mohapatra 2010 white button mushrooms

8/6/2019 Mohapatra 2010 White Button Mushrooms

http://slidepdf.com/reader/full/mohapatra-2010-white-button-mushrooms 1/7

E :

F o o

d E n g

i n e e r

i n g

&

JFS E: Food Engineering and Physical Properties

Postharvest Hardness and Color Evolutionof White Button Mushrooms ( Agaricus bisporus )DEBABANDYA MOHAPATRA , ZUBERI M. B IRA , JOE P. K ERRY , JESUS M. F RIAS, AND FERNANDA A. RODRIGUES

ABSTRACT: The quality evaluation of mushrooms was studied by storing fresh white button mushrooms ( Agaricus bisporus ) for 6 to 8 d, at various controlled temperature conditions (3.5 to 15 ◦ C) and measuring the instrumentaltextural hardness and color of the mushroom cap for different product batches. A nonlinear mixed effect Weibullmodel was used to describe mushroom cap texture and color kinetics during storage considering the batch variabil-ity into account. Storage temperature was found to play a significant role in controlling texture and color degrada-tion. On lowering storage temperature (i) the extent of the final browning extent in the mushroom after storage wasreduced and (ii) the rate textural hardness losses was slowed down. A linear dependence of the final browning index with temperature was found. An Arrhenius type relationship was found to exist between the temperature of storageandstorage time with respect to textural hardness.Theaverage batch energy of activation was calculated to be 207 ±

42kJ/mol in a temperature range of 3.5 to 20 ◦ C.Practical Application: This article evaluates how temperature abuse affects mushroom texture and color, applying methods that allowforthe consideration of thenatural product variability that is inherent in mushrooms. Its resultsapply to mushroom producers, retail distribution, and supermarkets for effective storage management.

Keywords: browning, mushroom, texture, variability

Introduction

M ushroom marketers often face difficulties in choosing a safestorage conditions on receiving different batches of mush-

rooms. Mushrooms may vary in their harvesting date and time,cultivated mushroom variety, harvest batches, storage conditionsadopted, and cold chain regime followed (Hertog and others 2007a; Aguirre and others 2008). Post-harvest, mushrooms immediately

start to soften and begin to brown in color due to enzymaticbreakdown of plant cells and loss of moisture through respiration(Burton and others 1987; Jolivet and others 1998; Brennan and oth-ers 2000; Zivanovic and others 2003; Zivanovic and Buescher 2004;Lespinardand others2009).This results in reduced product accept-ability, as consumer’s preference and demand is for white, unblem-ished and hard textured mushrooms. Additionally, bruising andstorage at elevated temperatures enhances the degradation pro-cess and reduces mushroom shelf life (Burton 1986). Consequently,monitoring cold-chain storage conditions that will preserve thequality of mushrooms is both critical and challenging (Aguirre andothers 2009).

Quality control during postharvest requires precise method-

ologies to estimate the acceptability of fresh produce of varying batches, growers,cultivation practices, and postharvest treatments.In an ideal situation, all products should arrive with the same ho-mogeneity as if it was from an experimental station unit, however,food retailers face an input of produce arising from different grow-ers, possibly harvested on different dates and locations and using

MS 20090761 Submitted 8/8/2009, Accepted 12/4/2009. Authors Mohapatra and Rodrigues are with Process and Chemical Engineering Dept. and au-thors Bira andKerry arewith NutritionalSciencesDept.,Univ. CollegeCork,Cork City, Ireland. Author Frıas is with School of Food Science and Envi-ronmental Health, Dublin Inst. of Technology, Cathal Brugha St., Dublin 1, Ireland. Author Mohapatra is currently with Faculty of Food Process-ing Technology and Bio-energy, Anand Agricultural Univ., Anand, Gujarat,India. Author Bira is also currently with Agricultural Research Inst., Ilonga,Kilosa, Morogoro, Tanzania. Direct inquiries to author Fr ´ ıas (E-mail: [email protected]).

very different cultural practices. Taken together, this has a signifi-cant effect on the homogeneity of the product and its’ time to reachthe limit of marketability (Schouten and others 2004; Hertog andothers 2007b). Moreover, there is biological variation contributedby micro nutrients, growing conditions, and so on for each batch of produce. Different units of an individual batch may behave differ-ently, even when stored under similar storage conditions (Brennanand others 2000; Hertog and others 2007a).

