MARINE TECHNOLOGY SOCIETY JOURNAL ...

MARINE TECHNOLOGY SOCIETY JOURNAL COPYRIGHT FORM

Title of Paper:

Author(s):

A.

Copyright Transfer Form

The undersigned, desiring to publish the above paper in the Marine Technology Society Journal, hereby transfers their copyrights in the above paper to the Marine Technology Society, Inc., known as MTS. In return for these rights, MTS hereby grants to the above authors, and the employers for whom the work was performed, royalty-free permission to perform the following: 1. Retain all proprietary rights other than copyright, i.e., patent rights. 2. Reuse all or portions of the above paper in other works. 3. Reproduce, or have reproduced, the above paper for the personal use of the author or for company use provided that the source and MTS copyright are indicated; the copies are not used in a way that implies MTS endorsement of a product or service of an employer; and the copies per se are not offered for sale. 4. Make limited distribution of all or portions of the above paper prior to publication. 5. In the case of work performed under U.S. government contract, MTS grants the U.S. government royalty-free permission to reproduce all or portions of the above paper, and to authorize others to do so, for U.S. government purposes. In exercising its rights under copyright, MTS will make all reasonable efforts to act in the interests of the authors and employers as well as in its own interest. In particular, MTS requires that: 1. The consent of the first-name author be sought as a condition in granting republication permission to others. 2. The consent of the undersigned employer be obtained as a condition in granting permission to others to reuse all or portions of the paper for promotion or marketing purposes. In the event the above paper is not accepted and published by MTS, this agreement will become null and void.

Authorized Signature

Title (If not Author)

Employer for Whom Work Performed Date Signed

B.

U.S. Government Employee Certification To be used only if all authors are government employees and prepared the manuscript as part of their job.

This will certify that all author(s) of the above paper are employed by the U.S. government and performed this work as part of their duties. Therefore, the paper is not subject to U.S. copyright protection.

Authorized Signature

Title (If not Author)

Government Organization

Date Signed

Thank you for helping MTS comply with U.S. copyright law, which requires that the transfer of copyrights from the author to the Marine Technology Society be confirmed in writing. Please note that if your paper is prepared as part of your job responsibilities, the rights to your paper initially rest with your employer. In that case, when you sign the copyright transfer form, we assume you are authorized to do so by your employer and that your employer has consented to all terms and conditions on this form. If not, it should be signed by someone so authorized. Note also that for jointly written papers, only one signature is required. However, we assume all authors have been advised and have consented to the terms of this form.

Marine Technology Society Journal Reprint/Journal Charge Information Charges: Select either requested (voluntary) price or wholesale price. Please allow four to five weeks for delivery Requested (Voluntary) Price: Reprints Mandatory page charges are a standard practice with many journals. Subscriptions and advertising revenues for professional journals defray only a small portion of publishing costs, and mandatory page charges pay for many expenses. The Marine Technology Society does not require mandatory page charges. However, the Marine Technology Society respectfully requests that that the author’s institution honor a voluntary reprint charge of $60 per printed page rather than pay the wholesale price. Reprints can be ordered in increments of 100 copies (or ten Journals) as in table, below. Page Length 100 copies or 10 Journals 200 copies or 20 Journals 300 copies or 30 Journals 400 copies or 40 Journals 500 copies or 50 Journals

1 $60

2 $120

3 $180

4 $240

5 $300

6 $360

7 $420

8 $480

9 $540

10 $600

11 $660

120

240

360

480

600

720

840

960

1080

1200

1320

180

360

540

720

900

1080

1260

1440

1620

1800

1980

240

480

720

960

1200

1440

1680

1920

2160

2400

2640

300

600

900

1200

1500

1800

2100

2400

2700

3000

3300

Wholesale Price: Reprints Below are the author’s prices for reprints. Prices include shipping, postage, and handling. Page Length 1-4 5-8 9-12 13-16 17-20 21-26

100 Copies $140 180 225 280 340 400

200 Copies $ 200 257 320 400 485 572

300 Copies $280 308 385 480 582 680

500 Copies $400 440 550 685 830 900

1000 Copies $600 660 825 1020 1245 1350

2000 Copies $1100 1095 1370 1695 2065 2200

If we are shipping internationally, please include the additional shipping and handling charges below to your order: 100 copies: 200 copies: 300 copies: 500 copies: 1000 copies: 2000 copies:

