Abstract
Apple fruit firmness is one of the main attributes indicating fruit quality at harvest. It is affected by numerous factors during the entire growing season. The effects of weather conditions during apple development are often mentioned as a result of their impact on attributes linked to fruit firmness: fruit size, calcium concentration, water content, etc. In this study, the effects of weather conditions on ‘McIntosh’ apple (Malus ×domestica Borkh. cv. McIntosh) firmness at harvest time were analyzed. Fruit were harvested at nine sites in Quebec and Ontario over 15 years (1996–2011). For each case, weather parameters were analyzed from full bloom until harvest, either in monthly subperiods from May until September or in terms of days from full bloom (DFB) until harvest. Regression results highlighted the negative effect of lower air temperature conditions from 31 to 60 DFB, higher air temperature conditions and precipitations from 61 to 90 DFB, and higher temperature conditions from 91 DFB until harvest on ‘McIntosh’ apple firmness level at harvest. Precipitation from 61 to 90 DFB alone explained 39% of ‘McIntosh’ apple firmness variation at harvest time. The prediction of apple firmness at harvest time could be helpful for producers to adjust their marketing and storage strategies according to apple quality level.
Firmness is the main attribute that gives an indication of fruit texture and it is often used by producers to evaluate harvest date (Trillot and Tillard, 2002). This quality index can be influenced by many preharvest factors such as season, orchard location, nutrition, and exposure to sunlight, which are independent of the fruit maturity level (DeEll et al., 2001). Apple cultivar, rootstock, thinning timing, and the use of compounds that alter partitioning of photosynthates have been shown to have an impact on the levels of firmness measured at harvest (Harker et al., 1997; Johnson, 1994). Fruit calcium concentration also seems to be involved in the maintenance of membrane integrity linked to firmness loss rate (Conway and Sams, 1987; Marmo et al., 1985).
Fruit size and maturity at harvest influence the storage behavior of apples, including their rate of firmness loss (Saure, 1996; Tromp, 1997). In fact, fruit size is highly related to cell size and number with larger fruit containing less but larger cells per unit of volume than smaller fruit (Blanpied et al., 1978). Knowledge of interactions between cell water relations and resulting turgor and tensile properties of cell walls could lead to a better understanding of physiological basis of cell strength, which is an indicator of apple firmness.
Many studies on the distinction between apple quality levels at harvest highlighted the important influence of weather conditions during fruit development (Calderón Zavala et al., 2004; Chapon and Westercamp, 1996; Smock, 1953). Depending on the development stage, weather conditions may have multiple effects on final firmness that has been reported to increase and reach desired levels earlier when the apples are exposed to higher temperatures during the first 6 weeks of fruit growth (Warrington et al., 1999). Higher air temperature conditions after fruit set, when the maturation process begins, has also been highlighted as a factor leading to decreased firmness levels at harvest (Calderón Zavala et al., 2004; Kondo and Takahashi, 1987; Tromp, 1997; Warrington et al., 1999). Low photosynthetic photon flux, characteristic of shading, has been negatively related to apple firmness at harvest (Corelli Grappadelli, 2003). Water tends to condense when exposed to lower temperatures, reducing turgor pressure in apples and slowing the maturation process. Lower air temperature conditions at harvest would thereby influence fruit firmness by varying water status of apples and delaying harvest time (Johnston et al., 2002). A study on the effects of weather conditions on apples in Quebec (‘McIntosh’, ‘Cortland’, ‘Spartan’, and ‘Empire’) highlighted the negative effect of high relative humidity during the entire growing season and the positive effect of long periods without rain during the period of 46 d after fruit set until harvest on firmness at harvest time (G. Bourgeois, unpublished data).
The overall objective of this study was to explain the variations in ‘McIntosh’ firmness values at harvest over multiple years and sites in Quebec and Ontario, Canada. The purpose of this approach was first to highlight the periods of apple development sensitive to specific weather conditions in reducing firmness levels at harvest. Second, characterization of these preharvest weather conditions in specific parameters was done to determine their direct or indirect effects on apple firmness. For future studies, the results of these analyses could permit the development of a spatiotemporal forecasting model based on preharvest weather conditions. Predictions of texture levels, both at harvest and after removal from storage, could provide a helpful tool to apple producers in their marketing strategies.
Materials and Methods
Experimental sites and years of harvest.