Modeling the quality kinetics of fresh products attempts to bet-ter understand the fate of quality during storage, taking not only the primary modeling variable (time) into account, but more im-portantly, the secondary variables that may be controlled during storage to optimally maintain the quality attributes of the product.Such information would be helpful to both producers and sellersin enabling them to optimize product storage conditions and inidentifying the significant factors affecting product shelf life. Mod-eling may also reveal theways in which variability affectsthe quality during operating storage conditions, which may in turn be used todefine limits beyond which the quality of product may be compro-mised within a certain tolerance (Lavelli and others 2006).

An assessment of fresh produce shelf life requires proper un-derstanding of the 2 phenomena affecting the process (i) biolog-ical metabolism, and (ii) underlying variability. Model building isemployed to assess the shelf life, normally based on experimentaldata that is generated through repetitive quality measurements, ei-ther by destructive or nondestructive methods carried out in real-situation or laboratory conditions. The repetitive measurementsform a longitudinal data structure which is well correlated withthe subject within a batch, but are independent of the intra batchvariability (Lammertyn and others 2003). Least squares regres-sion is commonly used to analyze the data by averaging repeatedmeasurements. Although this statistical method is robust to buildmodels within normal food experiments, it accumulates all the

variation in one error term and does not allow for the estimation of

C 2010 Institute of Food Technologists R Vol. 00, Nr. 0, 2010 — JOURNAL OF FOOD SCIENCE E1doi: 10.1111/j.1750-3841.2010.01518.xFurther reproduction without permission is prohibited



E

F o o d E n g n

e e r

n g &

Postharvest hardness and color evolution of white button mushrooms . . .

thedifferentpossiblesources of variation. While this is sufficient foruse with many experiments, it may be more desirable to estimateother and different sources of variability. In particular, postharvesttechnology is a field where this approach might prove to be inter-esting from a number of different perspectives, such as: (i) to beable to estimate the weight of different variability sources (withinbatch, between batches, between producers), which will help tomake clearer purchasing decisions, (ii) to identify if variability canbe reduced at any particular storage condition, and (iii) to eval-

uate through a scenario analysis if making an hypothetical opti-mization in the cold chain, this optimization will actually resultin an appreciable improvement of the shelf life taking account of product variability. Mixed-effects models may be useful for thosecases whereone hasto deal with within-subject,as well as between-subject variability, especially when having to deal with a biologicalcommodity. A mixed effects model has 2 components (i) fixed effectterm, which deals with the trend components, and (ii) random ef-fect term, which deals with subject specific intercepts and variance(Pinheiro and Bates 2000). Moreover, it allows for the presence of missing data and can allow for time-varying or unbalanced designs withunequal numbers of subjects across experimental groups (Pin-heiro and Bates 2000; Lammertyn and others 2003). Several studies

have been undertaken to predict the quality kinetics of fresh pro-duce using mixed effect models (Lammertyn and others 2003; Pi-agentini and others 2005; Latreille and others 2006; Schouten andothers2007; Aguirre andothers2009). A mixed effect model that ad-dresses a hierarchical level of variation has been employed by var-ious researchers (Fonseca and others 2002; Montanez and others2002; Ketelaere and others 2006). Mushrooms are known to havea very short shelf- life and susceptible to browning and moistureloss due to the enzymatic activity and lack of cell wall. The qual-ity deterioration is even faster at higher storage temperature con-ditions, due to enhanced metabolic activity. Therefore modeling the quality deterioration with respect to storage conditions pro-vides ample opportunity for the mushroom growers and marketersto modify the storage and handling conditions to have higher shelf life, thus reducing the economic loss. In this study, attempts weremade to model product instrumental texture and color character-istics to predict mushroom shelf life under different temperaturestorage conditions, taking batch variation into consideration, using a nonlinear mixed effect model.

