Structural trend and variability identification (STVI) method

This paper proposes an effective method for internal dynamic trend analysis on the mean value and variability trends on the basis of standard deviation. Thus, one is able to follow not only classical trends also coupled with it standard deviation trends, which provides additional information about the tendency of fluctuations around mean values. Thus, one is able to assess not only conventional trend analysis, but also trends concerning deviations from the mean value. The proposed method is referred to as the structural trend and variability identification (STVI), which is a new version of the existing innovative trend analysis (ITA) methodology. The STVI procedure provides to extract embedded trends of any duration along the hydro-meteorology time series records. In this paper, lag-one overlapping and non-overlapping classical and variability trends are presented for two finite durations of 10-year and 30-year, respectively. The application of the STVI method is presented for Danube River annual flows records of more than 160 years. The probability cumulative distribution function (CDF) for 10-year slopes confirms theoretically with Pearson CDF, which provides possibility of trend and variability future prediction possibility at different risk levels (return periods). It is found that in the future, there are structural trend possibilities at a set of 50% (2-year), 20% (5-year), 10% (10-year), 4% (25-year) and 2% (50-year) levels corresponding to slopes of 66.32 m ³ /year, 91.60 m ³ /year, 103.07 m ³ /year, 113.45 m ³ /year and 118.98 m ³ /year, respectively, for 10-year structural trends. Similar results are also presented for 30-year trend and variability slopes.