Activities Report

Diapause, Population Dynamics, and the Large-Scale Dispersal of Zooplankton.

Principal Investigators: Sharon Smith, Donald Olson and Nasseer Idrisi.

National Science Foundation funded project, Grant No. OCE-9911494.

NPZD-MICOM Coupled Model

A fully coupled biological/physical model has been developed for the Arabian Sea. The physical model is the Miami Isopycnic Coordinate Ocean Model (MICOM), configured for the Arabian Sea (Esenkov, 2000). In this model each isopycnic layer is governed by the shallow-water equations. Thermodynamic variables and the horizontal velocity are treated as "layer" variables that are vertically constant within layers but change discontinuously across layer interfaces. The first layer corresponds to an active mixed layer of Krauss and Turner type, so it is a non-isopycnic layer and is subject to diabatic forcing. The solution of global MICOM after 50 years of integration from an initialization with Levitus data were used for the initial and boundary conditions for the physical aspects of the regional model. The model has a 0.35∞ resolution in the horizontal and 15 isopycnic layers under the mixed layer. The model is forced by mean monthly fields of atmospheric temperature, humidity, and wind stress from COADS, net radiation from the Oberhuber atlas, and precipitation from the NOAA microwave sounding unit. The wind and thermohaline forces expand and shrink the mixed layer, causing mass and other properties to exchange with the interior layers. The deepening of the mixed layer is caused by entrainment from the isopycnic layers due to strong winds, while the shallowing is a result of heating in the absence of intense wind stirring. For more details about MICOM and implementation see Bleck et al. 1989 and 1992.

The NPZD biological model determines the nitrogen concentration in four compartments: nutrients (N), phytoplankton (P), zooplankton (Z), and detritus (D). The concentration of each variable is controlled by a local conservation equation given by

((è(Δpφ))/(èt))+—_s(uΔpφ -κΔp—_sφ)+(s((èp)/(ès))φ)_bot-(s((èp)/(ès))φ)_top=ΔpF φ ,

where φ is N, P, Z or D. The horizontal velocity (u) and the pressure (p) are given by the physical model. The expression s((èp)/(ès))φ represents entrainment/detrainment across an s surface. The system is coupled by the source-sink terms

F_P = GP-Φ_P Z-σ_PP,

F_Z = A_eA_m(Φ_P+Φ_D+Φ_Z)Z-Φ_ZZ-MZ ,

F_D = (Φ_P+Φ_Z)(1-A_e)Z –A_eΦ_DZ+σ_PP+MZ-ρ_DD-sD,

F_N = (1-A_m)A_e(Φ_P+Φ_D+Φ_Z)Z+ρ_DD -GP,

which represent the predator-prey dynamics in each layer. Phytoplankton growth is expressed as

G=P_maxf(T)Ī(N/(N+K_N)),

where the average light in each layer is

For the growth due to the nutrients supply, we uses the Michaelis-Menten form N/(N+K_N), where K_N is the half-saturation constant. Phytoplankton maximum growth rate P_maxf(T) is a function of temperature. The effect of a constant versus temperature-dependent function is evaluated and discussed in the next section. Phytoplankton losses are through zooplankton grazing and senescence. The zooplankton grazing rate contains total feeding preferences

where f_φ is the individual feeding preference. Notice that feeding and metabolism are functions of temperature g_maxf(T). The zooplankton concentration losses are due to mortality and self-predation. The sources for detritus are the senescence of phytoplankton, zooplankton mortality and fecal pellets produced by zooplankton grazing. The losses in this compartment occur through remineralization and zooplankton predation. Detritus is also allowed to sink in the model. The rate of sinking is given by

The concentration of nutrients is regenerated from zooplankton excretion and the remineralization of detritus. Nutrient loss is through phytoplankton uptake.

From U.S. JGOFS data in the Arabian Sea, it is noteworthy that phytoplankton concentration is very low at warmer temperatures (corresponding to the upper 50m), even though it is known that the nutrient supply is not a limiting for this region (Naqvi and Noronha 1991). As expected, the zooplankton concentration follows the same behavior due to the temperature-dependence function.

The observed variations of phytoplankton and zooplankton concentration with temperature suggest that a function of temperature is necessary to describe biological processes. Here, we used the function proposed by Thornton and Lessem (1978) to parameterize the maximum expected growth rate of phytoplankton, metabolism and feeding for zooplankton. This function is based on the premise that the biological process has a maximum reaction rate at some optimum temperature and diminishes asymmetrically as environmental temperature moves away from the optimum. This is important since many biological reaction rates are asymmetrical over the range of actual temperature (Thornton and Lessem 1978).