$ 56.00 $ 98.00 $131.00 $231.00 $403.00 $951.00

MTS Journal Reprints and Journal Orders

Revised 2009

page 1

Journal Copies and Discounts All co-authors receive one complimentary copy of the issue in which their paper appears. Costs for additional copies are as follows: Non-discounted price: @$20.00 1-5 copies: @$15.00 (a 25% discount) 6-49 copies: @$11.00 (a 45% discount) 50 + copies: @ $8.00 (a 60% discount) If we are shipping internationally, please include the additional shipping and handling charges below to your order: For orders up to $90: For orders over $90 and up to $180: For orders over $180 and up to $352: For orders over $352:


Shipping and Handling = $20 Shipping and Handling = $35 Shipping and Handling = $72 Shipping and Handling = $3 per copy

Revised 2009

page 2

Author’s Order Form Please fill out the form below and return to: Publications Manager, Marine Technology Society, 5565 Sterrett Place, Suite 108, Columbia, Maryland 21044, or fax to: 410-884-9060. Special requests, comments and concerns can be addressed to the Journal’s managing editor at: [email protected].

Yes, voluntary charges of $60 per page will be honored. Please indicate number of reprints _____________ or journals _______________. No, we can not pay voluntary charge. Please apply the wholesale cost and send _____ reprints. Please send

additional copies of the Journal at the author’s rate.

++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++ Author Information Name: ___________________________________________________________ Phone: __________________________________________________________ Volume Number:________

Issue Number: __________ Pages: ___________

Title of Article: _____________________________________________________ Total Price (from price list): ___________________________________________ Check Enclosed:

Yes

No

If you are paying by credit card, please highlight one of these: Visa/ Mastercard/ AmEx/ Diner’s Club: Expiration Date: ______________________

Account Number:

Name: ________________________________________________ Signature: ______________________________________________ Bill to:

______________________________________________ ______________________________________________ ______________________________________________

Send to:

______________________________________________ ______________________________________________ ______________________________________________


Revised 2009

page 3

1

2

3

4

PAPER

Dispersion of Passive Tracers in the Baltic Sea Deep Water as Applied to Dumped Chemical Weapons

5

AUTHORS

6

Victor Zhurbas Shirshov Institute of Oceanology of Russian Academy of Sciences and Marine Systems Institute, Tallinn University of Technology

7 8 9 10 11 12 13

Vadim Paka Shirshov Institute of Oceanology of Russian Academy of Sciences

42 43 44 45 46 47 48 49

15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41

Introduction

A

large amount of chemical weapons (CW)—around 32,000 t—was dumped in the Baltic Sea after World War II (HELCOM, 1996). The possible threat to the marine ecosystem and human health was poorly investigated, and the information available remained scattered within a number of national research projects and was incomplete. To better understand the possible threat, the European Union funded an international project in 2006, Modeling of Ecological Risks Related to Sea-Dumped Chemical Weapons (MERCW; www.mercw. org). Belgium, Denmark, Finland, Germany, and the Russian Federation were among the participants in the project that aim to study CW dump sites in the Baltic Sea and to assess the resulting risks to marine ecosystems and human health. The results of chemical, hydrographic, and geophysical investigations carried out mostly in a CW dumpsite in the southern Baltic Sea, east of the island of Bornholm, were presented by

Results of modeling of the migration of chemical warfare agents (CWAs) and their decay products from the initial chemical weapons dump site are presented. The aim was to find idealized sediment redistribution schemes in deep basins of the Baltic Sea corresponding to different wind conditions and to model the concentration of dissolved CWA in a continuous release scenario in the Bornholm dump site corresponding to real wind statistics. Keywords: numerical modeling, random walk, turbulent diffusion

Missiaen et al. (2010), Emelyanov 51 et al. (2010), and other MERCW 52 participants; however, to develop an 53 ecological risk assessment model, it 54 is necessary to supplement these site 55 investigations with a simulation 56 (modeling) of the migration of the 57 CW components and their decay 58 products from the initial dump site. 59 This paper presents the results of 60 such a simulation. The aim was to 61 find idealized sediment redistri62 bution schemes in deep basins of 63 the Baltic Sea corresponding to 64 different wind conditions and to 65 model the concentration of dissolved 66 chemical warfare agents (CWAs) in a 67 continuous-release scenario in the 68 Bornholm dump site corresponding 69 to real wind statistics. 70 To forecast possible scenarios of 71 propagation and dispersion of toxic 72 compounds from CW dump sites 73 in the Baltic Sea, several steps were 74 taken. First, a hydrodynamic model 75 capable of calculating velocities of cur76 rents and diffusivities in the Baltic Sea 77 deepwater as well as bottom stress was 78 developed. The bottom stress is re50

