Addressing short term heat peaks Introduction on the shelf life of - PowerPoint PPT Presentation

Addressing short term heat peaks Introduction on the shelf life of minced meat Material & K. Schmidt 1 , T. Lettmann 2 , R. Stamminger 2 Methods 1 University of Bonn, INRES IPE, Karlrobert-Kreiten Str. 13, 53115 Bonn CROP.SENSe.net & Nemaplot 2 University of Bonn, ILT, Nussallee 5, 53115 Bonn Modelling CCM, 4th International Workshop, Bonn, Germany Results Outlook Inverse objectives: Modelling the impact of high temperature peaks on microbial growth rates 1

Need for alternatives/extensions? MAP (“Modified Atmoshere Introduction Packaging”) extends shelf life of (minced) meat for a couple of Material & days. But how is the remaining Methods shelf life affected by careless consumer behavior during Modelling summer time with up to 50 ° C in a car? How are microorganism Results responding to heat events and Temperature response function how the decay kinetics of the 1.8 growth rate h-1, log cfu/g Outlook product are altered? 1.6 1.4 1.2 1.0 The common Arrhenius equation is 0.8 0.6 not applicable for such 0.4 0.2 temperature ranges 0.0 0 5 10 15 20 25 30 35 40 45 50 55 60 65 70 2 Temperature °C

Data source Material and methods Introduction MAP meat packages were stored both at constant temperatures in the range from 2 to 20 ° C and partly exposed Material & Methods to temperatures of 20/30/50 ° C, respectively for 3 hour periods at different points in time. Modelling Total viable counts (log cfu/g) were measured directly after delivery and continuously monitored in daily intervals until Results shelf life exceeded. Outlook The constant temperature trials are imposed to identify the basic influence of temperature on microbial growth. The variable temperature trials to identify response functions at higher temperatures. 3

Data source Experimental design Introduction Constant Variable temperatures temperatures Material & Methods 2 ° C 24h at 47h at 4 ° C Modelling 4 ° C 72h at 2 ° C 97h at 97h at 2 ° C 7 ° C Time 2 ° C 4 ° C Results 10 ° C 15 ° C Outlook 20 ° C 3h at 3h at 3h at 3h at 3h at 20 ° C 20 ° C 30 ° C 30 ° C 20 ° C Shelf life microbial growth was monitored until end exceeded 4

Basic model Introduction        N N        max 0 Log N log N  N = total viable counts   Material &  s    t 0   1 e Methods Modelling Results Microbial growth at constant temperature A logistical growth model is proposed as a first parsimonious Outlook 8 approach for modelling the 7 growth dynamics. log cfu/g More complex models, as 6 Richards, Gompertz or Baranyi 5 7°C equations are possible, if 4 deviance occur 3 0 50 100 150 200 time (hours) 5

Temperature response functions        N N        max 0 Log N log N    t  s    t 0   1 e d      Introduction   d T t dt r 0 Material & E Methods    A  Arrhenius function 1. d T k e RT r 0 Modelling     x x T T    opt     T T       T T max d T k e max opt    r max  T T  max opt Results with O„Neill function Outlook 2. 2   40      2 w 1 1    w   x 400 and         w Q 1 T T 10 max opt Exponential function       bT d T a exp 3. r 6

Temperature response functions Introduction Material & Methods 1.8 growth rate h-1, log cfu/g O'Neill <25°C 1.6 O'Neill >25°C Modelling 1.4 Arrhenius Exponential 1.2 Results 1.0 0.8 Outlook 0.6 0.4 0.2 0.0 0 5 10 15 20 25 30 35 40 45 50 55 60 65 70 Temperature °C 7

Thermal transition Ex.: heating up to 50 ° C for 3 hours & effective temperature the meat has been exposed to Introduction Modelling heat transfer of MAP foil Material & Methods heating 50 Modelling 40 Temperature °C MAP foil Results Meat surface 30 Meat core Outlook 20 10 0 0 2 4 6 8 10 12 Time (hours) 8

Model fitting Logistic growth with Logistic growth with Introduction O„Neill function Arrhenius function Material & Methods 8 8 7 Modelling 7 log cfu/g log cfu/g 2°C 6 2°C 6 4°C 4°C Results 7°C 7°C 5 10°C 5 10°C 15°C 15°C 20°C 20°C 4 4 Outlook R 2 =0.84 R 2 =0.86 3 3 0 100 200 300 400 0 100 200 300 400 time (hours) time (hours) 9

Biological times Introduction 9 8 Material & Methods log cfu/g 7 Model 2°C 6 Modelling 4°C 7°C 5 10°C Results 15°C 4 20°C Outlook 3 0 2 4 6 8 10 Biological time Transformation of the data to a time invariant scale permit the comparison of different temperatures 10