The boundary conditions for the biological variables were chosen as ≥open≤ in the south and east boundaries and closed at the north and west. By ≥open≤ we mean that the value at the boundary has the same tendency as the previous two adjacent grid points inside the domain.

Briefly, our analyses indicate:

1. Clear upwelling bloom conditions off the Somali and Omani coasts during the Southwest Monsoon.

2. Mild bloom conditions off the southeast tip of the Arabian Peninsula during the late Northeast Monsoon.

3. Including the temperature-dependence function allowed for the development of a deep chlorophyll maximum in the interior Arabian Sea, as observed from in situ data.

4. The temperature-dependence function constrained growth in the upper mixed layer, leading to oligotrophic/warm surface water conditions in the interior as observed from in situ and satellite imagery data.

A manuscript describing this work has been recently published:

Olascoaga, M. J., N. Idrisi and A. Romanou (2005).

Biophysical modelling of plankton dynamics in the Arabian Sea,

Ocean Modelling, 8, 55-80.

Lagrangian Particle Modeling.

Individual-based modeling of particles representing diapaused Calanoides carinatus was implemented successfully within the MICOM configuration for the Arabian Sea. Particles were tracked over space and time by means of a joint Markovian algorithm interpolated over time using the fourth-order Runga-Kutta method with subgrid spatial interpolation using a bilinear spline, solving for:

dx = (U + u)dt

du = -(1/T)udt + (K/T²)^1/2dw

where dx and du are the position and velocity displacements, respectively. The particle is advected at a time interval, dt (1 hour time-step), using the MICOM velocity field (U) plus a normally distributed random component (u) representing horizontal turbulent velocity. A turbulent time-scale (T) of 2.5 days was used for the particle advection scheme. A diffusion coefficient was calculated taking into account the velocity field variance (s), and it was scaled to a random increment (dw) taken from a normal distribution with zero mean and a range of 2dt.

The MICOM-plus-Lagrange code was run online for a nine month period from August to May, simulating the diapause period for C. carinatus. Three specific regions were chosen for release of 700 particles per run. The regions include offshore of Somalia, nearshore off Oman and in the Gulf of Aden. In each region, experimental runs were further subdivided into different subsurface depth layers. The launch depths ranged from layer 3 to layer 12, and included trajectory depths of 150 m (layer 3), 300 m (layer 5), 550 m (layer 8), 800 m (layer 10), and 1050 m (layer 12). These experiments were conducted to determine the extent of connectivity among the three different regions of interest.

An important aspect of predicting Lagrangian particle trajectories that incorporate random components to the set of solved equations is the accuracy of advecting the particles within the deterministic velocity field over time. Parameterizing particle displacement from one time-step to the next requires prescribing a turbulent time scale (T) that resolves subgrid-scale features as well as calculating the correct diffusion coefficient (K) for the region. Preliminary comparisons of model trajectory calculations with deep drifters drogued at 800 m in the northwest Indian Ocean indicate acceptable agreement for K (personal communication: Russ Davis, Scripps Institution of Oceanography, rdavis@ucsd.edu).

Associated with each diapaused particle are metabolic rates taken from measurements made from diapaused C. carinatus in the Benguela Current (Arashkevich et al. 1996). These metabolic rates are 10% of active metabolism. Also included with the particles are lipid reserves used by the particleπs metabolic requirements at each time-step. Individual variability is included in both metabolism and lipid reserve, where values corresponding to the range of natural variability are assigned to each particle at the beginning of each experimental simulation. These experiments allowed us to determine whether individual females in a population are able to survive until the next upwelling season and produce enough eggs to sustain a viable population.

Briefly, our analyses indicate:

1. That there is strong communication from the Somali coastal region to the Omani coastal region at depths between 150-500 m.

2. Particles are retained in both regions at depths greater than 1000 m.

3. Particles are lost to the interior at depths between 500-1000 m where there is a strong southeasterly current advecting particles to the equatorial region.

4. Particles launched in the Gulf of Aden are generally retained with weak communication with the Somali and Omani coastal regions in the upper 500 m.

5. Based on the amount of lipid reserve accumulated at the start of the diapause period, as determined from samples collected off the coasts of Somalia and Oman, there is sufficient energy available at the end of the 9-month period to allow for adequate egg production to sustain a viable population capable of growing at the onset of the following southwest Monsoon period.

A manuscript describing this work has been recently published:

Idrisi, N., M.J. Olascoaga, Z. Garraffo, D.B. Olson and S.L. Smith (2004).

Mechanisms for emergence from diapause of Calanoides carinatus in the Somali current. Limnology and Oceanography, 49, 1262-1268.

