"Acidity of Honeycrisp – Key to Eating Quality and Competitiveness of Northeastern Apple Growers"

Progress report for ONE24-449

Project Type: Partnership
Funds awarded in 2024: $29,851.00
Projected End Date: 07/31/2026
Grant Recipient: Cornell Cooperative Extension - Lake Ontario Fruit Program
Region: Northeast
State: New York
Project Leader:
Craig Kahlke
Cornell Cooperative Extension - Lake Ontario Fruit Program
Expand All

Project Information

Project Objectives:

This project seeks to measure the acidity of Honeycrisp and correlate high acid levels at harvest with very good to excellent eating quality after storage. In this study, we will sample fruit from a range of orchard blocks and separate them into low, medium and high acidity categories. At least three orchard blocks will be used per category. Acidity will be measured two ways – 1. The formal titratable acidity measurement that is time consuming, and 2. A newly acquired acidity meter that can assess samples quickly. In addition, fruit at harvest will also be treated or not treated with an inhibitor of ethylene production that slows down the rate fruit ripening. The fruit will be stored at the standard storage temperature of 38oF for up to 90 days. Changes of acidity in fruit from each of these categories will be measured at 30-day intervals during storage. Quality as well as informal and formal sensory evaluations (appearance, texture, mouth feel, juiciness, and overall acceptability) will be assessed after 90 days of storage. Outreach will consist of recommendations in fruit industry conferences and extension publications.

Introduction:

During the past decade, HC has continued to be the number one sought-after variety by consumers in the US. Until this crop year, demand and premiums have remained well above all other varieties, from grower returns to retail sales. Although all of the major apple-producing states have continued to plant, the low yield and low pack out of this variety continues to be a challenge for all of its growing regions in the US and Canada. Rootstock selection, microclimate, a high biennial bearing index, bitter pit, and multiple storage disorders are not even the full list of problems. With the overproduction of most commercial fresh market apple varieties, prices from the fall crop of 2023 through spring of 2024 have been low, across the region and in nearly all varieties. While HC returns have typically been as much as 3 ½ times higher than other varieties, it has fallen to levels dangerously close to the cost of production. If the “day of reckoning” has arrived, then researchers have to up their output at a higher level to ensure all loss and waste of this premium variety is minimized. 

In addition, increased production has also led to consumer dissatisfaction with fruit flavor. HC in the market can have crisp texture but disappointing flavor. The bland flavor has been associated with acidity – despite HC being a sweet fruit, acidity is the underlying positive flavor component. The challenge for the industry is to improve the flavor profile of this much wanted variety, especially with new club varieties entering the marketplace. Early research by Dr. Jim Mattheis (USDA Wenatchee) found that higher acidity at harvest was associated with higher acidity and better flavor after long-term storage. 

