Data and code from: Propagation of the Madden–Julian oscillation as a deterministic chaotic phenomenon

The Madden–Julian oscillation (MJO), a gigantic weather system marked by eastward travel of cumulus cloud clusters over the Indo-Pacific, often causes severe weather and climate events worldwide. Despite its frequent realization, the physics and predictability characteristics of MJO propagation are not well understood. We here reveal the chaoticity of MJO propagation, using 4,000-member ensemble global cloud-system-resolving simulations of two MJOs. Against previous linearized thinking, multiple regimes supporting MJO propagation emerge under a single atmosphere-ocean background, depending critically on the equatorial asymmetry of climatological sea surface temperature. Selection of the bifurcated regimes results from the stochastic modulation of moisture transport via westward-propagating equatorial waves, fostered or inhibited by tropical-extratropical interplay. Tracking the nonlinear processes we found gives metrics for predicting the timing of MJO propagation. , , # Data and code from: Propagation of the Madden–Julian oscillation as a deterministic chaotic phenomenon Dataset DOI: [10.5061/dryad.sxksn03dw](10.5061/dryad.sxksn03dw) ## Description of the data and file structure This dataset contains the simulation data, observational data, and scripts for creating the figures in Takasuka et al. (2025). It also provides NCL (NCAR command language) scripts for plotting the figures. ### Files and variables #### File: Takasuka_etal_2025_Science.tar.gz **Description:** All the data have the netCDF format. Variable information has been provided in the metadata of the files, so we do not repeat the details here. The directory structure is as follows. — NICAM This directory stores the three types of data. 1\) Atmospheric data composited for the Nov-1, Dec-1, and Dec-2 regimes. They exist at the directories of [Nov, Dec]/[Nov-1, Dec-1, Dec-2]/[case, com]. The directories of “case” and “com” store the 2d/3d variables for each case and composites cat...,马登-朱利安振荡（Madden–Julian oscillation, MJO）是一种巨型天气系统，以积云团在印度-太平洋区域向东传播为典型特征，常引发全球范围内的极端天气与气候事件。尽管其发生频率极高，但MJO传播的物理机制与可预报性特征仍未得到充分阐明。本研究通过针对两次MJO事件的4000成员集合全球云分辨模拟，揭示了MJO传播的混沌特性。与此前的线性化思维相悖，在单一海气背景场下，可支撑MJO传播的多类模态会因气候态海表温度的赤道非对称性发生显著变化。分叉模态的选择由向西传播的赤道波动对水汽输送的随机调制所驱动，而热带-热带外相互作用会对这一调制过程起到促进或抑制作用。本研究追踪得到的非线性过程可作为预测MJO传播时序的量化判据。 # 数据与代码来源：《马登-朱利安振荡的传播作为确定性混沌现象》 数据集DOI：10.5061/dryad.sxksn03dw ## 数据与文件结构说明 本数据集包含Takasuka等人2025年研究中的模拟数据、观测数据以及绘图脚本，同时提供了用于绘图的NCL（NCAR命令语言）脚本。 ### 文件与变量 #### 文件：Takasuka_etal_2025_Science.tar.gz **描述：** 所有数据均采用netCDF格式存储，变量信息已包含在文件元数据中，此处不再赘述。目录结构如下： — NICAM 该目录存储三类数据： 1) 针对11月1日、12月1日及12月2日模态合成的大气数据，存储于[Nov, Dec]/[Nov-1, Dec-1, Dec-2]/[case, com]目录下，其中"case"与"com"目录分别存储各案例及合成集的二维/三维变量……