THE USE OF ARTIFICIAL NEURAL NETWORKS (ANNs) FOR THE FORECAST OF PRECIPITATION LEVELS OF LAKE DOIRANI (N. GREECE)

摘要

Precipitation is a main part of the hydrologic cycle and constitutes the major source of the freshwater on the earth. Some of the most common and frequent natural hazards, such as droughts and floods, are related to the precipitation level. The knowledge of the precipitation's magnitude and the prediction of future values in an area are essential in order to plan and construct the appropriate hydraulic works. In this paper we make an effort to forecast the level of precipitation at the transboundary natural Lake Doirani which is situated at the borders of Greece and FYROM. In order to estimate future precipitation levels, we train an Artificial Neural Network (ANN) which will be able to predict future values, based on a time series of past precipitation, temperatures and lake levels. The forecast of the precipitation level can be used in order to prepare an effective management plan concerning the water resources of the lake.