-
An adjoint data assimilation approachPhysical and Biological
Controls on Calanus finmarchicus in the Georges Bank RegionGlOBEC
broad-scale surveysAcadia (Lynch et al., 1996, 1998)
-
Jan: abundances are low (C3C4>c5; centers are advected along
the bankMar-Apr: abundances are highMay: C2,C3 abundances
declineJun: abundances drop sharply on the crest; high centers
retain near the Southern Flank (C2, C3) and along the periphery of
the bank (C4,C5); C5>C4>>C3>C2Climatological C. f.
distributions (GLOBEC , 1995-1999)
-
QuestionsWhere are the off-bank sources in late winter?How are
the off-bank sources imported to the bank to initiate the growing
season?How do the biological processes & physical transports
result in the observed C. f. distributions?How do these animals
disappear from the crest of the bank?
-
R2F2F3
F4R5R3R4Molting from C1 and C2 mortalityMortality (R
-
Forward Model:Infer R_i and C_i (t=0) by minimizing:First guess:
R_i=0, C_i(t=0)=(Cobs on the bank; 0 off-bank)
-
ObservationsModel results
-
Inferred initial conditions
-
AbundanceInferred Source/sinksMolting fluxadvection
-
Inferred biological termsImportant Source regions: the GB, the
Georges Basin, Wilkinson Basin, the Browns banks
(Feb-Apr.)Mortality: Mar-April, Southern Flank (Food limitation,
Campbell et. al., 2001); May-Jun crest (predation, Bollens et al.,
1999 )
-
Inferred C3 controlsNet
bio-gain/lossJan-FebFeb-MarMar-AprApr-MayMay-JuneAdvection: loss:
Wilkinson, Georges Basins; Gains: northwest crest, Georges Basin
and the south tip of GSC.
Bio-gain: Feb-Apri ; Bio-loss: June;
Feb-Apri: Bio>Adv
June decline: Bio-losses;
-
Inferred C5 controlsJan-Feb: Advective convergence (Northern
Flank,Georges Basin)Feb-Apr: Bio, Adv both are important
Bio>AdvJune: Bio~AdvNet
bio-gain/lossJan-FebFeb-MarMar-AprApr-MayMay-June
-
Conclusions1. The Scotian Shelf, GOM are important sources of C4
and C5 in late winter. The convergence of advective C5 flux near
the northern periphery of the bank seems to be important to seed
the bank.2. Both biological reactions and physical advection are
important for the observed distributions. Physical transports
increase from winter to spring.
-
Conclusions3. Biological gains are major contributors for high
abundances in Spring; Biological losses are mainly responsible for
the decline of C2-C4 in June, while for C5, both biological loss
and advective transport are responsible for the June decline;4.
Molting fluxes largely exceed mortality rates (C2-C3, Jan.-Jun.).5.
Mar.15-Apr.15, mortality is high near the southern Flank;
May.15-Jun.15, mortality is high on the crest.6. Results are
inferred from modeled flow fields and data on the GB only. More
accurate inference about the controls in the off-bank region needs
data in those areas and improved estimate of circulation.
-
Future WorkIncorporate more data: N3C1(pump data)C2Adult stages;
data in other regions.Use 3D model to study vertical migration as
well as the controls already included. Explore the connection
between interannual, synoptic (e.g., storm driven) physical
variability & biological variability.Sensitivity study to rank
the physical & biological controlsExtend the study area to a
larger scale (e.g., North Atlantic).
-
Molting Flux (F)C: Concentration; D: stage duration;T:
temperature; CF: Food concentrationCampbell et al., 2001
-
C2 stage durationT limitedT & Chl limitedChla: OReilly &
Zetlin (1996); T (Lynch et al., 1996)
-
Climatological Jan-Feb flow (Lynch et al., 996)
-
Seasonal variations
-
ProcedureChoose a best first guess of control vector U.Integrate
the forward model and calculate the cost function.Run the adjoint
model to calculate the gradient of J to U.Use a descent algorithm
to find a new value of U.Repeat from the second step until a
satisfactory solution has been achieved.
-
The inversion reduces model-data misfit by ~90%
unconstrainedconstrained
-
Computed tendenciesIncreasing season: Jan-Apr;Decline season:
Apr-Jun;Centers are located along the bank periphery with advective
signatures.
-
Convergence of C5 advective flux (Jan.15)
-
Inferred C2 controlsJan-Feb: biological gains first appears near
the Northern edge of the Bank.Feb-May Biological gains/loses and
advective transport are important June: bio.-lose is mainly
responsible for the decline
-
Inferred C4 controlsNet
bio-gain/lossJan-FebFeb-MarMar-AprApr-MayMay-JuneHigh abundances in
spring: bio-gains; June-decline: bio-losses.
-
HypothesesGeorges Bank populations limited by C. f. supply from
GOM diapause stock during winter.GB dynamics influenced by growth
conditions on the bank during spring (food limited in April).GOM
sources to GB become more and more important during spring and
summer. Courtesy of http://globec.whoi.edu, phase1 projects
-
MARMAP Observations(1977-1987)Courtesy of www.at-sea.org
-
Figure 4. Estimates of predation impact by 5 predators, based on
feeding rates and abundances of predators and prey from Broad Scale
Survey samples. Impact is expressed as percent of the stocks of
Calanus and Pseudocalanus removed daily, assuming that the
predators are non-selective (E=0), and consume these species in the
same proportions as they occur. Prey of Clytia are nauplii, prey of
the other predators are copepodites and adults (Bollens et. al.,
online.sfsu.edu).