Abstract
Flowering and ripening times were recorded for the southern highbush blueberry (Vaccinium corymbosum) cultivars Emerald, O'Neal, and Star, along with the newer University of Georgia releases ‘Camellia’, ‘Rebel’, ‘Southern Splendour’, and ‘Suziblue’ over several years at three locations in Georgia. Nearby weather stations were utilized to record temperatures for calculating chill hours and heat units during each of the years. These data were used to calculate fruit development period for the different cultivars as a function of real time (days) and thermal time (heat units). Depending on cultivar and chill hours at a location, flowering time varied as much as 29 days from year to year. Overall, ‘Emerald’ was the earliest flowering, followed by ‘Rebel’. ‘Camellia’ was the latest to flower among the cultivars. Average fruit development period for the cultivars were: Southern Splendour, 56.2 days; Rebel, 65.3 days; Suziblue, 65.5 days; Star, 67.1 days; Camellia, 68.8 days; O'Neal, 71.9 days; and Emerald, 82.8 days. The shortened fruit development period of ‘Southern Splendour’ resulted in a ripening time similar to ‘Rebel’ and ‘Star’, even though flowering time of ‘Southern Splendour’ was often several days later. Using heat units instead of days did not greatly reduce the coefficient of variation associated with fruit development period. However, the mean absolute and maximum difference between observed and predicted time to ripening was improved for most cultivars when heat units were used. These data should be useful in estimating how these blueberry cultivars will perform under diverse environments.
INTRODUCTION
The time of flowering and ripening are important considerations for determining usefulness and adaptability of blueberry (Vaccinium sp.) cultivars to various regions. While blueberry flowering time depends greatly on chill hours and temperature (CitationNeSmith and Bridges, 1992), less is known about time to ripening. Typically, blueberry cultivars are described as having an average number of days from flowering to ripening, yet the variability can be tremendous (CitationBailey, 1947). The ability to more accurately estimate fruit development period (FDP), or the time from flowering to ripening, would be an important tool for marketing large volumes of fruit.
The rate and duration of plant development depends on temperature and time. A common means of normalizing temperature effects on development is the use of heat units (HU) or growing degree (CitationRitchie and NeSmith, 1991). The use of HU instead of real or calendar time tends to produce experimental results that are transferable and comparable, even though the results may be from specific locations and environments. Research has examined using HU for predicting FDP to a limited degree for some highbush blueberries, yet more work is needed (CitationCarlson and Hancock, 1991; CitationMainland, 2002). The objective of this research was to evaluate fruit development period for several southern highbush blueberry cultivars grown in diverse locations as a function of real time and heat units.
MATERIALS AND METHODS
The southern highbush blueberries used in this study were grown in selection test blocks at three locations in Georgia. Two of the sites, Alapaha and Homerville, Georgia, are in the Atlantic Coastal Flatwoods region of the state, which is the primary blueberry production region. In general, plants utilized from these sites were bushes 3 to 8 years old, grown in pine bark amended soil with irrigation. The third location was at Griffin, Georgia, which is in the Piedmont region of the state. Plants at this location were also 3 to 8 years old, and were grown with irrigation in soil that was amended with pine bark. Plant growth regulators, such as hydrogen cyanamide, were not used on any of the test plots. Plots at all locations were established in a completely randomized design, with 3 to 10 plants per plot. Cultivars evaluated at each site included Camellia, Emerald, O'Neal, Rebel, Southern Splendour, Star, and Suziblue. Of these, Camellia, Rebel, Southern Splendour, and Suziblue are new cultivar releases from the University of Georgia, and the others are older cultivars of considerable importance.
Flowering and ripening dates were recorded for plots of plants at the three locations from 2004 to 2008. Severe freeze damage across all sites prevented 2007 data collection. In order to determine dates of development, an overall estimation of the time of 50% flowering and ripening was made for an entire plot rather than for individual bushes or flower clusters. Such an estimate was used because it was most likely the kind of estimate growers would use.
Weather data was collected via the Georgia Automated Environmental Monitoring Network (http://www.georgiaweather.net/) for stations nearest the sites. These data were used to determine HU for each day from flowering to ripening using the equation:
where Tmax = daily maximum temperature, Tmin = daily minimum temperature, and Tbase = base temperature of 7°C (CitationNeSmith and Bridges, 1992). One constraint used in calculations was if Tmin was less than Tbase, then it was set equal to Tbase. The same HU formula was used for all cultivars at all locations.
Data analyses involved calculating means and coefficients of variation (c.v.) for the FDP of each cultivar in days and HU. Mean absolute difference (MAD) and maximum difference from the mean (DMAX) for predicted and observed data were also determined. Predicted data were values generated by “back-calculation” using the overall average HU value or overall average days for the FDP for each cultivar.
RESULTS AND DISCUSSION
Chill hours (<7°C) and flowering dates for the different cultivars at the three locations are listed in . There were considerable differences in bloom dates of cultivars at a given site from year to year. Some of this could be related to the variation in chill hours received; although, there were not huge variations in chilling received at a particular location from year to year. Hence, all of the variation in time to flowering could not be explained by chill hours received. CitationNeSmith and Bridges (1992) demonstrated that chill hours and heat units after dormancy requirements are met both control time of flowering. Dates of fruit ripening are presented in . As with time of flowering, there was variation in the date of ripening for a particular cultivar at a location from year to year.
