Myrmecological News 24 83-89 Vienna, March 2017 Forum The Global Ant Biodiversity Informatics (GABI) database: synthesizing data on the geographic distribution of ant species (Hymenoptera: Formicidae) Benoit Guénard (contact author), School of Biological micidae), have brought novel insights to macroecological Sciences, The University of Hong Kong, Hong Kong SAR; and evolutionary questions with a global perspective (DUNN Okinawa Institute of Science and Technology Graduate & al. 2009, WEISER & al. 2010, JENKINS & al . 2011, GUÉ- University, Okinawa, Japan. E-mail: [email protected] NARD & al. 2012, LUCKY & al. 2013). Other projects have Michael D. Weiser, Department of Biology, University of considerably improved knowledge on species distributions Oklahoma, OK, 73019, USA. at various spatial and taxonomic scales including: global Kiko Gómez, Castelldefels, Barcelona, Spain. scale (e.g., specimen records at ANT WEB 2015), biogeo- Nitish Narula & Evan P. Economo, Okinawa Institute of graphical regions (e.g., Neotropical: FERNANDEZ & SEN - Science and Technology Graduate University, Okinawa, DOYA 2004), continents (e.g., Europe: BOROWIEC 2014), Japan. countries (e.g., China: GUÉNARD & DUNN 2012, Costa Rica: LONGINO 2010, Fiji: SARNAT & ECONOMO 2012, Myrmecol. News 24: 83-89 (online 25 January 2017) licensed under CC BY 3.0 India: BHARTI & al. 2016, Japan: JAPANESE ANT DATA - BASE GROUP 2003); or sometimes more specifically on a ISSN 1994-4136 (print), ISSN 1997-3500 (online) given taxonomic group (e.g., Myrmica : RADCHENKO & 19 September 2016; accepted 29 September 2016 Subject Editor: Jens Dauber ELMES 2010) or on ecological groups (e.g., introduced spe- cies: WETTERER 2008). All these efforts provide valuable Abstract: The global distribution patterns of most vertebrate contributions that focus on specific facets of biodiversity groups and several plant groups have been described and ana- data, but thus far a comprehensive, global accounting of lyzed over the past few years, a development facilitated by the the known distributions of all ant species is still needed. compilation of important databases. Similar efforts are needed A long history of myrmecological research has accu- for large insect groups that constitute the majority of global bio- mulated a tremendous amount of biodiversity data in the diversity. As a result of this lack of information, invertebrate taxa are often left out of both large-scale analyses of biodiversity pat- form of published literature, museum collections, and spe- terns and large-scale efforts in conservation planning and prior- cimen record databases. However, the utility of these data itization. Here, we introduce the first comprehensive global data- for synthetic research has been limited due to its disaggre- base of ant species distributions, the Global Ant Biodiversity gation and fragmentation. The literature records are scat- Informatics (GABI) database, based on the compilation of 1.72 tered across a large number of often obscure and difficult million records extracted from over 8811 publications and 25 to access publications, some hundreds of years old, not existing databases. We first present the main goals of the data- found in normal University libraries, and often published base, the methodology used to build the database, as well as its in languages other than English. Moreover, the continued limitations and challenges. Then, we discuss how different fields evolution of ant taxonomy means that old literature records of ant biology may benefit from utilizing this tool. Finally, we need to be curated and reconciled with our latest taxono- emphasize the importance of future participation of myrmecolo- gists to improve the database and use it to identify and fill holes mic frameworks in order to remain relevant. Even if ag- in our knowledge of ant biodiversity. gregated these data are undoubtedly incomplete both for described species as well as the large number of yet un- Key words: Formicidae, ants, database, global distribution, described species (WARD 2014). species distribution, ecoinformatics, biogeography. The Anthropocene biodiversity crisis (CORLETT 2015) has created a pressing need for the development of a bio- Introduction diversity informatics framework that aggregates these di- Over the past few decades, researchers have generated data verse data types to understand and close the important geo- sets that describe species distributions and global diver- graphic (GUÉNARD & al. 2012) and taxonomic gaps (WARD sity patterns for several plant and vertebrate groups (e.g., 2014) of species distributions. Such databases can be used KIER & al. 2005, ORME & al. 2005, BUCKLEY & JETZ to monitor the distribution of known species, provide raw 2007, SCHIPPER & al. 2008, PIMM & al. 