Firmness data were obtained from ‘McIntosh’ apples (Malus ×domestica Borkh. cv. McIntosh) harvested in nine sites, five in Quebec (Saint-Paul d’Abbotsford, Franklin, Frelighsburg, Oka, and Mont Saint-Grégoire) and four in Ontario [Cobourg, Georgian Bay, Simcoe (Orchard 1), and Simcoe (Orchard 2)], over 15 years, from 1996 to 2011. Fruit firmness was evaluated by two measures on opposite sides of each apple. The peel was removed before the analysis using the electronic Effegi handheld penetrometer (Facchini, Alfonsine, Italy) from 1996 to 2005 and the Fruit Texture Analyzer fitted with an 11-mm tip (GÜSS, Strand, South Africa) from 2006 to 2011.
Weather data were obtained from the National Land and Water Information Service (Agriculture and Agri-Food Canada, 2007) and the weather apple network in Quebec (Bourgeois et al., 2008). Weather parameters available for analyses include: minimum, maximum, and mean daily air temperatures (°C) and daily precipitation (mm). Weather condition parameters were aggregated for each month from May to September, which represents the period of full bloom until harvest time. Temperature conditions were expressed in number of days for each month when mean air temperature was below 5 to 30 °C by steps of 5 °C. Precipitation conditions were expressed in number of days when cumulated precipitation was higher than the following values: 0.5 to 3.5 mm by steps of 0.5 mm, 5.0, and 6.0 mm.
To consider the weather condition effect on specific phenological development stages for each site and year of harvest, predicted growth stages of the ‘McIntosh’ apple were generated using a bioclimatic model for apple development (Bourgeois et al., 2013). The Biologische Bundesantalt, Bundessortenamt, and Chemische Industry (BBCH) phenological scale developed for the pome fruit in 1994 was used to describe the principal and secondary growth stages (Meier et al., 2001). Principal growth stages include bud development, vegetative growth, reproductive growth, development, and ripening of fruits, up to the senescence of the commercial product, allowing easy linkage between environmental data and particular developmental stages.
For each site and year of harvest, predicted phenological data were divided in number of DFB considering that BBCH stage 65 is the full bloom estimated date and BBCH stage 80 is the estimated harvest time based on apple diameter measurement. Subdivision of the DBF was established as the following: 0 to 30 DFB, 31 to 60 DFB, 61 to 90 DFB, and 91 DFB until harvest.
Principal component analysis (PCA) using the XLSTAT software (Addinsoft, Paris, France) was used to detect correlations between weather parameters, in terms of DBF, and firmness at harvest. The REG procedure and the STEPWISE selection method of SAS Version 9.2 (SAS Institute, Inc., Cary, NC) were used to analyze the effect of all the fruit quality attributes and weather conditions on firmness at harvest. Stepwise regression identified separately the variables that explained a significant portion of the variation in firmness levels at harvest. Regression results were then used to predict firmness levels at harvest and comparisons between accuracy levels of each model were assessed using statistical methods such as root mean square error (RMSE), forecasting efficiency (EF), and mean error (E) (Yang et al., 2000). The best prediction will have RMSE close to zero, EF close to one, and E close to zero. For each model, the percentage of well estimated (± 2.2 N of the observed firmness), over-, and underestimated cases of firmness were also calculated. This threshold represents the accuracy of the penetrometers used to measure the firmness level at harvest.
Results and Discussion
Observed firmness.
Firmness data obtained from ‘McIntosh’ apples harvested in nine regions over 15 years varied from 52.9 to 79.6 N with a mean of 67.7 ± 6.3 N. The three highest firmness values observed at harvest were Franklin in 1999 (75.2 N), Mont Saint-Grégoire in 2001 (77.4 N), and Simcoe in 2005 (79.6 N), whereas the three lowest values recorded in this study were Simcoe in 2006 and 2009 (52.9 N and 59.6 N, respectively) and Saint-Paul d’Abbotsford in 2010 (57.4 N).
Comparison of annual firmness levels at harvest in ‘McIntosh’ apples shows a slow decrease since the 1990s for Ontario and Quebec regions (Fig. 1). However, studies still have to be conducted on multiple comparisons between apple productions before and after the 1990s to detect the main factor explaining this decrease. Simple modifications in orchard management practices, apple tree loads per orchard, cropload, handling techniques, or fruit quality indices used to detect maturity levels might imply modifications of firmness levels at harvest (Harker et al., 1997). It has been suggested that the deterioration in textural properties of apples may be linked with development of more intensive growing systems within the orchard and greater standards for apple caliber, which lowers the overall firmness values at harvest (Harker et al., 1997). In England, this downward trend in apple firmness has been shown on ‘Cox’s Orange Pippin’ apples, where between 1966 and 1984, firmness after storage has declined by more than 10 N despite improvement in controlled-atmosphere regimes over the years (Horscroft, 1989). The interpretation of such data remains problematic as a result of variations in methods for measuring firmness and fruit size, storage technologies, and length of storage period as the fruit industry becomes more sophisticated (Horscroft, 1989).
