supplemental information 1. supplemental...

SUPPLEMENTAL INFORMATION

1. Supplemental Methods1.1 Description of the Ecosystem Model . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2

2. Supplemental FiguresSupplemental Figure 1. Chlorophyll concentrations and chlorophyll accumulation rates duringthe 2008 North Atlantic Bloom (NAB08) study . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4

Supplemental Figure 2. Culture-based example of issue regarding the correct identification ofbloom initiation during exponential growth . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6

3. Supplemental References Methods. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7

Supplemental Material: Annu. Rev. Mar. Sci. 2014. 6:167–94 doi: 10.1146/annurev-marine-052913-021325Resurrecting the Ecological Underpinnings of Ocean Plankton BloomsBehrenfeld and Boss

2

1. Supplemental Methods 1.1 Description of the Ecosystem Model In the main manuscript, we use a simple model that traces temporal changes in nitrogen

(N), phytoplankton (P), and herbivores (H) to illustrate ecosystem dynamics in the upper

ocean mixed-layer. Results from the model are presented in figure 4 (main manuscript).

Our model is based on the formulation of Evans and Parslow (1985), with slight

modifications following Moore et al. (2002) that include a parameterization of predator-

herbivore interaction. Modeled stocks (N, P, H) are in units of nitrogen (mmole N m-3).

The model consists of three equations (Evans & Parslow1985):

dN

dt MLD, Lat, yearday N

N Jb

P

mw N0 N

MLD.

dP

dt MLD, Lat, yearday N

N Jb

P c

1

P2H

P2 c2

mw P

MLD.

2

21 3 4 52

2

d.

d

H P wHc c H c H c H

t P c MLD

Note that in the text we use J)-b. Variables are described in Table 1, below,

along with assigned parameter values taken from Evans & Parslow (1985) and Moore et

al. (2002). The growth function is taken from Evans & Parslow (1985) with one change:

the effects of changing phytoplankton concentration on the light field are ignored (diffuse

attenuation of PAR is fixed at 0.1m-1). Entrainment of water (assumed to be lacking in P

or H) occurs when the mixed-layer-depth (MLD) increases: w+ = dMLD/dt > 0, otherwise

w+ = 0.


3

As in Evans and Parslow (1985), mixed layer deepening is assumed to dilute

phytoplankton and herbivores, but mixed layer shoaling is assumed to concentrate only

herbivores (motile) and to detrain phytoplankton (non-motile). Temporal changes in

MLD were described as: MLD = 75 + 50 x cos (2π (yearday-60)/365), which yields a

smooth MLD annual cycle ranging from 25 m to 125 m with a maximum mixing depth

on March 1st. For results shown in figure 4c and 4d of the main manuscript, an annual

cycle in incident sunlight was used for 47°N latitude and based on the light model of

Evans and Parslow (1985). As in Evans and Parslow (1985), the model is run for a 10

year period and we report the resultant annual cycle for year 10. The model was executed

using Matlab solver ODE23, which is an implementation of an explicit Runge-Kutta

scheme based on Bogacki and Shampine (1989).

The Matlab codes used to solve the above equations under a variety of scenarios used to

generate Figure 4 can be found at: http://misclab.umeoce.maine.edu/software.php.

Variable Value Parameter source (name)

N0 - nitrogen at depth 10 [mmole m-3] Evans & Parslow, 1985

J – nitrogen uptake half saturation 0.5 [mmole m-3] Evans & Parslow, 1985

b – phytoplankton respiration 0.07 [day-1] Evans & Parslow, 1985

m - diffusion velocity 3.0 [m day-1] Evans & Parslow, 1985

c1 - maximum herbivore growth rate 3.24 [day-1] Moore et al. 2002 (=z_umax)

c2 - grazing half saturation 0.44 [mmole2 m-6] Moore et al. 2002 (=z_grz2)

c3 - ingestion efficiency 0.5 unitless Evans & Parslow, 1985

c4 – herbivore non-predation loss 0.06 [day-1] Moore et al. 2002(=z_mort2)

c5 – herbivore quadratic mortality 1.0 [mmole-1 m3 day-1] Moore et al. 2002 (=z_mort)

