Temperature simulator of fish stored in tubs and boxes (FishT-TaB Simulator)

Introduction

This model was developed and applied by members of the EFSA Working Group on the transport/storage of fresh fishery products during the preparatory work on the BIOHAZ Scientific Opinion on the use of "tubs" for transporting and storing fresh fishery products (EFSA-Q-2019-00053).

The heat transfer modelling was applied to estimate the fish surface temperature under the temperature related processes of cooling and then keeping the chill temperature of the fish (‘cooling’ process) and/or keeping the chill temperature of the fish (‘keeping’ process) of fish in ice (in boxes) versus in water and ice (in tubs) under the same conditions of transport/storage.

Two criteria were considered to define the type of fish for the assessment:

• The fat content of the fish was considered by selecting fish with a low (such as plaice having a fat and water content of 1-4% and 79-81%, respectively) and high (such as Atlantic salmon having a fat and water content of 10-20% and 60-70%, respectively) fat content, referred to as ‘lean’ fish and ‘fat’ fish.

• Two different dimensions and geometries were considered, i.e. small flat fish versus bigger fish with a broad oval cross-section. These were categorised as follows; ‘small’ fish (e.g. a plaice of a size class 4 having a weight of 150-300 g) and ‘medium-sized’ fish (e.g. salmon with a length of 50 cm). The size of the latter was restricted by the size of the fish box.

The model was validated using data from experiments of the 'Qualitubfish’ project dealing with small lean fish (plaice) (Bekaert et al., 2016 - https://pure.ilvo.be/portal/files/4784279/ILVO_mededeling_221_Qualitubfish.pdf). Some conditions were based on the actual measurements carried out in the project experiments and, therefore, different from those in the case-studies described for the ‘abusive’ scenarios. 

For reasonably foreseeable ‘abusive’ scenarios, the time/Temperature (t/T) profiles inside the containers were modelled using mathematical heat transfer models. Such abusive scenarios assume that the initial fish temperature equals 0°C upon arrival at the first on-land establishment (‘keeping’ process) or 7°C (‘cooling-keeping’ process) and the outside temperature is mostly at 2°C but includes some abusive temperature peaks up to 6°C. For boxes, the fish is surrounded by air and two layers of ice (bottom and top), while for tubs the fish is in water below an ice layer on the top of the tub without mixing).

Description

Naming of the files correspond to the temperature related processes, i.e. either ‘cooling-keeping’ process (referred to a ‘Cooling’) and ‘keeping’ process (referred to as ‘Keeping’) as well as the fat content (i.e. fat fish referred to as ‘Fat’ and lean fish referred to as ‘Lean’) and dimensions of the fish (i.e. medium-sized fish referred to as ‘Medium’ and small fish referred to as ‘Small’) as defined in the case-studies of the ‘abusive’ scenarios assessed in the Scientific Opinion. Additionally, a file gathering the model validation conditions is included, resulting in the following cases:

• Keeping_LeanSmall (case study #1k)
• Keeping_LeanMedium (case study #2k)
• Keeping_FatSmall (case study #3k)
• Keeping_FatMedium (case study #4k)
• Cooling_LeanSmall (case study #1c)
• Cooling_LeanMedium (case study #2c)
• Cooling_FatSmall (case study #3c)
• Cooling_FatMedium (case study #4c)
• Validation_LeanSmall

Each case contains the code to simulate in COMSOL® a tub and a box under the same conditions without the solution (extension mph) or including the solution (extension mph including the word “solved”) and a report generated by COMSOL (*.pdf). In addition, it includes two videos showing the temperature changes with time in a box (Validation_BOX.webm) and in a tub (Validation_TUB.webm) for the validation case.

For each container, the model consists of a partial differential equation simulating the temperature dynamics (along the time) and distribution (on space) with different thermodynamic parameters for each considered material (e.g. air, water, ice, lean/fat fish, poly-ethylene container material). Mesh discretizations have been selected to allow simulations in reasonable time. More refine meshes were tested without substantial change of results, particularly for cases with thin or small regions. The partial differential equation was solved using the software COMSOL® (COMSOL Multiphysics Reference Manual, version 5.4", COMSOL, Inc, www.comsol.com).

More details about numerical methods, discretization in time and space and further details about the simplifications and assumptions can be found in Section 2.3.2 and Appendix B of the opinion.

Reference

Bekaert K, Deloof D, Vandermeersch G, De Witte B, Vlaemynck G and De Reu K (Institute for Agricultural and Fisheries Research (ILVO)), 2016. Project Qualitubfish. Opvolging van de kwaliteitsveranderingen van pladijs gedurende de opslag in tubs. Institute for Agricultural and Fisheries Research (ILVO) report 221; ISSN 1784-3197; 35 pp. Available at https://pure.ilvo.be/portal/files/4784279/ILVO_mededeling_221_Qualitubfish.pdf