It is the goal of the Avian Knowledge Network (AKN) to share observational data on bird species with as widespread an audience as possible. With this in mind, we recognize the need to assure data quality and validity, and that data sets available through the AKN have appropriate documentation (metadata) to judge data quality and potential usefulness. We must also assure the privacy rights of landowners and data contributors.

All contributing members of the AKN have complete authority over the use of the data they have provided, and can withhold data at any time from any party or application. Each data set contributed to the AKN has an associated level of access to that data (see below). It is recognized that the investment and time committed to the collection of a data set entitles the owner to the fundamental benefits of its use.

All users of any AKN data must clearly acknowledge the contribution of the members who are making data available. This acknowledgement shall include at most, co-authorship, and at least, recognition in any printed or Internet content, and will depend largely on the degree to which a given analysis or publication relies on any single data provider’s data. Additionally, appropriate branding (i.e. logos of contributing members) should be placed on any website, educational, outreach, presentation, or promotional materials on which the data are used. All researchers making substantial use of a data set are required to communicate directly with the investigators who acquired the data, or those to whom responsibility for the data has been entrusted, prior to publication. Consultation before and during analysis of the data is also essential and required. The AKN coordinator will require and facilitate the ethical and courteous use of data within the AKN archives. AKN data will not be modified except as needed to ensure compatibility with the recipient’s geographic information system. The AKN advisory board has the authority to mediate and decide any conflicts regarding data usage and publication.

All data contributed to the AKN is clearly documented using the descriptive metadata standards of the Bird Monitoring Data Registry (BMDR). The BMDR describes the data ownership, geographic coverage, protocols, and access rights. All data stored in the AKN conform to the Bird Monitoring Data Exchange (BMDE). The BMDE is a data exchange standard that extends the Darwin Core metadata schema for data integration (http://www.avianknowledge.net/organization-discovery/data_exchange). It is the responsibility of data users to understand sampling differences and/or protocol effects between projects before using any data from the AKN.

Location coordinates of data held within the AKN are defined by the project that collects the data. Descriptions of the locational precision for a project can be obtained by the BMDR. The spatial data standards of the AKN will adhere as much as possible to those standards being developed by the International Union of Biological Sciences Taxonomic Databases Working Group’s subgroup on spatial standards (http://www.tdwg.org).

Levels of Data Management and Access
For a participant to contribute data to the AKN the following is required: a) BMDR must be completed, b) data must conform to BMDE standards, c) data must be linked to GRID protocol, and d) data are uploaded to the AKN’s primary data warehouse. When completed, the following data access levels and sharing options are available to the data owner:

• Level 1--All data are stored in the AKN’s primary data warehouse. Backups of the warehouse are made using persistent data archive techniques. AKN data managers use all data backup options consistent with the goal of no data loss (backups with periodic data integrity testing). The warehouse serves as the primary archive of all AKN data, and no applications connect directly to the warehouse. Instead, data from the warehouse are ported to separate data views created specifically to optimize the performance of an application that connects to it. Data owners can specify how their data can be used in the data views, with the option that their data are not exposed to the public at all.
• Level 2--All of option 1 with the following addition: data can be used in certain publicly available, predefined visualizations (i.e. maps and graphs), but direct access to the data is restricted.
• Level 3--Data can be used in publicly available, predefined visualizations (i.e. maps and graphs). Additionally, data is made available to existing bioinformatic efforts (GBIF and ORNIS). These bioinformatic efforts only provide the data “marked-up” to Darwin Core. This means that only locations where positive observations and the number reported are available.
• Level 4--Data are used in publicly available, predefined visualizations (i.e. maps and graphs), and are available to existing bioinformatic efforts. Additionally, the complete BMDE data set is available upon request from the original data provider.
• Level 5--Data are used in publicly available, predefined visualizations (i.e. maps and graphs), and are available to existing bioinformatic efforts. Additionally, the complete BMDE data set is available for download directly via the Internet.