 |
| Titel |
0.1 Trend analysis of δ18O composition of precipitation in Germany: Combining Mann-Kendall trend test and ARIMA models to correct for higher order serial correlation |
| VerfasserIn |
Julian Klaus, Kwok Pan Chun, Christine Stumpp |
| Konferenz |
EGU General Assembly 2015
|
| Medientyp |
Artikel
|
| Sprache |
Englisch
|
| Digitales Dokument |
PDF |
| Erschienen |
In: GRA - Volume 17 (2015) |
| Datensatznummer |
250114169
|
| Publikation (Nr.) |
 EGU/EGU2015-14473.pdf |
|
|
|
|
|
| Zusammenfassung |
| Spatio-temporal dynamics of stable oxygen (18O) and hydrogen (2H) isotopes in precipitation
can be used as proxies for changing hydro-meteorological and regional and global climate
patterns. While spatial patterns and distributions gained much attention in recent years
the temporal trends in stable isotope time series are rarely investigated and our
understanding of them is still limited. These might be a result of a lack of proper trend
detection tools and effort for exploring trend processes. Here we make use of an
extensive data set of stable isotope in German precipitation. In this study we investigate
temporal trends of δ18O in precipitation at 17 observation station in Germany between
1978 and 2009. For that we test different approaches for proper trend detection,
accounting for first and higher order serial correlation. We test if significant trends in
the isotope time series based on different models can be observed. We apply the
Mann-Kendall trend tests on the isotope series, using general multiplicative seasonal
autoregressive integrate moving average (ARIMA) models which account for first and
higher order serial correlations. With the approach we can also account for the
effects of temperature, precipitation amount on the trend. Further we investigate
the role of geographic parameters on isotope trends. To benchmark our proposed
approach, the ARIMA results are compared to a trend-free prewhiting (TFPW)
procedure, the state of the art method for removing the first order autocorrelation in
environmental trend studies. Moreover, we explore whether higher order serial
correlations in isotope series affects our trend results. The results show that three out
of the 17 stations have significant changes when higher order autocorrelation are
adjusted, and four stations show a significant trend when temperature and precipitation
effects are considered. Significant trends in the isotope time series are generally
observed at low elevation stations (≈¤315 m a.s.l.). Higher order autoregressive
processes are important in the isotope time series analysis. Our results show that
the widely used trend analysis with only the first order autocorrelation adjustment
may not adequately take account of the high order autocorrelated processes in the
stable isotope series. The investigated time series analysis method including higher
autocorrelation and external climate variable adjustments is shown to be a better alternative. |
|
|
|
|
|
|