The Utility of Bitter Pit Prediction Models for Honeycrisp in Washington State

Written by Ines Hanrahan and Marcella Galeni, Washington Tree Fruit Research Commission, June 2019. 

 

 

 

At a Glance:

• Methods evaluated: Ethephon method; Passive method; Hot Water method; Penn State method.
• Time commitment: Inducing bitter pit takes three weeks; Penn State method (shoot growth combined with nutrient analysis) takes from two to three weeks.
• Accuracy: Bitter pit predictability of evaluated methods varied between years and orchards but is overall very low (0-33% correct prediction ≥90% confidence).
• Incidence rate: in-field bitter pit (0-22%); postharvest bitter pit (1-64%).
• Economics: The input to outcome ratio is best for the Hot Water method. The Penn State method is the most labor intensive and has an extra cost for the fruit mineral analysis.

Introduction:

All major apple varieties can develop preharvest and postharvest defects and disorders, rendering them not marketable or lowering their value drastically. The apple cv. Honeycrisp is especially susceptible to develop a multitude of pre- and postharvest physiological disorders, but due to its high value, this apple is often considered in replanting situations and has been one of the top varieties planted in Washington State in the second decade of this century (USDA, 2017).

Honeycrisp is prone to bitter pit. This poses a major problem in the cultivation of this apple. A high number of marketable fruits is lost every year to this particular damage. Bitter pit is a physiological disorder that begins in the orchards, but symptoms may appear and/or worsen in storage. In fact, clean fruit may be harvested and during storage, dark and sunken spots appear on the apple surface. If the apple is cut open, brown and corky tissue is revealed in the flesh directly under the skin. The ability to accurately predict bitter pit incidence in storage has been of high interest for the apple industry.

In other words, an easy and inexpensive test to be used close to harvest is needed to enable more accurate storage decisions, while providing producers with a reliable indication of potential storage losses due to bitter pit. For this reason, the Washington Tree Fruit Research Commission (WTFRC) compared four established methods used for this purpose in other growing regions in three years of study (Eksteen et al., 1977, Ostenson, 2012, Torres et al., 2015, Baugher et al., 2017).

Methods:

Between 2016 and 2018, nine orchards were utilized. In 2016 three orchards were sampled, with historically low, medium and high incidence of bitter pit. In that year, the Passive method, Ethephon method and Hot Water method were evaluated. In 2017, in addition to the three orchards sampled in the previous year, two new orchards were added. In three of the orchards all prediction methods were evaluated, in the remaining two sites, only the Penn State method was used. In 2018, four orchards were sampled, and the Passive method, Ethephon method and Penn State method were evaluated (Figure 1). Before the fruit were harvested, the amount of bitter pit (BP) present on the tree was assessed for each orchard.

Symptom free fruit were harvested between 4-6 feet on the west or south side of the tree. The apples were picked 2 weeks prior to the anticipated first commercial harvest. In all years, 20 representative trees were selected in each orchard. For the Penn State method, 20 apples were harvested, and five representative terminal shoot measurements were collected from each tree. For each, Ethephon method, Passive method and Hot Water method, 40 apples were harvested in every orchard. In all orchards untreated control (UTC) samples were collected. For the 2016 experiment, a total of 80 apples were harvested, and for the 2017-2018 experiments, 120 apples were harvested per location.

 

Within one day of harvest, fruit was prepared as follows:

