Based on the oxygen isotope ratio and microparticle record in ice cores recovered at Mt. Muztagata, Eastern Pamirs, the seasonal variations of atmospheric dust have been reconstructed for the past four decades. High dust concentrations and coarser particle grains have the similar trend with oxygen iso- tope value. Our statistical results indicate that 50%―60% high dust concentration samples occur dur- ing the season with high oxygen isotope values (summer), while low dust storm frequency during spring and winter. Back-trajectory analysis shows that the air mass hitting Muztagata predominately came from West Asia (such as Iran-Afghanistan Plateau) and Central Asia, which are the main dust source area for Muztagata. Dust storms in those source areas most frequently occur during summer (from May to August), while frequent dust storm events in northern China mainly occur during spring (March to May). Regions in the path of Asian dust transport, such as in Japan, the North Pacific, and Greenland, also show high dust concentrations during spring (from March to May). Our results indicate that dust storms have different seasonality in different regions within Asia.
Given the potential use of stable isotope in the paleoclimate reconstruction from lacustrine records as well as in the local hydrology cycle, it is crucial to understand the processes of stable isotope evolution in catchment in the Tibetan Plateau region. Here we present a detailed study on the water oxygen isotope based on 2 years observation including precipitation, river water and lake water in the Yamzho Lake, south of the Tibetan Plateau. Temporal variation of local precipitation δ18O shows an apparent "monsoon cycle". In monsoon season, δ18O in waters is lower. In non-monsoon season, δ18O in precipitation and lake water is higher and higher river δ18O exists in spring, probably reflecting the effect of land surface evaporation, together with the higher δ18O values in spring precipitation. It is also found that the surface lake water δ18O varies seasonally and annually. The lower lake water δ18O in the late summer is apparently related to the summer monsoon precipitation. The mean δ18O value of lake water in 2007 is 1.2‰ higher than that in 2004, probably due to the less monsoon precipitation in summer of 2007, as can be confirmed from the precipitation data at the Langkazi meteorological data. It is also found that an obvious shift of vertical lake water δ18O reflects the fast mixture of lake water. δ18O values of lake water are over 10‰ higher than those of precipitation and river water in this region due to the evaporation fractionation. The modeled results show that the evaporation process of the lake water is sensitive to relative humidity, and the present lake water δ18O reflects a relative humidity of 51% in the Yamzho Lake. It shows that the lake will take 30.5 years to reach present lake water δ18O given a large shift in the input water δ18O. The modeled results also reveal that surface lake water temperature and inflow δ18O have slight effect on the isotopic balance process of lake water in the Yamzho Lake.
