Click on author name to view affiliation information
Freeze events between January and April can result in major crop and economic losses for growers of low-chill, early-ripening varieties of blueberry (Vaccinium sp.) in Florida and Georgia. The objective of this research was to determine current responses by blueberry growers to freeze events. Blueberry growers in Florida and Georgia were surveyed about frost protection decision criteria. Growers had differing opinions on when to make the decision to frost-protect blueberry crops. Almost all (98.9%) of the respondents (n = 94) who reported using at least one method of active frost protection reported using irrigation. Farm size, as measured by blueberry acreage, did not influence decisions regarding the use of active frost protection measures. Blueberry growers, on average, reported that a loss of up to 30% to 39% of their crop could be tolerated and still produce a marketable crop. However, they may have been overly cautious at the early bud stages, with ≈40% and 55% of respondents protecting at the bud swell and tight cluster stages, respectively. Understanding the use of irrigation as a frost protection practice in the southeastern United States can aid in improving frost protection recommendations, helping growers maximize yield and saving water and money.
Blueberry acreage in Florida and Georgia has more than doubled since 2000. In 2016, blueberries were harvested from 4700 acres in Florida and 16,900 acres in Georgia. The two states accounted for 23.3% of the total U.S. blueberry harvested area (U.S. Department of Agriculture, 2017). Despite early-ripening varieties being susceptible to spring frost injury (Strik and Yarborough, 2005), low-chill, early harvest southern highbush blueberries [SHB (Vaccinium corymbosum interspecific hybrids)] have been extensively planted in this region to take advantage of pricing for entrance into early U.S. market windows (Brazelton, 2015; England, 2015). SHB varieties begin to flower in January in both Florida and Georgia, with frost events occurring from January until April. Depending on the floral bud development stage and when temperatures drop to freezing or below, damage can be significant (Warmund et al., 2008). As a result, frost-protection measures commonly are used to mitigate crop loss in Florida and Georgia (England, 2015; NeSmith, 2008).
Cost of orchard establishment can be $10,000 to $12,000 per acre or more (Fonsah et al., 2013; Singerman et al., 2016). Most growers use low-volume (usually drip) irrigation systems to address crop water requirements and overhead irrigation for frost protection. Frost protection, primarily overhead irrigation systems, can add another $2300 per acre to establishment costs (Singerman et al., 2016). Once established, operating costs for frost protection through overhead irrigation can be low (as compared with wind machines and heaters) (Poling, 2008). With relatively low operating costs, the main barrier for the use of frost protection is its limited effectiveness. In general, when temperature drops to –7 °C or wind speed increases to 16 km·h−1 or above, frost protection may not be effective. Other matters that growers should consider in frost-protecting their crops include increased evaporative cooling due to wind, the potential for poor ice formation, ice accumulation breaking and stripping flowers from the plant, and water restrictions due to regulatory requirements or physical limitations of the irrigation system (Perry, 1998; Poling, 2008; Snyder and Melo-Abreu, 2005).
Given these constraints, growers initiate frost protection in advance of a frost or freeze event when the temperature is anticipated to reach critical levels where lethal damage to reproductive tissue can occur (critical bud temperature values). Figure 1 illustrates the established critical temperatures for different blueberry bud stages (Patten et al., 1991; Spiers, 1978). Another problematic constraint is the information gaps related to variety, climate, and bud stage in recommendations provided to growers in Florida and Georgia. As a result, growers in Florida and Georgia must rely on critical bud temperature information developed for other regions.
Citation: HortTechnology hortte 28, 5; 10.21273/HORTTECH04008-18
Overall, the decision to start frost protection at a particular time depends on the cost, information available, and risk attitude of growers to adjust their practices in response to frost forecasts (Crane et al., 2010; Hu et al., 2006; Klockow et al., 2010; Kusunose and Mahmood, 2016; Mase and Prokopy, 2014; Stewart et al., 1984). The study by Stewart et al. (1984), which focused on frost-protection decisions of fruit growers in central Washington, found multiple on-farm weather resources and strategies available to growers to mitigate frost. Perceptions of the weather forecast quality and applicability of the forecast to growers’ production systems were crucial for the decision to frost-protect fruit crops.