Materials and Methods

C losed cup Agaricus Bisporus button mushrooms (white, close,uniform, clear, fresh, L value = 90 ± 5, a = 0.3 ± 0.8, b = 10 ±

2 and 0.331 ± 0.005 g solids/g water), sourced from the RenanireeMushroom Farm (Macroom, Ireland) and commonly destined forretail supermarket sales, were delivered to the laboratory using atemperature monitored distribution chain (6 ± 2 ◦ C, 80% ± 15%RH) in 7 kg crates without any individual packaging. Bruised anddamaged samples were discarded and samples for analysis weretaken at random from each batch of crates. Half of the mushroomsfrom the same batch were stored in temperature controlled coldrooms at different temperatures (5, 10, 15 ± 0.6 ◦ C) and the corre-sponding relative humidity was monitored (86% ± 7%). The otherhalf of the sample was kept in a domestic refrigerator that repro-duced the ideal storage temperature during retail and distributionof 3 to 4 ◦ C (3.5 ± 1.5 ◦ C, RH 92% ± 5%) and served as the con-trol sample to observe differences between ideal storage and thetemperature used for each individual batch tested. The temper-ature range of 3.5 to 15 ◦ C was chosen considering the practicaltemperature distribution chain of mushrooms, that is, during

postharvest handling, transportation, and storage. Texture and

color measurement were performed after the mushrooms reachedequilibrium temperature and every 24 h thereafter, until the end of the storage experiment, which varied between 6 to 8 d, depending on storage temperature, taking random samples from the lot. A to-tal of 14 batches of experiments were performed, covering a periodof 1 y of production.

Instrumental texture measurementTexture measurement is a complex measurement, especially in a

highly variable and anisotropic solid as mushrooms (McGarry andBurton 1994). Stored mushrooms were removed from storageand held at room temperature for 0.5 h before performing texturalassays. All such experiments were carried out using a texture pro-file analyser (Texture Expert Exceed, Stable Microsystems, Surrey,U.K.), with a 5 kg load cell following a modification of the methodproposed by Gonzalez-Fandos and others (2000). The crossheadspeed of the spindle for the pretest, posttest and test speed werekept at 1 mm/min. Only the mushroom caps were used for texturehardness measurements. To obtain a sample with the same tissueorientation and dimensions, a cylindrical sample of 10 mm diame-ter was bored out from the mushroom cap using a steel borer andcut to 10 mm length using a sharp knife and was then compressed

to 50% of the original height using a 35-mm aluminum cylindri-cal probe so as to achieve compression of the mushroom sample.Product hardness was the variable analyzed for each sample. Tests were performed on 5 replicate mushroom samples, from each stor-age condition, on each storage day, during the whole course of thetrial period, accounting for over 700 measurements.

Color measurementThe color of the mushroom cap was measured using a Minolta

Chroma Meter (Model CR-331, Minolta Camera Co., Osaka, Japan),using the Hunter Lab Color Scale. The color was measured at 3equidistant points on each mushroom cap using an aperture diam-eter of 4 mm. Total of 5 mushrooms were randomly selected from

each batch per day for the color measurement, accounting for over2800 measurements of color. Mushroom color has been commonly measured using the L value of the Hunter scale (Jolivet and others1998; Brennan and others 2000; Cliffe-Byrnes and O’Beirne 2007);however, some studies have pointed to changes in other param-eters of the hunter scale ( a ∗ and b ∗) related to browning (Burton1989; Vizhanyo and Felf oldi 2000; Aguirre and others 2008). To cap-ture this variation in a single index that would be related to a turntowards brown color, the browning index (BI) was calculated using the following expression (Maskan 2001; Bozkurt and Bayram 2006):

BI = 100 ×X − 0.31

0.17,

where

X =(a ∗+ 1.75 L )

(5.645 L + a ∗− 3.012b ∗),

L , a ∗, b ∗ values represent the lightness, redness, and greenness of the sample.

Mathematical modeling Themathematical model to predict mushroom shelf life was car-

ried out using the data generatedfrom measurement of the texturalhardness and color (as indicated by the browning index).

Model building was performed using the following procedure:1. An analysis of variance (ANOVA) of the quality parameters

clearly showed that they were all affected by temperature and

E2 JOURNAL OF FOOD SCIENCE — Vol. 00, Nr. 0, 2010



E :

F o o

d E n g

i n e e r

i n g

&


storage time ( P < 0.05). The primary modeling of the data was thenperformed using suitable mathematical models for individual tem-peratures and batch experiments. After a graphical, the first-ordermodel, the biexponential model, the logistic model and the Weibullmodel were used as candidate models to describe the kinetics of texture and browning. The most appropriate model which gavemaximum determination coefficient R 2 , a low standard error, lower Akaike’s InformationCriterion (AIC) and Bayesian Information Cri-terion (BIC) was chosen. The AIC and BIC are model discrimina-

tion criteria used for selection nonlinear models, which considerthe goodness of fit of the model and the number of parameters em-ployed. The smaller the value of the AIC and BIC the better a modelperforms (Pinheiro and Bates 2000).

