respectively. The program for temperature increase involved an initial temperature of 5°C for 5 min, with an increase of 20°C·min −1 up to 250°C, where the temperature remained for 2.5 min. The chromatogram obtained was analysed, and each peak was checked by determining the percent area on the chromatogram, the retention time, the spectrum and the base peak and then referring to the characteristic mass spectra of compounds listed on the MS library (Topi, 2020). MS (EI + , 70 ev) was performed by the full-scan method with a range from 20 to 400 m·z −1 . Identification of volatile compounds was achieved by matching mass spectra with those of chemical standards and commercial libraries.

Assessment of skin color
The color of the skins of 30 fruits per replicate was measured by the L*, a*, b* chroma space mode with a multifunction colorimeter (NR60CP), Shenzhen 3nh Technology Co., Ltd. (Shenzhen, P. R. China), and an area of 8 mm 2 was measured. In the CIE L*, a*, b* uniform color space, L* indicates brightness, a* indicates chromaticity from red (+) to green (-), and b* indicates chromaticity from yellow (+) to blue (-).

Measurement of texture properties
The texture characteristics of 60 fruits per replicate were measured by a Rapid TA texture analyzer (Shanghai Teng Da Instrument Technology Co., Ltd., Shanghai, China). A nondamaged and nonrotted fruit was placed on the test platform equipped with a texture analyzer and a P/36 R cylindrical probe for texture profile analysis (TPA) tests. The program of the TPA measurement mode was set to a pretest speed of 1 mm·s −1 , a test speed of 1 mm·s −1 , and a posttest speed of 1 mm·s −1 . The compression level was 15% of the original height, the two-compression pause time was 4 s, and the trigger force was 5 g. The parameters obtained from this test result were hardness, springiness, chewiness and resilience (Contador et al., 2016). The puncture test without two compressions used a P/2E 2-mm needle probe to perform a single puncture on the cheek area of the fruit. The test program was set to puncture test with a penetration depth of 3 mm, and other parameters were the same as for the TPA test. The parameters of skin strength and flesh firmness were measured (Shimomura et al., 2016).

Measurement of total soluble solids, titratable acidity and weight loss
After evaluating the texture characteristics, the fruits were used to measure the TSS content with an HT113AT handheld digital refractometer (Shenzhen, Guangdong, P. R. China), and the results are expressed in percentage (%). The sample of 20 mL cherry juice was titrated with 0.1 mol·L −1 NaOH to pH 8.2 to determine TA, and the result was expressed in percentage (%) of malic acid. A total of 60 fruits per replicate were weighed by a balance with a precision of 0.0001 g, and weight loss is expressed as the percentage loss relative to the initial weight (Zhao et al., 2019).

Statistical analyses
Preliminary statistical analyses of all the experimental data were performed using Microsoft Excel 2010 software, and the data were further analyzed using SPSS Statistics Version 21.0 software (Chicago, IL, USA). Analysis of variance (ANOVA) was carried out under different rain shelter and shelter-free conditions, and the differences between the means were compared by Duncan's multiple comparisons test (p < .05). The data are presented as the mean values (means) ± standard deviations (SDs) of three replicates. The data were processed according to principal component analysis (PCA) to evaluate the comprehensive scores of the fruit samples under different rain shelter and shelter-free conditions throughout the storage period (Liang et al., 2015).

