过去500年南北半球环状模数据

（1）数据内容：1500-2000年年平均的北半球环状模指数和南半球环状模指数；（2）数据来源及加工方法：该数据由作者自主生产，基于PAGES2k代用资料数据集，利用机器学习模型（随机森林、极端随机树、轻量梯度提升机、CatBoost）重建而产生。（3）数据质量描述：该数据集与多个器测数据在器测时段内有较高的一致性，重建效果更好。数据可用于研究多时间尺度（年际、年代际、多年代际）上南北半球主要大气环流的变化规律及机理。

1. Data Content: Annual mean Northern Hemisphere Annular Mode (NAM) index and Southern Hemisphere Annular Mode (SAM) index spanning from 1500 to 2000 CE.
2. Data Source and Processing Methodology: This dataset was independently developed by the authors, and reconstructed based on the PAGES2k proxy dataset using multiple machine learning models including Random Forest, Extremely Randomized Trees, Light Gradient Boosting Machine (LightGBM), and CatBoost.
3. Data Quality Description: This dataset exhibits high consistency with multiple instrumental datasets during the instrumental period, with superior reconstruction performance. It can be used to explore the variation patterns and underlying mechanisms of dominant atmospheric circulations over both Northern and Southern Hemispheres across multiple time scales, including interannual, interdecadal and multidecadal scales.

搜集汇总

背景与挑战

背景概述

该数据集包含1500-2000年的北半球和南半球环状模指数，通过机器学习方法重建，数据质量高，适用于年际、年代际和多年代际的大气环流研究。