Transcript Measuring biodiversity

Measuring biodiversity
Dr RJ (Bob) Scholes
Chair, Global Terrestrial Observing System
CSIR Environmentek
PO Box 395, Pretoria 0001, South Africa
[email protected]
This paper will cover…
• Theory of biodiversity observations
• Existing approaches and systems
• Approaches that may satisfy the goals
Biodiversity: 3 aspects x 3 levels
(Noss 1994)
Landscape
patterns
Physiognomy/ha
bitat structure
Population
structure
Genetic
structure
At any level, diversity has at
least two components…
• How many different types of things are
present
– Elephant, rhino and lion is less diverse than
– Elephant, rhino, lion, leopard and buffalo
• How evenly they are represented
– 1000 elephants and 1 lion is less diverse than
– 500 elephants and 500 lions
‘Academic’ ways of measuring
biodiversity
Species level
• Richness: Total number of species in an area (
diversity)
• Species turnover along a gradient ( diversity)
Ecosystem level
• Number of different habitats or ecosystems (
diversity)
Genetic level
• Genetic homology
• Cladistic distance
‘Policy’ ways of measuring biodiversity
• ‘Extinction based’ (IUCN)
– Threatened species (Red Data Books)
• ‘Area based’ (Millennium goals)
– Area under protection
– Area of a key habitat (eg Forest cover)
• ‘Richness based’
– Indicator groups or species eg CI Rapid Biodiversity
Assessment
• Complementarity –based
– Various conservation optimisation tools, eg CPLAN
• Various spatial representations
– Hotspots, last wild places
Royal Society Report
2003
• ‘… no sound basis exists for assessing
performance against these targets.’
• ‘The fate of organisms not yet
recognised by science cannot be
measured’
• Lack of baselines
• Biodiversity measures must be related
to the objectives of measurement
Attributes of a good indicator
•
•
•
•
•
•
•
Does it measure what it says it does?
Sensitive, but not oversensitive
Scale appropriate in time and space
Well-understood model
Reliable data available
Monitoring systems in place
Understandable by policymakers
(NRC 2000)
Natural Capital Index
RIVM/UNEP-WCMC/GEO-3
NCI = ecosystem quality x ecosystem quantity
Example: NCI for The Netherlands
SAMA* Biodiversity Intactness Index
• Based on impacts on populations,
rather than extinctions
• Considers a range of impacts
– Protected, sustainably used, unsust
used, partially transformed, transformed
• Scale independent
• Applicable now, but amenable to
incremental improvement
Millennium Ecosystem Assessment
Strengthening Capacity to Manage Ecosystems Sustainably for Human Well-Being
*Southern Africa Millennium Ecosystem Assessment
SAMA Algorithm
m
t
n
m
t
j 1
i1
k 1
j 1
i 1
B  (   Cijk AjkRij ) /(  AjRij )
B = biodiversity intactness index
Cijk = populations of i under use k/ popn when protected
Ajk= Area of land use k in ecosystem j
Rij = Richness of taxon i in ecosystem j
i= taxon, from 1 to t
j= ecosystem, from 1 to m
k= land use type, from 1 to n
Needs: Land cover, richness, impact matrix
Worked example
South Africa, biome resolution
Taxa
Ecosystems
Plants
Forest
Fynbos
Grassland
Nama Karoo
Savanna
Succulent Karoo
Thicket
Mammals
Birds
Reptiles
Amphibians
0.772469
0.701956
0.688667
0.93602
0.816195
0.923047
0.825957
0.746174
0.683595
0.654478
0.885709
0.76951
0.878428
0.774175
0.926298
0.904151
0.899739
0.975683
0.939368
0.972191
0.942372
0.861815
0.825405
0.814188
0.943799
0.88569
0.93975
0.886054
0.780285118
0.713922301
0.698459825
0.936551519
0.820343291
0.92433773
0.828344218
0.789639
0.70769093
0.70942015
0.93890049
0.82664655
0.92494938
0.84208857
0.81082
0.768388
0.937475
0.880225 0.815859638
0.8087939
WWF Ecoregion database
867 biodiversity-based regions of the world
Based on best available information
Species lists for birds, mammals, reptiles, plants,amphibia
Global land cover products
• Many are now available
– DISCover, FAO-FRA, GLC 2000, Modis…
• Global coverage, resolution < 1 km
– 20 m products available for key areas
• Methods and results convergence
– GOFC/GOLD (a GTOS panel)
• Mid 1990s baselines feasible, year 2000
baseline in hand
• Reliable expectation of year 2010 repeat
What GTOS can offer
Global Terrestrial Observing System
• Biodiversity is one of the five mandated
topics covered by GTOS
– Land, freshwater, cryosphere
– Close collaboration with GOOS on coasts
– GOOS covers open ocean
• TEMS database
– ‘biodiversity module’
• Biodiversity network: methods harmonised
• GOFC/GOLD
– Land cover dynamics, especially forests
TEMS: Terrestrial Ecosystem
Monitoring Sites
Who, what, where
Web directory of 1,600 sites and 55 networks in
110 countries that carry out long-term terrestrial
ecosystem monitoring of 110 variables
http://www.fao.org/gtos/tems
TEMS Biodiversity module
Variables specific to biodiversity:
•Colonization of habitat by invasive species
•Habitat conversion
•Habitat fragmentation
•Indicator species
•Pollinator species
•Species Richness
•Threatened Species
Many of the other 115 variables in TEMS are also directly linked
to Biodiversity.
The CBD and WSSD goals
CBD CoP VII/26, WSSD Implementation Plan
• ‘..significantly reduce the rate of loss of
biodiversity by 2010…’
• This is a ‘double differential’ problem
– Change in a rate
• Requires at least 3 snapshots in time to
solve
Rate 1
Rate 2
Biodiversity
rate 2 < rate 1
1990
2000
2010
Global strategy for plant conservation
CBD CoP VI/9 April 2002
1. Accessible list of plant species…
2. Assess status of all species…
3. Understand conservation needs for
threatened species…
4. 10% of each ecological region, 50% of
species conserved in situ
5. 90% threatened species cons ex situ
6. 30% of production lands managed
consistent with conservation goals
UN Millennium Goals
• ‘…reverse the loss of environmental
resources…’
• Proposed indicators:
– Proportion of land area covered by forest
– Proportion of land area protected for
biodiversity conservation
• These indicators are measurable, but
not necessarily sensitive to the goal
Pragmatic issues…
• For the purpose of evaluating progress
towards the goals, biodiversity
measurements
– Don’t have to be perfect, just agreed
– Need to be based on ‘activity’ rather than
‘stock’ measurements (cf UNFCCC)
– Satellite-based land cover measurements,
coupled with ‘sparse’ in-situ information in
an explicit way (a ‘model’) could do the job
for terrestrial systems
A proposal
Agree to develop an approach based on
1. Land cover/use in 1995, 2000, 2005 and 2100
[GOFC/GOLD + WCMC]
2. Richness within ecosystem units [WWF +
Taxonomy initiatives+ NGOs + nations]
3. An impact matrix (land use x taxa, per biome)
derived from site data, models and expert
judgement [ Diversitas + GTOS]
For test by 2005, and retrospective application by 2010