Identification of volatiles in sweet cherry fruit
The overall sweet cherry flavor is a combination of aroma and taste sensations including sugars and organic acids. Sweet cherry volatiles have been studied intensively, and alcohols, aldehydes, esters, terpenoids, ketones and lactones have been described to date (Hayaloglu & Demir, 2016;Sun et al., 2010;Wen et al., 2014). However, the fruit variety, ripeness, pre-and postharvest environmental conditions and analysis methods used all affect the volatile characteristics of sweet cherries (El Hadi et al., 2013). In this study, SPME/GC-MS was used to determine the volatile components of the mature stages of sheltered and shelter-free fruits. The results are shown in Table 1, and the total ion chromatograms of the sweet cherry fruits are shown in Figure S1. In the present study, a total of 68 volatiles (Table 1) in 10 chemical groups (i.e., acid, alcohols, aldehydes, alkane, benzol, esters, ether, ketones, olefin, and phenol) were separated and identified in the sweet cherry Summit cultivars under arched-shelter, umbrella-shelter and shelter-free conditions. Among these compounds, aldehydes had the largest numbers, with 12, 13, and 16 compounds in the arched-shelter, umbrella-shelter and shelter-free conditions, respectively, with relative contents accounting for 70.29%, 76.53% and 79.12%, of the total detected volatiles, respectively. The next most abundant were alcohols, with 12 compounds in the arched-shelter, umbrella-shelter and shelter-free conditions, and the relative contents accounted for 24.31%, 20.43% and 12.57%, respectively. Other categories had fewer compounds. From the aroma abundance data of some volatiles, it was noted that benzyl alcohol, acetaldehyde, 2-methyl-propanal, hexanal, (E)-2-hexenal and benzaldehyde were the main aroma substances for the Summit cultivar. The highest content among the aroma substances in the arched-shelter, umbrella-shelter and shelter-free conditions was (E)-2-Hexenal, at 30.68%, 31.17% and 34.71% abundance, respectively. Hexanal followed, at 14.74%, 17.18% and 16.31%, respectively. The relative contents of benzaldehyde were also higher, at 10.1%, 11.58% and 6.38%, respectively. Thus, the proportions of the same aroma compounds in the sweet cherries in the arched-shelter, umbrella-shelter and shelter-free conditions were quite different. These results may be due to the differences in cultivation type and/or maturation level of the fruits. According to Selli et al. (2012), aroma compounds in Dwarf Cavendish banana grown from open-field and protected cultivation area were different. At the same time, Xi et al. (2016) found that the composition and content of aroma volatiles in apricot fruits presented different patterns during development and ripening.

Changes in fruit skin color during storage
Color is one of the most important indicators for evaluating the quality of fresh sweet cherries; it is mainly related to the content and composition of anthocyanins (Bernalte et al., 2003;Kayesh et al., 2013), which range from a few mg to approximately 700 mg per 100 g (bright color to dark cherries). As shown in Table 2, with the extension of storage, the skin chromaticity parameters L*, a*, and b* of sweet cherries fruit all decreased, indicating that the skin color of the fruit gradually darkened, and the red and yellow gradually decreased. Previous studies have reported that the skin color parameters L*, a*, b*, chroma, and hue angle of sweet cherry fruits showed downward trends during postharvest storage (Esti et al., 2002;Sharma et al., 2010;Zhang et al., 2020). In the early stage of storage, the skin color of shelter-free fruit was the brightest (largest L* value) and this fruit was redder and yellower (higher a* and b* value) than those under the arched-shelter and umbrella-shelter conditions. Differences in cultivation methods and varieties will affect the fruit quality of sweet cherries. According to Børve and Stensvand (2007), the skin color of uncovered sweet cherry fruits was brighter but less red than that the fruit under cover shields. In addition, Zadravec et al. (2009) found that the color parameter L* was significantly increased but the a* and b* values were significantly reduced by covering sweet cherry 'Hedelfinger'. However, during the entire storage, the rate of decline in the skin chromaticity parameters of shelterfree fruit was the highest, among which the rates of decline of L*, a*, and b* values were 25.28%, 93.01%, and 95.45%, respectively, while the rates of decline of arched-shelter fruits were 6.82%, 90.67%, and 91.62%, respectively, which were the lowest. Therefore, on the 16 th day of the storage, the skin chromaticity parameters L*, a*, and b* of archedshelter fruits were significantly higher than those of shelterfree fruits, but there was no significant difference in the skin color of umbrella-shelter fruits. These results showed that sweet cherry fruits under arched-shelter conditions could better maintain their skin color during storage at room temperature and had better storability.