2. The secondary modeling of the data considered two compo-nents: (i) dependence of the texture and browning primary modelparameters was described following equations 3 and 4 proposedin the temperature dependence section below, and (ii) batch vari-ation would be expected to follow the hypothesis of Hertog andothers (2007a) that each individual product and batch has pertur-bation at the initial state at which it is processed. Extra randomeffects were introduced following this and its addition tested us-ing a log-likelihood ratio test. A likelihood-ratio test is a statistical

test for making a decision between two models where the hypoth-esis is based on the value of the log-likelihood ratio of the 2 mod-els following a chi-square distribution (Bates and Watts 1988). Thelog-likelihood ratio test is a conservative test that will check for sta-tistical significance of adding further nested random effects to amodel (Pinheiro and Bates 2000). The test requires that the 2 mod-els must be nested, this is, that if one of the models can be trans-formed into the other by fixing one parameter.

3. Finally prediction plots using the Best Linear Unbiased Pre-diction (BLUP), which depict the model prediction of each individ-ual experiment considering the random effects assigned to it in themodel (Pinheiro and Bates 2000), were made to confirm the suit-ability of the candidate models.

4. An iterative procedure was used to find the best candidatesecondary model that could describe, with a minimum set of pa-rameters, that data that resulted from the experimentation.

Modeling textureThe best candidate primary model to describe the texture and

browning kinetics, in a similar way as with Kong and others (2007).The textural hardness of the mushrooms was described by the

Weibull model as follows:

H = B H + ( A H − B H )e − e lk H × t β H (1)

where H is the textural hardness of the mushroom cap, A H , and B H

are the initial and final hardness of mushroom cap during storage, t

isthe timeof storage (d), lk H is thenatural logarithm of therate con-stant of the reaction and β H is the dimensionless shape parameter.The shape parameter accounts for upward concavity of the curve(β H < 1), a linear curve ( β H = 1) as in case of first-order kinetics,and downward concavity ( β H > 1) (Pinheiro and Bates 2000).

Modeling colorThe browning index of the mushroom caps was analyzed using a

modified Weibull model, to force the rate constant parameter to bepositive:

B I = A B I + (B B I − A B I )e − e lk BI × t β BI (2)

where BI is the browning index, A BI is the upper asymptotic value

of the Weibull curve, B BI , is the initial value of the browning index,

t is the time of storage is days, lk BI is the log rate constant of thereaction, and β BI is the shape factor for browning index.

Temperature dependenceThe temperature dependence of the rate constant was modeled

following an Arrhenius relationship

k = k ref e −Ea R

1T −

1T r (3)

where k ref is the rate constant at the reference temperature T ref (5 ◦ C), Ea is the energy of activation of the process, and R is theuniversal gas constant (8.314 KJ Mol − 1 K − 1). In this way k ref and E a

are easy to interpret parameters and allow for comparison of thetemperature dependence of this process with other quality factors(chemical or not).

The temperature dependence of the A , B , and β parameter fol-lowed a polynomial relation:

y = a + b × T + c × T 2 (4)

where y is the parameter A , B , or β and a and b and c are the inter-cept, linear, and quadratic dependence of the parameter with tem-perature, respectively. Parameters statistically nonsignificant ( P >

0.05) were dropped from the model building.

Statistical analysisOn the basis of the primary models generated, the secondary

models were developed by including the random effect terms thataddressed batch and individual variance effects on quality evolu-tion. The nonlinear mixed modeling was performed using the nlmelibrary (Pinheiro and Bates 2000) from the R 2.9.1 software (R De-velopment Core Team 2007), for textural hardness and browning index.

Results and Discussion

Textural hardnessThe textural hardness kinetics of button mushrooms stored atdifferent temperatures are shown (Figure 1). It was evident that the

Time [Days]

H a r

d n e s s

[ N ]

20

40

60

80

86420

3.5°C5°C10°C15°C

Figure 1 --- Average textural hardness kinetics of mush-rooms at different storage temperatures 15 ◦ C, ∇ 10 ◦ C,+ 5 ◦ C, ◦ 3.5 ◦ C (control). Error bars represent 95% con-dence intervals based on the t -distribution for eachtime/temperature combination.