‘McIntosh’ apple firmness levels at harvest (N) as a function of years of harvest, for all sites and years of harvest included in this study, with a tendency curve plotted to the data.
Citation: HortScience horts 48, 4; 10.21273/HORTSCI.48.4.474
‘McIntosh’ apple firmness levels at harvest (N) as a function of years of harvest, for all sites and years of harvest included in this study, with a tendency curve plotted to the data.
Citation: HortScience horts 48, 4; 10.21273/HORTSCI.48.4.474
‘McIntosh’ apple firmness levels at harvest (N) as a function of years of harvest, for all sites and years of harvest included in this study, with a tendency curve plotted to the data.
Citation: HortScience horts 48, 4; 10.21273/HORTSCI.48.4.474
Mean predicted dates and corresponding BBCH stages for each subperiod of fruit growth (expressed as DFB) for all sites and years of harvest included in this study are shown in Table 1. By relating dates to fruit development, monthly analyses results and phenologically based weather parameters could be interpreted together.
Subperiods of days from full bloom with corresponding BBCH stages and dates as a mean for all sites and years of harvest for ‘McIntosh’ apples included in this study.
Predicted dates corresponding to each subperiod of DFB varied slightly between the cases corresponding to the lowest and the highest recorded firmness values at harvest. Overall estimated number of days elapsed since 1 Jan. for each BBCH stage, the earlier occurrences of the first four phenological stages for cases when firmness levels at harvest are lower might have an impact on this fruit quality parameter. This variation is less noticeable for the BBCH stage 80.
Principal component analyses.
The first PCAs were conducted on the effect of all monthly weather parameters on firmness level at harvest. A matrix of correlation highlighted parameters with significant negative effects, including the number of days with mean daily temperature below 20 °C in September, below 25 °C in June, and the number of days with daily cumulated precipitation above 6.0 mm in August (r = –0.39, –0.49, and –0.40, respectively) and positive effects on firmness levels at harvest, including the number of days when mean daily temperature was above 25 °C in June (r = 0.49).
Another PCA included the number of days when daily cumulated precipitation was above 0.5 mm and the number of days when monthly mean air temperature was below 15 °C from May to September. For low air temperature, the months of May and June were divided into four weekly subperiods. No correlation was significant between any weather parameter and firmness levels at harvest despite aggregations to reduce the number of parameter included in the analysis. Therefore, the number of days when mean air temperature was below 15 °C was aggregated for July, August, and September, and the number of days when daily cumulated precipitation was above 0.5 mm was aggregated for July and August. Once again, no significant correlation was found between any weather parameter and firmness levels at harvest.
The same test was conducted on effects of weather parameters per phenological subperiod in DFB. Weather parameters calculated for each developmental subperiod included in the PCA were mean minimum air temperature (°C), number of days when mean air temperature was below 15 and 20 °C, and cumulated precipitation (mm). The matrix of correlation generated by this test highlighted a significant negative effect of the cumulated precipitation from 61 to 90 DFB (r = –0.63) and positive effects of mean minimum air temperature from 0 to 30 DFB and from 31 to 60 DFB (r = 0.44 and 0.42, respectively) on ‘McIntosh’ firmness levels at harvest. PCA conducted with the same parameters according to DFB, although without the parameter of the number of days when mean air temperature was below 20 °C, highlighted the same three correlations as mentioned in the earlier PCA.
Regression analyses.
Multiple linear regression [Eq. (1)] of the weather conditions effect on ‘McIntosh’ apple firmness level at harvest for certain month during the fruit development (from 14 May until end of September) for all sites and years of harvest observed in this study.
The predicted values of firmness at harvest obtained with the regression equation [Eq. (1)] including only weather conditions during the growing season, in monthly values, were considered adequate with 52% of the predictions located in the threshold of ± 2.2 N of the observed firmness value at harvest (RMSE = 4.30, EF = 0.51, and E = 0.26). When plotted together, predicted and observed firmness data presented a higher underestimation for high observed values (between 60.9 and 77.4 N) than an overestimation of low observed values (between 52.9 and 69.4 N) (Fig. 2). Thirteen cases of intermediate firmness values (between 59.6 and 79.4 N) were well estimated using Eq. [1].