Changes in fruit TPA parameters during storage
Hardness is one of the important indicators used to assess the texture, to reflect the storability of the fruit and to measure the effect of storage (Yang et al., 2007). The force required for the fruit to deform under the action of external force is called hardness. Fruits with high hardness are strong during storage and transportation. During storage at room temperature, the fruit hardness of sweet cherries showed an overall downward trend (Figure 1(a)). In addition, the hardness of the arched-shelter fruits was always the highest during storage, followed by umbrella-shelter and shelterfree fruits; there was a significant difference between the arched-shelter and shelter-free fruits, but the hardness values of umbrella-shelter and shelter-free fruits were significantly different only on the 12 th day. According to Cline et al. (1995), the fruit hardness of sweet cherry on Colt and F.12/1 rootstock covered by plastic rain shelters was significantly higher than that of uncovered plants, which was similar to our research results. On the 16 th day during storage, the fruit hardness reduction rates in the archedshelter, umbrella-shelter and shelter-free conditions were 57.86%, 58.62%, and 61.56%, respectively. These results showed that the storage and transportation performance of the sweet cherry fruits in the arched-shelter condition were the largest, followed by the umbrella-shelter and shelter-free conditions. Springiness refers to the ability of the fruit to recover when the pressure is removed, and springiness is determined by the extensible and somewhat elastic cell walls in the fruit (Singh et al., 2013). The fruit springiness of sweet cherry gradually decreased with the extension of the storage period (Figure 1(b)). On the 0 th day of storage, the fruit in the shelter-free condition had the highest springiness, followed by the fruit in the arched-shelter and umbrella-shelter conditions, and there were significant differences between these groups. The springiness of the shelter-free fruit decreased rapidly during storage; however, the arched-shelter fruit still maintained high springiness on the 4 th day, and its springiness on the 12 th day was equivalent to that of the shelter-free fruit on the 8 th day. On the 16 th day of storage, the springiness of fruit under the arched-shelter condition was the largest, followed by the umbrella-shelter and shelter-free conditions, and there were significant differences between them.
Chewiness is the energy required to simulate the chewing of food by teeth and comprehensively reflects the resistance of the fruit to chewing. The chewiness of the sweet cherry fruits showed a downward trend throughout storage ( Figure  1(c)), and the decline was greater in the early stage of storage.
The chewiness of arched-shelter fruit in the early storage period was significantly higher than that of umbrellasheltered and shelter-free fruits. In the late storage period, the chewiness levels of arched-shelter and umbrella-shelter fruits were significantly higher than that of shelter-free fruit.
The resilience of the fruit reflects the ability of the fruit to quickly recover from deformation after being compressed. If the fruit tissue is greatly damaged, the resilience tends to zero. As shown in Figure 1(d), the resilience of sweet cherry fruits showed an overall downward trend during storage, and there were certain fluctuations in the middle and late stages. The fruit resilience of arched-shelter sweet cherry fruits during the entire storage period was significantly higher than that of shelter-free fruits and was significantly higher than that of the umbrella-shelter fruits on the 0 th, 4 th, and 16 th days of storage. The fruit resilience under the umbrella-shelter condition was significantly higher than that of the arched-shelter and shelter-free conditions only on the 8 th day of storage.
The parameters of fruit TPA, such as hardness, springiness, chewiness, and resilience, can reflect the tactile sensation of the human mouth to the pulp and can reflect the density and firmness of the fruit (Guiné et al., 2011;Sams, 1999). This study found that the above TPA detection parameters exhibited certain differences between the sweet  cherry fruits under the arched-shelter, umbrella-shelter and shelter-free conditions, and they all decreased to varying degrees during storage. TPA parameters of sweet cherry fruits appears with higher values when the arched-shelter and umbrella-shelter are used. It is speculated that rain shelters might prevents the leaves and fruits from frequent contact with rainwater and reduce the water content in the soil when there is an abundant of rainwater during the fruit development stage (Jin et al., 2018). However, The texture of fruits and vegetables is mainly determined by the moisture content and tissue structure characteristics (Lahaye et al., 2018). The hardness, springiness, chewiness and resilience parameters of apple and Chinese bayberry also showed downward trends during storage, which was consistent with our research results (Pan & Tu, 2005;Yang et al., 2007). However, the research results of loquat fruit were contrary to ours because the fruit's hardness, adhesiveness, cohesiveness, springiness, chewiness and resilience all increased with the storage time (Song et al., 2010). This difference may be due to the lignification of the pulp tissue of loquat during storage, which makes its texture parameters show different trends. The decrease in fruit hardness after harvest is mainly due to the hydrolysis of pectin and cell wall substances in the middle gum layer caused by the action of cell wall enzymes (Payasi et al., 2009). In addition, during the ripening and softening processes, the fruit cell wall pectin gradually degrades, which causes the integrity of the cell wall structure to be damaged, and the fruit texture appears to soften (Vicente et al., 2007). Compared with the umbrellashelter and shelter-free fruits, the arched-shelter sweet cherries had better texture characteristics during the harvest period and still had highest texture parameters on the 16 th day of the storage, indicating that they had better storage characteristics. Bugaud et al. (2007) reported that the main geographical factors affecting the texture of ripe bananas were climatic factors, that is, the daily temperature and accumulated rainfall during the growth of banana bunches. The loss of texture quality is not only related to the aging process but also to water loss, reduced satiety and wound healing, including leakage of osmotic solutes. The differences and changes in texture are related to changes in cell size, intercellular adhesion, starch/sugar conversion, water loss, cell wall composition and cell wall strength (Toivonen & Brummell, 2008).