In general, using COADS monthly forcing and realistic parameterization of the NPZD model and metabolism of C. carinatus, we were able to reproduce the general features in response to monsoonal forcing in the Arabian Sea. These included upwelling bloom conditions off the Somali and Omani coasts during the southwest Monsoon. Oligotrophic/warm surface water conditions were reproduced in the interior. We observed a depth-dependent retention or exchange between important upwelling regions of most of the particles released during the non-upwelling period. Calculations on metabolic rates for C. carinatus imply that diapaused individuals should be able to successfully survive to the next growing season with sufficient energy to reproduce.

Presentations

Modeling diapause as a life-history strategy of an important copepod grazer (Calanoides carinatus) in the Somali Great Whirl. Nasseer Idrisi, G. Peng, O. Esenkov, D. Olson, and S. L. Smith. Gordon Research Conference, Coastal Ocean Modeling, 1999 New London, NH.

Biophysical modeling of plankton dynamics off Somalia and Oman. M. Josefina Olascoaga, N. Idrisi, A. Romanou, D.B. Olson, and S.L. Smith. (Poster) presented at the Layed Ocean Modelling Meeting, February 6-8, 2002, RSMAS/Univ. of Miami, Miami, FL.

Translocation of diapausing Calanoides carinatus in the mesopelagic/deep layers in the Arabian Sea: Modeling Lagrangian particle drift in an isopycnic ocean model. Nasseer Idrisi, M.J. Olascoaga, D.B. Olson, and S.L. Smith. Oral presentation at the Layed Ocean Modelling Meeting, February 6-8, 2002, RSMAS/Univ. of Miami, Miami, FL.

Biophysical modeling of plankton dynamics off Somalia and Oman. M. Josefina Olascoaga, N. Idrisi, A. Romanou, D.B. Olson, and S.L. Smith. (Poster) presented at the Ocean Sciences Meeting, February 11-15, 2002, Honolulu, HI.

Translocation of diapausing Calanoides carinatus in the mesopelagic/deep layers in the Arabian Sea: Modeling Lagrangian particle drift in an isopycnic ocean model. Nasseer Idrisi, M.J. Olascoaga, D.B. Olson, and S.L. Smith. Oral presentation at the Ocean Sciences Meeting, February 11-15, 2002, Honolulu, HI.

Modeling plankton dynamics in the Arabian Sea. Joint oral presentation by J.M. Olascoaga, and N. Idrisi. Marine Biology and Fisheries divisional seminar series, RSMAS/Univ. of Miami, 26 March, 2002.

Behavioral/Physical Mediation in Space and Time: Patterns of Dispersion and Fate of Deep Hibernating Copepods. Nasseer Idrisi, M.J. Olascoaga, Z. Garraffo, D.B. Olson, S.L. Smith. Invited presentation at the Biodiversity Symposium, Sponsored by NSF, Univ. of Michigan, Ann Arbor, MI, Sept. 30, 2002

Lagrangian modeling of diapaused copepods in intermediate to deep layers in the Arabian Sea. Nasseer Idrisi and M.J. Olascoaga LAPCOD 2002, Key Largo, Florida, Dec. 12-16, 2002

Cited literature

Anderson, L. A., Robinson, A. And Lozano, C. (2000) . Physical and biological modeling in the Gulf Stream region: I. Data assimilation methodology. Deep-Sea Res. I. 47: 1787-1827.

Arashkevich, E., A. Drits, and A. Timonin. (1996). Diapause in the life cycle of Calanoides carinatus (Kroyer), (Copepoda, Calanoida). Hydrobiologia, 320: 197-208.

Bleck, R., Hanson, H. P., Hu, D. and Kraus, E. B. (1989). Mixed layer-thermocline interaction in a three-dimensional isopycnic coordinate model, Journal of Physical Oceanography 19: 1,417-1,439.

Bleck, R., Rooth, C., Hu, D. and Smith, L. T. (1992). Salinity-driven thermocline transients in a wind- and thermohaline-forced isopycnic coordinate model of the north atlantic, Journal of Physical Oceanography 22: 1,486-1,505.

Esenkov, O. E (2000) . A Numerical Study of the Dynamics of the Somalia Current, PhD thesis, University of Miami.

McCreary, J. , Kohler,K. E., Hood, R. R. and Olson, D. B. (1996) . A four-component ecosystem model of biological activity in the Arabian Sea. Prog. Oceanog. 37: 193-240.

Naqvi, S. W. A. and Noronha, R. J. (1991). Nitrous oxide in the Arabian Sea. Deep-Sea Res. 38: 871-890.

Thornton, K. W. and Lessem, A.S. (1978). A temperature algorithm for modifying biological rates, Trans. Am. Fish. Soc. 107: 284-287.