‘McIntosh’ apple firmness at harvest predicted with Eqs. [1] and [3] as a function of observed values for each site and years of harvest with their corresponding linear functions and the 1:1 ratio line.
Citation: HortScience horts 48, 4; 10.21273/HORTSCI.48.4.474
‘McIntosh’ apple firmness at harvest predicted with Eqs. [1] and [3] as a function of observed values for each site and years of harvest with their corresponding linear functions and the 1:1 ratio line.
Citation: HortScience horts 48, 4; 10.21273/HORTSCI.48.4.474
‘McIntosh’ apple firmness at harvest predicted with Eqs. [1] and [3] as a function of observed values for each site and years of harvest with their corresponding linear functions and the 1:1 ratio line.
Citation: HortScience horts 48, 4; 10.21273/HORTSCI.48.4.474
During the month of June, equivalent to the period of 20 to 40 DFB, as a mean of all sites and years included in this study (Table 1), ‘McIntosh’ apples are normally between fruit set (BBCH = 71) and 50% of final diameter achieved (BBCH = 75). At this period of fruit development, apple cell enlargement occurs at the highest rate, increasing the water and mineral needs of the fruit (Chapon and Westercamp, 1996). Lower air temperature conditions at this subperiod of development may reduce respiration rate (Blanke, 1995). Fruit cells require high amounts of energy to synthesize proteins, ethylene, and aromatic volatiles, and this energy is mainly provided by respiration. The resulting glucose is then used for further ripening in the presence of oxygen through glycolysis and the citric acid cycle, releasing carbon dioxide and energy (Von Mollendorff, 1996). Nutrient uptake depends on respiration as well, diminishing under conditions of low soil temperature, and possibly also related to lower air temperature, implying lower uptake and partitioning of calcium. Lower growth rate and calcium content of apples both imply lower firmness level at harvest, explaining the highest variation of firmness level at harvest resulting from lower air temperature conditions during this period of the apple fruit growth. In fact, the highest firmness level at harvest, Simcoe in 2005 with 79.4 N, corresponded to the lowest number of days when mean air temperature was below 25 °C during the month of June with only 20 d. Mean value over all sites and years of harvest in this study was 29 d below 25 °C in June.
Fruits absorb calcium only during a short period after full bloom, rarely more than 6 weeks, then depending on fruit growth, calcium is diluted in the apple (Faust, 1986). The period of 31 to 60 DFB overlaps with this maximal calcium absorption, which may be altered by low air temperature conditions as well. Firmness has been linked to higher air temperature conditions during the first 6 weeks of fruit growth (Warrington et al., 1999). Calcium has many roles in cell integrity, influencing cell wall mechanical properties and adhesion between cells, both influencing turgor by their indirect effect on membrane permeability, osmotic gradient across plasma membrane, and cell wall movement (expansion and contraction) (Harker et al., 1997). Calcium deficiency has been linked to increase in membrane permeability as well as to increased senescence breakdown incidence during storage (Bramlage et al., 1979; Saure, 1996). Soil moisture excess encourages vegetative growth, large fruit size, and, if extreme, it may even affect time of fruit maturation (Bramlage, 1993). High soil moisture conditions are either the result of high irrigation or intense rainfall. However, precipitation events are closely related with low light intensities and low air temperatures, which make it difficult to separate the direct and indirect effects of each weather parameter on apple quality retention (Bramlage, 1993). It is possible that lower air temperature conditions between 31 and 60 DFB affect negatively apple firmness at harvest by increasing its final caliber, diluting even more the calcium concentrations absorbed by the apple (Faust, 1986).