Vol. 00, Nr. 0, 2010 — JOURNAL OF FOOD SCIENCE E3



E

F o o d E n g n

e e r

n g &


while cap hardness could be maintained with storage at 3.5 ◦ C,higher temperatures produced a decline in textural hardness that was more pronounced with the increase in storage temperature.If storage temperature was increased to 10 ◦ C, after 4 dthe mush-rooms would have a texture different ( P < 0.05) from the control at3.5 ◦ C and if increased at 15 ◦ C after the 2nd day of storage.

The estimated fixed and random effect parameters of the finalmodel are outlined in Table 1 with 95% confidence intervals, allparameters being significant ( P < 0.05). Initial models were built

considering within-lot and within-batch variability similar to Mo-hapatra and others (2008). When performing individual fits in eachbatch, it was observed that the standard deviation of the estimatedpower terms was very low compared to the average (2.2 ± 0.2). Inthis way, the random effect associated to the β term was removedfrom the model.

As indicated in Figure 1, the kinetics, and therefore the rate con-stant, of texture decay was found to be dependent on the storagetemperature. To study this, an Ahrrenius plot with the random ef-

Table 1 --- Parameter estimates of the Weibull model forpredicting the textural hardness of mushrooms ( T ref 5 ◦ C).

Parameter Low 95% CI Estimate Up 95% CI

Fixed parametersB H [N] 13 16 18AH -B H [N] − 55 − 49 − 44β H [] 1.8 2.2 2.7lk H [] − 4.7 − 3.9 − 3.0Ea [KJ Mol − 1 ] 290 207 124

Random parameters: variability between batchesσ BH [N] 0.5 2.0 7.9σ AH − BH [N] 7 10 15σ lkH [] 0.83 1.20 1.74

fects associated to the k parameter of a model without temperaturedependence was built (see Figure 2) which confirmed this depen-dence. From the slope of the linear regression of Figure 2 an energy of activation of 190 ± 40 KJ/mol could be estimated. This value wasused as an initial estimate for the 1-step estimation of the modelparameters.

The activation energies at the 95% confidence level and the es-timates of the initial and final values of hardness and the power

1/Temperature [K−1]

l n ( k ) [ D a y s

]

0

2

4

0.003600.003550.00350

Figure 2 --- Arrhenius plot of the individually tted κ param-eter for each batch studied.

0

20

40

60

80

100

120

76543210

Storage Days

A

H a r d n e s s ,

N

0

10

20

30

40

50

6070

80

876543210

Storage Days

B

H a r d n e s s ,

N

0

20

40

60

80

100

120

9876543210

Storage Days

C

H a r d n e s s ,

N

Figure 3 --- Typical textural hardness ki-netics of mushroom batches at differ-ent storage temperatures with theirrespective control and best linear un-biased predictors (BLUP) of the modeldescribed in Table 1 (A) ♦ 15 ◦ C (ob-served), − 15 ◦ C (BLUP), (B) ◦ 10 ◦ C(observed), − 10 ◦ C (BLUP), (C) 5 ◦ C(observed), − 5 ◦ C (BLUP), 3.5 ◦ C (ob-served), --- 3.5 ◦ C (BLUP)

E4 JOURNAL OF FOOD SCIENCE — Vol. 00, Nr. 0, 2010



E :

F o o

d E n g

i n e e r

i n g

&


term for the final model are shown (Table 1). The activation energy for the loss of mushroom hardness (207 ± 42 kJmol − 1) value was well within the range of other quality characteristics for other re-portedforms of stored vegetables (Giannakourou and Taoukis 2003;

Time [Days]

B r o w n

i n g

I n d e x

[ ]

15

20

25

30

86420

3.5°C5°C10°C15°C

Figure 4 --- Average browning index kinetics of mushroomsat different storage temperatures: 15 ◦ C, ∇ 10 ◦ C, + 5◦ C, ◦ 3.5 ◦ C (control). Error bars represent 95% Gaussiancondence intervals based on the t -distribution for eachtime/temperature combination.

