Global Observation Systems for Biodiversity
RJ (Bob) Scholes Council for Scientific and Industrial Research South Africa
[email protected] Decline and loss of biodiversity
– The great majority of plant and vertebrate species are declining in distribution, abundance or both – Humans have increased the species extinction rate by 50 to 1,000 times the rates typical in the fossil record – 10–30% of mammal, bird, and amphibian species are currently threatened with extinction – Diversity of genes and populations, and extent of near- natural ecosystems is currently declining in most places in the world Outline
• What is biodiversity? • What is an ‘observation system’? • Who needs it? • How might it be organised? • How can remote sensing help? Biodiversity The variety of life on earth
composition
land cover species richness
ecosystem genes services populations ecosystems structure function Structure of 21st century science
Research Observation Diversitas, IGBP, IHDP etc GEOSS, CEOS, IGOS Observation
Assessment IPCC, MA, Imoseb etc
Policymakers UNFCCC, CBD, CCD, etc Observing systems …are more than just ways of collecting data • Complete chain from observation to use • Seamless continuum from observations to products So what is the problem? • The picture with respect to biodiversity is patchy – Geographical gaps (tropics, southern hemisphere) – Topical gaps (invertebrates, marine organisms) – Inconsistency in space and time • The delivery pipeline is blocked – Many more data are collected than are used – Key constraint is ‘interoperability’ – Requires • Data sharing policies and protocols • Harmonisation of methods A political process to help fix the problem Global Earth Observing System of Systems Tokyo Brussels Washington DC Eath Observation summit G8 Evian Earth Observation summit Earth Observation summit nd st rd World Summit on SustWorld Summit Dev 2 1 3
Sep Jun Apr Feb 2002 2003 2004 2005 IPTT GEOSS The End GEOSS’ 9 Societal Benefit Areas • Natural Hazards and Disasters • Human Health • Energy Resource Management • Weather • Water • Climate • Ecosystems • Agriculture • Biodiversity Benefits of integration
Health Climate Water
Biodiversity Ecosystems
Agriculture Users of a biodiversity observation system
• International treaty processes – CBD, CCD, CITES, Ramsar, CMS • Biodiversity and conservation NGOs – IUCN, WWF, CI, WCS, TNC etc • National and local conservation agencies and biodiversity custodians • Researchers Information needs revealed by MA • Genuinely global databases – Consistent, reliable, all ecosystems • Time series of change • Information beyond species richness – Presence/absence is an insensitive indicator • Functional biodiversity – Ecosystem services: particularly support, regulation and spiritual/recreational services • Linked, georeferenced socio-economic data The concept of sample hierarchies GTOS Global Hierachical Observation SysTem
High intensity, process-oriented research 100
Representative, quasi-continuous stations 1000 Models
Systematic, periodic plot samples 10 000
Wall-to-wall, frequent coverage: images, maps & models Tier 1: Intensive research sites
Skukuza, Kruger Park, South Africa Tier 2: Stations Tier 3: plots
Kalahari transect Tier 4: Wall-to-wall observational products Land cover and use are central
Merged land-cover, WWF protected areas Ecoregions Effect of ecosystem use on population abundance
1.2
1 1.2 0.8 1 0.6 0.8 0.4 BIRDS
0.6 0.2
0.4 0 PLANTS Protected Light use Degraded Cultivated Urban Plantation
0.2 Average expert impact estimate 0 Protected Light Use Degraded Cultivated Urban Plantation Scholes & Biggs (2005) Nature 434:45-48 Interoperability Biodiversity shows the way
Broadly: making information in one system available to other systems • Sharing agreements, transparency, standards Narrowly: Ways of coding information so that it can be queried and interpreted by other systems • ABCD: Access to Biodiversity Collection Data • Darwin Core • EML: Ecological Metadata Language An architectural sketch
Change Gazeteer community ecosystem Remote maps •Location •Name •Location sensed •Location Ecosystem •Date Images Surfaces •Polygon •Species 1 •Date services •Species n •Attributes Conservation plans Spp list specimen population models •Species •Species Distri- •binomial Research •Location •Location bution •authority outputs •Date •Date GIS maps •synonyms •Source •Abundance Resource trends Taxonomy fingerprint use abiotic Biodiversity •Species •Species •climate •offtake Tabula- trends •Genus •Location •substrate •value tions •Family •Date •Topo/ •users •Class etc •Genes bathy Responsibilities GEOSS Change Gazeteer community ecosystem maps •Location Images •Name •Location •Location Ecosystem •Date IGOL, GITAN Surfaces •Polygon •Species 1 •Date services •Species n •Attributes Conservation plans Spp list specimen population models •Species •Species Distri- •binomial GBIF, CoML•Location •Location bution •authority Resource •Date •Date GIS maps •synonyms trends •Source •AbundanceFAO,
Taxonomy fingerprint IUCNuse abiotic Biodiversity •Species •Species •climate BCoL •offtake Tabula- trends •Genus •Location •substrate •value tions •Family GenBank•Date •Topo/ •users •Class etc •Genes bathy Ongoing and near future activities
• GEOSS Work Package 60 – Ecosystem classification – Site networks • Mid-decade land cover (for 2010 target) • Linking of digital collection records (GBIF) • Interpolation of species distributions Not so impossible dreams… • Remote sensing of biological communities and individual organisms – Hyperspectral (100’s of narrow bands) – Hypertemporal (daily to weekly observations) – Hyperspatial (~ 1 m resolution) • Population dynamics of many species, under a range of pressures • Extensive genetic information for important functional groups • Fully integrated volunteer networks Remote sensing as a cube… spectral al ati sp
t em p or al Hyperspectral imagery may be able to distinguish individual dominant species
Multi-Sensor MicrosatelIite Initiative
To fly on ZASat ~ Dec 07 Hyperspectral instrument by Belgium 200 bands, 10 nm wide, 400-2400 nm 15 m GSD 15 km swath ~ 4 revisits/year ‘Hypertemporal’ imagery is our best chance to obtain ecosystem function information GPP = LΣ(FPAR*PAR)
2001 0.40
0.35
Modis data 0.30
0.25
also 0.20
0.15
AVHRR, Greenness
Envisat 0.10
etc 0.05
0.00
-0.05 28-Jul-01 16-Sep-01 5-Nov-01 25-Dec-01 13-Feb-02 4-Apr-02 24-May-02 date ‘Hyperspatial’ imagery tells us about fine structure
PAN image Segmented image, colourised to show (QuickBird 1 m GSD) structurally different covers Other remote sensing of structure… •SAR – Sees through clouds (very important in the tropics) – 20-30 m resolution – Demonstrated ability to map biomass 10-50 Mg/ha – Potential for mapping wetlands • Lidar – Vertical and horizontal profiling of vegetation – Intuitive ecological interpretation – No space-based instruments yet •BDRF Summary • Biodiversity observations are abundant for species at point locations, but poorly accessible • Ecosystem-scale time-series products are essential, but rare • Remote sensing has much to offer in contributing to an integrated biodiversity observation system, especially – Land and sea ecosystem type, extent and fragmentation – Ecosystem function indicators