Tmin61–90 varied between 13.9 and 18.0 °C and accounted for 17% of ‘McIntosh’ firmness level variation at harvest. At this period of apple development, from mid-July to mid-August, as a mean over all sites and years included in this study, ‘McIntosh’ apples have normally achieved between 70% and 90% of their final caliber (Table 1). Apples are in cell expansion process, although fruit growth rate starts to decrease as apple maturation process sets in (Chapon and Westercamp, 1996). For the same subperiod of apple development, cumulated precipitation (Prec61–90), which varied between 10 and 150 mm, demonstrated a negative effect on firmness at harvest and accounted for the highest variation of 39% of firmness values recorded for all cases included in this study. Considering that these two weather parameters during this period of fruit development accounted for 56% of firmness variation together, we may consider this subperiod as the most sensitive to firmness variations observed at harvest. During this period, cell wall thickening and increase in attractive forces between the walls of adjacent cells remain important, as the primary wall integrity of individual cells in determining the overall physical properties of the fruit (Harker et al., 1997). Higher air temperature conditions and increased precipitation occurrences during this period might decrease firmness level at harvest by affecting cell wall integrity or the vacuolation process implied in the course of cellular expansion to maintain cell turgor (Harker et al., 1997; Warrington et al., 1999). Respiration rate decreases slowly during this period of fruit growth until a constant low value is maintained for the month before harvest (Fidler, 1973). Apples exposed to higher air temperature conditions start showing abnormal metabolism, resulting in loss of membrane integrity and structure with disruption of cellular organization and rapid deterioration of the produce (Wills et al., 1981). This might be the reason for decreased firmness at harvest linked to higher air temperature conditions during this period of fruit growth. Because ‘Delicious’ apples under a certain level of water stress showed lowered firmness loss rate, indicating an alteration in the physiological mechanism of fruit softening (Ebel et al., 1993), it is possible that the reverse effect could be observable as well. Increased precipitation during this period of fruit growth might imply a faster rate of firmness loss leading to lowered firmness at harvest.
Tmin91-H varied from 11.6 to 15.2 °C and accounted for 9% of ‘McIntosh’ firmness level variation at harvest. Ninety-one DFB until harvest represents the last month before harvest, between mid-August and mid-September, where final fruit maturation processes occur such as soluble solid accumulation, slow fruit diameter increase, background fruit color change, and decrease in acidity (Devlin and Witham, 1983). Higher air temperature conditions during this period of the apple fruit development may lead to increase in respiration rate, which would result in premature ethylene production. Advanced maturity at harvest has been linked to reduced firmness level, which also influences fruit quality during storage (Chapon and Westercamp, 1996). Increased air temperatures, or its indirect relation with high solar radiation and low precipitation conditions, may also be involved with calcium deficiency as a result of reduced transportation of ions to and from cells (Westwood, 1993). This deficiency increases membrane permeability, thereby leading to water loss by evapotranspiration and reduced cell turgor (Saure, 1996).
By simulating the predicted values of firmness at harvest with the regression equation [Eq. (2)], including only weather conditions during the growing season in DFB values, the prediction was considered adequate with 48% of the predictions located in the threshold of ± 2.2 N of observed firmness values at harvest (RMSE = 3.04, EF = 0.75 and E = –0.47). The linear regression fitted to the predicted firmness values at harvest as a function of observed data revealed a more accurate slope (0.75), closer to the 1:1 ratio line, for predictions using Eq. [2] than predictions using Eq. [1]. Twelve cases of intermediate firmness values (between 52.9 and 73.4 N) were well estimated using Eq. [2]. Although this represents more over-estimated cases than for Eq. [1], with 32% instead of 16%, and less under-estimation than for Eq. [1], with 20% instead of 32%, Eq. [2] remains a better option for firmness prediction, because it shows higher accuracy than Eq. [1] based on RMSE and EF values. In addition, the R2 value of the fitted regression for Eq. [2] (0.76) is much higher than the fitted regression in Eq. [1] (0.48), implying also a greater accuracy for predictions using phenological stages instead of monthly weather conditions.
Multiple linear regression [Eq. (3)] of the weather conditions effect on ‘McIntosh’ apple firmness level at harvest with standardized weather parameters for periods of the fruit development, in days from full bloom (DFB), for all sites and years observed in this study.
By simulating the predicted values of firmness with the regression equation including only weather conditions during the growing season, in DFB values, the prediction is considered adequate with 44% of the predictions that are located in the threshold of ± 2.2 N of the observed firmness value at harvest (RMSE = 3.17, EF = 0.73, and E = 0.40). Only 11 cases included in this study showed accurate predictions using Eq. [3], which is the lowest from all regression predictions, although the regression fitted with these data showed the slope closest to 1. However, in comparison with predictions from Eqs. [1] and [2], its R2 value (0.74) is less important than for the relation illustrating predictions using Eq. [2].
When comparing Eq. [2] and Eq. [3], the second one would be more easily converted to a dynamic model for the purpose of forecasting firmness levels of ‘McIntosh’ apples at harvest. However, Eq. [3] contains a slightly nonsignificant parameter and a lower R2 value than Eq. [2]. As for predictions of firmness values using both regressions, only one additional case was well predicted using Eq. [2] instead of Eq. [3]. Therefore, differences between both regressions are not very important, so Eq. [3] remains more appropriate for use for developing a forecasting model.
