FiCli, the Fish and Climate Change Database, informs climate adaptation and management for freshwater fishes

Inland fishes provide important ecosystem services to communities worldwide and are especially vulnerable to the impacts of climate change. Fish respond to climate change in diverse and nuanced ways, which creates challenges for practitioners of fish conservation, climate change adaptation, and management. Although climate change is known to affect fish globally, a comprehensive online, public database of how climate change has impacted inland fishes worldwide and adaptation or management practices that may address these impacts does not exist. We conducted an extensive, systematic primary literature review to identify peer-reviewed journal publications describing projected and documented examples of climate change impacts on inland fishes. From this standardized Fish and Climate Change database, FiCli (pronounced fick-lee), researchers and managers can query fish families, species, response types, or geographic locations to obtain summary information on inland fish responses to climate change and recommended management actions. The FiCli database is updatable and provides access to comprehensive published information to inform inland fish conservation and adaptation planning in a changing climate.

www.nature.com/scientificdata www.nature.com/scientificdata/ mixed, or unknown benefit to the species, population, or assemblage (Fig. 3). This unique synthesis currently comprises projected and documented studies between 1985 and 2018 and can serve to focus species, population, assemblage, or ecosystem research priorities because different filters can be applied to identify knowledge gaps. Informed predictions about climate change effects for under-or unrepresented species can then be generated using projections and observations among conspecifics or surrogate species or ecosystems. The FiCli is a living database that will be updated by its curators as new studies are published, and users also have the opportunity to inform curators of peer-reviewed publications that should be considered for addition to the database.

Methods
Database, R-Shiny and Decision Path tool. The FiCli database was generated from a review of primary scientific literature from 1985 to 2018. The year 1985 was a conservative choice for representing the earliest publications of effects of climate change on fishes, based on initial searches. The database initially included primary journal articles on climate change effects for North American inland fishes (Lynch et al.) 8 and was then expanded to include global coverage for publications that explicitly assessed the projected and documented effects of climate change on inland fishes (Myers et al.) 10 . These papers were partitioned into studies documenting effects of climate change on fishes ("documented effects, " generally based on long-term datasets) or through projections of future effects ("projected effects, " often by applying modeling approaches). Some studies included both documented and projected effects.
We present a brief description of the criteria used for inclusion of papers in the database based on previous literature reviews (Lynch et al., Myers et al.) 8,10 . Our literature review included multiple rounds of review by team members to ensure appropriateness for inclusion in the dataset. We then re-reviewed all papers in the database to identify suggested management recommendations and a binary response category to illustrate whether the species or population exhibited, or were projected to exhibit, a positive, negative, mixed, unknown, or no effect of climate change. For example, increases in growth and distribution were considered to positively benefit the species, while earlier or later spawning time was considered to be an unknown effect. This binary response category is a coarse indicator of direction of response to climate change and was determined based on the interpretation of results presented in the reviewed paper and our best judgement. For ease of interpretability, the management recommendations were binned into eleven overarching categories (Table 1).
In order to allow public access to the database, the FiCli database is available as a R-Shiny application supported by the U.S. Geological Survey (USGS), National Climate Adaptation Science Center (available here: https://ficli.shinyapps.io/database/) and does not require installation of the R program. All data and the R-Shiny application code are made available by USGS (https://doi.org/10.5066/P9VKJ23R). The online FiCli database is dynamic, and new papers will continue to be added on a rolling basis as they are published. A web upload form on the R-shiny site allows user-submitted database entries that will be reviewed by the FiCli team and added as appropriate. Future plans include: (1) expansion and re-release of the FiCli database to include studies with a primarily marine focus, and (2) inclusion of papers that use experimental approaches to describe fish response to changes in temperature and flow in the context of climate change; such studies were initially excluded from ours and previous literature reviews (Lynch et al., Myers et al.) 8,10 .

Data Records
The FiCli database includes summarized data on study systems, species, climate change responses, and adaptation and/or management actions into standardized categories that can inform species vulnerability assessments and adaptation planning (Online-only Table 1). Information was extracted from each peer-reviewed publication that the author team identified for the systematic literature review. A series of filters have been embedded into the application for customizable searches (Online-only Table 1). A graphic description of the results can be generated within the R-shiny application (Fig. 2), and detailed summaries can be exported as Bibtex or Ris files. The original raw data tables used to create the FiCli database are also available on Figshare (Krabbenhoft et al.) 12 .

technical Validation
The literature compilation and summary presented in the FiCli database have been reviewed and validated by multiple approaches. Each publication included in the database has been reviewed for content at two levels: (1) a cursory inspection for applicability and (2) a detailed review to extract relevant information entered into summary categories (i.e., database columns). The information included in each reference entry (i.e., database rows) was then reviewed by another author of the database.
Many of the references included in the database formed the basis of two recent primary literature synthesis publications by this group of authors. The first publication (Lynch et al.) 8 synthesized climate trends that influence North American inland fish populations and assemblages and included case studies illustrating the discerned trends. The second publication (Myers et al.) 10 was a global synthesis of documented and projected climate change effects on inland fishes. Thus, the literature references provided in the database have undergone review for accuracy The USGS, an affiliation of multiple authors, requires internal review of science products, including manuscripts for publication and data incorporated into publications. The database contents were reviewed and made publicly available through a Data Management Plan required of all USGS science products, and the manuscript was reviewed internally in adherence to the Fundamental Science Practices process. The USGS review process is intended to ensure and enhance the quality, accuracy, and availability of all science products released to the public and scientific community.
The references included in the database were derived through a systematic literature review, and information included in the database has been intensively reviewed. Though the literature included was considered thorough, it is unlikely to be exhaustive, especially as new science is published. Semi-annual updates to the database are planned. The most recent update date will be provided in the metadata. Further, we summarize key information extracted from individual publications on the documented or projected impacts of climate change on inland fish globally, but we have not reviewed each publication for accuracy and rely on the journal peer-review process to ensure scientific quality, accuracy, and integrity.
In conclusion, climate change will continue to impact inland fish populations and assemblages, productivity, and overall biodiversity (Diaz et al., Till et al.) 13,14 . The full extent of the consequences of climate change on inland waters is yet to be known, but inland fish and fisheries are changing with implications for human use of these natural resources. The FiCli database will make the most up-to-date peer-reviewed information of climate change impacts on inland fish easily accessible and will serve to provide information to managers and policy makers on actions to enhance climate change adaptation in inland waters.

Code availability
Code is available for download at https://doi.org/10.5066/P9VKJ23R. There are no restrictions to the access or use of this code. Code was implemented in R (version 3.6.1; https://r-project.org) using the package shiny (version 1.3.2).

Binned Adaptation and Management Recommendations Example from Literature
Prioritize populations based on vulnerability Promote thermal-tolerance diversity when prioritizing metapopulations for conservation (Anderson et al.) 15 Manage flows Work with dam managers to optimize reservoir release schedules (Segurado et al.) 16 Restore connectivity Maintain connectivity to inlet streams to promote cooling (