Determining harvest maturity at the correct timing for HC has proven difficult and inconsistent. In most varieties tested in maturity programs, internal ethylene concentrations (IECs), as measured in ppm by a gas chromatograph, can be correlated with maturity. In most varieties, the start of measurable (0.2-0.5 ppm) IEP signals the start of the ripening phase, in which the fruit are very near harvest for long-term, controlled atmosphere (CA) storage. In addition, there can be a steady or exponential rise in IEC in which fruit can become overmature and unsuited for CA and must be packed and marketed quickly. In HC, IEC does not correlate with maturity. There is no correlation with when to harvest for any storage regime, length, or marketing window using IEC in HC. Currently, a number of other quality indicators are used to try and determine ideal harvest timing for storage/market destination. One of the primary harvest indicators used in HC is color. In concert with acceptable flavor, color that “jumps out” when the yellow background color makes the red appear almost fluorescent/iridescent is ideal. The major flaw with using color is that weather patterns in the days/weeks prior to maturity can affect proper color. For instance, cooler and cloudier weather patterns can wash out existing color or cause poor color development altogether. Another quality indicator that can help time harvest is firmness. Again, weather patterns during a large part of the growing season or close to harvest can affect firmness.  Above average precipitation can cause fruit to be softer than target levels for CA storage. Drought or below average precipitation can cause fruit to be firmer than normal, and trick growers into harvesting too late if they are waiting for firmness to get into target range for a “normal” season. SSC (%), measured by a refractometer, indicates the overall sweetness of the fruit. As with the above quality indices, weather patterns can affect the SSC and perceived sweetness and flavor. Above average precipitation can cause fruit to have lower SSC and not taste as good. Conversely, drought or below average precipitation can cause fruit to have higher SSC than normal, and make for sweeter tasting fruit. Another quality indicator to help time correct harvest dates is the starch pattern index (SPI). A sample of apples are cut in half, and sprayed with an iodine solution. The conversion of starches to sugars causes the purple-colored iodine solution to turn ripe. Varying stages of ripeness (variety specific) show patterns in the fruit. At Cornell, we use a score of 1-8, with 1 being completely unripe (purple) and 8 being completely (over)ripe (white). Each variety (X) has a set standard (1<X<8) between for suggested harvest from on the early side (midterm to long-controlled atmosphere (CA) storage) to later harvest (to be packed relatively quickly and stored short term in regular air). While SPI can be a guideline to when to harvest for a planned storage/marketing plan, weather patterns can also swing it out of alignment with other maturity/quality indicators. For instance, a droughty season can cause a small amount of carbohydrates to be deposited into the fruit. When these fruit enter the maturation phase, the conversion of starch to sugars happens relatively quickly, and is not a great indicator of when to harvest and does not reflect true maturity. Another relatively new way to help gauge maturity is the delta absorbance (DA) meter. The loss of chlorophyll in the peel can be an effective indicator of fruit ripening. This vis–NIR portable spectrometer measures delta absorbance. It provides a non-destructive relative chlorophyll content index (Ziosi et al. 2008). Absorbance of radiation of wavelengths 670and 720 nm is converted to absorbance based on the Beer–Lambert Law and the index of absorbance difference (IAD) and represents the absorbance peak of chlorophyll-a and the background spectrum, respectively. The lower the number, the more mature the apple is (less chlorophyll remaining). Past research has shown that DA readings in our region for HC are often not in alignment with other maturity/quality indicators. Therefore, we aim to show that TA will be a tool in the toolbox that can be most in alignment with good eating quality. This is turn should increase farm income for growers of HC in the Northeast region of the US and Canada.

Cooperators

Click linked name(s) to expand/collapse or show everyone's info
  • Tyler Baker
  • Scott Henning
  • Chris Watkins

Research

Materials and methods:

Prior to the need for sample collection, a meeting was held with the cooperators to make sure we were on the same page for how we were going to treat the fruit at harvest and the methods that we were going to follow.

 

It was our hope that we would get the used autotitrator up and running prior to harvest but we were unable due to the fact that there were some parts that need to be replaced and that company who services the machine has changed hands several times.  However as outlined in the grant proposal we were able to purchase an additional ATAGO pocket brix/acid meter for the cooperators at Lake Ontario Fruit, Inc. (LOFI). Prior to harvest, I demonstrated the use of the meter to the QA crew at LOFI.

 

Also prior to harvest, I contacted the potential growers from whose farms we would be collecting samples.  We have been collecting Honeycrisp fruit for various research projects from some of the same farms and orchard blocks for several years in a row. Upon explaining the promise of measuring acidity and tying high acid levels to better storage quality the growers were happy to let me take samples and were looking forward to the results.

 

At the correct maturity timing for Honeycrisp harvest (September 6-12, 2024) for these potential research blocks, myself and my assistants collected 28 samples from 14 orchard blocks from 9 separate farms.  The 28 samples we're divided into half treated with 1-MCP, and half untreated.  Each sample consisted of 120 apples at harvest. 100 were taken 2 LOFI and treated or not treated with 1-MCP, and conditioned at 50F for seven days before being put into a 38 F room in regular air storage for the duration. The remaining 20 fruit per sample were taken to the harvest maturity lab at Orleans CCE.  Standard maturity measurements were taken, which included DA meter, firmness, total soluble solids (brix), and starch pattern index (SPI). For the brix and acidity readings, the juice from the hole punches taken for firmness was used. After calibration with distilled water, several approximately 5 ml of the sample juice were placed on the ATAGO brix/ acid meter on the button was pushed giving us the brix. After the meter was cleaned and dry with distilled water and dried with Kimwipes™, approximately 1gram of juice was placed in the beaker on the scale provided with the ATAGO test kit. Distilled water was added to bring the total volume weight to 50 grams. After a quick stir, approximately 5 ml of the solution was placed on the meter with the button pushed giving us the percent malic acid. Another button was pushed giving us the brix:acid ratio. All results were recorded.