2014). These ef- data to accelerate taxonomy and inventory work, and pro- forts in "biodiversity informatics" have facilitated unpre- vide tools that will facilitate field research for ecologists cedented understanding of patterns of biodiversity (e.g., and conservation biologists. KREFT & JETZ 2007), the processes shaping them (e.g., Here we introduce the Global Ant Biodiversity Infor- KERKHOFF & al. 2014), and have considerable implications matics (GABI) database, an attempt to summarize the for conservation (JOPPA & al. 2013). While insects repre- knowledge of all ant species distributions, and in their sent more than two-thirds of described species (ZHANG aggregate form, geographic patterns of ant biodiversity. We 2013), they have been mostly left out of global-scale com- describe the overall goals of the project, the data sources prehensive analyses (DINIZ -FILHO & al. 2010, but see FO- and compilation methodology, the organization of the data, LEY & al . 2007). Recent attempts, using community-level and review some potential scientific uses of the data. The data and generic distribution records of an ecologically data can be viewed through a custom-built web-mapping dominant and ubiquitous group, ants (Hymenoptera: For- interface, antmaps.org (JANICKI & al. 2016). Additional mechanisms to distribute the data are under development, ever represent a small proportion of our data and are and pending its completion and full documentation in the added on the basis of trust of known experts. We do literature, we intend for the GABI data to be an open re- hope that more direct information could be collected in source available to all. We expect GABI to become an im- that manner in the future through direct communication portant new tool for biologists for the study and under- with local experts. Toward that end, the release of the standing of ant distribution at all taxonomic levels; and website antmaps.org (presented in details in JANICKI & open a new era of more rapid and penetrating macroeco- al . 2016) and of the option "Report Data Issue" has faci- logical, macroevolutionary, and taxonomic investigation litated the communication about personal records. of the Formicidae. With this new tool, we hope that ant Organization and presentation of the data biologists will gain a better understanding of ant species distributions and use it to accelerate the discovery and de- Taxonomic information: Taxonomy is an evolving field scription of new taxa and new species distribution records of biology, and the taxonomic frameworks used for ant in the mold of what already exists for vertebrate groups like biology remain dynamic at all levels (WARD 2007, 2014). birds. One major challenge for the construction of a global data- base is to ensure that the current taxonomy of the taxa used Goals of the GABI project represents their most updated status. Thanks to tools de- Compile all known information on the geographic distri- veloped previously (e.g., BOLTON 2014, 2015), taxonomic bution of all ant species in a single database, including a changes have been kept up to date along the construction comprehensive accounting of publications, specimen data- of GABI, with continuous synchronization with newly pub- bases, museum collections, and other collection records. lished literature. – Identify erroneous and dubious records that may exist To ensure the traceability of every record and of their in the literature or other data sources. changes over time, the information relative to the original – Provide range maps for each species, genus and sub- presentation of the taxon has been kept separately within family, distinguishing native and exotic ranges. the database. In a second step the current valid name was – Provide a first draft of species checklists for all regions updated in separate fields using the community website of the world. BOLTON (2015). The validity of each taxonomic combina- – Provide the raw material for enhanced biogeographic tion has been checked manually. This process permits for analysis of ant biodiversity patterns. the correction of species misspelling that could artificially – Provide the raw material for integrating ant biodiversity inflate the species richness in a given region. As a result, information into global conservation efforts. a total of 38,239 combination pairs of original taxonomic – Develop an interactive framework for the continued description*valid species names have been retrieved. For growth and improvement of the database, with the par- instance, records of Amblyopone pallipes were changed to ticipation of the global myrmecological research com- Stigmatomma pallipes (HALDEMAN , 1844) following taxo- munity. nomic changes of YOSHIMURA & FISHER (2012), resulting in the pair Amblyopone pallipes