The development and application of smart technologies in various fields is increasing every year. Different monitoring systems and sensors generate a large amount of data sets which allows to solve various tasks on data prediction and classification. This paper deals with data sets generated by a new tree monitoring system Open image in new window which evaluates in particular the sap flow density flux describing water transport in trees. The main task consists in prediction of the values of this characteristic which reflects the tree life state based only on observable air temperature during the predictable time interval and subsequent classification of trees according to some prespecified classes. The Fourier series based model is used to fit the data sets with periodic patterns. The multivariate regression model defines the functional dependencies between sap flow density and temperature time series. The paper shows that Fourier coefficients can be successfully used as elements of the feature vectors required to solve different classification problems. Artificial multilayer neural networks are used as classifiers. The quality of the developed model for prediction and classification is verified by numerous numerical examples.