VertNet: A New Model for Biodiversity Data Sharing

Abstract

Biodiversity is in a crisis caused by multiple human impacts on the environment [1]–[3]. The immediate and critical tasks in addressing this crisis are to examine global biodiversity patterns and document changes through time and space in order to understand the factors contributing to loss of biodiversity [4]. Meeting this challenge has emerged as a global priority [5]–[7] that requires approaches to mobilize data across broad geographic and taxonomic ranges. The community of vertebrate natural history collections has begun to meet this challenge by establishing social and technological infrastructures that provide open access to species occurrence data through broad participation and funding from the US National Science Foundation (NSF) and the Global Biodiversity Information Facility (GBIF). One result is VertNet, a publicly accessible database of vertebrate biodiversity data from natural history collections around the world. VertNet currently consists of four existing global vertebrate networks: Mammal Networked Information System (MaNIS) (http://manisnet.org – mammals [8]); Ornithological Information System (ORNIS) (http://ornisnet.org – birds); HerpNET (http://herpnet.org – amphibians and reptiles); and FishNet 2 (http://www.fishnet2.net – fishes). These networks collectively mobilize over 52 million records from over 70 institutions, which represent about 70% of all the vertebrate species occurrence data that are accessible through GBIF. VertNet was created to develop the tools and infrastructure necessary to make the data in these distributed networks available in a standard format to maximize their potential for understanding and protecting biodiversity. GBIF and VertNet work synergistically to enhance biodiversity data mobilization efforts. GBIF has identified the important role that VertNet will play in its new emphasis on decentralization of services and applications [5]. In particular, VertNet provides important data maintenance services, including data cleaning and indexing, thus removing development and deployment burdens for many fundamental tasks from GBIF. Data from VertNet are currently accessed through the four networks by a broad audience at a high frequency of about 2.5 million records per week. The networks continue to grow, even those whose extramural funding have expired; MaNIS, for example, has grown from 17 contributing institutions under the original grant to 38 institutions, with a waiting list of 31 more that will be added to the network as time and resources permit. The same enthusiasm is mirrored across the other three vertebrate networks. This rapid growth demonstrates an important sociological shift from skepticism to enthusiasm for data sharing. Ironically, the success of these networks has become their biggest challenge, straining original architectures and demanding a scalable and more sustainable solution. Below, we provide perspectives on the sociological and technical developments that brought vertebrate biodiversity networks to this point and discuss solutions to the immediate and anticipated challenges.