Investigation of relationship between southern oscillation index (SOI) and reconstructed precipitation of central Zagros


Faculty of Geography, University of Tehran


1- Introduction
Iran’s precipitation mainly affects by air masses that arrive to it from adjacent area directly or affects as teleconnection pattern due to remoteness from water sources. Due to irregular arrival of the air masses to Iran and far distance to teleconnection centers, occurrence of precipitation variations is one of the prominent characteristics of Iran’s climate. Also due to short period of instrumental data in the meteorological stations, there is no possibility for understanding of the climatic variations during the last centuries. Reconstruction of climatic variations based on current evidences, shows a more realistic understanding of the climatic variations of the region. Trees are the useful climatic evidences that experience different climatic situation of temperature and precipitation during their life period and the effects of these climatic variations is reflected in their tissues. In temperate climates, trees add a ring annually to their previous rings. Therefore, by studying of the tree-ring widths, useful climatic information can be obtained over the last centuries of the region. In the previous studies, relationship between the Southern Oscillation Index (SOI) and Iran’s precipitation only has been studied in a relatively short period due to the lack of instrumental data. In this study, we present a longer period of relationship between monthly precipitations of the region with Southern Oscillation Index (SOI) by reconstruction of precipitation based on Oak tree rings in the central Zagros region over the last centuries (1705-2010).
2- Methodology
In the present paper, three sites of Oak species with a short distance to the meteorological stations were selected in Lorestan (Shineh), Kermanshah (Faryadras) and Ilam (Dalab) provinces. In Shineh and Faryadras sites Quercus infectoria species and in Dalab site Quercus Persica species used for sampling. Two increment cores at different sides of each tree were extracted by a Suunto increment borer at breast high. In each site 10 to 14 trees were sampled. After air drying, the surface of cores was prepared by a razor blade. The ring widths were measured with a LINTAB5 measuring system with a resolution of 0.01 mm, and all cores were cross-dated by visual and statistical tests (sign-test and t-test) using the software package TSAP-Win. Raw ring widths were standardized by ARSTAN program to remove non climatic signals. Three site chronologies were obtained from the three sites. To increase the length of the chronology and due to the same trends and high correlations between the three site chronologies, all growth curves were combined and the regional chronology was calculated. The values of mean sensitivity (MS), signal to noise ratio (SNR) and first auto correlation (AC1) were calculated for the regional chronology. Mean monthly precipitation of Khorramabad (1951-2010), Kermanshah (1951-2010) and Ilam (1987-2010) meteorological stations were used to calibrate the tree-ring/climate relationship. Due to the high correlation (p