The Effect of Serial Correlation on Nonparametric Trend Determination at Zenith Total Delay Time Series

GNSS technique is important in climate studies because of obtaining Zenith Total Delay which is significant parameter to determine weather conditions with analysis of GNSS observations. Interchanges of long term data can be detected via trend analysis produced by ZTD time series data. In this study, trend and serial correlation effect of time series of (IGS Repro1) were investigated using between 1995-2010 reprocessed ZTD data in the framework of COST Action ES1206, GNSS4SWEC (Advanced Global Navigation Satellite Systems Tropospheric Products for Monitoring Severe Weather Events and Climate). Mann-Kendall test was performed for nonparametric method. The autocorrelation coefficient was calculated to set the serial correlation. Trend-Free Prewhitening method has been applied to eliminate the correlation. The effect on the trend determination of serial correlation and the performance of Trend-Free Prewhitening method were tried to made examination by making re-trend analysis in series which eliminated correlation. As a result of the performed methods, serial correlation is substantial whole ZTD data. At the same time, there are trends in LAMA, ONSA and PENC stations after applying Trend-Free Prewhitening. In this work, MATLAB programming language was handled in all of the tests