TABLE 1 Chill Hours and Dates of 50% Flowering during Multiple Years for Several Southern Highbush Blueberry Cultivars at Trial Sites Near Alapaha, Griffin, and Homerville, Georgia
TABLE 2 Dates of 50% Ripening during Multiple Years for Several Southern Highbush Blueberry Cultivars at Trial Sites Near Alapaha, Griffin, and Homerville, Georgia
When comparing cultivars, average fruit development period (FDP) ranged from a low of 56.2 days for ‘Southern Splendour’ to a high of 82.8 days for ‘Emerald’ (). FDP varied considerably for individual cultivars when days were used. For example ‘Emerald’ FDP ranged from a low of 71 days to a high of 90 days. The use of HU to calculate FDP resulted in less variability among cultivars to a degree, although, the average c.v. across all cultivars was not always less for HU than days.
TABLE 3 Mean and Coefficient of Variation (c.v.) for the Fruit Development Period Expressed as Days and Heat Units (HU) from 50% Flowering to 50% Ripening for Several Southern Highbush Blueberry Cultivars Grown at Three Locations in Georgia from 2004 through 2008
The increased accuracy in predicting FDP using HU can be seen more clearly in the analyses in . The MAD was lower for most cultivars when using HU. In fact, the average MAD for the FDP across cultivars was lowered from 3.7 days to 2.8 days when using HU. Likewise, the DMAX was less when using HU. The average DMAX across cultivars was 10.1 days when using real time, and it was 6.4 days when using HU. The data indicate that HU are valuable in determining the time from flowering to ripening.
TABLE 4 Mean Absolute Difference and Maximum Difference from the Mean between Observed and Predicted Time of Ripening for Several Southern Highbush Blueberry Cultivars Using Average Days and Heat Units (HU) from the Time of Flowering (Predicted Data Were Values Generated by “Back-Calculation” Using the Overall Average HU Value or Overall Average Days for the FDP for Each Cultivar from )
Others have shown benefits of using HU for predicting ripening dates in highbush blueberries, yet several models and complex calculations were used (CitationCarlson and Hancock, 1991; CitationMainland, 2002). The current data for southern highbush blueberries suggest that the single HU model worked reasonably well across the cultivars, although, each cultivar had its own HU requirement. An advantage of the single, simple model used is that the data for temperatures (daily minimum and maximums) are readily available from most weather stations. Hence, growers could keep up with HU accumulated.
The additional variation in the data of time from bloom to ripening (not accounted for by HU) could be associated with several factors. For example, ripening can be slowed by heavy fruit set or poor leaf bud development (CitationNeSmith and Krewer, 1999; CitationNeSmith, 2002; CitationWilliamson et al., 2001). Also, if cross pollination is abundant, berry ripening can be at a faster rate (CitationBrevis and NeSmith, 2004). Thus, these additional factors need to be accounted for in order to more completely assess variability in FDP.
In summary, FDP in southern highbush blueberries varies considerably among cultivars and over years. Some of the variability in FDP could be accounted for by using a simple HU model. Thus, cultivars can be categorized based on their HU requirement for the FDP. Information on FDP and HU can be used to more reliably estimate days to ripening for southern highbush blueberries. Also, the information on cultivars can be used to determine environments suitable for producing southern highbush blueberries to achieve desired market windows.
LITERATURE CITED
- Bailey , J.S. 1947 . Development time from bloom to maturity in cultivated blueberries . Proc. Amer. Soc. Hort. Sci. , 49 : 193 – 195 .
- Brevis , P. and NeSmith , D.S. February 2004 . ‘Brightwell’: A more self-fertile rabbiteye blueberry cultivar , February , 15 – 16 . Tulsa , OK : Southern Region of the American Society for Horticultural Science Annual Meeting .
- Carlson , J.D. and Hancock , J.F. Jr. 1991 . A methodology for determining suitable heat-unit requirements for harvest of highbush blueberries . J. Amer. Soc. Hort. Sci. , 116 : 774 – 779 .
- Mainland , C.M. 2002 . Blueberry fruit set and intervals from blossoming to ripening . Acta Hort. , 574 : 189 – 192 .
- NeSmith , D.S. 2002 . Response of rabbiteye blueberry (Vaccinium ashei Reade) to the growth regulators CPPU and gibberellic acid . HortScience , 37 : 666 – 668 .
- NeSmith , D.S. and Bridges , D.C. 1992 . Modeling chilling influence on cumulative flowering: A case study using ‘Tifblue’ rabbiteye blueberry . J. Amer. Soc. Hort. Sci. , 117 : 698 – 702 .
- NeSmith , D.S. and Krewer , G. 1999 . Effect of bee pollination and GA3 on fruit size and maturity of three rabbiteye blueberry cultivars with similar fruit densities . HortScience , 34 : 1106 – 1107 .
- Ritchie , J.T. and NeSmith , D.S. 1991 . “ Temperature and crop development ” . In Modeling plant and soil systems , Edited by: Hanks , R.J. and Ritchie , J.T. 5 – 29 . Madison , WI : American Society of Agronomy (Monograph 31) .
- Williamson , J.G. , Maust , B.E. and NeSmith , D.S. 2001 . Timing and concentration of hydrogen cyanamide affect blueberry bud development and flower mortality . HortScience , 36 : 922 – 924 .