 

As mentioned in the previous paragraph, the fruit where conditioned at 50F for a week before placing them in their recommended storage temperature of 38F. This conditioning of Honeycrisp is standard procedure to reduce the likelihood of the post harvest disorder soft scald/soggy breakdown. The researchers were curious whether or not the conditioning. An effect on the rate of acid reduction in the fruit.  Therefore, extra samples were harvested for three of the 14 orchard blocks.

At approximately 30 and 60 days postharvest, 20 apple samples each were taken from each of the 28 sample bags. These were brought back to the harvest maturity lab and tested for firmness, brix, total malic acid, and brix:acid ratios with the equipment and methods described earlier. All data was recorded.

 

At approximately 120 days postharvest, the remaining samples were taking out of storage. There was some loss to rots, and these were culled out prior to final use of  the samples from 10-20 for final acidity measurements at the harvest maturity lab. The remaining usable apples were tested for firmness, brix, total malic acid, and brix:acid ratios with the equipment and methods described earlier. All data was recorded. After this testing was concluded, a subset of samples representing at least two samples each from the three acid categories (high, medium, and low) corresponding to the malic acid levels described earlier were taken to the sensory analysis (SA) lab in the care of Dr. Robin Dando on the Cornell University campus. Craig worked with members of the  Dando lab to set up the SA. The SA is a scientifically validated voluntary taste testing by students. 125 students completed the SA. The SA Add a computer that accompanied the samples. Each sample also had a picture of a “typical apple”, whole. Each participant in the SA scored these samples, from the whole apple picture and its visual appearance, including taste, texture, aroma, and off flavors. Each sample had a random code assigned to it. At the time of this report, the PI is still waiting for the data analysis from the SA.

Research results and discussion:

The principle measurement for this entire study us the acidity, in the form of malic acid. From unpublished data from Dr. Jim Mattheis of USDA, there are three main acid categories for apples that are eaten for the fresh market.  High = >0.6% malic acid. Medium  is 0.4% or greater, but 0.6 or less. Low is under 0.4%. At harvest, the 14 orchard blocks had approximately 43% of the samples (6/14) at the high acid level, 36% of the samples (5/14), and approximately 21% of the samples (3/14) at the low acid level. Since half the samples just after harvest were treated with the ethylene blocker 1-MCP, and half were left untreated, there were 28 total samples for the subsequent analyses. Past research and observations have shown that 1-MCP helps retain acidity, even in samples stored in regular atmosphere. The primary benefit for most apple varieties is the retaining of firmness. However, Honeycrisp is one of the only apple varieties in which you do not see a significant difference in firmness between treated and untreated 1-MCP samples. Nonetheless, because of the maintenance of acidity benefits,  most storages treat their Honeycrisp with 1-MCP.  

Please see the data table below for the summary of the malic acid data for all the samples at all sampling periods.  One of the things that I learned during this first year of this two year project is that Honeycrisp orchard blocks that have very uniform trees and good pruning and good crop loads, have more even maturity (which is not something new) but tend to have higher acid levels. In contrast, trees with variable crop loads or heavy crop loads and less dwarfing rootstocks tend to have lower acid levels at harvest. There were some samples that retained their same acid level categories throughout the 120 day period, but all declined over timet. It is known that acid levels will always decline over time. The reason for a rise in some of the samples at some samplings is because at each sampling, we are doing destructive sampling and not sampling the same apples. We are sampling apples from the same harvest during subsequent testing periods, but not the same apples so there will be some inherent variation.

