Dryad is a nonprofit, curated, discipline-agnostic data repository built around a journal-integrated submission workflow: a researcher deposits the dataset underlying a specific manuscript, a dedicated Dryad curation team reviews it for completeness and reusability, and the published dataset receives a DataCite DOI that links back to the article. Dryad traces to 2007 planning meetings at the National Center for Ecological Analysis and Synthesis (NCEAS) and the National Evolutionary Synthesis Center (NESCent), launched in 2008 with NSF funding under the working name DRIADE (Digital Repository of Information and Data for Evolution), and grew directly out of the 2011 Joint Data Archiving Policy (JDAP) adopted by a group of evolutionary-biology and ecology journals requiring that data supporting published results be archived in a public repository. It has since broadened well beyond its ecology/evolution origins to accept research data from any discipline.
This guide covers what actually distinguishes Dryad’s model from a generalist deposit-and-forget repository, what it costs to publish there under the current fee schedule, and how to reason about whether it is the right choice for a specific dataset in a data management plan (DMP) — rather than defaulting to it because it is one of the names most commonly listed alongside Zenodo and Figshare.
How Dryad’s submission and curation workflow works
The defining operational feature of Dryad, relative to a self-service repository, is human curation before publication:
- Manuscript-linked submission. Deposits are normally tied to a specific paper. Dryad integrates directly with journal manuscript-tracking systems — including Editorial Manager and eJournalPress — plus preprint and indexing infrastructure such as openRxiv, PubMed Central, and Europe PMC, so a dataset can be submitted alongside a manuscript and later linked automatically once the paper is published.
- Curator review. A dedicated Dryad curation team checks each submission for file readability, adequate metadata and documentation, and whether the files actually support reuse and reproduction of the study’s results — not just that files were uploaded. This is a real editorial step, not an automated intake check, and it is the reason publication is not instantaneous the way it typically is on a purely self-service repository.
- Optional pre-publication privacy for peer review. Authors can keep a submitted dataset private during the journal’s peer-review process before it goes public alongside (or shortly after) the article.
- CC0-default licensing. Dryad publishes datasets under a CC0 (public domain dedication) default, on the premise that maximally unrestricted reuse rights are the norm for data supporting a published scientific claim, distinct from the more varied licensing norms common on general-purpose repositories.
What Dryad costs: the current fee schedule
Dryad charges a Data Publishing Charge (DPC) scaled to dataset size, rather than being free at the point of deposit for every author. Two paths determine who actually pays it:
- Sponsored authors. Researchers affiliated with a Dryad institutional, publisher, or society partner have their DPC covered by that partner, typically up to 10 GB per dataset; a separate large-dataset fee applies above that threshold, billed to the author or the sponsoring partner depending on the specific partner agreement.
- Unsponsored authors. Authors with no covering partner pay the DPC directly by credit card at submission. Under the fee schedule effective May 6, 2025, the charge scales with dataset size: $150 for up to 5 GB, $180 up to 10 GB, $520 up to 50 GB, $808 up to 100 GB, $1,750 up to 250 GB, $3,086 up to 500 GB, $6,077 up to 1 TB, and $12,162 up to 2 TB (Dryad’s own published rate card — verify the current figures directly at datadryad.org/help/requirements/costs before budgeting, since fee schedules change).
Additional charges to budget for: a nonrefundable Private for Peer Review Fee (credited toward the eventual DPC once the dataset is made public), and a $199 administration fee if an author requests an invoice rather than paying by card. Dryad also operates a fee waiver program for authors who genuinely cannot pay — waivers are not approved for submissions over 10 GB, so very large unsponsored datasets need either partner sponsorship or an alternative repository. Because whether a given author is sponsored depends on their institution’s or journal’s specific partnership with Dryad, checking partner status before submission — via Dryad’s own institutional and publisher partner pages — is worth doing before assuming a deposit will be free.
Software, code, and large supplemental files: the Zenodo integration
Dryad is built to curate and preserve research data, not software. Since a 2020 integration, Dryad supports concurrent submission of associated software and supplemental materials through Zenodo: the data goes into Dryad’s curated workflow while code/software is deposited to Zenodo, with the two records cross-linked so a reader following either DOI can find both halves of the research output. This division of labor is a genuine structural difference from a single generalist repository (Zenodo or Figshare) that accepts data, code, and other output types side by side without that split — worth knowing before assuming Dryad alone covers a project’s entire software-plus-data submission.
