Automated retrieval, preprocessing, and visualization of gridded hydrometeorology data products for spatial-temporal exploratory analysis and intercomparison
Automated retrieval, preprocessing, and visualization of gridded hydrometeorology data products for spatial-temporal exploratory analysis and intercomparison
复制标题
DOI：
10.1016/j.envsoft.2019.01.007
复制
发表时间：
2019-06
期刊：
影响因子：
--
通讯作者：
J. Phuong;C. Bandaragoda;E. Istanbulluoglu;C. Beveridge;R. Strauch;Landung Setiawan;S. Mooney
中科院分区：
文献类型：
--
作者：
J. Phuong;C. Bandaragoda;E. Istanbulluoglu;C. Beveridge;R. Strauch;Landung Setiawan;S. Mooney
空间分布的时间序列数据支持一系列环境建模和数据研究工作。任何此类努力的关键第一步是获取插值水文气象数据。标准化的工具，以促进这一进程的分析还没有现成的流域规模的研究。在这里，我们介绍网格水文气象观测站（OGH），这是一个开源Python库，通过提供网络基础设施组件来获取和管理从区域和大陆尺度网格水文气象产品处理的分布式数据，填补了这一关键软件空白。我们的方法包括注释元数据，使网格数据产品在软件中可扩展和可用，从而实现使用数据的模型的互操作性和再现性。本文介绍了OGH的设计，架构和应用程序，使用四个常用的用例与网格化的时间序列数据在流域尺度。OGH及其相关注释通过Anaconda Cloud在conda-forge包存储库中分发。Freshwater Initiative Observatory存储库（https：//github.com/Freshwater-Initiative/Observatory）中提供了每个示例用例的教程Observyter笔记本。这些示例旨在利用HydroShare提供的计算资源和软件库（（https：//www.hydrosare.org/resource/87dc5742cf164126a11ff45c3307fd9d））。
Spatially-distributed time-series data support a range of environmental modeling and data research efforts. A critical first step to any such effort is acquiring interpolated hydrometeorological data. Standardized tools to facilitate this process into analyses have not been readily available for watershed scale research. Here, we introduce the Observatory for Gridded Hydrometeorology (OGH), an open source python library that fills this critical software gap by providing a cyberinfrastructure component to fetch and manage distributed data processed from regional and continental-scale gridded hydrometeorology products. Our approach involves annotating metadata to make gridded data products discoverable and usable within the software, enabling interoperability and reproducibility of models that use the data. This paper presents the design, architecture, and application of OGH using four commonly practiced use-cases with gridded time-series data at watershed scales. OGH and its associated annotations are distributed via Anaconda Cloud within conda-forge package repository. The tutorial Jupyter notebooks for each example use-case are available within the Freshwater Initiative Observatory repository (https://github.com/Freshwater-Initiative/Observatory). The examples are designed to utilize the compute resources and software libraries provided by HydroShare ((https://www.hydroshare.org/resource/87dc5742cf164126a11ff45c3307fd9d)).