Non Parametric Statistical Tests for the Analysis of Multiple-sensor Time Series of Remotely Sensed Data

The present paper presents an application of non- parametric statistical tests for the analysis of time series of remotely sensed data, acquired by different sensors of the same series. A set of points, extracted from the NOAA AVHRR GAC pathfinder dataset, which covers the period 1981-2001, was used as test dataset. Two tests, the Kruskal-Wallis and the Friedman test, have been used to address two significant issues of the analysis of long time series. The first experiment was aimed at determining whether observations acquired by different sensors of the same series, over the same stable target, belong to the same statistical population or not. The second experiment, tested whether seasonal phenological parameters of the vegetation, extracted from the same data, but after the application of different pre-processing algorithms, are significantly different.