Biological time & model fitting Introduction 2°C 2/12°C 9 2/30/12°C 2/30/2°C Material & 2/50/12°C Methods 8 2/50/2°C 2/20/2/20 2/20°C Modelling 7 4/20/4/20°C log cfu/g 4/20°C Results 4/20/6°C 6 4/30/6°C 4/20/12°C Outlook 5 4/30/12°C 7/20/6°C 7/50/6°C 4 7/20/12°C 7/50/12°C 2/20/12°C 3 2/20/6°C 0 2 4 6 8 10 2/50/6°C Biological time 2/50/12°C 11

Validation 8 8 4/20/4/20 °C 4/30/6 °C Introduction 7 log cfu g-1 7 log cfu g-1 6 6 Material & Methods 5 5 4 4 0 20 40 60 80 100 120 140 160 180 200 0 20 40 60 80 100 120 140 160 180 200 Modelling 8 8 4/20/12 °C 2/50/2 °C 7 7 Results log cfu/g log cfu/g 6 6 5 Outlook 5 4 4 0 20 40 60 80 100 120 140 3 0 20 40 60 80 100 120 140 2/20/12 °C 8 8 7/50/12 °C 7 7 log cfu/g log cfu/g 6 6 5 5 4 4 3 12 0 20 40 60 80 100 120 140 160 180 200 0 50 100 150 200 250 300 Time (hours) Time (hours)

Reduced model Transformed logistic growth model without lag phase and 95% confidence intervals (CI) for constant Introduction temperature data Material & Methods Modelling 9 8 Results log cfu/g 7 Model Outlook 2°C 6 4°C 7°C 5 10°C 15°C 4 20°C R²=0.88 CI 3 0 1 2 3 4 5 13 Biological time

Temperature response function Introduction 0.20 growth rate h-1, log cfu/g Material & 0.15 Methods O'Neill w. lag 0.10 Arrhenius w. lag Modelling Exponential no lag O'Neill no lag Arrhenius no lag 0.05 Results 0.00 Outlook 0 5 10 15 20 25 Temperature °C The embedded temperature response function varies with the chosen model 14

Initial density problem Introduction 9 8 Material & Methods log cfu/g 7 Model 2°C 6 4°C Modelling 7°C 5 10°C Results 15°C 4 20°C CI Outlook 3 0 1 2 3 4 5 Biological time The variance found in the beginning maintains throughout the whole microbial dynamics. When shelf life times exceed, the range of a „ biological shelf life 15 time “ is in a proportional uncertainty

Conclusions, part I  A logistic growth model can be sufficiently fitted to microbial Introduction growth data. More complex models did not significantly improve the accuracy. Material &  All tested temperature response functions were applicable for Methods the range up to 20 ° C.  The O„Neill function only is suitable to address heat peaks and Modelling higher temperatures. Results  The growth model & embedded response function must be seen as one unit for any further application. Outlook  Some deviance in the comparison of prediction and observation indicate that such deterministic, continuous models are of limited use for prediction. The model behaviour is not flexible enough. 16

Conclusions, part I Introduction  The usually large variance in the initial microbial densities (cfu/g) values are maintained throughout the growth dynamics Material &  Any efforts in a more precise modelling are eliminated by this Methods uncertainty, which makes the prediction of an expiry time Modelling vague. Results Outlook 17

Estimating TVC by spectroscopy Use of sensor technology, here Introduction hyperspectral reflectance, to estimate initial microbial density Material & Methods of minced meat, N 0 . Modelling K. Schmidt, A.-K. Mahlein, U. List, J. Kreyenschmidt Results Working hypotheses: Outlook  Total viable counts affects the reflection signature of meat and can be classified  Potential disturbances by other factors can be discriminated  The specific signatures can be fitted to a common model, changes in parameter vectors can be associated to fine tuned scaling of microbial density 18

Background Using sensor technology to detect plant stress & phenotyping Introduction Material & Methods 0.6 0.5 Modelling Reflection 0.4 Results 0.3 Healthy 0.2 medium stress Outlook heavy stress 0.1 0.0 400 500 600 700 800 900 1000 Wavelength nm Objectives: Estimate the current stress conditions (or microbial contamination) with a reasonable confidence on the basis of spectral reflection of the visual (VIS) and near-infrared (NIR) wave lengths. 19

Material & Methods Introduction 3 fresh and 3 older (stored for 24h at 20 ° C) minced meat packages Material & have been scanned with an ASD Methods Field Spec sensor, Wavelength 400 to 1050 nm (VIS, NIR). Before Modelling scanning, half of the meat was Results separated for lab analysis. Outlook 20