1. All fruit were washed in 74°F (24°C) tap water and left to air dry.
2. Ethephon method (modified after Eksteen et al., 1977) (all years):
   7. A plastic box was filled with 2 gal (7.6 l) of water (temperature: 77°F) and 0.5 oz (15 ml) of Ethephon (1 gal= 0.25 oz) was mixed in.
   8. Every apple was dipped (2 sec) in the prepared solution.
   9. The samples were laid to dry on trays layered with paper towels.
3. Passive method (modified after Torres et al., 2015) (all years):
   1. Fruit was placed on trays and kept at room temperature (± 72°F).
4. Hot Water method (modified after Ostenson, 2012) (2016-2017):
   1. A cooler was filled with hot water and adjusted to ~ 120°F (49°C).
   2. The apples were submerged in the hot water and held under water with warmed up hard plastic ice packs. If temperature dropped below 115°F (46°C), hot water (~125°F) was added.
   3. After 30 minutes, the samples were taken out of the water and laid to dry on trays layered with paper towels.
5. All apples included in the experiment were placed in apple boxes and stored at room temperature for 3 weeks.
6. The fruit was evaluated after one, three, five, eight, eleven, fourteen, and twenty-one days.
7. Penn State method (modified after Baugher et al., 2017) (2017-2018):
   1. A commercial fruit dryer was preheated to 160°F (71°C).
   2. A fruit peeler was used to remove a 3/8” wide (1cm) strip of peel from around the circumference at the calyx end of the fruit.
   3. Samples were put on a drying tray and dried for 9 hours at 160°F.
   4. After 9 hours the samples were stored in Ziploc® bags and sent to a laboratory for nutrient analysis.
8. The UTC fruit was stored following the Honeycrisp storage recommendations (Hanrahan & Blakey, 2017) and apples evaluated after two, four, six, eight and twelve weeks of cold storage.

 

Results and Discussion:

Logistics

Between 2016 and 2018, we estimated the amount of labor needed, the time to prepare fruit for each method and the necessary time to accomplish the entire predictive evaluation. All methods were completed one week after commercial harvest. While the Ethephon, the Passive and the Hot Water method took up to three weeks after the initial harvest to obtain the results, the Penn State method took two days for preparation (nutrient analysis and shoot growth measurement) and approximately two weeks to obtain the results, considering the samples were sent to a laboratory for nutritional analysis (Figure 2). Theoretically, all methods provide results back in time before storage rooms are closed, since the majority of Honeycrisp get stored for one week at 50°F after commercial harvest, before transfer to the final storage temperature at 36°F.

 

 

Natural Bitter Pit Development Pattern

One aspect of bitter pit prediction that is not considered in any of the current methods involves the amount of fruit left in the orchard, due to bitter pit incidence at the time of harvest.

There was less fruit lost to bitter pit in the field, than in storage, in the three years of study (Figure 3). A range between 0-22% of fruit was left in the field due to bitter pit, but after 12 weeks in storage, 1-64% of the fruit harvested symptom-free, developed bitter pit. No correlation was found between the amount of fruit with visible bitter pit symptoms at the time of harvest and the amount of bitter pit that developed subsequently in storage. This means that one cannot predict storage bitter pit potential based on field symptom expression alone.

Based on our results it takes 10 weeks (approx. 2.5 months) for all bitter pit to express in storage under commercial RA (refrigerated air, 36⁰F) conditions. We do not have data on how long this process would take in CA (controlled atmosphere) or when 1-MCP is applied to fruit. However, if fruit is sold earlier than 10 weeks of storage (RA, no 1-MCP), the risk of additional symptom expression in transport and during display at supermarkets exists.

 

Presence of bitter pit in storage is typically detected after two weeks, with a fast-increasing rate of symptoms within six weeks. Subsequently the expression of bitter pit slows down until it stabilizes at 10 weeks in cold storage. We have recorded the final, full expression of symptoms at ten weeks of storage (Figure 4).

 

Accuracy of Bitter Pit Prediction Methods

In order to determine the accuracy of bitter pit prediction methods, we compared each method with the actual bitter pit incidence after storage, considering a range between 90 to 100% accuracy level.

In 60% of the orchards we observed a higher amount of bitter pit developing naturally in storage, than predicted by any of the methods tested (Figure 5).

In 2016, the Hot Water method predicted the precise amount of bitter pit for orchard 2.

In 2017, the Ethephon and Hot Water method predicted the exact amount of bitter pit developed in storage for orchard 2.

In 2018 there was nearly no bitter pit appearing in any of the methods tested. The Penn State method alone anticipated the appropriate incidence of symptoms in storage for orchard 9.

None of the methods tested showed a similar pattern of predictive power in all years. In 2016 and 2017, we selected the same three orchards, with historically different levels of bitter pit. In 2016 there was less bitter pit in storage than in 2017. Another factor for year to year variability could be the change of sampling size. We harvested 80 apples in 2016 and 120 apples in 2017. Increasing sample size may have led to a more accurate incidence determination, but it did not lead to an increase in prediction accuracy for any of the tested methods.

