Abstract

<div> <p>Torrential rainfall and floods have had devastating impacts on civilizations throughout history. Thus, understanding long-term characteristics of extreme precipitation is necessary to identify physical mechanisms involved in such events and to be able to assess, not only the past, but also the present and future risk of extreme precipitation for society. However, the scarce spatial and temporal distribution of existing datasets on extreme precipitation complicates a detailed study of such events in the paleo climate context.  </p> </div><div> <p>In this study, we employ the newly produced seamless simulations from the Community Earth System Model v1.2.2 that covers the period from 1501 BC to 1849 AD to analyze the daily extreme precipitation before the preindustrial period. We explore the statistical characteristics of extreme precipitation and their association with natural external forcing, such as changes in the orbital parameters, solar cycle, insolation, and volcanic eruptions. For this, we applied to the simulations an extreme value analysis  by adopting a peak-over-threshold method (Coles et al., 2001). The 99th percentile of daily precipitation anomalies with respect to 1501BC - 1849AD are taken as the extreme values and these extremes are fitted to the Generalized Pareto Distribution to create time-stationary and covariate models (GPD models) at each grid point.  </p> </div><div> <p>The stationary GPD model shows that over the mid-latitudes, high scale and negative shape parameters predominate in the Pacific while the opposite condition occurs in the Atlantic sector. Over the Southern Ocean, low scale and negative shape parameters are more common. The covariate GPD models indicate some connection between the external forcing and extreme precipitation. The changes in the orbital parameters are slightly connected to the extreme precipitation over the tropical Atlantic and southern Indian oceans. Among all the forcing, the volcanic eruptions are the most influential in the extreme precipitation during the past 3350 years. The return periods of extreme precipitation decrease over the tropical Pacific, and the mid-latitude oceans and lands after volcanic eruptions, indicating that such eruptions likely increase the occurrence of extreme precipitation in these regions. Over the regions where a decrease in extreme precipitation is followed after the eruptions, such as India, Australia, and eastern Asia, the return periods decrease after volcanic eruptions.  </p> </div><div> <p>Overall, our study provides a long-term continuous view on the global extreme precipitation, which elucidates some complementary information to the currently available proxy and instrumental observations on extreme precipitation events. </p> <p> </p> </div><div> <p>Coles, S., Bawa, J., Trenner, L., & Dorazio, P. (2001). An introduction to statistical modeling of extreme values (Vol. 208, p. 208). London: Springer. </p> </div>

Full Text
Published version (Free)

Talk to us

Join us for a 30 min session where you can share your feedback and ask us any queries you have

Schedule a call