Abstract
Water stable isotope ratios of precipitation predominantly reflect moisture source and precipitation intensity. Trees can incorporate and store the isotopic signals into annual tree-ring cellulose records, permitting reconstruction of the temporal changes of hydroclimate over decades to millennia. This is especially valuable in the Himalayan region where the understanding of monsoon dynamics is limited by the lack of a dense and representative observational network. Here, tree ring δ18O records were analyzed from two distinct physiographic sites along the upper Kali Gandaki valley in the central Nepal Himalayas, representing the wet High-Himalayas and the Trans-Himalayan dryland to the north. Empirical correlations and regression analyses were compared to an in-situ calibrated oxygen isotope fractionation model, exploring the relationships between tree ring δ18O and seasonal variability of hydroclimatic forcing at the different locations. For this purpose, gridded precipitation data from the Asian rain gauge dataset APHRODITE, as well as high resolution onsite observations (relative humidity, air temperature, δ18O of precipitation and radial tree growth) were used.
We found that two distinct sets of meteorological values, reflecting pre-monsoon and monsoon conditions, are needed to reproduce the measured tree ring δ18O values from the High-Himalayan site, but a single set of values (monsoonal) performs best for the Trans-Himalayan site. We conclude that Trans-Himalayan trees capture long-term changes in strength of the Indian summer monsoon. In contrast, High-Himalayan tree ring δ18O records a more complex hydro-climatic signal reflecting both pre-monsoon and monsoon seasons with very contrasting isotopic signatures of precipitation. The unravelled difference in the two hydroclimatic proxy records offers an opportunity to reconstruct first-order hydroclimate conditions and to gain new insights on monsoon timing and large scale atmospheric water source determination across the Himalayan orographic region.