Estimation of stream temperature in firtina creek (Rize-Turkiye) using artificial neural network model

Area of Science:

• Environmental Science
• Ecology
• Aquatic Biology

Background:

• Water temperature is a critical environmental factor influencing aquatic ecosystems and organism health.
• Elevated water temperatures pose a significant mortality risk to sensitive species like salmonid fishes.
• Monitoring and predicting adverse temperature shifts are essential for aquatic habitat management.

Purpose of the Study:

• To develop and validate an artificial neural network (ANN) model for predicting stream temperatures.
• To utilize the ANN model for assessing suitable habitats for Sea Trout (Salmo trutta labrax) under adverse environmental conditions.
• To provide a tool for proactive management of aquatic ecosystems facing temperature fluctuations.

Main Methods:

• An artificial neural network (ANN) model was developed and tested.
• The model used historical water temperature, dissolved oxygen, and pH data.
• Meteorological data, including air temperature and rainfall, were incorporated as inputs.

Main Results:

• The ANN model demonstrated capability in predicting stream temperatures in the Firtina Creek.
• The model's predictions can help identify potential temperature-related risks to aquatic life.
• The study successfully simulated habitat suitability under various environmental scenarios.

Conclusions:

• Artificial neural network models are effective tools for stream temperature prediction.
• Accurate temperature forecasting is vital for the conservation of endangered aquatic species like Sea Trout.
• This approach supports informed decision-making for managing aquatic ecosystems under climate change.