Can North America Grow Tasty Chestnuts? A Descriptive Sensory Analysis of Chestnut Cultivars

While some nuts, such as almonds and peanuts have extensive lists of descriptors, much less information is available on chestnuts.

Consumers purchase foods for a variety of reasons. Sometimes produce is purchased because of persuasive advertising, placement or the presentation of the commodity on shelves and in displays, visual characteristics of the product, packaging, or pricing. However, repeat purchases most often influenced by sensory characteristics such as appearance, aroma, flavor, and texture.

The primary human senses used to evaluate food are sight, hearing, touch, taste, and smell. Humans often use more than one of the senses to judge food and each sense may not be used independently. For example, nuts with dark-colored kernels are often perceived as having a stronger flavor than those with light-colored ones. Also, human senses can be fooled. For example, a red-colored drink with an orange flavor is often judged by consumers to be cherry-flavored.

Because consumers often lack experience in describing food, trained experts are often used to describe the sensory characteristics of foods. While some nuts, such as almonds and peanuts have extensive lists of descriptors, much less information is available on chestnuts. In Europe, chestnut sensory studies generally assess overall color, appearance, flavor, taste, texture and absence of pellicle intrusions. One Italian study also evaluated ease of chestnut peeling, aroma, firmness, sweet, salty and bitter attributes. Because different chestnut species and cultivars are grown and marketed in North America, a study was conducted to evaluate texture and flavor attributes of commonly grown chestnut cultivars using descriptive sensory analysis.

For this study, cultivar ‘Qing,’ ‘Eaton,’ ‘AU-Homestead’ and ‘Colossal’ chestnuts were harvested at the University of Missouri Horticulture and Agroforestry Research Center near New Franklin, Missouri. Marrone di Castel del Rio chestnuts were provided by a commercial grower in Isleton, California. Colossal chestnuts were also obtained from the same source in California and another producer in Owosso, Michigan to assess site variation. Immediately after harvest, chestnuts were washed, dried for 4 hours, sealed in polyethylene bags and stored at 40°F until sensory characteristics were evaluated by six highly trained panelists at the Kansas State University Sensory Analysis Center. Because panelists had limited experience with chestnuts, three orientation sessions were used to identify chestnut sensory attributes. Also, panelist’s understanding and the ability to use the identified sensory characteristics were carefully assessed during these orientation sessions.

After the preliminary sessions, panelists evaluated the samples in a room with natural and fluorescent lighting at 72 ºF and 55 % relative humidity. Before serving, chestnuts were cured for 3 days at 72 ºF. Nuts were scored and oven-roasted at 400ºF for 25 minutes. Five chestnuts were then placed in styrofoam bowls and covered with a lid for 10 min before serving. Panelists were instructed to use at least three samples during evaluations. Distilled, deionized water and unsalted crackers were served as palate cleansers. Each of the attributes was evaluated using a 0 to15 point intensity scale (0.0 = none to 15.0 = extreme). Eight samples were presented to the panelists during each session. All samples were evaluated in triplicate over six 1.5-hour sessions.

Twenty-three sensory attributes were identified to describe the chestnut cultivars in this study (Table 1). Chestnut cultivars were characterized by a slight nutty flavor (5.4 to 6.0 mean intensity ratings). Slightly lower impressions of brown, toasted and beany flavors were also perceived in chestnut samples (4.0 to 4.9 ratings). Lower, but detectable levels of hazelnut-like, almond-like, maple, buttery, caramelized, musty/dusty, musty/earthy, oily, sweet, sour, bitter and astringent flavors (< 3 ratings) were perceived. Mustard flavor was present at low sensory threshold (≥ 1 rating) in only one sample and ratings for floral/fruity, raw impression and fermented were very low (< 1 rating). These results demonstrate that cultivars commonly grown in North America have a complex flavor profile. 

Results from this study also indicated that six sensory characteristics of the chestnuts varied among cultivars (Table 2). ‘AU-Homestead’, ‘Eaton’ and ‘Qing’ chestnuts had higher peelability ratings than all other cultivars. These cultivars, in addition to ‘Colossal’ from Missouri and ‘Peach’, were also easier to peel than ‘Colossal’ harvested in California. Initial firmness ratings of ‘Colossal’ from Michigan and Missouri and ‘Marrone di Castel del Rio’ chestnuts were lower than all other cultivars. ‘Marrone di Castel del Rio’ chestnuts had a higher dissolvability rating than those of ‘AU-Homestead’, ‘Colossal’ from California, ‘Peach’ and ‘Qing’.

Three flavor attributes differed significantly in intensity ratings among cultivars. ‘Marrone di Castel del Rio’ chestnuts had a higher mustard intensity rating than those of all other cultivars. ‘Marrone di Castel del Rio’ and ‘Colossal’ from Michigan chestnuts were sweeter than those of ‘Peach’ and ‘Qing’. ‘AU-Homestead’ and ‘Peach’ chestnuts had a higher sour intensity rating than ‘Colossal’ chestnuts from Missouri. 

The reason ‘Colossal’ chestnuts from California were distinguished from those grown in Michigan and Missouri may be attributed to extreme climatic conditions before harvest. In Isleton, California the mean maximum daily temperature in August (no precipitation) and September was 91ºF (0.5 inches rainfall). In contrast, the mean maximum daily temperature where other ‘Colossal’ chestnuts were grown was 78 ºF and 62ºF in August and September, respectively, in Owosso, Michigan and 82ºF and 76ºF, respectively, in New Franklin, Missouri. Precipitation at both locations was over 2 inches for each month. Thus, high temperatures before harvest may have altered the development of flavors, resulting in relatively lower ratings of maple, brown, toasted, buttery, caramelized, and floral/fruity flavors for Colossal chestnuts from California as compared to those samples from the other two cooler locations.