Sample delivery
Samples at Harvest, delivered to Lake Ontario Fruit, Inc. for 1-MCP treatment and storage.
A sample in the lab ready for acidity testing.
A sample in the lab ready for acidity testing.
Summary of All Acidity Data, Year 1    
    Total Acidity (Converter to Malic Acid), % (from 1:50 dilution)  
ID Trt Harvest 30 Days PH 60 Days PH 90 Days PH AVG    
A-1 NO 1-MCP 0.65 0.52 0.58 0.45 0.55    
A-1 1-MCP 0.65 0.33 0.62 0.39 0.50    
A-2 NO 1-MCP 0.87 0.68 0.78 0.75 0.77    
A-2 1-MCP 0.87 0.56 0.58 0.82 0.71    
A-3 NO 1-MCP 0.56 0.37 0.31 0.30 0.39    
A-3 1-MCP 0.56 0.21 0.41 0.34 0.38    
A-4 NO 1-MCP 0.68 0.37 0.66 0.45 0.54    
A-4 1-MCP 0.68 0.37 0.53 0.44 0.51    
A-5 NO 1-MCP 0.64 0.63 0.74 0.47 0.62    
A-5 1-MCP 0.64 0.58 0.73 0.68 0.66    
A-6 NO 1-MCP 0.43 0.53 0.70 0.73 0.60    
A-6 1-MCP 0.43 0.47 0.55 0.46 0.48    
A-7 NO 1-MCP 0.38 0.34 0.89 0.31 0.48    
A-7 1-MCP 0.38 0.38 0.51 0.46 0.43    
A-8 NO 1-MCP 0.56 0.54 0.59 0.62 0.58    
A-8 1-MCP 0.56 0.49 0.52 0.49 0.52    
A-9 NO 1-MCP 0.29 0.00 0.00 0.14 0.11    
A-9 1-MCP 0.29 0.23 0.26 0.19 0.24    
A-10 NO 1-MCP 0.64 0.43 0.59 0.59 0.56    
A-10 1-MCP 0.64 0.50 0.48 0.44 0.52    
A-11 NO 1-MCP 0.47 0.50 0.42 0.38 0.44    
A-11 1-MCP 0.47 0.41 0.43 0.38 0.42    
A-12 NO 1-MCP 0.47 0.46 0.35 0.35 0.41    
A-12 1-MCP 0.47 0.35 0.29 0.31 0.36    
A-13 NO 1-MCP 0.86 0.55 0.69 0.51 0.65    
A-13 1-MCP 0.86 0.27 0.72 0.61 0.62    
A-14 NO 1-MCP 0.30 0.44 0.30 0.32 0.34    
A-14 1-MCP 0.30 0.35 0.27 0.24 0.29    
AVG ALL 0.56 0.42 0.52 0.45 0.49    
MIN ALL 0.29 0.00 0.00 0.14 0.11    
MAX ALL 0.87 0.68 0.89 0.82 0.77    
AVG No 1-MCP NA 0.45 0.54 0.46 0.50    
AVG 1-MCP NA 0.39 0.49 0.45 0.47    
                 
Average all 4 Samplings 1-MCP no 1-MCP   Sensory Analysis Ithaca 12-4-24 (1-MCP)
High #>.6 2,5,13 2,5,6,13   High #>.6 2,5  
Medium #>.4, <.6 1,4,6,7,8,10,11 1,4,7,8,10,11,12   Medium #>.4, <.6 1,6  
Low # < 0.4 3,9,12,14 3,9,14   Low # < 0.4 9,11*  
          *11 actually averages medium  
                 

 

Research conclusions:

Again, this is just the first year of a two year project. We set out what we wanted to do and what we had outlined in the funded project proposal. We were able to successfully measure malic acid levels in a wide range of Honeycrisp samples. In addition, they divided up nicely into three acid categories. As expected, the acidity declined over time. We were able to observe some differences in planting systems and crop load that probably can be correlated with acid levels. As mentioned above, one of the things that we learned during this first year of this two year project is that Honeycrisp orchard blocks from high density planting systems (like tall spindle or super spindle) that have very uniform trees and good pruning and good crop loads, have more even maturity (which is not something new) tend to have higher acid levels. In contrast, trees with variable crop loads or heavy crop loads and less dwarfing rootstocks (center leader planting systems) tend to have lower acid levels at harvest. There were some samples that retained their same acid level categories throughout the 120 day period, but all declined over time. One of the recommendations I will make to growers this year is to try to prune as uniformly and narrowly as possible and to thin the trees to the correct crop load, which varies by rootstock, planting system, in row spacing, and between row spacing.

Participation Summary
9 Farmers participating in research

Education & Outreach Activities and Participation Summary

Participation Summary:

Education/outreach description:

If our hypothesis correlating the highest acid categories to the best eating quality prove to be true, we will begin by sharing the results of this research with colleagues in Eastern NY. If they can repeat our research and draw similar conclusions, more growers can be helped, and should lead to increases in pack out that should garner more return customers and higher prices. Other regions in the Northeast that grow HC commercially could follow with confirmational testing, such as in New England and the provinces of Ontario and Quebec in Canada. Better tasting fruit from across a large region can only strengthen demand and better returns from grower to retailer.

 

Any opinions, findings, conclusions, or recommendations expressed in this publication are those of the author(s) and should not be construed to represent any official USDA or U.S. Government determination or policy.