Supplemental Material: Annu. Rev. Mar. Sci. 2014. 6:167–94doi: 10.1146/annurev-marine-052913-021325Resurrecting the Ecological Underpinnings of Ocean Plankton BloomsBehrenfeld and Boss

Supplemental Figure 1. Chlorophyll concentrations and chlorophyll accumulation rates

during the 2008 North Atlantic Bloom (NAB08) study (Mahadevan et al. 2012). (a) The

NAB08 study spanned a period of April-through-May, which encompassed the climax (year date

128) in bloom chlorophyll concentration. Throughout this period, various data sources recorded

chlorophyll concentrations, including satellite measurements, autonomous sensors, and in situ

measurements. Data in panel a (green line) is a spline fit to this combined data set. (b) A daily

specific rate of accumulation in chlorophyll (rchl) can be estimated from the chlorophyll time-

series (panel a) following: rchl = ln(Chlt2/Chlt1)/(t2-t1), where Chlt2 and Chlt1 are the observed

chlorophyll concentrations at time points t2 and t1, respectively (note, effects of dilution by

mixed layer deepening are not accounted for in this calculation, so values for rChl are

conservative). The time-series for rChl during the NAB08 study is shown in panel b, with periods

4


of increasing chlorophyll identified by blue shading and decreasing chlorophyll indicated by

yellow shading. This assessment shows periods of positive rChl throughout the NAB08 study.

Satellite data show that positive values of rChl may also be observed as early as February and

March for this location (Behrenfeld 2010, Behrenfeld et al. 2013). The values of rChl shown in

panel b are calculated from the spline-fit data in panel a, which gives a more smoothed time

course than if raw values had been used in the calculation (i.e., the actual daily values for rChl

exhibit considerably greater variability).

5


Supplemental Figure 2. Identifying bloom initiation during exponential growth. Using a

threshold chlorophyll concentration, such as 5% above the annual median concentration (Siegel

et al. 2002, Henson et al. 2009), can result in incorrect conclusions regarding bloom initiation. In

panel a, the blue line represents chlorophyll concentration in an exponentially growing

phytoplankton culture. Chlorophyll concentration exceeds by 5% the median for this time-

course on day 6 (red circle). Based on this criterion for identifying initiation, one would have to

conclude that growth conditions on day 6 crossed a critical threshold that first permited biomass

accumulation. If chlorophyll concentrations during exponential growth are instead plotted on a

logarithmic scale (panel b), the slope of the resultant line is proportional to growth rate. This

rescaling clearly reveals that the culture started blooming on day 2 (red circle), which now

correctly identifies bloom initiation.

6


3. Supplemental Materials References

Behrenfeld MJ. 2010. Abandoning Sverdrup's Critical Depth Hypothesis on phytoplankton

blooms. Ecology 91:977-89

Behrenfeld MJ, Doney SC, Lima I, Boss ES, Siegel DA. 2013. Physical-ecological interactions

of the subarctic Atlantic annual plankton bloom. Glob. Biogeochem. Cycles 27:

Bogacki P, Shampine LF. 1989. "A 3(2) pair of Runge-Kutta formulas". Appl. Math. Letters

2:321–5.

Evans GT, Parslow JS. 1985. A model of annual plankton cycles. Biol. Ocean. 3, 327-47.

Henson SA, Dunne JP, Sarmiento JL. 2009. Decadal variability in North Atlantic phytoplankton

blooms. J. Geophys. Res. 114:C04013

Mahadevan A, D'Asaro E, Lee C, Perry MJ. 2012. Eddy-driven stratification initiates north

Atlantic spring phytoplankton blooms. Science 337:54-58

Moore JK, Doney SC, Kleypas JA, Glover DM, Fung IY. 2002. An intermediate complexity

marine ecosystem model for the global domain. Deep-Sea Res. II 49:403-62.

Siegel DA, Doney SC, Yoder JA. 2002. The North Atlantic spring bloom and Sverdrup's critical

depth hypothesis. Science 296:730-33

7

Supplemental Material: Annu. Rev. Mar. Sci. 2014. 6:167–94doi: 10.1146/annurev-marine-052913-021325Resurrecting the Ecological Underpinnings of Ocean Plankton BloomsBehrenfeld and Boss

supplemental information 1. supplemental...

Documents