Forecasting water temperature in northern Adriatic lagoons: a functional data approach

Abstract

Objectives: The aim of this thesis is to model the connection between weather variables such as air temperature and water variables such as water temperature, salinity and dissolved oxygen. This is being done to establish a method to predict water quality parameters which are relevant for shellfish culture.

Methodology: A basic analysis will be carried out using R. Afterwards an functional ARX model will be developed in order to predict the water temperature based on the air temperature and previous values for the water temperature. An addition of the variables salinity and the dissolved oxygen content is optional.

Experimental: Model was tested on data obtained from several station in northern Adriatic lagoons which have significant shellfish farms.

Research field: This thesis is focused on Blue Growth, more precisely Precision Shellfish Farming (PSF).

Expected results: The resulting model is supposed to be able to predict water temperature developments based on previous values and air temperature values in order to aid farmers to enhance their farming practice and increase efficiency.