14

ABSTRACT

quired to parameterize vertical fluxes of matter through the interface between bottom sediments and nearbottom water (i.e., resuspension). The circulation model is based on the Princeton Ocean Model (POM), in which the vertical grid size is logarithmically refined towards the bottom in order to resolve the bottom boundary layer (BBL). Second, a diffusion model was worked out to calculate matter transport in the bottom sediments water body system. Input parameters for the diffusion model are 4D fields of current velocities and diffusivities along with bottom fluxes of a matter and settling velocity of suspended particles. Matter transport was calculated by solving a semi-empiric equation of turbulent diffusion in two ways by integrating numerically (a) a finite difference analogue of the diffusion equation provided the tracer’s concentration and (b) a stochastic analogue of the diffusion equation (i.e., diffusion is described by a random walk of Lagrangian particles) provided particles trajectories. Finally, the hydrodynamic and diffusion models were

January/February 2012 Volume 46

Number 1

1

79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107

120

applied to calculate propagation of toxic compounds from the Baltic Sea dump sites under different scenarios. Note that calculation of the bottom stress using POM does not take into account the effect of surface waves on bottom friction considered in detail by Kuhrts et al. (2004), Jönsson et al. (2005), and Seifert et al. (2009). However, the dump sites are deep enough that wave action does not extend significantly to the nearbottom layers.

121

Hydrodynamic Model

108 109 110 111 112 113 114 115 116 117 118 119

122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153

FIGURE 1 A Model domain with bottom topography of the Baltic Sea. Digital bottom topography was taken from Seifert et al. (2001). Labeled are the basins of the Baltic Sea: (1) Arkona Basin, (2) Bornholm Deep, (3) Słupsk Furrow, (4) Gulf of Gdaņsk, (5) East Gotland Basin.

Regional Implementation of POM To calculate velocities of sea currents, diffusivities, and bottom stress in the Baltic Sea deepwater, the POM, developed by Blumberg and Mellor (1980, 1983), was used. The model domain covers longitudinal and latitudinal ranges of 10°40′24°20′E, 54°00′-60°30′N, including the whole Baltic Sea closed at sounds with an exception of the Gulf of Bothnia and the eastern part of the Gulf of Finland. Dimensions of the finite difference grid was Nx × Ny = 411 × 391 (Figure 1). The value of the grid step was 2′ and 1′ in longitude and latitude, respectively, which corresponds to approximately 2 km in both directions. There were 30 σ-layers in vertical direction. Above the BBL, the thickness of σ-layers was uniform, having a value 0.038D, wherein D is the total thickness of the water column (D = sea depth + surface elevation). Since we are mostly interested in adequate resolution of flow in BBL, the thickness of σ-layers within BBL is taken logarithmically decreasing toward the bottom, so that the closest to bottom σ-layer is 0.002D thick.

2

154

We expect that the existence of the gulfs does not affect considerably the mesoscale bottom circulation in vicinities of the dump sites. That is why it is ap157 propriate to exclude the Gulf of Bothnia and the eastern part of the Gulf of 158 Finland from the model domain, placing artificial shorelines at 24°20′E and 159 60°30′N. Again, the model does not include water exchange between the Baltic 160 and North seas as well as river runoff, because the above processes are not ex161 pected to affect considerably the mesoscale variability of circulation in the deep 162 layers of the Baltic proper. 163 When formulating the initial conditions, we tried to reproduce principal fea164 tures of thermohaline fields typical for different Baltic basins in different seasons 165 using simple analytical parameterizations (Zhurbas et al., 2003). The temperature 166 field in summertime was taken horizontally homogeneous with linearly stratified 167 upper warm layer (T = 18°C at the sea surface and T = 16°C at the base of seasonal 168 thermocline z = 30 m) and T = const = 5°C in the lower layer z > 30 m. In win169 tertime, there is no upper warm layer, so that the whole sea has a uniform tem170 perature T = 5°C. 171 The initial salinity was parameterized by a simple analytical form 155

156

8 S0 at z ≤ z0 > > > > < z z0 0:2 at z0 < z ≤ z1 S ðx; y; z Þ ¼ S0 þ ðS1 S0 Þ z1 z0 > > > > : S ðx; y; z1 Þ þ 0:005ðz z1 Þ at z > z1

172 173

Marine Technology Society Journal

ð1Þ

wherein z is directed downward and parameters z0(x, y), z1(x, y), S0(x, y), and S1(x, y) are used to describe major features of salinity field in the Baltic basins.

