Transcript Silwood park

Silwood Park
2 June 2009
QuickTime™ and a
TIFF (Uncompressed) decompressor
are needed to see this picture.
Biodiversity and DGVMs
I. Colin Prentice, QUEST
QuickTime™ and a
TIFF (Uncompressed) decompressor
are needed to see this picture.
QuickTime™ and a
TIFF (Uncompress ed) dec ompres sor
are needed to s ee this pic ture.
What (some) DGVMs can do
(reasonably) well
• Global carbon and water cycle properties
(yes), feedbacks (maybe)
• Broad-scale biogeographic patterns
• Global patterns of trace gas emissions
(including fires)
QuickTime™ and a
TIFF (Uncompress ed) dec ompres sor
are needed to s ee this pic ture.
Carbon and water cycling in LPJ
• Photosynthesis from simplified Farquhar
model with optimized Vcmax
• Stomatal conductance maintains constant
ci/ca when Emax * W/Wmax not exceeded, else is
adjusted to maintain this rate of ET
• ET from Monteith (1995): Priestley-Taylor,
modified by stomatal conductance
QuickTime™ and a
TIFF (Uncompress ed) dec ompres sor
are needed to s ee this pic ture.
PN Foster et al. (work in progress)
Xu-Ri et al. (work in progress)
Limitations of DGVMs (1)
• PFTs are too few (5-10) => “plasticity
dilemma”
• Attribution of generic “literature values” for
traits
• No fundamental explanation for niche space
of each PFT
• => Divergent predictions of “biome shifts”
(vegetation structural changes)
QuickTime™ and a
TIFF (Uncompress ed) dec ompres sor
are needed to s ee this pic ture.
Limitations of DGVMs (2)
• Poor representation of functional diversity
within communities
• Life history trade-offs (e.g. leaf economics
spectrum) not modelled
• => Compositional responses to environmental
variability and change probably wrong
• => Potential for wrong biogeochemical
responses as well?
QuickTime™ and a
TIFF (Uncompress ed) dec ompres sor
are needed to s ee this pic ture.
Can DGVMs reliably predict
responses to climate change on
decadal-centennial time scales?
QuickTime™ and a
TIFF (Uncompress ed) dec ompres sor
are needed to s ee this pic ture.
Can DGVMs reliably predict
responses to climate change on
decadal-centennial time scales?
• Probably not
QuickTime™ and a
TIFF (Uncompress ed) dec ompres sor
are needed to s ee this pic ture.
Can DGVMs reliably predict
responses to climate change on
decadal-centennial time scales?
• Probably not
• Iconic example: “Amazon die-back”
QuickTime™ and a
TIFF (Uncompress ed) dec ompres sor
are needed to s ee this pic ture.
Relevant scientific advances for next
generation DGVMs
• Rediscovery of functional morphology
• Link between plant hydraulics and
photosynthesis
• Recognition of biophysical constraints and
trade-offs
• Application of economic approaches to plant
function and co-existence
• Comparative data to test hypotheses and
assign functional properties to PFTs
QuickTime™ and a
TIFF (Uncompress ed) dec ompres sor
are needed to s ee this pic ture.
Next steps for DGVMs
Define a small set of model parameters for a
plant
• Define a pathway from widely measured traits to
fundamental variables in models
•
QuickTime™ and a
TIFF (Uncompress ed) dec ompres sor
are needed to s ee this pic ture.
The List (to be developed.....)
wood anatomy/density
leaf hardness
leaf size/dissection
13C
bark thickness
.....
QuickTime™ and a
TIFF (Uncompress ed) dec ompres sor
are needed to s ee this pic ture.
sapwood permeability
wilting point
boundary-layer conductance
ci/ca “operating point”
fire resistance
.....
Conclusion: including biodiversity in
DGVMs
• Characterize functional variation among species in
terms of processes
• A longer list of PFTs: sample from a continuum of
traits, allowing for trait covariation
• Exploitation of quantitative trait data at species level
[TRY data base]
• New benchmarks for DGVMs: compositional
transects, climate responses... [BBS project]
QuickTime™ and a
TIFF (Uncompress ed) dec ompres sor
are needed to s ee this pic ture.