Biodiversity trend indicators from OBIS

544 views
426 views

Published on

Trendylyzer.

Published in: Technology
0 Comments
0 Likes
Statistics
Notes
  • Be the first to comment

  • Be the first to like this

No Downloads
Views
Total views
544
On SlideShare
0
From Embeds
0
Number of Embeds
0
Actions
Shares
0
Downloads
6
Comments
0
Likes
0
Embeds 0
No embeds

No notes for slide
  • we only have one planet - We only have one ocean
  • We observed that Marine Species are moving much faster to the Poles than terrestrial species. With a loss in native species and a gain in new species.
  • Ocean acidification will affect cold water corals most in the North Atlantic
  • By the end of the century pteropod shells will be degraded by 50 to 80%.
  • Species are important for ecosystem functioning.
  • What is also interesting is that many taxonomic groups reached their maximum Nr of observations many years ago.
  • When looking at each term separately, there is a high variation in ranks, with the same fish species listed in the top 10 down below to rank 5000.
  • So when you look at Nr of records, the fulmar had high peaks in the 1990s.
  • But when you combine all the terms, the Fulmar is losing its common species status among the other seabirds.
  • We digged further plotting the 20 most common mammal species per decade, and gave rainbow colors to the 1990s list. Grey are those that are not common in 1990. The nice outcome is that there is still a rainbow in the other decades.
  • We validated our approach by taking 1000 random samples of 20 mammal species and compared it with 20 common species in that decade. => the set of common species in common between decades are in excess compared to the random set of species.
  • We repeated this for ten different taxonomic groups
  • We started this modeling approach by asking FishBase to provide us a list of common and rare fish species, which we then used for training and testing the Neural Network. The good news is that 80% of the classification was in agreement between the Function and the Expert classification.
  • Biodiversity trend indicators from OBIS

