Long-term trend analysis using discrete wavelet components of annual precipitations measurements in Marmara region (Turkey) |
| |
| Authors: | Turgay Partal Murat Küçük |
| |
| Institution: | Istanbul Technical University, Civil Engineering Department, Hydraulics Division 34469, Maslak/Istanbul, Turkey |
| |
| Abstract: | The purpose of this study is to determine the possible trends in annual total precipitation series by using the non-parametric methods such as the wavelet analysis and Mann-Kendall test. The wavelet trend (W-T) analysis is for the first time presented in this study. Using discrete wavelet components of measurement series, we aimed to find which periodicities are mainly responsible for trend of the measurement series. We found that some periodic events clearly affect the trend of precipitation series. 16-yearly periodic component is the effective component on Bal?kesir annual precipitation data and is responsible for producing a real trend founded on the data. Also, global wavelet spectra and continuous wavelet transform were used for analysis to precipitation time series in order to clarify time-scale characteristics of the measured series. The effects of regional differences on W-T analysis are checked by using records of measurement stations located in different climatic areas. The data set spans from 1929 to 1993 and includes precipitation records from meteorological stations of Turkey. The trend analysis on DW components of the precipitation time series (W-T model) clearly explains the trend structure of data. |
| |
| Keywords: | Long-term trend analysis The Mann-Kendall test Discrete wavelet transform Precipitation |
| 本文献已被 ScienceDirect 等数据库收录！ |
