METODO SPEDITIVO PER LA STIMA DEI DATI MANCANTI E SUA APPLICAZIONE AI DATI PLUVIOMETRICI GIORNALIERI DEL BACINO DEL CANDELARO (PUGLIA)

The environmental time series always show the presence ofgaps, but to deal with data it needs to fi ll in the gaps estimating the lack values . At present, a lot of statis tical techniques are available to carry out that fi lling in, they goes fro m the simplest, e.g. the use of the sample mean, to the most sophisticated ones, e.g. the multiple imputation. The present paper proposes a brand new methodology fo r the lack data estima tion, that methodology tries to merge the obvio us advantages of the simplest techniques (e.g. their vocation to be easily implemented) with one of the strenght point of the newest techniques, namely the uncertainty assessment associated to the estimated value. The proposed method has been applied to the daily rainfall data coming f rom the Candelaro river basin (Puglia) and is composed by two stages; the fi rst one looks fo r the sites more similar to that whose time series presents a gap; the second stage carries out the actual estimation of the gap using a regressive non parametrical method.