Piagentini and others 2005). The estimated power term (2.2 > 1)suggested that the kinetics had a downward concavity feature thatmade texture kinetics depart from conventional first-order kinet-ics. The best fitted values for mushroom textural hardness whenstored under different temperature-time for different batches of mushrooms are shown (Figure 3). It can be seen that the model de-scribes thekineticsand thedifferences between abuse storage tem-perature and control. Despite the natural variability, mushroomsabusedsuffera decrease in hardness that is apportioned to thetem-

perature abuse and that the model built in the present study is ableto reproduce.

The random effect terms in Table 1 suggest that the final valueof the mushroom hardness at the end of storage ( σ BH ) did not vary much among batches, compared to the variation in initial texturalhardness ( σ A − BH ), which is 5 times higher. The structure of the bestmodel fit and the estimated parameters point to the interesting hy-pothesis that as a result of storage, the variation between batchesof mushrooms will decrease. The variation of the reaction rate con-stant between batches showed a coefficient of variation of over the30%, (Table 1). This is characteristic of the high variability associ-ated to fresh produce for retail in general and in particular of mush-rooms (Aguirre and others 2009).

Browning index The kinetics of the average browning index for different temper-

atures of storage is shown in (Figure 4). From a graphical inspectionsimilar conclusions can be drawn as with the texture in respect tothe effect of temperature abuse during the storage of mushroomscan be concluded, with time and temperature having a significant

0

5

10

15

20

25

30

35

40

45

76543210

Storage Days

A

B r o w n

i n g

I n d e x

0

5

10

15

20

25

30

35

40

45

876543210

Storage Days

B

B r o w n

i n g I n

d e x

0

5

10

15

20

25

30

35

9876543210

Storage Days

C

B r o w n

i n g

I n d e x

Figure 5 --- Typical browning index kinet-

ics of mushrooms at different storagetemperatures tted to Weibull model: (A)♦ 15 ◦ C (observed), − 15 ◦ C (predicted), (b)◦ 10 ◦ C (observed), − 10 ◦ C (predicted),(C) 5 ◦ C (observed), − 5 ◦ C (predicted),

3.5 ◦ C (observed), --- 3.5 ◦ C (predicted).It can be seen that mushroom storagetemperature has an effect on the aver-age browning kinetics and how inher-ent mushroom variability inuences thewhole process.




E :

F o o

d E n g

i n e e r

i n g

&


Burton KS. 1986. Quality: investigations into mushroom browning. Mushroom J158:68–70.

Burton KS. 1989. The quality and storage life of Agaricus bisporus. Mushroom Sci12:287–93.

Burton KS, Frost CE, Atkey PT. 1987. Effect of vacuum cooling on mushroom brown-ing. Int J Food Sci Tech 22:599–606.

Cliffe-Byrnes V, O’Beirne D. 2007. Effects of gas atmosphere and temperature onthe respiration rates of whole and sliced mushrooms ( Agaricus bisporus ): impli-cations for film permeability in modified atmosphere packages. J Food Sci 72:197–204.

Fonseca SC, Oliveira FAR, Fr ıas JM, Brecth JK. 2002. Application of mathemati-cal modeling and computer simulation to the design of modified atmospherepackages accounting for product variability. In: Balsa-Canto E, Mora J, Banga JR,

Onate E, editors. Computational techniques in food engineering. Chapter 5. p 70–84.Giannakourou MC, Taoukis PS. 2002. Systematic application of time temperature in-

tegrators as tools for control of frozen vegetable quality. J Food Sci 67(6):2221–8.

Giannakourou MC, Taoukis PS. 2003. Systematic application of time temperatureintegrators as tools for control of frozen vegetable quality. Food Chem 83:33–41.

Gonzalez-Fandos E, Gim enez M, Olarte C, Sanz S, Sim on. 2000. Effect of packag-ing conditions on the growth of micro-organisms and the quality characteristicsof fresh mushrooms ( Agaricus bisporus ) stored at inadequate temperatures. J ApplMicr 89:624–32.

Hertog MLATM, Lammertyn J, Scheerlinck N, Nicola ı BM. 2007a. The impact of bio-logical variation on postharvest behavior: the case of dynamic temperature condi-tions. Postharvest Biol Technol 43:183–92.

Hertog MLATM, Scheerlinck N, Lammertyn J, Nicola ı BM. 2007b. The impact of bi-ological variation on postharvest behavior of Belgian endive: the case of multiplestochastic variables. Postharvest Biol Technol 43:78–88.