Changes in fruit puncture test parameters during storage
The puncture test can relatively accurately measure the local tissue characteristics of the fruit, especially the characteristics of the fruit skin (Camps et al., 2010). As shown in Figure 2(a), the skin strength of sweet cherry fruits under the three conditions all showed downward trends with the extension of storage, which were consistent with the results of other researchers' experiments on apples (Ling et al., 2014). In addition, during the entire storage period, the skin strength of arched-shelter fruit was always the highest and was significantly different from that of the shelter-free fruit. However, the fruit skin strength of umbrella-shelter fruit was significantly higher than that of shelter-free fruit on the 0 th, 4 th, and 16 th days of storage. Early in the storage period, the skin strength of the arched-shelter fruit decreased rapidly but then decreased slowly in the later period. The skin strength of the umbrella-shelter fruit decreased slowly during the first 4 days of the storage, then accelerated, but then decreased slowly after the 8 th day. The skin strength of shelterfree fruits decreased rapidly in the early stage of storage, decreased slowly in the middle stage, but began to decline more rapidly in the later stage.
The flesh firmness determined by the puncture test is not affected by the shape and size of the fruit (Shiu et al., 2015). The flesh firmness of the sweet cherry fruit was similar to the change in fruit hardness measured by TPA during storage, and both showed downward trends (De et al., 2016). However, the declines in different periods were not exactly the same, which may be because the shape and size of the fruit affected the TPA measurement results (Alvarez et al., 2002). The flesh firmness of the arched-shelter and umbrellashelter fruits decreased rapidly in the first 4 days of storage and then decreased slowly. The flesh firmness of shelter-free fruits decreased rapidly in the first 8 days of storage and then slowed down slightly on the 12 th day, but the decline accelerated again on the 16 th day. Therefore, on the 16 th day of the storage, the decrease rates of the flesh Figure 2. Changes in puncture test parameters skin strength (a) and flesh firmness (b) of sweet cherry fruit during storage. The results presented are the means ± SDs of three replicates. Different letters (on the same day) indicate significant differences (p < .05).

Changes in fruit TSS, TA and weight loss during storage
TSS is an important indicator of the nutritional quality of sweet cherry fruits. From Figure 3(a), it can be seen that the TSS content of sweet cherry fruits showed an increasing trend during storage, which is similar to the trend of the change in the TSS contents of three sweet cherry varieties ("Van", "Bing" and "0900 Ziraat") during the shelf life, as reported by Unal and Akbudak (2008). This may be due to the conversion of starch to soluble sugars, the hydrolysis of cell wall polysaccharides (Comabella & Lara, 2013), and water loss leading to an increase in dry matter (Unal & Akbudak, 2008). Temperature, atmospheric conditions and fruit ripening stage will all affect the TSS ratio. During the entire storage period, the TSS of arched-shelter fruit was significantly higher than that of shelter-free fruit and higher than that of umbrella-shelter fruit, with significant differences on the 0 th and 12 th days of storage. The TSS of sweet cherry fruits under umbrella-shelter conditions was significantly higher than that of fruits under shelter-free conditions from 8 to 16 days of storage, indicating that the sweet cherry fruits under arched-shelter and umbrella-shelter conditions had higher TSS content and better nutritional quality than fruits under shelter-free conditions. In addition, the fruits can maintain a higher TSS content during storage. Research by Børve and Stensvand (2007) found that the sugar content (soluble solids) of sweet cherry fruits covered by rain shields was significantly higher (p = .01) than that of uncovered fruits in the Sekse 98 orchard, which was similar to our results. The increase of TSS of fruit covered by rain shelters may again be related to the water relations of the tree and lower supply of moisture to the fruit, similar to the speculation of TPA increase (Jin et al., 2018).
The organic acid content of fresh fruit is usually evaluated by TA. The TA of umbrella-shelter and shelter-free fruits decreased even more with increasing storage time, which were 22.15% and 26.01%, respectively, while the archedshelter fruits only decreased by 19.49% (Figure 3(b)). The reduction of fruit acid usually shortens the potential storage or shipping life of sweet cherries, and organic acid can be used as the carbon source of the tri-carboxylic cycle as the main part of the respiration process (Wang et al., 2014). Above results indicated the arched-shelter fruit maintains the TA content very well during storage.
Water content affects the freshness and taste of the fruit, and weight loss is an important indicator that affects the quality of fruit storage. As shown in Figure 3(c), with the prolongation of storage time, the fruit water loss, i.e., the weight loss of sweet cherry fruits, gradually increased. This may be due to the respiration and transpiration of the fruit  during storage, which transfers the moisture of the fruit to the surrounding air (Yaman & Bayoιndιrlι, 2002;Zhao et al., 2019). The sweet cherry fruits under arched-shelter conditions had less weight loss than fruits under umbrella-shelter and shelter-free conditions and were significantly different from the fruits under shelter-free conditions. In addition, the weight loss levels of umbrella-shelter fruits on the 4 th, 12 th, and 16 th days of the storage were significantly lower than those of the shelter-free fruits. On the 16 th day of the storage, the weight loss of the shelter-free fruits was 24.12%, but the weight loss of arched-shelter and umbrellashelter fruits was significantly lower than that of the shelterfree, at 20.49% and 21.86%, respectively. This showed that compared with shelter-free fruits, arched-shelter fruits and umbrella-shelter fruits had less weight loss during storage and had better storage characteristics.