In previous regression results, the parameter of cumulated precipitation from 61 to 90 DFB remained the factor explaining most of the variation between firmness values for all sites and years included in this study with 39% of additional variation. When plotted with corresponding cumulated precipitation values from 61 to 90 DFB for each site and year, firmness values at harvest seem slightly separated in two aggregates (Fig. 3). In fact, a gap of cumulated precipitation values between 100 and 130 mm explains this division between both groups of firmness values, dividing high (64 N or greater) and low (less than 64 N) firmness values. The effect of precipitation conditions above ≈120 mm might induce a greater reduction of firmness observed in ‘McIntosh’ apples at harvest. This distinction is much easier to visualize when precipitation values are squared.
‘McIntosh’ apple firmness at harvest as a function of cumulated precipitation from 61 to 90 DFB for each site and year of harvest with a quadratic function applied to the plotted data and its corresponding equation and R2 value. DFB = days from full bloom.
Citation: HortScience horts 48, 4; 10.21273/HORTSCI.48.4.474
‘McIntosh’ apple firmness at harvest as a function of cumulated precipitation from 61 to 90 DFB for each site and year of harvest with a quadratic function applied to the plotted data and its corresponding equation and R2 value. DFB = days from full bloom.
Citation: HortScience horts 48, 4; 10.21273/HORTSCI.48.4.474
‘McIntosh’ apple firmness at harvest as a function of cumulated precipitation from 61 to 90 DFB for each site and year of harvest with a quadratic function applied to the plotted data and its corresponding equation and R2 value. DFB = days from full bloom.
Citation: HortScience horts 48, 4; 10.21273/HORTSCI.48.4.474
By simulating the predicted values of firmness with the regression equation including only weather conditions during the growing season, in DFB values, the prediction is considered adequate with 64% of the predictions that are located in the threshold of ± 2.2 N of the observed firmness value at harvest (RMSE = 3.18, EF = 0.73, and E = –0.28). This regression overestimated fewer cases than Eqs. [2] and [3] with only 16% of overestimation compared with 32 and 24%, respectively. When plotted according to the observed firmness values for all sites and years included in this study, the regression fitted to this relation revealed a slope of 0.71, which is close to 1, although not as much as for predictions of firmness using Eq. [2] or [3]. As for the R2 value for the regression fitted to the predicted values using Eq. [4] in function of observed firmness data, its value of 0.73 is also lower than for predictions with Eqs. [2] and [3].
The use of the number of DFB has proven to increase the accuracy of the effect of weather conditions on firmness level at harvest for ‘McIntosh’ apples. Predictions of BBCH stages linked to DFB values were used to evaluate the maturity stages of apples linked to these effects of weather conditions on firmness. However, this scale is based on fruit caliber after full bloom until harvest, which limits the ability to link them to maturation stages. Knowledge of the diameter values corresponding to each BBCH stage of the ‘McIntosh’ development could add concrete information for producers to relate more easily with which stages of apple growth are more sensitive to firmness loss as a result of highlighted weather conditions in this study. By including the combined effect of caliber and firmness measurements at harvest, the compromise between these two parameters could lead to a better understanding of weather condition effect on the volume of apple by distinguishing their effects on the number of cells and the volume per cell in an apple, representing cell division and cell expansion processes (Harker et al., 1997).
Also, according to Tong et al. (1999), lower winter air temperatures, below critical air temperature levels for apple trees, may damage the tree vascular system, affecting fruit maturity and ripening during the growing season. It would be interesting to analyze the effects of weather conditions outside fruit growth season as well to verify the indirect effect of winter conditions on fruit firmness at harvest. Acquiring information on thinning practices in orchards, timing and intensity, could differentiate cases of high variations in apple firmness levels between orchards. In fact, it was shown that thinning during the period of 5 to 15 d after bloom was identified as a critical time to influence fruit firmness through thinning on ‘Cox’s Orange Pippin’ apples (Johnson, 1994). Because it is also known that calcium and phosphorus sprays and nitrogen fertilizer treatments influence final firmness at harvest, obtaining information on nutrient spraying practices could also help discriminate the cases when firmness was strongly influenced by these practices (Webster and Lidster, 1986).
