Ocean Biogeographic Information System Edward Vanden Berghe.

Ocean Biogeographic Information System

Edward Vanden Berghe

‘Mission’

• OBIS publishes primary data on marine species locations online through www.iobis.org – It facilitates data discovery and exploration by

• Searching by species, higher taxa, time, location, depth, data set

• Mapping observed occurrences • Modelling of potential environmental range

– Integrates data over marine themes• Microbes to whales• Genetics and morphology• Poles to equator…

– Enables data capture for re-use• Provides the biogeographic context for

ocean research

Global loss of species from LMEsBiodiversity issue

Worm et al. 2006 in science

Why do this?

• Proper management of natural resources requires properly managed data and information– Several organisations sharing fisheries data

• OBIS model makes data and information management more efficient– Share responsibilities, tools, standards…– Share data across different organisations and

countries• OBIS is a way of ensuring data is not lost

– Archaeology and rescue for historic data– Repositories for new data

• Assist in data discovery– Links with EoL, BOLD…

OBIS as part of GBIF

• Same technology– DiGIR, investigating IPT

• Same structure– Darwin Core, OBIS Schema– Investigating expansion

• Same philosophy and terms of use

OBIS as part of UNESCO

• Adopted by the Intergovernmental Oceanographic Commission of UNESCO

• As part of their International Oceanographic Data and information Exchange programme

• So no longer an project-based activity, but intergovernmental

GBIF OBIS IODE

OBIS in context

• Data integration component of CoML– Capturing and integrating data– Support the 2010 synthesis

• Marine component of GBIF– Fully inter-operable with GBIF standards– Extending with marine-specific elements

• Biodiversity component of IOC– Under IODE programme– Contributing to GOOS

• Partner with FAO, UNEP (WCMC)• Hosted by Rutgers University IMCS• Funded by Sloan Foundation

Distribution of cod, Gadus morhua, shown as ‘c-squares’ map

Predicting distribution of invasive species, Pterois volitans

OBIS records viewed

0

20000

40000

60000

80000

100000

120000

140000

160000

180000

200000

Mar-2006 Oct-2006 Apr-2007 Nov-2007 Jun-2008 Dec-2008 Jul-2009 Jan-2010

Standards

• Biogeography: GBIF/TDWG– Darwin Core, Extended to OBIS Schema– ABCD

• Metadata: discovery metadata– Global Change Master Directory – NASA– MEDI – IODE; FGDC – US Gov?

• Taxonomy: World Register of Marine Species (WoRMS) – Contribution to Species 2000 and Catalogue of

Life• Geography

– Polygon sets• EEZs, FAO areas, IHO…

– Gazetteer

Extensions to Darwin Core

• Begin and end point– Position, depth, date/time– Accommodate trawls

• Sample size– Quantitative information

• In collaboration with GBIF working on system to extend standard– Ecological information, EML

Standards: taxonomy

• Aphia is general species register maintained at VLIZ– Consists of several overlapping subsets

• defined geographical (ERMS, NWARMS…) • defined taxonomic (Porifera,

Platyhelminthes…)• defined thematic (HABs, invasive species)

• Exposed through www.marinespecies.org

• WoRMS = Aphia + external GSDs– Algaebase, Hexacorallia, FishBase…

OBIS number of records

• 699 datasets• 20.1 million distribution records • 147,000 names, 107,000 taxa• Among the largest provider to

the Global Biodiversity Information Facility

• Who’s providing data?– Regional OBIS Nodes– Census of Marine Life

Map of CoML field projects

CoML & Ocean Observing

An example of how CoML technologies can be

implemented in earth and ocean observing systems

7 seals tracked during 2-3 month summer feeding migrations

Some animals dive 1000m

Some CoML Discoveries

Athorybia rosacea – a siphonophore from the

Sargasso Sea

Palinurus barbarae – a lobster from around Madagascar

Promachoteuthis sloani – a squid from the mid-Atlantic Ridge

Ceratoserolis – an isopod from the Weddell Sea

Kiwa hirsuta – a crab from around Easter

Island (nicknamed the Yeti crab)

HMAP data extends timeline

1

10

100

1,000

10,000

100,000

1,000,000

1600 1700 1800 1900 2000 2100

Location of RONs

Role of the regional nodes

• Ensuring true global cover for OBIS• Regional nodes are closer to the providers

of the data– Local visibility for global OBIS data– Mobilise data from region– Technical assistance, also with standardisation– Specialised information products and services

• Data available on the regional network are also available on the global network– Increased global visibility for local data and data

providers– Data sharing

Public Accessibility

IOBIS

Global Datasets e.g.FAO, HexacoralliaFishbase &Seamounts

Data Found ByIOBIS Not FromGlobal Databases

Online Providers

Regional Node With Local DatasetsAnd Online Providers

RegionalSubset

All data

Regional Node With Local Datasets Only

GBIF

Global NodeRegional NodeProviderGlobal DatabaseRegional DatabaseProductsAll DataSubset of Data

OBIS Regional Node Architecture

RMB - March 14, 2004

Still a lot of work…

• We don’t know the total biodiversity– New species are discovered

• Selective sampling in geography– Mostly in surface waters– Temperate zones

• Selective sampling in taxonomy – Mostly big things, vertebrates

New species are discovered

Data from http://marinespecies.org

Geographical bias

Bias in depth: all data

Bias in depth: deeper than 2500m

Taxonomic bias

Taxon # species # in OBIS %Cetaceans 133 117 88Seals… 45 36 80Fish 24139 21258 88

Echinoderms 6199 1624 26Decapods 8227 3796 46Bryozoans 6000 1096 18

Analysis of OBIS data

• First attempts at diversity pattern on a global scale, with a large number of taxa– Previously either local or on one taxon

(e.g. commercial large fish like tuna, forams…)

– ‘Safety in numbers’• Results not affected by idiosyncrasies of

single taxon or study

• Results very preliminary, and need data cleaning and further checking– E.g. by artificially removing datasets

from analysis

Global pattern of sampling effort

Pattern in number of species

Corrected for bias: ES(50)

Large Marine Ecosystems

Latitudinal gradient ES(50)

es(50)

0

10

20

30

40

50

60

-100 -50 0 50 100

Marine fish to be discovered

Mora et al (2007). The completeness of taxonomic inventories for describing the global diversity and distribution of marine fishes. Proc. R. Soc. B, published on line

Percentage completeness 1 100

Plans for the future

• More data and analysis• Develop thematic portals, on issues

of direct societal relevance– Invasive species, HABs…

• Develop demonstrator projects– Species distributions, hotspots…

• Support CoML scientists– Integration across themes– 2010 Synthesis– Publications: theme section(s)

Get in touch

• www.iobis.org

• [email protected], [email protected]