Where Dryad’s trust infrastructure sits: Merritt and CoreTrustSeal
Dryad’s underlying preservation copy is held in Merritt, the California Digital Library’s digital preservation repository — Dryad and the California Digital Library have been operating partners since 2013. It is Merritt’s technical layer, not a separate Dryad-branded certification, that holds CoreTrustSeal certification as a trustworthy digital repository. This is a useful distinction when a funder or journal’s data policy specifically asks for evidence of a trustworthy or certified repository: the certification is real and does cover the infrastructure preserving Dryad-deposited data, but it sits one layer down from the Dryad brand itself, and CoreTrustSeal certifications are reviewed and renewed on a rolling multi-year cycle — check the current status directly against CoreTrustSeal’s own certificate registry rather than relying on any secondary summary, including this one.
Dryad vs. Zenodo vs. Figshare vs. an institutional repository
All four are commonly cited in a DMP’s appropriate-repository line, but they are not interchangeable:
- Dryad — curated by humans before publication, manuscript-linked workflow, CC0-default, discipline-agnostic, charges a size-based DPC (waivable, or partner-sponsored). Strongest fit when a journal or funder specifically expects a curated, reviewed deposit tied to a paper.
- Zenodo — self-service, CERN/OpenAIRE-operated, free at point of deposit, accepts data, software, and a wide range of other output types without Dryad’s curation review or size-tiered charge, and is EU-funded infrastructure rather than a US nonprofit. See the Zenodo entry for its own workflow.
- Figshare — commercial (Digital Science), self-service, discipline-agnostic, also sold directly to institutions as branded repository infrastructure.
- Institutional or discipline-specific repositories — may be required or preferred by a specific funder or field regardless of the generalist options above; see discipline-specific repository and generalist repository for the underlying distinction, and re3data.org (also summarized at Re3data) for locating one.
For a deeper look at what CoreTrustSeal certification actually requires and whether pursuing it is worthwhile for a repository operator (rather than a depositing researcher), see the CoreTrustSeal certification guide.
Deciding whether Dryad fits a specific data management plan
Dryad is usually the right default when: the dataset directly underlies a specific journal article (especially at a journal that already partners with Dryad, making the deposit sponsored/free), the field expects curated rather than self-service deposit, and CC0 licensing is acceptable for the data in question. It is a weaker fit when: the dataset is large (multi-hundred-GB to multi-TB) and unsponsored, since DPCs scale steeply with size and waivers cap out at 10 GB; the output is primarily software or code rather than data (route that to Zenodo, or through Dryad’s Zenodo integration alongside the data); or a funder, field, or institution mandates a specific discipline-specific or institutional repository instead. A DMP should name the specific repository and, where cost is a live constraint, note whether the submitting author’s institution or target journal is a Dryad sponsoring partner — that single fact determines whether the DPC line item is zero or several hundred dollars. See data availability statement and embargo for two related mechanics that often need to be decided alongside repository choice.
Frequently asked questions
Is Dryad free to use?
Not universally. Authors affiliated with a Dryad institutional, publisher, or society partner typically have their Data Publishing Charge sponsored (usually up to 10 GB). Unsponsored authors pay a size-tiered DPC directly, starting at $150 for up to 5 GB, with a fee waiver available for smaller datasets from authors who cannot pay.
Does Dryad accept software and code?
Dryad’s curation workflow is built for research data, not software. Since 2020 it supports concurrent, cross-linked submission of associated software to Zenodo alongside a Dryad data submission, rather than hosting code itself.
Is Dryad CoreTrustSeal certified?
Dryad’s preservation copy is held in Merritt, the California Digital Library’s repository, and it is Merritt’s technical layer that holds CoreTrustSeal certification — not a separate Dryad-branded certificate. Always confirm current status against CoreTrustSeal’s own registry, since certifications are reviewed on a rolling renewal cycle.
What license does Dryad use for published data?
Dryad publishes datasets under a CC0 (public domain dedication) default, reflecting a norm of maximally unrestricted reuse for data supporting a published research claim.
How is Dryad different from Zenodo?
Dryad is curated (human review before publication), manuscript-linked, CC0-default, and charges a size-based fee for unsponsored deposits. Zenodo is self-service, free at the point of deposit, accepts a broader mix of output types including software, and is operated by CERN/OpenAIRE rather than being a US-based nonprofit.