Decrease in firmness observed since the last decade or so is either the result of increased intensity in apple production systems and fruit size standards for fresh fruit market or variations in weather conditions (Harker et al., 1997; Johnson, 1992; Tromp, 1997). The importance of preharvest weather condition effects on apple firmness at harvest is often mentioned in studies on apple quality at harvest, although information on their impact on fruit development is limited (Calderón Zavala et al., 2004; Green, 1963). This study highlighted the combined negative impact of lower air temperature conditions after cellular division, higher air temperature conditions and precipitations during the end of cellular expansion process, and higher air temperature conditions preceding harvest time on ‘McIntosh’ apple firmness level at harvest. The parameter of precipitation conditions during cellular expansion explained by itself 39% of firmness variation with a higher impact on decreasing firmness when daily precipitation was higher than 120 mm. Using the regression including only standardized weather parameters, it would be interesting in future research to develop forecasting models for apple firmness using only weather conditions during the growing season. The ability to predict firmness indices before harvest could help producers in establishing appropriate marketing strategies. They could either invest in long-term storage in years of high firmness values at harvest or sell fruit on the fresh fruit market if firmness has already started to decrease. In another perspective, these predictions could also permit the determination of temporal and spatial conditions required to maintain higher apple firmness values at harvest.
Literature Cited
Agriculture and Agri-Food Canada 2007 National land and water information service. Daily 10km gridded climate dataset; 1961–2003, Canada, Version 1.0 (CSV) Generated using ANUSPLIN. Version 4.3
Blanke, M.M. 1995 Regulation of respiration in apple, avocado and citrus orange fruit Acta Hort. 398 139 146
Blanpied, G.D., Bramlage, W.J., Dewey, D.H., LaBelle, R.L., Massey, L.M., Mattus, G.E., Stiles, W.C. & Watada, A.E. 1978 A standardized method for collecting apple pressure test data Plant Sci. 1 1 6
Bourgeois, G., Plouffe, D., Beaudry, N., Choquette, D., Chouinard, G. & Bellerose, S. 2013 Using the DJPheno software for model calibration and validation of apple phenological stages. Proc. of the IXth Intl. Symp. on Modelling in Fruit Res. and Orchard Mgt Acta Hort. (in press)
Bourgeois, G., Plouffe, D., Chouinard, G., Beaudry, N., Choquette, D., Carisse, O. & DeEll, J. 2008 The apple CIPRA network in Canada: Using real-time weather information to forecast apple phenology, insects, diseases and physiological disorders Acta Hort. 803 29 34
Bramlage, W.J. 1993 Interactions of orchard factors and mineral nutrition on quality of pome fruit Acta Hort. 326 15 28
Bramlage, W.J., Drake, M. & Baker, J.H. 1979 Changes in calcium level in apple cortex tissue shortly before harvest and during postharvest storage Commun. Soil Sci. Plant Anal. 10 417 426
Calderón Zavala, G., Lakso, A.N. & Piccioni, R.M. 2004 Temperature effects on fruit and shoot growth in the apple (Malus domestica) early in the season. XXVI Intl. Hort. Congr.: Key Processes in the Growth and Cropping of Deciduous Fruit and Nut Trees Acta Hort. 636 447 453
Chapon, J.-F. & Westercamp, P. 1996 Entreposage frigorifique des pommes et des poires. Tome 2: ‘Conduite de la conservation.’ Centre technique interprofessionnel des fruits et légumes (Ctifl), Paris, France
Conway, W.S. & Sams, C.E. 1987 The effects of postharvest infiltration of calcium, magnesium, or strontium on decay, firmness, respiration, and ethylene production in apples J. Amer. Soc. Hort. Sci. 112 300 303
Corelli Grappadelli, L. 2003 Light relations, p. 195–216. In: Ferree, D.C. and I. J. Warrington (eds.). Apples; botany, production and uses. CABI Publishing, Cambridge, MA
DeEll, J.R., Khanizadeh, S., Saad, F. & Ferree, D.C. 2001 Factors affecting apple fruit firmness—A review J. Amer. Pomol. Soc. 55 8 27
