Data Sharing and Archival
Now that you have planned your project and data management, collected data, integrated imagery, or generated model output, it is time to share your data products with the scientific community and public by archiving them at the ORNL DAAC.
What the ORNL DAAC archives
The ORNL DAAC archives data and model products that were generated with funding from the NASA Carbon Cycle and Ecosystems Focus area and related programs (Terrestrial Ecology , Carbon Monitoring System (CMS), Interdisciplinary Science (IDS), etc.)
For data that do not fall into this category, the ORNL DAAC User Working Group, EOSDIS, and the ORNL DAAC will evaluate relevance of the data to the DAAC's mission and funding source to determine if the data are appropriate for archival at the DAAC.
The DAAC uses a Data Scope and Acceptance Policy to prioritize datasets for the archival process.
If you are interested in archiving your data at the ORNL DAAC, begin the process by filling in this short Submit Data interest form. The steps that you and the DAAC take to archive data are described below.
Other archives might be the right place for your data, if the ORNL DAAC is not.
Archiving Data Associated with a Manuscript
Many journals now require that data associated with a manuscript be archived before the manuscript can be published.
The ORNL DAAC will archive data associated with a manuscript as long as the data is appropriate based on funding agency and relevance. Investigators should initiate a data submission to the ORNL DAAC when the manuscript is submitted to the journal. We will work with the investigator to archive the data so that it will be available upon publication of the manuscript.
View our diagram outlining the Manuscript Submission Process.
What the ORNL DAAC expects for a data submission
The archived holdings at the ORNL DAAC are organized into datasets that include all the information associated with a single research effort. For submission purposes, a complete dataset includes:
What a dataset contains
A dataset contains data that typically share the same investigator(s) and methods over possibly several sites or years.
Save your data in a well organized file structure using stable file formats and descriptive files names
Document(s) describing data
Provide detailed documentation to assist future workers using your dataset.
Any files that are associated with your dataset should be included. Examples of supplemental material include published papers, field notes, a list of related datasets, or even an archived website.
Answers to data provider questions
You will be asked to fill out a short online form to help us better understand your dataset.
The questions can be previewed here: Data Provider Questions.
The ORNL DAAC also archives model products that contain the methodological detail of numerical modeling studies. Because numerical models evolve continuously over time, an archived version of a model must contain a complete description of the model. Before considering archiving your model product with us, check the ORNL DAAC recommendations for model archival.
If you are interested in archiving your dataset or model product at the ORNL DAAC, fill out this short Submit Data interest form.
Data Management Best Practices for Archival
Compiling documentation and metadata should be straight forward if you managed and defined your data throughout your project. Basic quality assurance checks, as listed below, are the final step before submitting your dataset to the ORNL DAAC.
Data management best practices
Compiling documentation and metadata and basic quality assurance checks are the final step before archiving your data.
Perform basic quality assurance
Check that there are no missing values for key parameters. Scan and/or plot for impossible and anomalous values. Perform and review statistical summaries.
Jump to Perform basic quality assurance for more information.
Consider what a future investigator needs to know in order to obtain and use your data.
Jump to Prepare documentation for more information.
Recommendations for Model Archival
Model products provide the methodological detail of numerical modeling studies needed to ensure the long-term reproducibility of experimental results. Because numerical models evolve continuously over time, the most complete description of many numerical models is the version of the model used in a particular manuscript. The ORNL DAAC provides an archive for such models. The best practices for data management provided on these pages can and should be applied to model products to ensure other researchers can easily use your model.
Archiving a Benchmark Model
For long-term, stable archival benchmark models are required to include model name and version, the associated documentation, and example inputs and outputs.
Benchmark Model Requirements
Click a benchmark model requirement to display more information.
Benchmark model metadata should include a unique model name and version number, an archival date, and contact information.
Compile the complete source code for the model, including code compilation (build) utilities for at least one operating system type.
Provide example model inputs, detailed documentation describing the application of the model using the example inputs, and example outputs.
We also recommend that a benchmark model include documentation of model process representation and, as appropriate, a description of model lineage.
Archiving a Manuscript Model
An archival product for models associated with a published research manuscript must include model code, boundary conditions, parameterizations, and any analysis routines required to reproduce the published results.
Manuscript Model Requirements
Click a manuscript model requirement to display more information.
The model for archival should be explicitly related to at least one published manuscript, with one manuscript author serving as the scientific point of contact.
Compile the complete source code for the model, including build utilities for the platform on which published results were generated.
Provide all model inputs required to generate the key research results reported in the associated publication.
We also recommend including all post-processing or analysis routines required to generate the targeted research results, detailed instructions describing the process of model execution, and, as appropriate, a description of model lineage.
Your data should be archived in an appropriate archive with similar data. The ORNL DAAC might not be the right place for your data. Other data archives available are:
- Other NASA EOSDIS data centers
- Knowledge Network for Biodiversity
- Long-Term Ecological Research Network
- National Phenology Network
- National Oceanographic Data Center
- U.S. Geological Survey Science Data Catalog
You may want to archive your data in an archive that you commonly visit, or your funder might specify an archive.
More details of our Data Management Best Practices can be found in Best Practices for Preparing Environmental Data Sets to Share and Archive, published by ORNL DAAC in 2010, and Environmental Data Management Best Practices Part 1 Tabular Data and Part 2 Geospatial Data, from the NASA Earthdata webinar series and presented by ORNL DAAC staff (hosted on YouTube).