Browsing by Author "Karataş, B."

Now showing 1 - 2 of 2

Examination of Behavioral Traits of Monocultures and Polycultures of Two Different Trout Species (Oncorhynchus Mykiss, Salmo Sp.) at Different Ratios Depending on Various Factors
(ABADER (Adıyaman Bilimsel Arastırmalar Dernegi), 2021) Karataş, B.; Arabaci, M.
The present study aimed to determine both the effects of monocultures of rainbow trout (R) and brown trout (B) and polycultures (R 75%-B 25%; R 66%-B 34% and R 50%-B 50%) on behavior. In the study, the area used vertically by the fish in the tank, the mobility rate of fish in the tank, the rate of tendency of fish in eating as soon as they were fed, whether the fish test the feed, the interspecies feed competition, the time they start to take the first feed, the duration of the feed consumption of the fish and the feed area of the fish have been considered as behavioral evaluation criteria. The trout were monitored with a camera to determine their behavior. Considering all behavioral criteria, the best polyculture rate was determined as R 66%- B 34%. Brown trout were found to be more mobile and exhibit more relaxed behavior compared to other groups in polyculture. In addition, interspecies feed competition was mostly encountered in this group. As a result, in this study, in which two different trout species were monocultured and treated at different polyculture ratios, the main factor causing behavioral change in fish was found to be the different stocking rates of fish to each other in the same tank. Different rates applied in polyculture caused unpredictable changes in behavior in both species. The ratio of fish used in polyculture was found to be a considerable factor affecting the final product and their behavior for aquaculture. © 2021, ABADER (Adıyaman Bilimsel Arastırmalar Dernegi). All rights reserved.
Using Different Machine-Learning Algorithms To Predict Dissolved Oxygen Concentration in Rainbow Trout Farms
(Central Fisheries Research Institute, 2026) Karataş, B.; Çakmakçı, C.; Yücel, E.S.; Demir, M.; Şen, F.
Dissolved oxygen (DO) is a vital parameter in intensive rainbow trout aquaculture, directly influencing fish growth, health, and survival. As such, accurate monitoring and prediction of DO levels are crucial for ensuring sustainable and efficient aquaculture practices. This study assessed and compared the predictive performance of four machine learning algorithms Multivariate Adaptive Regression Splines (MARS), Random Forest (RF), Extreme Gradient Boosting (XGBoost), and Averaged Neural Networks (avNNet) in estimating DO concentrations based on a range of water quality parameters. A total of 120 samples were collected from 12 rainbow trout farms across Türkiye. The input variables included suspended solids, electrical conductivity, turbidity, nitrate, nitrite, ammonia, ammonium, orthophosphate, pH, water temperature, and total phosphorus. DO levels ranged between 8 and 15 mg/L. Model performance was evaluated using Mean Absolute Error (MAE), Root Mean Square Error (RMSE), Mean Absolute Percent Error (MAPE), and the coefficient of determination (R²). All models demonstrated strong predictive accuracy, with XGBoost achieving the best overall performance (MAE: 0.44, RMSE: 0.58, MAPE: 0.04, R²: 0.78), followed by RF, avNNet, and MARS. These findings highlight XGBoost as a robust predictor of dissolved oxygen levels in aquaculture systems, which may contribute to improving water quality management and increasing productivity in rainbow trout aquaculture. © 2026, Central Fisheries Research Institute. All rights reserved.