Ice Anatomy: A Benchmark Dataset and Methodology for Automatic Ice Boundary Extraction from Radio-Echo Sounding Data

The measurement of ice thickness is of great importance for the accurate estimation of glacier volume and the delineation of their bedrock topography. In particular, this is a crucial factor in forecasting the future evolution of glaciers in the context of a changing climate. In order to derive the ice thickness, the travel time of electromagnetic waves in radargrams acquired by radio-echo sounding (RES) systems is analyzed. This can only be achieved by identifying the ice surface and underlying ice bottom in corresponding radargrams. Manually identifying these two reflection horizons in RES data is a laborious and time-consuming process. Consequently, scientists are attempting to automate this task through the use of techniques such as deep learning. Such automation can significantly reduce the time between a field campaign and the calculation of the glacier's ice thickness distribution. In this paper, we present the first benchmark dataset for delineating the ice surface and bottom boundaries in RES data, to facilitate straightforward comparisons of deep learning models in the future. The ``IceAnatomy'' dataset comprises radargrams and the corresponding manual picks, amounting to a total of over 45,000km of observations. The RES data originates from three sources: FAU, CReSIS, and AWI. The dataset comprises different RES systems as well as different pre-processing methods. In addition, the data was acquired over a large range of geographical and glaciological settings, featuring different thermal regimes present in Antarctica and the Southern Patagonian Icefield. This diversity ensures that the models' behaviors can be analyzed in different scenarios. We define a standardized train-test split for each source in the dataset. This allows us to introduce not only a baseline model trained on the entire training set (the ``omni'' model), but also three source-specific baseline models. The source-specific models are trained exclusively on the subset of the training data acquired by the specified source. The baseline models provide an initial benchmark against which subsequent models can be compared. The source-specific models demonstrate more accurate results than the omni model. For the FAU, CReSIS, and AWI test sets, the source-specific models achieve Mean Meter Errors of 2.1m, 23.1m, and 4.9m for the ice surface and 9.1m, 78.2m, and 29.3m for the ice bottom. In relation to the mean measured ice thickness, these errors equate to 1.2%, 3.1%, and 0.3% for the ice surface and 4.9%, 10.4%, and 1.5% for the ice bottom. For more information, please read the following paper: [Coming soon. Currently under review.] Please also cite this paper if you plan on using the dataset. For the implementation of a baseline model please visit: [Coming soon]冰厚测量对于精准估算冰川体积、刻画其基岩地形具有重要意义。尤为关键的是，在气候变化背景下，冰厚是预测冰川未来演化的核心影响因素。为获取冰厚数据，需分析无线电回声测深（radio-echo sounding, RES）系统采集的雷达回波图（radargrams）中电磁波的传播时间。这一过程需先在对应雷达回波图中识别冰面与冰底这两类反射层位。手动识别无线电回声测深数据中的这两个反射界面，是一项耗时耗力的工作。为此，科研人员正尝试借助深度学习等技术实现该任务的自动化。此类自动化流程可大幅缩短野外考察与冰川冰厚分布计算之间的时间间隔。 本文提出首个用于识别无线电回声测深数据中冰面与冰底边界的基准数据集，以方便未来深度学习模型的直接对比研究。"IceAnatomy"数据集包含雷达回波图及配套的人工拾取层位标注，总观测里程超过45000千米。该数据集的无线电回声测深数据来源于三个渠道：FAU、CReSIS与AWI，涵盖不同的RES系统与预处理方法。此外，数据采集覆盖了广泛的地理与冰川学环境，包含南极与南巴塔哥尼亚冰原的不同热状态特征。这种数据多样性可确保在不同场景下分析模型的性能表现。 我们为数据集中的每个来源定义了标准化的训练-测试划分方案，这使得我们既能构建基于全部训练集训练的基准模型（即"全源"模型），也能构建三个单源基准模型。单源基准模型仅使用指定来源的训练数据进行训练。这些基准模型可为后续模型的性能对比提供初始参照标准。相较于全源模型，单源基准模型的预测精度更高。在FAU、CReSIS与AWI的测试集上，单源基准模型针对冰面的平均米级误差分别为2.1米、23.1米与4.9米，针对冰底的平均米级误差分别为9.1米、78.2米与29.3米。相对于实测冰厚平均值，上述误差分别对应冰面的1.2%、3.1%与0.3%，以及冰底的4.9%、10.4%与1.5%。 如需了解更多信息，请参阅以下论文： [即将发表，目前处于审稿阶段。] 若您计划使用本数据集，请一并引用该论文。 如需基准模型的实现代码，请访问： [即将上线]