174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221

These parameters are taken constant within each basin (i.e., the salinity is horizontally homogeneous within each basin). In accordance with observations (Zhurbas et al., 2004), the following parameter values were used: S0 = 7.5 psu (the whole Baltic Sea), z 0 = 35 m, z 1 = 48 m, S 1 = 18 psu (the Arkona Basin and west of it), z0 = 50 m, z1 = 90 m, S1 = 16 psu (the Bornholm Deep), z0 = 65 m, z1 = 90 m, S1 = 13 psu (the Słupsk Channel), z0 = 70 m, z1 = 100 m, S1 = 11.5 psu (the rest of the Baltic Sea). To simulate an inflow event in the framework of our model, we increase initial salinity in western basins of the Baltic Sea to S1 = 22 psu in the Arkona basin, S1 = 18.5 psu in the Bornholm Deep, S1 = 16 psu in the Słupsk Furrow. The boundaries between the Arkona/ Bornholm/Słupsk/Gdaņsk basins were taken at the sills or shallowest connections. Nevertheless, a discontinuity of salinity remained at the boundaries, which may have resulted in incorrect values of salinity generated by the model. In order to remove the discontinuity, spatial smoothing of the initial salinity field has been applied. The smoothing procedure consists of a low-pass, filtering the values of z0, z1, S 0 , S 1 over approximately 10 × 10 nearest grid nodes. The model runs started from the motionless state and performed with horizontally homogeneous wind stress of different scalar and direction. It is worth noting that application of initial salinity conditions (equation 1) and horizontally uniform wind forcing is aimed to characterize some typical wind/current patterns in the Baltic Sea environment and ignores such potentially important episodic events as the major inflows that should be treated separately.

222

Passive Matter Transport Model

The POM code available through the Internet does not explicitly contain a 224 facility to calculate transport of a passive matter. It is implied by the POM authors 225 that such a facility can be supplemented to the model independently by an expe226 rienced POM user. Similar to equations for temperature and salinity transport 227 presented in the Users Guide for POM (Mellor, 2004), an equation for passive mat228 ter transport can be written as 223

∂CD ∂CuD ∂CvD ∂C ðw ws Þ ∂ KC ∂C ∂ ∂C ∂ ∂C þ þ þ ¼ þ AC þ AC ∂t ∂x ∂y ∂σ ∂σ D ∂σ ∂x ∂x ∂y ∂y

ð2Þ

wherein C is the mass concentration of a passive matter, ws is the settling velocity, 230 KC and AC are vertical and horizontal apparent diffusivities, respectively. If C is 231 supposed to be the concentration of resuspended bottom sediments, the turbulent 232 flux of the matter through the bottom interface due to sediment’s erosion, FC, can 233 be parameterized as 229

FC ðx; y; t Þ ¼

M ½τ ðx; y; t Þ τ 0 ; τ ≥ τ 0 0; τ < τ0

ð3Þ

wherein τ and τ0 are the bottom stress and its threshold value above which the 235 sediment erosion emerges; M and τ0 are some empirical parameters that depend 236 on physical and chemical properties of sediments (Mehta, 1988). In the case of 2 237 fine sediments, M and τ0 vary within a range of τ0 = 1-16 dyn/cm (or 0.1−2 −4 238 1.6 N m ) or τ0/ρw = (1-16) × 10 m2 s−2 for normalized wind stress; ρw is −6 239 the water density and Mτ0 = 10 to 5 × 10−5 g/cm2. 240 Solution technique to integrate equation (2) numerically is similar to that of 241 used in the POM code to solve equations for temperature and salinity. Sub242 routines ADVT and PROFT used to integrate equations for temperature and 243 salinity, are modified to take into account, settling velocity and nonzero turbu244 lent flux of a matter at the bottom (note that the “genuine” POM code does 245 account for the surface flux of temperature and/or salinity but does not account 246 for the bottom fluxes). 234