Research shows that more studies examining the determinants of frost-protection decisions are needed, given the potentially high economic impact of frost and the need for more efficient frost-protection systems and opportunities to save water and energy by developing precision frost-protection technologies (Chevalier et al., 2012; Simnitt et al., 2017). The objective of this project was to identify, by survey, the prevalence of frost-protection systems, the use of information resources to mitigate damage caused by frost events, and the decision timing for operating a frost-protection system by Florida and Georgia blueberry growers.
A paper-based survey was developed that contained 19 questions directed to identify freeze-protection practices and decision criteria. The target population for the survey was commercial blueberry growers of Florida and Georgia. Questions focused on decision management based on bud stage, meteorological information, and informational resources. Further, questions were asked to identify varieties under frost protection, equipment used to frost-protect crops, and areas of frost protection. Before distribution to growers, the survey instrument was pretested, and input from the pretests was incorporated into the final survey language and design. The Florida and Georgia blueberry surveys were submitted to the University of Florida and University of Georgia Institutional Review Boards, respectively, where they received exempt status.
The Florida blueberry survey was distributed and collected at the Fall Blueberry Short Course of the Florida Blueberry Grower’s Association in Oct. 2015. In Georgia, surveys were administered and collected in collaboration with extension activities at county meetings in southern Georgia between Nov. 2015 and Jan. 2016.
Survey responses were coded and analyzed using SAS (version 9.4; SAS Institute, Cary, NC). Summary statistics, including mean and standard deviation, were calculated for each question, and questions were cross-tabulated and tested for significance using the Pearson’s χ2 test and the Kolmogorov–Smirnov test.
After we eliminated nonqualifying respondents, the final number in the survey was 106. Of the 106 respondents, 94 used active frost protection, with 46 from Florida (48.9%) and 48 from Georgia (51.1%). Most respondents were from farms of less than 51 acres (Table 1). Taking the midpoint for each of the acreage response intervals, we estimate that these Florida and Georgia growers manage about 1696 and 2850 acres, respectively, or about 37.7% and 19% of the total blueberry acreage in their respective states.
Most growers had experience with blueberry production, with the average number of years growing blueberries being 10.1 years in Florida and 13.1 in Georgia. A significantly larger (Wilcoxon rank-sum test, P = 0.05) proportion of Florida respondents (≈40%) had grown blueberries for 6 years or less, compared with less than 10% of Georgia respondents, reflecting the relative youth of Florida’s blueberry industry.
In both states, growers reported producing a mix of varieties, mostly SHB. The median number of varieties selected by respondents from the questionnaire list was three; in addition, many respondents, including those growers with rabbiteye blueberry (Vaccinium ashei) varieties, selected the “other” option (Table 2). The SHB varieties Farthing and Emerald were popular in both states; in addition, the majority of respondents in Florida reported growing Jewel, whereas the majority in Georgia reported growing Star.
Of the respondents who reported using at least one method of active frost protection, almost all (98.9%) used irrigation for frost protection (Table 3). The emphasis was on using a frost-protection system that minimizes production losses with the least amount of damage to the plants. The water source for irrigation was statistically different between Florida and Georgia, with Florida growers using more groundwater and Georgia growers using more on-farm ponds. (Table 4). Factors contributing to the predominance of well water as a source for freeze protection in Florida are the ease of access (with shallow aquifers available in most blueberry-producing Florida regions) and the water-permitting procedures (with permits usually making allowances for freeze protection that are not counted against the water permitted for supplemental irrigation). In Georgia, most growers who freeze-protect blueberry crops draw water from artificial ponds to reduce well water drawdown.
