The GBIF France new portal : Atlas of Living France (ALF)

Authors : Marie-Elise Lecoq4, Anne-Sophie Archambeau4, Eric Chenin4 , Fabien Cavière4, Sophie Pamerlon4, Régine Vignes-Lebbe4

4 GBIF France, 43 rue Bufon, 75005 Paris - gbif(at)gbif.fr

With more than 500 million records documenting species occurrences shared and accessible through GBIF.org, the Global BiodiversityInformation Facility (GBIF) is a major actor in biodiversity data discovery. The launch of the new GBIF website in November 2013 saw animprovement in the reliability, access and ftness for use of the published data. With this amount of data, it is not possible to ofer as muchdetail as in national portals, especially in terms of map visualization, feld search, data quality, etc.

GBIF France1 launched its new portal2 totally based on Atlas of Living Australia3 modules at the end of May 2015 after 3 months ofconfguration and installation. We have been able to match this generic infrastructure to our local needs (e.g., links to diferent Frenchstructures such as Service du Patrimoine Naturel for species pages) and requests (e.g., adding a map to the result page of an occurrencesearch), developing our own design by adding the portal to our national website, and translating the ALA portal into French.

Process of publishing data into ALF

Those PHP scripts (a) use theGBIF API (b) to insertinformation on institutions,collections, data resourcesand contacts using the UUIDprovided by the GBIF into theMySQL instance (d). We usethe Client URL RequestLibrary named CURL to callthe GBIF web services. GBIFweb services return JSONfles (c)

Collection pages

PHP

MySQL

The metadata registry isstored on a MySQLinstance

We download archives fromGBIF web services (2)through the loading tool (1)proposed by the software.It stores the DwC-Archive(3) locally and creates, onthe MySQL instance (4), ametadata record for eachresource.

DwC-A

Institutio

n pages

Contact p

ages

Biocache command line tool is a java executablewith a scala implementation. The entire processtakes four steps : loading, processing, sampling andindexing data into the Apache Cassandra (NoSQLDatabase).

Data re

source pages

JSON

(a)

(b)

(d)

(c)

(1)

(2)

(3)

(4)

Reference: 1 www.gbif.fr 2 www.gbif.fr then click on « Consulter » 3 http://www.ala.org.au/

Occurre

nce  searc

h engine

ALF system

GBIF.org