247

248

Particle Transport Model

Suppose that in a moment ti a particle is located in coordinates (xi , yi , zi), wherein the velocity components and horizontal and vertical eddy diffusivities are 251 (Ui , Vi , Wi ), Khi , and Kvi , respectively. Values of (Ui , Vi , Wi ), Khi , and Kvi are implied 252 to have been found by means of interpolation of the hydrodynamic model results to the 253 space-time point (xi , yi , zi , ti). Coordinates of the same particle (xi +1, yi +1, zi +1) in a 254 subsequent moment ti +1 = ti + τ, where τ is a small time increment, can be ex255 pressed as 249 250

xiþ1 ¼ xi þ Ui τ þ xi0 yiþ1 ¼ yi þ Vi τ þ yi0 ziþ1 ¼ zi þ ðWi wS Þτ þ zi0

ð4Þ


Number 1

3

256 257 258 259 260

where ðxi0 ; yi0 ; zi0 Þ are the components of random displacement of the particle due to turbulent velocity fluctuations and wS is the settling velocity. The random displacements ðxi0 ; yi0 ; zi0 Þ can be expressed through eddy diffusivity and time increment; such approach is typical for random walk models. In the simplest case of uniform diffusivity field, these expressions are 0

0

ðxi ; yi Þ ¼ ð2⋅Khi ⋅τ Þ

1=2

ð5Þ

A

262 261 0

zi ¼ ð2⋅Kvi ⋅τ Þ 263 264 265 266

1=2

ð6Þ

A

wherein Khi and Kvi are values of Kh and Kv in the space-time point (xi, yi, zi, ti), A is the Gaussian random value with zero mean and unite variance. It was shown (Thomson, 1984; Hunter et al., 1993; Visser, 1997) that the random walk model (equations 3-6) is equivalent to a continuous equation ∂q ∂2 ¼ 2 Kv⋅q ∂t ∂z

267 268 269

ð7Þ

wherein q(z, t) is the concentration (or probability) of particles. Equation (4) is a simplified Fokker-Plank equation (Risken, 1984), and at Kv(z) = const it becomes identical to the Fickian diffusion equation

∂q ∂ ∂q ¼ Kv : ∂t ∂z ∂z 270 271 272 273 274 275 276 277 278 279 280 281 282 283 284 285 286 287

ð8Þ

In general, the case of Kv(z) ≠ const, equation (7) and its random equivalent equation (6) describe processes whose properties may be quite different from the classic diffusion. For example, Visser (1997) reported results of numerical experiments on random walk of particles that were initially uniform distributed in a layer with reflective boundaries. If the random walk of particles was described by model (6) and the diffusivity was taken spatially variable (Kv(z) ≠ const), the particles being uniformly distributed in the beginning would soon accumulate toward a part of the layer where the diffusivity is minimum. Therefore, application of the ‘naive’ random walk model (equations 3-6) to the case of nonuniform diffusivity results in transformation of initially uniform distribution of particles to a nonuniform one, which is in conflict with traditional diffusion concept and is impossible in the framework of diffusion equation (equations 5-8). In application to the marine environment, it means that the “naive” random walk model (equations 3-6) will forecast unrealistic removal of particles from the BBL where the vertical eddy diffusivity is maximum and further accumulation in abovelaying layers. In order to avoid unrealistic accumulation of particles Hunter et al. (1993) and Visser (1997) proposed following random walk scheme ∂Kv ⋅ τ þ 2Kv zi þ 0:5 ∂z z¼zi

j

∂Kv zi ¼ ∂z

j

4


0

⋅τ ⋅τ z¼zi

1=2 A;

ð9Þ

which is equivalent to the Fickian diffusion equation (7). In principle, equations similar to equation (9) can be also applied to the random horizontal displacements xi0 and yi0 instead of equation (5). However, in our case, this does not seem urgent because we do not expect the presence of stable horizontal gradients of lateral eddy diffusivity in marine environment. A suspended particle transport model based on equations (4), (5), and (9) and coupled with the abovedescribed circulation/turbulence model was developed for the Baltic Sea. As for the boundary conditions for the random walk model, reflection of particles at the sea surface and sea bottom and absorption of particles at the shoreline have been applied. Note that in sediment transport problems, a more sophisticated boundary condition at the bottom is used (e.g., Kuhrts et al., 2004). Namely, matter exchange between sediments and water column is parameterized/controlled by the value of shear stress or friction velocity u* at the bottom. If u* < uD, wherein uD is the deposition threshold, deposition of suspended matter takes place and downward matter flux is q b wS , wherein qb is the suspended matter concentration by the bottom. As a consequence the total number of suspended particles is decreased. If uD < u* < uR, wherein uR ≥ uD is the resuspension threshold, the matter flux across the bottom vanishes (suspended particles are reflected from the bottom and the total number of particles is unchanged). Finally, if u * > u R resuspension process takes place and upward matter flux at the bottom is proportional to (u* − uR), so that the total number of suspended particles increases due to injection of new particles from bottom sediments. Since the objective of this work is