Sprinkler type also varied by state: wobbler sprinklers in Florida vs. impact sprinklers in Georgia (Table 4). The main difference in the choice of sprinkler type may be that Georgia tends to have longer, more-frequent freezes. Impact sprinklers have changeable nozzles to deliver greater volumes of water, which may be needed in more severe freezes, or to turn the pressure up at the pump to deliver more water without damaging the sprinklers.
Growers were asked to evaluate the importance of climatic and crop factors on the decision to initiate irrigation for frost protection (Table 5). Five factors (air temperature, dew point, wind speed, bud stage, and freeze duration) identified in the literature (Perry, 1998; Poling, 2008; Stewart et al., 1984) as important were presented to the growers. Results show that growers placed air temperature, dew point, wind speed, and bud stage as very important indicators, at 77.9%, 74.2%, 77.3%, and 85.1%, respectively, to activate frost protection. By stage 5 (late pink stage) (Fig. 1), less than 10% of the respondents would not frost protect and the majority would begin using the system when the temperature reached 31 °F. Although an irrigation system, once started, should be run until the temperature rises above freezing and the ice is melting or risk crop damage (Snyder and Melo-Abreu, 2005), only 56.8% of the respondents thought the duration of a freeze event was very important.
Location was the only farm characteristic that significantly affected the ranking of importance of the factors related to irrigation initiation. Specifically, wind speed was rated less important by Florida growers [62.8% of Florida respondents rated it as “very important,” compared with 91.1% of respondents in Georgia (χ2 test, P = 0.01)] since windy, advective freeze events could be a greater challenge for growers in Georgia. The analysis of the other farm characteristics (years in blueberry production and size of operation) did not reveal any statistically significant relations (Pearson’s χ2 test, P = 0.05).
Respondents revealed that they regularly access weather information, with 81% gathering information from on-farm meteorological devices. In addition, growers access university-supported systems frequently (Table 6). In Florida and Georgia, the Florida Automated Weather Network (FAWN) and Georgia Automated Environmental Monitoring Network (GAEMN), respectively, were used occasionally or regularly by ≈90% of respondents. FAWN and GAEMN were created to provide weather data from areas throughout Florida and Georgia underserved by the National Weather Service (Chevalier et al., 2012). This suggests growers are interested in comparing their meteorological monitoring devices with an off-farm source, which may add confidence to their decision process. This conclusion is similar to results found by Crane et al. (2010) that most farmers in Georgia used three sources of weather information to cross-check data from various sources. It should be noted that the survey only asked growers how often they received weather information from these sources and not what sources they use when determining the risk of a future freeze event.
Crane et al. (2010) reported that social networks are important for interpreting the weather information, which includes other growers, extension agents, and business interactions with suppliers and buyers. Our findings are consistent with these results, with the top-ranked resource consulted when deciding to turn on irrigation for frost protection being other growers (62%) (Table 7), who may be considered one of the most trusted sources of information by growers (Mase and Prokopy, 2014). This finding supports the “train the trainer” approach to extension programs (Warner et al., 2014), where extension agents rely on communicating with the opinion leaders who then communicate with the rest of the farming community, is reasonable.
State extension services also were consulted about irrigation initiation by more than 20% of growers. Although the use of extension service in general was similar in the two states, extension website resources specifically were significantly more popular in Georgia than in Florida [34% and 10.2% of respondents, respectively (χ2 test, P = 0.01)]. It is routine for Georgia extension agents to distribute University of Georgia extension publications to clients. In contrast, 51.3% of respondents in Florida, compared with 26.1% in Georgia, relied on other resources.
The growers acquire weather information from similar sources (e.g., on-farm weather stations and FAWN or GAEMN), and they also have similar responses toward the importance of following temperature, dew point, and bud stage as criteria for frost protection decision management. However, the survey identifies that growers have differing opinions on when they make the decision to frost protect.
In total, when asked to select only one, 28% of respondents reported using dry bulb air temperature compared with 72% using wet bulb temperature to determine when to use frost protection, with significantly more growers in Florida using the dry bulb temperature in irrigation initiation [38.5%, compared with 18.6% in Georgia (χ2 test, P = 0.05)].