 

The Penn State method was used for two years only (2017, 2018). We harvested 20 apples per orchard, instead of the 60 apples suggested by Penn State Extension, in order to cut down on handling and sample preparation time. The vastly reduced sample size may have contributed to the fact that this method did not accurately predict bitter pit showing up in storage. Amongst all orchards tested, the Penn State method accurately predicted the same number of apples with symptoms in storage only for orchard 9 (Figure 6).

 

Overall, the Ethephon method accurately predicted bitter pit incidence in storage approximately 10% of the time, the Hot Water method 33%, the Penn State method 11%, while the Passive method never correctly predicted presence of bitter pit in storage (Figure 7). All methods tested underestimated the actual bitter pit risk. The Ethephon method 90% of the time, the Passive method 100%, the Hot Water method 67%, and the Penn State method 56% of time.

The Ethephon, Passive and Hot Water method never overestimated the amount of bitter pit in storage, and the Penn State method overestimated 33% of the time. Therefore, the Hot Water method was the most accurate between all methods tested, when using a 90% accuracy level. It correctly predicted actual bitter pit incidence in storage more often than the other methods.

 

Conclusion

None of the methods adequately predicted bitter pit incidence amounts in storage. Our data indicates that the Hot Water method correctly predicted development of bitter pit in storage more often than the other methods. Also, bitter pit expression in the field is not a good predictor of presence of bitter pit in storage. Major problems when utilizing the Penn State method are the missing mathematical formula used to calculate percentages of bitter pit, which cannot be surmised from the published table (Marini, R., Baugher, T., 2017), and the 2-fold higher cost compared to all other methods due to laboratory and sample preparation costs. In conclusion, the use of any of these methods to predict the potential of bitter pit development in storage is not recommended, because of lack of accuracy in prediction within all of the methods tested.

 

Acknowledgement:

WTFRC team: Kevin Köpcke, Felix Schumann, Manoella Mendoza, Mackenzie Perrault

We would like to thank the cooperating growers, Corina Serban (formerly Stemilt-Wenatchee), Garrett Bishop (G.S. Long, Yakima) for their generous donation of experimental fruit and laboratory costs.

Contact:

Ines Hanrahan

hanrahan@treefruitresearch.com

 

 

 

 

 

Literature Cited

Baugher, T.A., Marini, R., Schupp, J.R. and Watkins, C.B. (2017) Prediction of Bitter Pit in ‘Honeycrisp’ Apples and Best Management Implications. HortScience 52(10):1368-1374. <https://fruit.wisc.edu/wp-content/uploads/sites/36/2017/11/Prediction-of-Bitter-Pit-in-Honeycrisp.pdf>

Eksteen, G.J., Ginsburg, L. and Visagie, T.R. (1977) Post-harvest prediction of bitter pit. Dec. Fr. Gr. 27(1), 16-20. <https://www.researchgate.net/publication/265274782_Existing_Pre-harvest_Predictions_and_Models_for_Bitter_Pit_Incidence>

Hanrahan, I., Blakey, R., 2017. Honeycrisp storage recommendations revisited. WSU Tree Fruit. <https://treefruit.wsu.edu/article/honeycrisp-storage-recommendations-revisited/>.

Marini, R., Baugher, T.A., 2017. Fruit Disorders: New Tool to Assess the Potential for Bitter Pit in Honeycrisp. PennState Extension. <https://extension.psu.edu/fruit-disorders-new-tool-to-assess-the-potential-for-bitter-pit-in-honeycrisp>.

Ostenson, H., 2012. Learn to store Honeycrisp. Good Fruit Grower. Article by Hansen, M., 2012. <http://www.goodfruit.com/learn-to-store-honeycrisp/>.

Torres, E., Recasens, I., Peris, J.M., Alegre, S., 2015. Induction of symptoms pre-harvest using the ‘passive method’: An easy way to predict bitter pit. Postharvest Biology and Technology volume 101, 66-72. <https://www.researchgate.net/publication/268695547_Induction_of_symptoms_pre-harvest_using_the_’passive_method’_An_easy_way_to_predict_bitter_pit>

USDA, 2017. Washington Tree Fruit Acreage Report. <https://www.nass.usda.gov/Statistics_by_State/Washington/Publications/Fruit/2017/FT2017.pdf>.