Data release for 'Ensemble Forecasting of Major Solar Flares: Methods for Combining Models'

This is a release of the data that were used for validation in the paper 'Ensemble Forecasting of Major Solar Flares: Methods for Combining Models' by J. A. Guerra, S. A. Murray, D. S. Bloomfield, and P. T. Gallagher, that has been submitted to the Journal of Space Weather and Space Climate.

The naming scheme for the files is in the format:

class_type_metric.dat

'class' denotes whether the forecast is for M- or X- class flares.

'type' is what kind of forecast, i.e., the original ensemble members, an ensemble created from probabilistic validation metrics, or an ensemble created from categorical validation metrics.

'metric' specifies the metric used to create the ensemble in the case of 'probabilistic' or 'categorical' types as above (see paper for further details), or in the case of the original ensemble members the name of the operational forecasting method.

The data files are in the format:

obs,prob

'obs' denotes whether or not a flare was observed within 24 hours of the forecast issue time (1 for yes and 0 for no).

'prob' gives the probabilistic forecast value (between 0.0 and 1.0).

These data files can easily be read into the R verification package to replicate the results presented in the paper.

本数据集为J. A. Guerra、S. A. Murray、D. S. Bloomfield与P. T. Gallagher合著，已提交至《空间天气与空间气候期刊》(Journal of Space Weather and Space Climate)的论文《强太阳耀斑集成预报（Ensemble Forecasting）：模型融合方法》(Ensemble Forecasting of Major Solar Flares: Methods for Combining Models)中用于验证环节的实验数据。

文件命名遵循`class_type_metric.dat`格式，各字段含义如下：`class`用于指代预报针对的耀斑等级，即M级或X级太阳耀斑；`type`表示预报类型，具体包含原始集成成员、基于概率验证指标构建的集成预报，以及基于分类验证指标构建的集成预报；`metric`在“概率”或“分类”类型的集成场景下，用于指定构建集成时所采用的评估指标（详细说明请参见原论文）；若为原始集成成员，则指代业务化预报方法的名称。

数据文件采用`obs,prob`格式，其中`obs`表示预报发布后24小时内是否观测到耀斑，1代表观测到，0代表未观测到；`prob`代表概率化预报值，取值范围为0.0至1.0。

用户可将此类数据文件轻松导入R语言验证包(R verification package)，以复现原论文中展示的实验结果。