    1. 1. Trendylyzer: a Long-Term Trend Analysis on Biogeographic Data from OBIS Ward Appeltans, Peter Pissierssens - IOC-UNESCO Gianpaolo Coro, Angela Italiano, Pasquale Pagano - ISTI- CNR Anton Ellenbroek - FAO Thomas Webb - University of Sheffield IMDIS 2013 International Conference on Marine Data and Information Systems Lucca (Italia), September 23-25, 2013
    2. 2. One Ocean – One Planet The Ocean represents 71% of the Earth’s surface and 99% of the living space on the planet by volume Marine Plankton provides 50% of the O2 we breathe The ocean regulates climate by storing heat and absorbing 30% of the CO2 we produce Over 3 billion people depend on marine and coastal biodiversity resources for their livelihoods.
    3. 3. Climate change (rising temperature, acidification, melting sea ice..) Overfishing (30% of fish stocks overexploited) Pollution (micro plastics, 500 dead zones) Invasive species (Pests, Fouling, toxic algae) Habitat destruction (mangrove deforestation, bottom-trawling) 60% of the world's major marine ecosystems have been degraded or are being used unsustainably. Ocean is under threat
    4. 4. Ocean is under threat Effect on biodiversity?
    5. 5. Polar species are disappearing and warmer water species are moving towards the Poles Kaschner K, Tittensor DP, Ready J, Gerrodette T, et al. (2011) Current and Future Patterns of Global Marine Mammal Biodiversity. PLoS ONE 6(5): e19653. doi:10.1371/journal.pone.0019653 http://www.plosone.org/article/info:doi/10.1371/journal.pone.0019653 Average of 72 km (marine) <-> 6 km (terrestrial) per decade (Poloczanska, E.S. et al (2013). Global imprint of climate change on marine life. Nature Clim. Change. http://dx.doi.org/10.1038/nclimat e1958) Projected effects of climate change on marine mammal species richness.
    6. 6. Corals are bleaching van Hooidonk et al. 2013. Temporary refugia for coral reefs in a warming world. Nature Clim. Change http://dx.doi.org/10.1038/nclimate1829 50% 100%
    7. 7. Cold-water corals are dissolving Ocean acidification will affect cold water corals most in the North Atlantic present 2099 Tittensor, D. P. et al. (2010), Seamounts as refugia from ocean acidification for cold-water stony corals. Marine Ecology, 31: 212–225. doi: 10.1111/j.1439-0485.2010.00393.x
    8. 8. Shells are dissolving Work in progress as part of the GEOWOW project. Unpublished figures. Copyright Steeve Comeau and James Orr Pteropod data from OBIS, Aragonite saturation data from James Orr, Model from Steeve Comeau By the end of the century pteropod shells will be degraded by 50 to 80%.
    9. 9. 80% of the pollution at sea comes from land
    10. 10. 90% of large predator fish stocks are gone Myer & Worm (2003). Rapid worldwide depletion of predatory fish communities. Nature. 423
    11. 11. Based on the IUCN Red List: Only 2% of known global extinctions are marine Only 3% of known endangered (and critical endangered) species are marine Few global marine extinctions are documented, but many local extirpations, and decreasing populations Are we losing species?
    12. 12. High biodiversity => more stable ecosystem functioning (biological insurance) Species composition is one of the most basic metrics of biodiversity Most documented trends are on few taxonomic groups (large commercial species, plankton) Species diversity and Ecosystem Functioning
    13. 13. Ocean Biogeographic Information System OBIS is the world’s largest open access, online data system on the diversity, distribution and abundance of marine species 120,000 marine species 35 million distribution records 1,130 datasets >450 data providers (56 countries) Data legacy of 10-year Census of Marine Life Adopted by UNESCO-IOC, project of IODE, hosted by IOC project office for IODE in Oostende (Belgium) www.iobis.org
    14. 14. Marine Species Diversity The knowns and unknowns? 700K – 1 million marine species 230K described 120K in OBIS 12K OBIS/year 5,4K IUCN assessment Appeltans et al (2012). The Magnitude of Global Species Diversity, Current Biology 22
    15. 15. Only 1/3 to 1/4 of marine species are known 2/3 of known marine species have < 3 observations Only 5% of known species diversity is observed per year Most data from large vertebrates (fish, birds, mammals) Most data from Northern Hemisphere (least biodiversity rich) Most data from coastal areas (mid waters almost unexplored) Very little historical data - only 100 marine species have a yearly record between 1955 and 2005 Knowledge gaps
    16. 16. 1850 1900 1950 2000 Xenacoelomorpha Trematoda Thaliacea Tardigrada Staurozoa Solenogastres Sipuncula Seagrasses Scyphozoa Scaphopoda Rotifera Rhodophyta Rhabditophora Reptilia Remipedia Radiozoa Pycnogonida Priapulida Porifera Polyplacophora Polychaeta Pisces Phoronida Ostracoda Ophiuroidea Ochrophyta Nemertea Nematoda Myzozoa Myxozoa Myriapoda Monoplacophora Monogenea Mesozoa Merostomata Maxillopoda Mangroves Mammalia Malacostraca Loricifera Kinorhyncha Hydrozoa Holothuroidea Hexapoda Hemichordata Heliozoa Haptophyta Gnathostomulida Gastrotricha Gastropoda Fungi Foraminifera Euglenozoa Entoprocta Echiura Echinoidea Cycliophora Cubozoa Ctenophora Cryptophyta Crinoidea Clitellata Ciliophora Choanozoa Chlorophyta Chaetognatha Cestoda Cercozoa Cephalopoda Cephalochordata Cephalocarida Caudofoveata Bryozoa Branchiopoda Brachiopoda Bivalvia Bigyra Aves Asteroidea Ascidiacea Arachnida Appendicularia Anthozoa Amoebozoa Acanthocephala Maximum Observations in OBIS
    17. 17. Progress! Adding more records to OBIS 0 5 10 15 20 25 30 35 40 Apr-01 Sep-02 Jan-04 May-05 Oct-06 Feb-08 Jul-09 Nov-10 Apr-12 Aug-13 #rrecordsinmillions post CoML 37 million soon
    18. 18. Progress! Adding more datasets 0 200 400 600 800 1000 1200 Apr-01 Sep-02 Jan-04 May-05 Oct-06 Feb-08 Jul-09 Nov-10 Apr-12 Aug-13 post CoML 244 new datasets soon
    19. 19. Progress! Adding new species
    20. 20. Progress! We are filling in the gaps
