SMAP HDF Data at the NASA NSIDC DAAC: Strengthening Support for Visualization and Analysis of Multidimensional Data

The NASA National Snow and Ice Data Center Distributed Active Archive Center (NSIDC DAAC) distributes and supports Hierarchical Data Format (HDF) data products from several remote sensing observatories including the NASA Soil Moisture Active Passive (SMAP) mission. Launched on 31 January 2015, SMAP provides global high-resolution mapping of soil moisture and freeze-thaw state. These measurements enable a better understanding of the connections between carbon, energy, and hydrologic fluxes on earth, as well as improved drought, flood, and weather forecasting capabilities that can directly impact agricultural productivity. Given this wide range of research and policy applications, the NASA NSIDC DAAC supports a diverse user community employing a variety of geospatial software tools to analyze and visualize SMAP data. SMAP data products are produced in HDF5 format, organized into hierarchical groups containing multiple data sets. These data sets can be heterogeneous in terms of their data type and array dimensions. While the HDF5 model allows for this heterogeneity, this complex data format can pose problems when read into many software applications. Although SMAP data include rich file-level metadata, these metadata only partly adhere to CF conventions and ISO 19115 standards, adding additional interoperability challenges. In the case of SMAP, issues such as incompletely filled geolocation arrays and Level 1 and 2 (swath) 1-dimensional structure were revealed through user feedback and NASA Earth Observing System Data and Information System (EOSDIS) software development. As a result, SMAP products are now better supported in ArcGIS through multidimensional data tools including the mosaic dataset and OPeNDAP workflows. These issues underscore the need for ongoing collaboration between the NSIDC DAAC, software development teams, data providers, end-users, and the data and metadata standards communities in order to ensure interoperability across data products and tools.