Devlin, R.M. & Witham, F.H. 1983 Plant physiology. 4th Ed. Willard Grant Press, Boston, MA
Ebel, R.C., Proebsting, E.L. & Patterson, M.E. 1993 Regulated deficit irrigation may alter apple maturity, quality, and storage life HortScience 28 141 143
Faust, M. 1986 Preharvest factors affecting storage quality of apples. The Deciduous Fruit Grower. Frigair 86 Conf., Pretoria, Gauteng
Fidler, J.C. 1973 Conditions of storage, p. 1–51. In: Fidler, J.C. (ed.). The biology of apple and pear storage. Cmwlth. Agricultural Bur., UK
Green, G.C. 1963 The effect of weather and climate upon the keeping quality of fruit. Technical Note No. 53. Rept. of Working Group of the Commission for Agricultural Meteorol. World Meteorol. Organization, Geneva, Switzerland
Harker, F.R., Redgwell, R.J., Hallett, I.C. & Murray, S.H. 1997 Texture of fresh fruit Hort. Rev. 20 121 224
Horscroft, J. 1989 Production factors influencing the textural qualities of apples. PhD thesis, Wye College, Univ. London, London, UK
Johnson, D.S. 1992 The effect of flower and fruit thinning on the firmness of ‘Cox’s Orange Pippin’ apples at harvest and after storage J. Hort. Sci. 67 95 101
Johnson, D.S. 1994 Influence of time of flower and fruit thinning on the firmness of ‘Cox’s Orange Pippin’ apples at harvest and after storage J. Hort. Sci. 69 197 203
Johnston, J.W., Hewett, E.W. & Hertog, M. 2002 Postharvest softening of apple (Malus domestica) fruit: A review New Zeal. J. Crop Hort. 30 145 160
Kondo, S. & Takahashi, Y. 1987 Effects of high temperature in the nighttime and shading in the daytime on the early drop of apple fruit ‘Starking Delicious’ J. Jpn. Soc. Hort. Sci. 56 142 150
Marmo, C.A., Bramlage, W.J. & Weis, S.A. 1985 Effects of fruit maturity, size, and mineral concentrations on predicting the storage life J. Amer. Soc. Hort. Sci. 110 499 502
Meier, U., Bleiholder, H., Weber, E., Feller, C., Hess, M., Wicke, H., van den Boom, T., Lancashire, P.D., Buhr, L., Hack, H., Klose, R. & Stauss, R. 2001 Growth stages of mono-and dicotyledonous plants; BBCH monograph. 2nd ed. Federal Biological Res. Ctr. for Agr. and For
Saure, M.C. 1996 Reassessment of the role of calcium in development of bitter pit in apple Aust. J. Plant Physiol. 23 237 243
Smock, R.M. 1953 Some effects of climate during the growing season on keeping quality of apples J. Amer. Soc. Hort. Sci. 62 272 278
Tong, C., Krueger, D., Vickers, Z., Bedford, D., Luby, J., El-Shiekh, A., Schackel, K. & Ahmadi, H. 1999 Comparison of softening-related changes during storage of ‘Honeycrisp’ apple, its parents, and ‘Delicious’ J. Amer. Soc. Hort. Sci. 124 407 415
Trillot, M. & Tillard, S. 2002 Le pommier. Monographie. Centre technique interprofessionnel des fruits et légumes (Ctifl). France
Tromp, J. 1997 Maturity of apple cv. Elstar as affected by temperature during a six-week period following bloom J. Hort. Sci. 72 811 819
Von Mollendorff, L.J. 1996 Physiological changes in fruit during development and ripening, p. 33–38. In: Combrink, J.C. (ed.). Integrated management of post-harvest quality. INFRUITEC; An Inst. of the Agricultural Res. Council, Stellenbosch, Western Cape
Warrington, I.J., Fulton, T.A., Halligan, E.A. & De Silva, H.N. 1999 Apple fruit growth and maturity are affected by early season temperature J. Amer. Soc. Hort. Sci. 124 468 477
Webster, D.H. & Lidster, P.D. 1986 Effects of phosphate sprays on McIntosh apple fruit and leaf composition, flesh firmness and susceptibility to low-temperature disorders Can. J. Plant Sci. 66 617 626
Westwood, M.N. 1993 Temperate-zone pomology; Physiology and culture. 3rd Ed. Timber Press Inc., Portland, OR
Wills, R.H.H., Lee, T.H., Graham, D., McGlasson, W.B. & Hall, E.G. 1981 Physiology and biochemistry, p. 17–37. In: Postharvest: An introduction to the physiology and handling of fruit and vegetables. AVI Publishing Company Inc., Australia
Yang, J., Greenwood, D.J., Rowell, D.L., Wadsworth, G.A. & Burns, I.G. 2000 Statistical methods for evaluating a crop nitrogen simulation model, N_ABLE Agr. Syst. 64 37 53
