4.3 Implication of typhoon precipitation isotopic composition for paleoclimate and paleotempestology records
The paleoclimate proxies, such as δ18O values of stalagmites and δD values of selected plant-derived organic compounds (e.g., leaf wax) in lacustrine sediments are often related to the δ18O or δD of local precipitation during formation, and thus can be used to indicate precipitation conditions in historical periods. The δ18O value of stalagmite is mainly controlled by drop-water and temperature in the cave, and also influenced by physical processes such as evaporation and infiltration related to the δ18O value of precipitation (Lachniet, 2009). Whereas δD of leaf wax in sediments and δD and δ18O values of tree ring cellulose are significantly controlled by ambient environment water (e.g., precipitation and atmospheric moisture) during plant growth, recording important information about regional precipitation (Hou et al., 2008; McCarroll and Loader, 2004). Therefore, linking these proxies to the δD and δ18O values of precipitation during the atmospheric water cycle helps us to further reconstruct the paleoclimatic conditions.
To date, a number of successful studies have been carried out to reconstruct the evolution of the ancient monsoon in the corresponding regions from paleoclimate proxies, such as the reconstruction of the East Asian Summer Monsoon in a varied time scales using the δ18O values of stalagmites (Zhang et al., 2013; Yang et al., 2020). However, there is still a paucity of studies that use these paleoclimate proxies to reconstruct extreme weather events (e.g., typhoons) from historical periods which is known as paleotempestology. Studying the δD, δ18O of precipitation during typhoons can provide a good scientific reference for paleotempestology reconstruction of typhoon activity. The results of this study show that the δD and δ18O values of typhoon-related precipitation are significantly lower than those of normal precipitation in the same period, which may associate with more negative δ18O in speleothem and tree ring and more negative δD in plant-derived compounds, indicating larger precipitation amount, lower temperature, or most importantly, larger proportion of the moisture being transported from tropical seas. At the same time, there are significant decrease in temperature, water vapor pressure, and atmospheric humidity and increases in precipitation amount in the study area (Fig. 2), and these meteorological factors control the δ18O of tree rings during tree growth period (from May to September, consistent with the main period of typhoon occurrence) (Anderson et al., 2002) and δD values of leaf waxes (Hou et al., 2008). Thus, by using the δD and δ18O values in the paleoclimate proxy we can reconstruct the precipitation events (including typhoon-related precipitation), and then roughly invert the frequency of typhoon events in the historical period. Unfortunately, the stage changes of typhoon precipitation may be difficult to be recorded in these paleoclimate proxies due to their small-time scales, leading to great difficulties in the study of paleo-typhoon meteorological processes. In addition, the intensity and migration route of typhoons may have a significant effect on the D-excess values of precipitation (Table 2). Therefore, it may be worthwhile to consider and further investigate whether the D-excess values of paleoclimate proxies can be used to reconstruct the intensity and approximate route of typhoons in historical period.