An architecture for consolidating multidimensional time-series data onto a common coordinate grid
In this paper, consolidating measurement data for use by data models or in inter-comparison studies frequently requires transforming the data onto a common grid. Standard methods for interpolating multidimensional data are often not appropriate for data with non-homogenous dimensionality, and are hard to implement in a consistent manner for different datastreams. In addition, these challenges are increased when dealing with the automated procedures necessary for use with continuous, operational datastreams. In this paper we introduce a method of applying a series of one-dimensional transformations to merge data onto a common grid, examine the challenges of ensuring consistent application of data consolidation methods, present a framework for addressing those challenges, and describe the implementation of such a framework for the Atmospheric Radiation Measurement (ARM) program.
- Research Organization:
- Pacific Northwest National Laboratory (PNNL), Richland, WA (United States)
- Sponsoring Organization:
- USDOE Office of Science (SC), Biological and Environmental Research (BER)
- Grant/Contract Number:
- AC05-76RL01830
- OSTI ID:
- 1336507
- Alternate ID(s):
- OSTI ID: 1341743
- Journal Information:
- Earth Science Informatics, Journal Name: Earth Science Informatics Vol. 10 Journal Issue: 2; ISSN 1865-0473
- Publisher:
- Springer Science + Business MediaCopyright Statement
- Country of Publication:
- Germany
- Language:
- English
Similar Records
A Scientific Data Processing Framework for Time Series NetCDF Data
Multi-Resolution Indexing for Hierarchical Out-of-Core Traversal of Rectilinear Grids