Principal component analysis for fruit quality
PCA is a statistical analysis method that uses linear changes to simplify multiple variables into a small number of comprehensive variables by examining the correlations between multiple variables, and PCA is widely used in biological trait analysis and product quality analysis (Alizadeh Behbahani et al., 2017;Kallithraka et al., 2015). The PCA of 11 indicators of sweet cherry fruit quality showed that there were two principal components (PC1 and PC2) with characteristic values greater than 1, and their cumulative variance contribution rate reached 94.370% (Table 3). Therefore, these components could represent the information characteristics of the overall data. As shown in Table 3 and Figure 4, the variance contribution rate of the PC1 was 85.614%, which was mainly determined by these indexes with higher absolute load values, such as skin strength, hardness, flesh firmness, chewiness, resilience, springiness, a*, b*, TA and weight loss; the variance contribution rate of PC2 was 8.756%, which was determined by TSS and L*, with a higher absolute value of the load. Figure 5 shows that the comprehensive quality scores of sweet cherry fruits exhibited a downward trend with changes in storage characteristics, indicating that the quality of sweet cherry fruit gradually deteriorated. On the 0 th day of storage, the comprehensive quality scores of the archedshelter and shelter-free fruits were similar, and the umbrellashelter fruits had the lowest comprehensive scores, but with the extension of the storage period, the comprehensive score of sweet cherry fruit quality declined rapidly, with the comprehensive score of fruit quality of shelter-free fruit declining the most. However, the comprehensive scores of fruit quality of arched-shelter, umbrella-shelter and shelterfree fruits tended to be negative within 4-8 days, while the comprehensive scores of shelter-free fruits first decreased to a negative value, followed by the umbrella-shelter and arched-shelter fruits. Therefore, the sweet cherry fruits under arched-shelter conditions had best quality and storability and had the longest shelf life, while the umbrellashelter fruits had a better storability and longer shelf life than the shelter-free fruits.

Conclusions
A total of 68 volatiles were identified and classified in sweet cherry fruits under arched-shelter, umbrella-shelter and shelter-free conditions. In all three conditions (shelter-free, arched-shelter and umbrella-shelter), the principal volatiles were benzyl alcohol, acetaldehyde, 2-methyl-propanal,    hexanal, (E)-2-hexenal and benzaldehyde. The above volatiles are the main contributors to the aroma of sweet cherry fruits. In addition, evaluation of the quality properties of sweet cherry fruits under arched-shelter, umbrella-shelter and shelter-free conditions during storage showed that the skin color of fruits gradually darkened, and the degree of browning increased. The fruit quality properties of archedshelter and umbrella-shelter fruits during storage were better than those of shelter-free fruits, and the storability of fruits was also better. PCA indicated that the quality of sweet cherry fruits under shelter-free conditions deteriorated fastest during storage, followed by fruits under umbrella-shelter and arched-shelter conditions, with fruits under archedshelter conditions having the best quality. This study provides reference values for the selection of rain shelter covering methods for sweet cherries in rainy areas and for the improvement of the postharvest quality and shelf life.

Disclosure statement
No potential conflict of interest was reported by the author(s).