Jolivet S, Arpin N, Wicher HJ, Pellon G. 1998. Agaricus bisporus browning: a review.Mycol Res 102:1459–83.

Ketelaere BD, Stulens J, Lammertyn J, Cuong NV, Baerdemaeker JD. 2006. A method-ological approach for the identification and quantification of sources of biologicalvariance in postharvest research. Postharvest Biol Technol 39:1–9.

Kong F, Tang J, Rasco B, Crapo C. 2007. Kinetics of salmon quality changes during thermal processing. J Food Eng 83:510–20.

Lammertyn J, De Ketelaere B, Marquenie D, Molenberghs G, Nicolaı BM. 2003. Mixedmodels formulticategorical repeatedresponse: modeling thetime effectof physicaltreatments on strawberry sepal quality. Postharvest Biol Technol 30:195–207.

Latreille J, Mauger E, Ambroisine L, Tenenhaus M, Vincent M, Navarro S, Guinot C.2006.Measurementof the reliability of sensorypanelperformances.Food QualPref 17:369–75.

Lavelli V, Pagliarini E, Ambrosoli R, Minati JL, Zanoni B. 2006. Physicochemical, mi-crobial, and sensory parameters as indices to evaluate the quality of minimally-processed carrots. Postharvest Biol Technol 40:34–40.

Lespinard AR, Goni SM, Salgado PR, Mascheroni RH. 2009. Experimental determi-nation and modelling of size variation, heat transfer and quality indexes during mushroom blanching. J Food Eng 92(1):8–17.

LukasseLJS, Polderdijk JJ. 2003.Predictive modeling of post-harvest quality evolutionin perishables, applied to mushrooms. J Food Eng 59:191–8.

Maskan M.2001.Kineticsof colourchange ofkiwifruits duringhot airandmicrowavedrying. J Food Eng 48:169–75.

McGarry A, Burton KS. 1994. Mechanical Properties of the mushroom, Agaricus bis-porus. Mycological research 98(2):241–245.

Mohapatra D, Fr ıas JM, Oliveira FAR, Bira ZM, Kerry J. 2008. Development and vali-

dation of a model to predict enzymatic activity during storage of cultivated mush-rooms ( Agaricus bisporus ) J Food Eng 86:39–48.Montanez JC, Fr ıas JM, Fonseca SC, Luna R, Fernandez P, Oliveira FAR. 2002. Mathe-

matical modeling accounting for food product variability usingmixed effects mod-els. In the Proceedings of International Conference on Simulation in Food andBio-Industry, Cork, Ireland. p 35–9.

Muskovics G, Felf oldi J, Kov acs E, Perlaki R, K allay T. 2006. Changes in physical prop-erties during fruit ripening of Hungarian sweet cherry ( Prunus avium L.) cultivars.Postharvest Biol Technol 40:56–63.

PiagentiniAM, Mendez JC, Guemes DR,Pirovani ME. 2005. Modellingchangesof sen-sory attributesfor individual and mixed fresh-cutleafy vegetables. Postharvest BiolTechnol 38:202–12.

Pinheiro JC, Bates DM. 2000. Mixed-effects models in S and S-plus. New York:Springer-Verlag.

R Development Core Team. 2007. R: a language and environment for statistical com-puting. R Foundation for Statistical Computing, Available from: http://www.R-project.org.

Schouten RE, Jongbloed G, Tijskens LMM, Van Kooten O. 2004. Batch variability andcultivar keeping quality of cucumber. Postharvest Biol Technol 32:299–310.

Schouten RE, Huijben TPM, Tijskens LMM, Van Kooten O. 2007. Modeling quality attributes of truss tomatoes: linking color and firmness maturity. Postharvest BiolTechnol 45:298–306.

Vizhanyo T, Felf oldi J. 2000. Enhancing colour differences in images of diseasedmushrooms. Comp Elec Agric 26:187–98.

Zivanovic S, Buescher R, Kim SK. 2003. Mushroom texture, cell wall composi-tion, color, and ultrastructure as affected by pH and temperature. J Food Sci68(5):1860–5.

Zivanovic S, Buescher R. 2004. Changes in mushroom texture and cell wall composi-tion affected by thermal processing. J Food Sci 69(1):44–9.


mohapatra 2010 white button mushrooms

Documents