The mustard flavor may be related to undetected microorganisms under the pericarp (shell) or within other tissue even though chestnuts with visible defects were discarded. In other studies, microorganisms have been isolated from the pericarp surface and cotyledons (edible “nut”) of chestnuts. In a Michigan survey, several microorganisms negatively impacted fresh chestnuts. A post-harvest disinfectant treatment can be used for long-term storage of chestnuts. However, in this study, chestnuts were not disinfected to prevent the introduction of non-chestnut flavors. 

While this study provides a sensory profile of various chestnut cultivars, it does not provide information on how well consumers like one particular cultivar as compared to others. This is a different type of experiment where untrained volunteers taste and rate how much they like the cultivar. This type of study is more difficult to conduct. For example, a review board must approve university surveys using human subjects and participant consent forms must be signed. Also, a small incentive, such as $5 or a prize is often used to entice participation in the study. Also, a controlled environment for testing must be used, samples randomized, and data statistically analyzed. This type of study requires more fresh product, time and labor than the previous descriptive type, but can provide very useful information to producers and sellers. However, for further expansion and growth of the North American chestnut industry, it is imperative that this type of information is known. 

Table 1. Descriptive terms and definitions used in the sensory analysis of chestnuts.

Attribute

Definition

Tactile

Peelability

Ease of peeling the shell and pellicle away from the nut

Texture

Initial firmness

Force required to initially bite through one piece of nutmeat using the incisors

Dissolvability

Degree to which the sample dissvolves or remains semisold when manipulated against the roof of the mouth with the tongue after seven chews

Flavor

Nutty

Intensity of all nutty characteristics including sweet, oily, light brown, slightly musty and/or buttery, earthy, woody, astringent and bitter flavors

Hazelnut-like

Sweet, light brown, oil somewhat woody aromatic associated with hazelnuts

Almond-like

Sweet cherry pit-like nutty aromatic associated with almonds

Maple

Sweet aromatic characterized as caramelized, woody and slightly green

Brown

Rich, full aromatic with a degree of darkness generally associated with canned pinto beans

Toasted

A brown, baked aromatic impression 

Buttery

Aromatics commonly associated with natural, fresh, slightly salted butter

Caramelized

Aromatic of a round, full-bodied, medium brown sugar

Raw

An unprocessed or an uncooked impression

Beany

Slightly brown musty/earthy, musty/dusty, slightly nutty and starchy aromatics associated with beans

Fermented

Sweet, slightly brown, overripe aromatics associated with fermented fruits, vegetables, or grains with yeasty notes

Mustard

Sweet, woody sour, vinegar-like, somewhat pungent, slightly horseradish-like aromatics associated with prepared mustard

Floral/fruity

Aromatics associated with flowers and non-citrus fruits

Musty/dusty

Aromatic associated with dry, brown soil

Musty/earthy

Aromatics of a damp basement or soil or decaying vegetation

Oily

Light aromatics associated with vegetable oil such as corn or soybean oil

Sweet

Basic taste associated with sucrose

Sour

Basic taste associated with citric acid

Bitter

Basic taste described as harsh with the taste simulated by solutions of caffeine

Astringent

Sensation of drying, drawing-up or puckering of any of the mouth surfaces


 Table 2. Mean sensory attribute intensity ratings for chestnut cultivars.

Descriptor

‘AU Homestead’

‘Colossal’ California

‘Colossal’ Michigan

‘Colossal’ Missouri

‘Eaton’

‘Marrone di Castel del Rio’

Peelability

14.86a

8.08d

9.36cd

11.67b

14.83a

8.72cd

Initial firmness

7.00a

6.86a

5.81b

5.94b

7.33a

5.86b

Dissolvability

6.50bc

6.39c

6.97ab

6.75abc

6.81abc

7.08a

Nutty

5.75

5.78

5.92

6.03

5.72

5.58

Hazelnut-like

1.92

1.94

2.19

2.06

2.06

1.61

Almond-like

1.94

1.81

1.86

2.06

1.92

1.92

Maple

1.61

1.78

1.94

1.92

1.72

1.50

Brown

4.78

4.42

4.72

4.89

4.86

4.69

Toasted

4.33

4.17

4.50

4.61

4.50

4.39

Buttery

2.72

2.56

2.78

2.94

2.72

2.92

Caramelized

2.36

2.53

2.72

3.14

3.17

2.42

Raw impression

0.28

0.47

ND

0.06

0.17

0.11

Beany

4.39

4.11

4.31

4.28

4.31

4.22

Fermented

ND

0.14

ND

ND

0.25

ND

Mustard

0.17b

0.33b

0.33b

0.58b

0.17b

1.58a

Floral/fruity

0.44

0.39

0.61

0.67

0.64

0.44

Musty/dusty

2.44

2.56

2.28

2.36

2.47

2.36

Musty/earthy

1.89

2.19

2.00

2.17

2.28

2.22

Oily

2.64

2.53

2.69

2.69

2.53

2.72

Sweet

2.50bcd

2.64abcd

2.78ab

2.67abc

2.69abc

2.92a

Sour

1.92ab

1.72bc

1.72bc

1.64c

1.72bc

1.78bc

Bitter

2.39

2.44

2.19

2.42

2.31

2.25

Astringent

1.75

1.75

1.75

1.86

1.81

1.89

Values with different letters in a row are statistically significant at P ≤ 0.05 by Fisher’s least significant difference (LSD) test. ND = Not Detected.

Source 

Related Events

Related Articles

Related Resources