Transcript Accounting for biodiversity in marine ecosystem models

Accounting for biodiversity in
marine ecosystem models
Jorn Bruggeman
S.A.L.M. Kooijman
Dept. of Theoretical Biology
Vrije Universiteit Amsterdam
Interspecific differences quantified by traits





How to capture biodiversity in models?
Species-specific models are incomparable
Approach: one omnipotent species
Parameter values determine the species
Species-determining parameters: traits
Ecosystem diversity
Phototrophs and heterotrophs: a section through diversity
bact 1
heterotrophy
bact 3
?
bact 2
?
?
mix 1
mix 2
mix 3
mix 4
?
phyto 1
?
phyto 2
?
phyto 3
phototrophy
phyto 2
Infinite diversity  continuity in traits
Species = investment strategy


Why not ‘just’ do everything well?
Good qualities must be paid for
–
–


Budget is limited  make choices!
Usefulness of qualities depends on environment
–

costs for directly associated machinery
(photosynthesis, phagocytosis)
costs for containment if qualities conflict
(nitrogen fixation requires anoxic environment)
No photosynthesis in dark environments
Species define niche by choosing qualities to invest in (‘strategy’)
Cost-aware phytoplankton population
+
+
structural biomass
nutrient
+
+
+
κL
light harvesting
+
structural biomass
nutrient
+
κN
+
nutrient harvesting
Functional group: phytoplankton

Discretized trait distribution
–

15 x 15 trait values = 225 ‘species’
Start with homogeneous distribution, low densities
Realistic setting

Bermuda Atlantic Time-series Study (BATS)
–

10 years of monthly depth profiles for physical/biological variables
Turbulent water column model (1D)
–
–
–
–
General Ocean Turbulence Model (GOTM)
upper 250 meter
k-ε model for turbulence parameterization
realistic forcing with meteorological data (ERA-40)
Biota: chlorophyll
Modeled light harvesting equipment
 chlorophyll
BATS measured chlorophyll
averaged over 10 years
Succession: average trait values in time
Modeled light harvesting equipment
 cell-specific chlorophyll
Modeled nutrient harvesting equipment
 surface-to-volume  1/cell length
Trends

Cell-specific chlorophyll increases with depth
–
–

High-chlorophyll species do better in low-light deep
Thus: succession (‘shade flora’), not photo-acclimation (Geider)
Seasonal succession: large  small species
–
–
–
Small species fare better in oligotrophic environment
Bloom start with high nutrient level, large species
Small species gain upper hand as bloom proceeds (Margalef)
Conclusions and perspectives

Trait-based approach demonstrates diversity in space and time
–
–

Description of BATS
–
–

Qualitatively ‘reasonable’ with current (5 parameter!) model
Space for improvement; parameter fitting with base no-trait model
Aim: collapse trait distribution
–

increase in chlorophyll content with increasing depth
decrease in cell size between start of bloom and winter
Loss of state variables  fitting becomes possible
Future: traits for ecosystems
–
–
–
heterotrophy
predation/defense
body size