    21. 21. Data e-infrastructure Initiative for Fisheries Management and Conservation of Marine Living Resources (i-Marine) Trendylyzer is a tool developed within iMarine iMarine is an initiative aimed at establishing and operating an e-infrastructure supporting the principles of the Ecosystem Approach to Fisheries Management and Conservation of Marine Living Resources. Research Infrastructures CP & CSA funded by the European Commission under the FP7 Capacities Programme - eInfrastructure Unit DG CONNECT (1 Nov 2011 - 30 April 2014)
    22. 22. Trendylyzer 22
    23. 23. Trendylyzer
    24. 24. Trendylyzer global North Sea Celtic-Biscay Shelf ?
    25. 25. Discovering trends in pelagic North Sea species Plankton regime shift Herring recovered after the fish ban 25
    26. 26. Major problems: Data gaps: temporal, geographic and taxonomic scope Sampling bias Few long-term time series data Discovering trends in biodiversity
    27. 27. Major problems: Most species are rare, or difficult to observe Still many new species discoveries creates noise in trends Which species are most common, and do have enough data? + common species may also have a bigger impact on ecosystem functioning compared to uncommon species Discovering trends in biodiversity
    28. 28. Questions: Are e.g. the 20 most common species for each taxonomic group always common? What is the natural variation in common species? Do we observe higher variation in species poor regions? Are there common species endangered (on the IUCN Red List of threatened species)? Biodiversity indicator based on Common Species?
    29. 29. For each species: 1. Nr of observations 2. Nr of individuals per observation 3. Nr of observations per dataset 4. Nr of datasets 5. Nr of geographical cells 6. Temporal frequency of the observations Normalizing => relative commonness. Create score or rank by taxonomic group Defining Common Species
    30. 30. High variation among terms, so each term separately will yield different ranks Pisces, by different ranks, 2000s n records s ranks n records2 s ranks2 n datasets n 1 deg n MEOW 5000 1000 500 200 100 20 10 9 8 7 6 5 4 3 2 1 Lim lim Lim lim Lim lim Lim lim Lim lim Lim lim Lim lim Gad mor Gad mor Gad mor Gad mor Gad mor Gad mor Gad mor Clu har Clu har Clu har Clu har Clu har Clu har Clu har Mer mer Mer mer Mer mer Mer mer Mer mer Mer mer Mer mer Mel aeg Mel aeg Mel aeg Mel aeg Mel aeg Mel aeg Mel aeg Ple pla Ple pla Ple pla Ple pla Ple pla Ple pla Ple pla Eut gur Eut gur Eut gur Eut gur Eut gur Eut gur Eut gur Pla fle Pla fle Pla fle Pla fle Pla fle Pla fle Pla fle Hip pla Hip pla Hip pla Hip pla Hip pla Hip pla Hip pla Thy atu Thy atu Thy atu Thy atu Thy atu Thy atu Thy atu Xip gla Xip gla Xip gla Xip gla Xip gla Xip gla Xip glaSpr spr Spr spr Spr spr Spr spr Spr spr Spr spr Spr spr Thu alb Thu alb Thu alb Thu alb Thu alb Thu alb Thu alb Scy can Scy can Scy can Scy can Scy can Scy can Scy can Mic kit Mic kit Mic kit Mic kit Mic kit Mic kit Mic kitOph elo Oph elo Oph elo Oph elo Oph elo Oph elo Oph elo Mer mer Mer mer Mer mer Mer mer Mer mer Mer mer Mer merTri min Tri min Tri min Tri min Tri min Tri min Tri minCal lyr Cal lyr Cal lyr Cal lyr Cal lyr Cal lyr Cal lyr Aci tra Aci tra Aci tra Aci tra Aci traTri esm Tri esm Tri esm Tri esm Tri esm Tri esm Tri esmThu ala Thu ala Thu ala Thu ala Thu ala Thu ala Thu ala Tra tra Tra tra Tra tra Tra tra Tra tra Tra tra Tra traSco sco Sco sco Sco sco Sco sco Sco sco Sco sco Sco scoMic pou Mic pou Mic pou Mic pou Mic pou Mic pou Mic pou Pol vir Pol vir Pol vir Pol vir Pol vir Pol vir Pol vir Squ aca Squ aca Squ aca Squ aca Squ aca Squ aca Squ aca Enc cim Enc cim Enc cim Enc cim Enc cim Enc cim Enc cimGly cyn Gly cyn Gly cyn Gly cyn Gly cyn Gly cyn Gly cyn Myo sco Myo sco Myo sco Myo sco Myo sco Myo sco Myo sco Lop pis Lop pis Lop pis Lop pis Lop pis Lop pis Lop pis Thu ori Thu ori Thu ori Thu ori Thu ori Rhi typ Rhi typ Rhi typ Rhi typ Rhi typZeu fab Zeu fab Zeu fab Zeu fab Zeu fab Zeu fab Zeu fab Mul sur Mul sur Mul sur Mul sur Mul sur Mul sur Mul surGym cer Gym cer Gym cer Gym cer Gym cerMol mol Mol mol Mol mol Mol mol Mol mol Mol mol Mol molCyc lum Cyc lum Cyc lum Cyc lum Cyc lum Cyc lum Cyc lum Pte vol Pte vol Pte vol Pte vol Pte vol Pte vol Pte vol
    31. 31. Results Fulmar PEAK based on Nr of records
    32. 32. Results Fulmar Losing its common species status among the seabirds? ?
    33. 33. Results Grey = not a common species in 1990 20 most common mammal species per decade
    34. 34. Results The top 20 common marine mammal species for each decade that are in common with the 1990 top 20 common are not just a random sample
    35. 35. Results 20 most common species per decade for 10 different taxonomic groups
    36. 36. Results
    37. 37. Results
    38. 38. Conclusion Work in progress!! We detected changes in common species composition: Changing order Species dropping in and out These trends will indicate shifting geographic/taxonomic focus of surveys, together with real changes in the abundance of some common species. => need to involve experts to validate the trends Biodiversity trends
    39. 39. Future The next step is now to relax the terms of the Commonness Score Function by giving a weight to each term using pattern recognition models such as artificial neural networks.
    40. 40. A preliminary test was done using a Bayesian Neural Network 40 fish species (20 rare and 20 common) for training 40 fish species (20 rare and 20 common) for testing The classification was obtained from FishBase (expert based) =>80% agreement between Function and Expert classification Future
    41. 41. For Policy Making, biodiversity trend indicators only make sense when combined with environmental and socio-economic data. Future Global map (A) of cumulative human impact across 20 ocean ecosystem types. B S Halpern et al. Science 2008;319:948-952
    42. 42. Indicators can serve national, regional and global assessments of the Ocean • 2014: CBD 4th Global Biodiversity Outlook • 2014: GEF Transboundary Water Assessment • 2015: 1st UN World Ocean Assessment • 2018: 1st IPBES assessment 42
    43. 43. One Planet – One Ocean THANK YOU This works is part of iMarine (283644) with funding from the European Commission under Framework Programme 7

    ×