Regional Climate Model Evaluation System powered by Apache Open Climate Workbench v1.3.0: an enabling tool for facilitating regional climate studies

• Published in: Geoscientific Model Development Vol. 11; no. 11; pp. 4435 - 4449
• Main Authors: Lee, Huikyo, Goodman, Alexander, McGibbney, Lewis, Waliser, Duane E., Kim, Jinwon, Loikith, Paul C., Gibson, Peter B., Massoud, Elias C.
• Format: Journal Article
• Language: English
• Published: Katlenburg-Lindau Copernicus GmbH 05-11-2018; Copernicus Publications
• Subjects: Aeronautics; Business metrics; Climate change; Climate models; Climate studies; Climatic analysis; Climatic data; Components; Computer programs; Computer simulation; Data analysis; Data processing; Datasets; Earth; Environmental assessment; Evaluation; Frameworks; Interfaces; Loaders; Microprocessors; Open source software; Performance measurement; Regional analysis; Regional climate models; Regional climates; Regional development; Satellite observation; Satellites; Software; Source code; Statistical analysis; Sustainability; Sustainable development; Toolkits; United States > US
• ISSN: 1991-9603; 1991-962X; 1991-959X; 1991-9603; 1991-962X
• DOI: 10.5194/gmd-11-4435-2018

The Regional Climate Model Evaluation System (RCMES) is an enabling tool of
the National Aeronautics and Space Administration to support the United
States National Climate Assessment. As a comprehensive system for evaluating
climate models on regional and continental scales using observational
datasets from a variety of sources, RCMES is designed to yield information on
the performance of climate models and guide their improvement. Here, we
present a user-oriented document describing the latest version of RCMES, its
development process, and future plans for improvements. The main objective of
RCMES is to facilitate the climate model evaluation process at regional
scales. RCMES provides a framework for performing systematic evaluations of
climate simulations, such as those from the Coordinated Regional Climate
Downscaling Experiment (CORDEX), using in situ observations, as well as satellite and reanalysis data
products. The main components of RCMES are (1) a database of observations
widely used for climate model evaluation, (2) various data loaders to import
climate models and observations on local file systems and Earth System Grid
Federation (ESGF) nodes, (3) a versatile processor to subset and regrid
the loaded datasets, (4) performance metrics designed to assess and quantify
model skill, (5) plotting routines to visualize the performance metrics,
(6) a toolkit for statistically downscaling climate model simulations, and
(7) two installation packages to maximize convenience of users without Python
skills. RCMES website is maintained up to date with a brief explanation of
these components. Although there are other open-source software (OSS)
toolkits that facilitate analysis and evaluation of climate models, there is
a need for climate scientists to participate in the development and
customization of OSS to study regional climate change. To establish
infrastructure and to ensure software sustainability, development of RCMES is
an open, publicly accessible process enabled by leveraging the Apache
Software Foundation's OSS library, Apache Open Climate Workbench (OCW). The
OCW software that powers RCMES includes a Python OSS library for common
climate model evaluation tasks as well as a set of user-friendly interfaces
for quickly configuring a model evaluation task. OCW also allows users to
build their own climate data analysis tools, such as the statistical
downscaling toolkit provided as a part of RCMES.