Examining only the subset of respondents who indicated using wet bulb temperature in determining when to frost-protect crops, we found that the decision when to turn the system on and off varied (Fig. 2). The decision to turn the system on and off was grower-dependent regardless of input information and was influenced by region. The difference among growers is especially noticeable for tight cluster and early pink floral stages, when some growers turn their systems on when temperatures reach 34 to 36 °F, whereas others do not protect. For these two bud stages, responses of Florida and Georgia growers differed greatly. For the tight cluster stage, the majority of Georgia respondents selected “do not protect” (66.7%), whereas only 22.7% of respondents from Florida selected the same answer (χ2 test, P = 0.05). Similarly, for the early pink stage, 25% of the Georgia responses were “do not protect” compared with 4.8% of such responses in Florida. The willingness of some growers to protect at any stage could be due to an information deficit, where growers are unfamiliar with critical bud temperatures. However, given that 85.1% of respondents ranked bud stage as “very important” in the decision to activate irrigation, it could alternatively reflect grower preference for caution in protecting their investment, especially in Florida with higher early season prices.
Citation: HortTechnology hortte 28, 5; 10.21273/HORTTECH04008-18
There is also a range of temperatures used by growers to turn the system off (Fig. 3), which affects the duration of frost-protection irrigation. The decision of when to turn the frost protection on/off can significantly affect water use and potential cost savings to the growers.
Citation: HortTechnology hortte 28, 5; 10.21273/HORTTECH04008-18
Moreover, when asked how many times they had turned on irrigation due to cold temperatures in the past season, the number of events for which growers reported initiating frost protection differed among respondents. From Fall 2014 through Spring 2015, growers activated irrigation an average of 5.2 times (the median was 4), with a range from 0 to 20 times during the season. Georgia respondents reported engaging irrigation systems 5.4 times on average (the median was five), whereas for the Florida respondents the average was 5.0 times (the median was three) (Fig. 4). State-based differences in the number of events for which growers initiated irrigation for frost protection were statistically inconclusive. Although analysis of variance and median scores tests failed to reject the hypothesis that the responses from Georgia and Florida were drawn from a distribution with the same mean and median (P = 0.1), the two-sided Wilcoxon rank-sum test (P = 0.1) and the asymptotic Kolmogorov–Smirnov two-sample test (P = 0.5) rejected the hypothesis that the responses were from the same distribution. We further identified five smaller general regions with similar climatological conditions in the two states based on information from AgroClimate (2018) and even within these smaller regions, the range of the responses regarding the number of frost-protection irrigation events was large (Table 8).
Citation: HortTechnology hortte 28, 5; 10.21273/HORTTECH04008-18
The differences in the frost-protection practices may be attributed to growers’ perceptions concerning frost-protection risks and the impact of risk on yields (Fig. 5). Approximately 25% of respondents said that they could lose 20% to 29% of floral buds and ≈40% said they could lose 30% to 49% of buds and still have sufficient yield to produce a marketable crop.
