Application of the Machine Learning Methods to Assess the Impact of Physicochemical Characteristics of Water on Feed Consumption in Fish Farms

Aug 6, 2026

Upload by Ivy

Machine learning methods, a prominent branch of artificial intelligence, support aquaculture by addressing complex challenges in managing production. This study evaluates how changes in the physicochemical characteristics of water relate to feed consumption using machine learning approaches. Eleven water parameters, including temperature, pH, dissolved oxygen, electrical conductivity, salinity, nitrite nitrogen, nitrate nitrogen, ammonium nitrogen, total phosphorus, total suspended solids, and biological oxygen demand, are analyzed. Temperature emerges as the most important driver for feeding. Outlet measurements for pH, EC, TP, TSS, salinity, and nitrite nitrogen rank as more informative than inlet values. Regression models using Random Forest, Gradient Boosting Machine, and eXtreme Gradient Boosting achieve higher predictive success, highlighting machine learning’s value for understanding fish feeding dynamics and improving farm management decisions.

LanguageEnglish

CategoryResearch & Report

FormatPDF

Pages9

Reading Time23 min

AccessFree to download

Introduction

Aquaculture environmental impacts

Importance of sustainability and resource monitoring

Artificial intelligence and machine learning in fish farms

Materials and Methods

Data and physicochemical variables

Machine learning models

Results

Feature importance and key parameters

Model performance for feed consumption prediction

Discussion

Implications for aquaculture management

Journal of Agricultural Sciences (Tarim Bilimleri Dergisi) 2025, 31 (1) : 71 – 79 DOI: 10. 15832/ankutbd.1470111 Journal of Agricultural Sciences (Tarim Bilimleri Dergisi) J Agr Sci-Tarim Bilie-ISSN: 2148-9297 [jas.ankara.edu.tr](jas.ankara.edu.tr) Application of the Machine Learning Methods to Assess the Impact of Physicochemical Characteristics of Water on Feed Consumption in Fish Farms Nedim Özdemira, Mustafa Çakirb, Mesut Yilmazc, Hava Şimşeka, Mükerrem Atalay Orald, Okan Orale* aFaculty of Fisheries, Muğla Sıtkı Koçman University, Muğla, TURKEY bIskenderun Vocational School of Higher Education, İskenderun Technical University, İskenderun, TURKEY cFaculty of Fisheries, Akdeniz University, Antalya, TURKEY dFaculty of Maritime, Akdeniz University, Antalya, TURKEY eFaculty of Engineering, Akdeniz University, Antalya, TURKEY ARTICLE INFO Research Article Corresponding Author: Okan Oral, E-mail: [okan@akdeniz.edu.tr](okan@akdeniz.edu.tr) Received: 17 April 2023 / Revised: 25 July 2024 / Accepted: 31 July 2024 / Online: 14 January 2025 Cite this article Özdemir N, Çakir M, Yilmaz M, Şimşek H, Oral M A, Oral O (2025) . Application of the Machine Learning Methods to Assess the Impact of Physico-chemical Characteristics of Water on Feed Consumption in Fish Farms. Journal of Agricultural Sciences (Tarim Bilimleri Dergisi), 31(1):71-79 . DOI: 10. 15832/ankutbd.1470111 ABSTRACT Machine learning methods, which are one of the subfields of artificial intelligence and have gained popularity in applications in recent years, play an important role in solving many challenges in aquaculture. In this study, the relationship between changes in the physico-chemical characteristics of water and feed consumption was evaluated using machine learning methods. Eleven physico-chemical characteristics (temperature, pH, dissolved oxygen, electrical conductivity, salinity, nitrite nitrogen, nitrate nitrogen, ammonium nitrogen, total phosphorus, total suspended solids, and biological oxygen demand) of water were assessed. Among all the measured physico-chemical characteristics of water, temperature was determined to be the most important parameter to be evaluated in fish feeding. Moreover, pH2, EC2, TP2, TSS2, S2 and NO2-N parameters detected in the outlet water are more important than those detected in the inlet water in terms of feed consumption. Through regression analysis carried out using machine learning methods, the models developed with Random Forest, Gradient Boosting Machine and eXtreme Gradient Boosting algorithms exhibited higher success rates in predicting feed consumption compared to the other models. The present study highlights the pivotal role of machine learning methods in enhancing our understanding of fish feeding dynamics based on physico-chemical characteristics of water, thus contributing significantly to aquaculture management practices. Keywords: Aquaculture, Feed intake, Artificial intelligence, Rainbow trout, Sustainability 1. Introduction Fish farms release various amounts of waste into the aquatic environment, which increases the necessity of examining and analysing the impact of aquaculture on the environment (Ahmad et al. 2022) . The polluted environment primarily harms biodiversity, disrupts ecological balances and prevents sustainability by affecting production (Leaf & Weber 1998; Sharma & Birman 2024) . The total trout production of Turkey was 145649 tons in 2022, and Muğla province contributed to this production with 18.2% of the total amount (Çöteli 2023). The Eşen River is at the heart of intensive trout production in Muğla province (Sezginet al. 2023; Koçer et al. 2010; Pulatsü & Yıldırım 2011) . For sustainability, it is essential to ensure that natural resources are used effectively and efficiently, in a balanced way, and in harmony with nature (Qin et al. 2024; Moldan et al. 2012) . It is necessary to know the potential and structure of natural resources well and to observ

FAQs

Which water physicochemical parameters are used to assess feed consumption?

The study evaluates eleven parameters: temperature, pH, dissolved oxygen, electrical conductivity, salinity, nitrite nitrogen, nitrate nitrogen, ammonium nitrogen, total phosphorus, total suspended solids, and biological oxygen demand.

What water parameter is identified as most important for fish feeding?

Temperature is determined to be the most important parameter for evaluating fish feeding among all measured physicochemical characteristics.

Which machine learning algorithms perform best in predicting feed consumption?

Random Forest, Gradient Boosting Machine, and eXtreme Gradient Boosting models show higher success rates in predicting feed consumption compared to other models.

Aug 6, 2026

Upload by Ivy