288 289 290 291 292 293 294 295 296 297 298 299 300 301 302 303 304 305 306 307 308 309 310 311 312 313 314 315 316 317 318 319 320 321 322 323 324 325 326 327 328 329 330 331 332 333 334 335

336 337 338 339 340 341 342 343 344 345 346 347 348 349 350 351 352 353 354 355 356 357 358 359 360 361 362 363 364 365 366 367 368 369 370 371 372 373 374 375 376 377 378 379 380 381 382 383

pathways of suspended particles transport rather than matter exchange between sediments and water column, we apply a simple condition of particle refraction at the bottom which conserves the total number of particles. To study pathways of suspended particles transport in the bottom layer of the Baltic Sea, a number of simulation experiments have been performed. In each experiment, 100 particles with a given settling velocity value were released at a given position just above the sea bottom and the trajectories of the particles were calculated using the random walk model (equations 4, 5, and 9) and 3D velocity/diffusivity fields obtained from hydrodynamic model runs at given wind conditions. Note that in each random walk experiment, the 3D velocity/diffusivity fields were taken frozen (i.e., time independent throughout the random walk period of 90-360 days). Of course, particle tracks calculated for such long periods using frozen velocity/diffusivity fields in the random walk model will differ from “real” tracks calculated using time-dependent velocity/ diffusivity fields simulated by the hydrodynamic model at a realistic, synoptically variable wind forcing. The point that we seek is a dependence of suspended particles pathways on circulation patterns established in the Baltic Sea deep water at different wind conditions, and random walk experiments with frozen velocity/ diffusivity fields seem to be most suitable for this purpose. The settling velocity and random walk period in the simulation experiments varied in the range of w S = 0.5-4.0 m/day and ΔT = 90-360 day, respectively. The sites of particle release were chosen in the bottom layer

of the Bornholm Channel, Bornholm Deep, and Słupsk Furrow (Figure 1). 386 Note that in the case of the Bornholm 387 Deep the particles were released just in 388 the center of an area where CW had 389 been dumped after World War II 390 (HELCOM, 1996). 384 385

391

Verification of the Model

Since we are studying transport of 393 suspended particles released just 394 above the sea bottom, before analyzing 395 results of the simulation experiments 396 we have to make sure that (a) the 397 hydrodynamic model does resolve 398 the BBL of enhanced turbulent dif399 fusivity and (b) the random walk 400 model does not display unrealistic re401 moval of particles from the BBL and 402 further accumulation in above-laying 403 layers. 404 Model test runs studied the ability 405 of the random walk model (9) to 406 describe adequately vertical transport 407 of particles in the Baltic Sea BBL. 408 The point is that the random walk 409 model (9) is strictly equivalent to the 410 continuous diffusion equation (8) 411 only if the time increment τ is infini412 tesimally small. Since one cannot deal 413 with infinitesimals in practice, we ap414 plied a straightforward way to choose 415 proper value of τ based on the com416 parison of the random walk model per417 formance at different values of τ and 418 different vertical profiles of Kv in the 419 Baltic Sea BBL simulated in the hydro420 dynamic model. 421 Using test experiment results, one 422 may conclude that at τ = 0.1 s and dif423 fusivity profiles simulated in the Baltic 424 Sea BBL the random walk model (9) 425 preserves uniformity of vertical distri426 bution of particles within a few percent 427 accuracy. All random walk experi428 ments discussed in the next sec429 tion were performed at the time step 430 of τ = 0.1 s. 392

