P2C2: Leveraging Bayesian Approaches to Link Reconstructed, Observed, and Projected Meteorological Drought while Accounting for Inherent Data Biases

该项目旨在调查过去、现在和未来的气象干旱。具体地说，该项目将通过测试两个基本问题来解决不兼容的数据偏差问题，例如来自这三个数据源的数据偏差(时间分辨率)：(1)基于代理的干旱重建中的误差是否与全球气候模型(GCM)得出的气候预测中的误差显著不同，特别是对于高阶统计量，以及(2)统一的统计建模框架是否能够明确地考虑固有的数据源差异，以估计从工业化前的过去、通过仪器时期到未来的干旱严重程度。研究人员将应用新的统计方法(分层贝叶斯模型)来合并干旱重建、观测和预测，并生成从2000年前到本世纪末(2100)的网格干旱时间序列。潜在的更广泛的影响包括开发一个可视化网站，允许公众查看北美任何网格单元的2100年干旱指数时间序列，其中还将包括模型和气候的不确定性。将组织一个基于会议的研讨会和动手代码演示，以培训科学家在该项目中开发的统计建模框架。该奖项反映了NSF的法定使命，并通过使用基金会的智力优势和更广泛的影响审查标准进行评估，被认为值得支持。

This project aims to investigate past, present and future meteorological droughts. Specifically, this project will address the issue of incompatible data biases e.g. (temporal resolution) from these three data sources by testing two fundamental questions: (1) do errors in proxy-based drought reconstructions significantly differ from those in Global Climate Model (GCM)-derived climate projections, particularly for higher order statistics, and (2) can a unified statistical modeling framework explicitly account for inherent data source differences to estimate drought severity from the pre-industrial past, through the instrumental period, and into the future. The researcher will apply novel statistical methods (Hierarchical Bayesian modelling) to merge drought reconstructions, observations, and projections, and generate a gridded drought time-series beginning 2000 years ago and extending to the end of this century (2100). The potential Broader Impacts include developing a visualization website that will allow the public to view 2100 years of drought index time series for any gridded cell, in North America which will also include model and climate uncertainty. A conference-based workshop with hands-on code demonstrations will be organized to train scientists on the statistical modelling framework developed in the project.This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.