Developing a new tool based on a quantile regression mixed-TGC model for optimizing gilthead sea bream (Sparus aurataL) farm management

Abstract

In this work, a seasonal quantile regression growth model for the gilthead sea bream (Sparus aurata L) based on an aggregation of the quantile TGC models with exponent 1/3 and 2/3, named the “Quantile TGC‐Mixed Model”, is presented. This model generalizes the proposal of Mayer, Estruch and Jover (Aquaculture, 358‐359, 2012, 6) in the sense that the new model is able to describe the evolution of weight distribution throughout an entire production cycle, which could be a powerful tool for fish farm management. The information provided by the model simulations enables us to estimate total fish production and final fish size distribution and helps to design and simulate production and sales plan strategies considering the market price of different fish sizes, in order to increase economic profits. The most interesting alternative in the studied case results in sending all production when 0.25 quantile fish reach 600 g, although on each fish farm it would be necessary to evaluate optimum strategy depending on its own quantile regression model, the production cost and the market price.

Farm‐scale models in fish aquaculture – An overview of methods and applications

“…34 In some FSMs, variability is represented among individuals or classes of individuals. 30,57,63,72,77 A common method is to use probability distributions, usually Gaussian, for one or more parameters of the growth model (e.g. assimilation efficiency) and to perform as many simulations as there are individuals in the cohort (e.g.…”

“…The fourth step of FSM development is model application. The FSMs reviewed were developed for various purposes, including estimating or comparing production 31,34,76 ; socioeconomic outputs 51,69 or environmental impacts 57,68 of scenarios; optimising production, resource use or facilities 30,35,67,72 ; verifying compliance with effluent-discharge regulations 43 ; and facilitating site selection. 63 In summary, the most frequent applications are assessment of production or ecological carrying capacity (with a focus on eutrophication impacts), as well as economic assessment or optimisation.…”

“…worst‐case, average, best‐case) are sometimes used to resolve variation in growth among individuals; however, results of these scenarios remain determined by the parameter values and data set beforehand 34 . In some FSMs, variability is represented among individuals or classes of individuals 30,57,63,72,77 . A common method is to use probability distributions, usually Gaussian, for one or more parameters of the growth model (e.g.…”

The Quantile Regression Mixed Growth Model Can Help to Improve the Productivity in Gilthead Sea Bream (Sparus aurata) and European Sea Bass (Dicentrarchus labrax) Growing in Marine Farms

“…Finally, a new Thermal-unit Growth Coefficient (TGC) model based in the sum of daily temperature (Equation 3) was reported by Cho & Bureau [4] for salmonids, and adapted by Mayer et al [5] in gilthead sea bream using the effective temperature (> 12ºC). 1 1…”

Bioeconomic modeling of optimal harvest time in Nile tilapia (Oreochromis niloticus) considering size heterogeneity and minimum marketable size

“…Continuous models compared to discrete models offer the advantage of predicting future values because they can simulate population variation based on weight and time by using a normal distribution, which is relevant for sowing and selective harvesting (Arnason et al, 1992). Other authors have applied for the same purpose, the Generalized Linear Model Method (Briceño et al, 2010) and quantile regressions (Mayer et al, 2009;Estruch et al, 2017;Jover, 2017).…”