Citation: HortTechnology hortte 28, 5; 10.21273/HORTTECH04008-18
Of the respondents who use irrigation for frost protection, approximately 69% operate diesel pumping systems, 18% operate electric pump systems, and 13% operate both diesel and electric pumping systems on their farms (Table 9). Many variable inputs are involved in identifying the cost of operation for an irrigated frost-protection system. Inputs include sprinkler rating (inches per hour), number of sprinklers, pump efficiency rating, gallons of diesel consumed to deliver an acre-inch of water, and the price of a gallon of water. Pump efficiency and gallons consumed per acre-inch of water information can be found in Martin et al. (2011), and sprinkler rating can be calculated from information in Tyson et al. (2006). As an example, sprinkler nozzles are rated for 0.15 inch/h, and the pump demand is 67.9 gal/min per acre in a 60 × 60-ft irrigation pattern. To calculate the water used by a sprinkler head, the following relationship can be used:
In a 60 × 60-ft pattern of sprinklers, there are 12.1 sprinklers in an acre (3600 ft2 ÷ 43,260 ft2 = 12.1). In a system pumping at 45 psi with 0 ft of lift, 0.95 gal of diesel will be consumed per acre-inch of water applied. Pump efficiency is considered at 90% efficiency and 1.11 is used as the multiplier, and fuel cost is $2.50 per gallon of diesel (price as of June 2017) (U.S. Department of Energy, 2018). Given these assumptions, the irrigation costs are:
To identify the operational cost using electricity, convert diesel to electricity by using a conversion factor of 14.12 gal/kWh (1.00 diesel factor × 14.12 electrical factor = 14.12):
If frost protecting 10 acres for 6 h, the cost of frost protection equals $478.20 or $270.00 for diesel or electric systems, respectively. Although growers are divided on the importance of freeze duration and the majority responded that marketable yield could be 51% to 70% bud survival, this suggests growers may delay frost protection to conserve water resources or expenses.
Given that the average number of frost events reported by the growers in both states was about five (for 2014–15), delaying switching the system on by an hour for each event means a reduction in irrigation by 5 h in the entire season. If sprinkler nozzles are rated for 0.15 inch/h, the pump demand is 67.9 gal/min per acre in a 60 × 60-ft irrigation pattern, or 4074 gal/h per acre, which for an irrigation run time of 5 h accounts for 20,370 gal/acre. Blueberry acreage varied among respondents, from less than five acres to more than 75 acres. For a grower with five acres of blueberry, a 5-h reduction in irrigation could translate into 101,850 gal of water-use reduction, whereas for a farmer with 75 acres, this would translate into 1,527,750 gal. For comparison, domestic per-capita water-use rate is 87 gal/d (Marella, 2014), meaning that a family of three people would use 31,777 gal over 1 year. Therefore, even a 5-h reduction in frost-protection irrigation can result in a significant water use savings. Such water-use reduction can be important in the areas like Florida, where traditional water supply sources are not able to sustain increasing water demand driven by the growing population (Economic and Demographic Research, 2018).
To meet consumer demands for blueberry, production in Florida and Georgia increased from 2000 to 2015 × 293% and 274%, respectively. Even though the price per pound decreased for Florida and Georgia by 34% and 15%, respectively (USDA, 2013, 2017), prices remain relatively high, making growers especially concerned about the risk of losing crops due to freeze. Our survey showed that Florida and Georgia growers frost-protect ‘Farthing’ and ‘Emerald’ (Lyrene, 2008; Williamson et al., 2014), which are early flowering varieties in both regions, susceptible to spring frosts. Further, growers are willing to frost-protect crops for the duration of the event and are not overly concerned about the length of time. The majority of growers are willing to wait until the temperature is above the 35 °F wet-bulb temperature before turning the irrigation system off, regardless of operational cost.
A descriptive study of fruit growers in Yakima Valley, WA, found that growers exchanged information when monitoring frost events (Stewart et al., 1984). Similarly, in our survey, growers were more inclined to consult with other growers for information over other resources. As Stewart et al. (1984) noted, growers tend to communicate with one another during the vigilant period of a freeze despite business competition.
As a result of the risks of freeze-related crop damage and the benefits and drawbacks of irrigation for frost protection, precision is important. The blueberry frost survey provides additional information about frost-protection use, decision criteria related to initiation and termination, sources of information used, and types of irrigation systems. This information can be used by extension to design a strategy to more efficiently disseminate frost warnings and irrigation advice to growers. The survey results also provide a baseline of irrigation frost-protection habits against which to compare future intervention, such as the dissemination of more precise irrigation recommendations.
Contributor Notes
This research was supported by a Florida Department of Agriculture & Consumer Services Specialty Crop Block Grant.
Corresponding author. E-mail: elizabethconlan@gmail.com.