Results of Numerical Experiments

431 432

Simulated Bottom Currents and Stress

433 434

Since the resuspension fluxes of bottom sediments are parameterized through the bottom stress (see equations 3-6), it is important to simulate and analyze variations of the latter in the vicinity of a dump site at different hydrometeorological conditions. One might expect that in the course of major inflow events when lateral changes in salinity and density in BBL are extremely large, the bottom layer velocities and, therefore, stress have maximum values, which make the resuspension process more probable. Northerly and easterly winds are known to contribute to eastward transport of inflow water in the bottom layer of the Bornholm Strait and the Słupsk Furrow. Figure 2 presents velocity of currents at 1 m distance above the bottom simulated after 5 days of northerly and easterly wind stress of 0.2 N m−2 at summertime stratification. It is seen from Figure 2 that the highest bottom current velocities in the open sea are found at four locations: the Bornholm Strait, the Słupsk Sill, the eastern end of the Słupsk Furrow (at a sill between the Słupsk and Gotland/Gdaņsk basins), and the Klaipeda Bank located 10 nm SE of the East Gotland dump site. Of course, in these four locations, we expect to have the highest bottom stress. A pattern of normalized bottom stress simulated under conditions of inflow and no inflow followed by the northerly wind is shown in Figure 3. Note that the inflow event does not result in drastic changes in the bottom stress pattern: there is surprisingly little increase of areas of enhanced


Number 1

5

435 436 437 438 439 440 441 442 443 444 445 446 447 448 449 450 451 452 453 454 455 456 457 458 459 460 461 462 463 464 465 466 467 468 469 470 471 472 473 474 475 476 477

FIGURE 2 Velocity of currents in the bottom layer (1 m above the bottom) simulated at northerly (middle) and easterly (bottom) wind of 0.2 N m−2 stress along with a bottom topography map (top). Polygons and circle show position of the Bornholm and East Gotland dump sites.

478 479 480 481 482 483 484 485 486 487 488 489 490 491 492

bottom stress. Note that bottom stress values exceed the lower limit f o r t h e r e s u s p e n s i o n t h r e s h o l d 4 2 2 only in τmin 0 =ρw ¼ 1×10 m s four open-sea locations mentioned above (red spots in Figure 3). The value of simulated bottom stress in locations of the Bornholm and Gotland dump sites is found to be more than an order of magnitude below the lower limit for the resuspension threshold, so the resuspension of bottom sediments due to near-bottom currents seems to be unlikely in those areas at any hydrometeorological conditions

6

imaginable. However, we do not ex494 clude that generally accepted value for 495 the lower limit of resuspension threshmin 496 old, τ0 ¼ 0:1N m2 , is strongly 497 overestimated if one is dealing with a 498 nonconsolidated sediments. 499 Figure 4 presents patterns of nor500 malized bottom stress simulated 501 under conditions of no inflow with 502 summertime stratification followed 503 by 5 days of the northerly, southerly, 504 easterly, and westerly wind of the −2 505 force 0.2 N m . In general, bottom 506 stress patterns in Figure 4 are similar 507 to that of Figure 3. The only sub493


stantial difference is the absence of high bottom stress area in the Klaipeda Bank at easterly and westerly winds.

508

Runs for Finite Difference Diffusion Model

511

To test the performance of a diffusion model based on numerical integrating of a finite difference analogue of equation (4) imbedded in POM code, an instant source of a passive matter was placed at the bottom in the Bornholm Strait (actually the initial concentration of tracer is vertically uniform within the nearest to the bottom σ-layer of 0.002D ≈ 0.1 m thickness). The instant source centered at (14°30′E, 55°15′N) has the Gaussian shape in horizontal plane with length scale parameter of 6 km (the distance at which the matter concentration being taken unity in the source center goes down a factor of exp(1/2)). The reason for deploying the passive matter source in the Bornholm Strait was as follows. Based on distribution of bottom sediments, Emelyanov et.al. (1995) suggested a two-branch pattern of bottom layer currents in the Bornholm Basin. They asserted that the deep flow entering the Bornholm Basin from the Bornholm Strait undergoes a bifurcation. The two branches of the flow go to the east, so that the northern (southern) branch follows the northern (southern) slope of the basin. Finally, the branches merge before entering the Słupsk Furrow. Such a pattern is not easy to test using measured and/or simulated currents because the velocities in the bottom layer are expected to be very small and variable in space and time. However, if we place a source of passive matter at the bottom in the Bornholm Strait and let it diffuse and be carried by currents, the matter distribution

513

509 510

512

514 515 516 517 518 519 520 521 522 523 524 525 526 527 528 529 530 531 532 533 534 535 536 537 538 539 540 541 542 543 544 545 546 547 548 549 550 551 552 553 554

FIGURE 3 Normalized bottom stress simulated after 5 days of 0.2 N m−2 northerly wind at no-inflow (top left) and inflow (other panels) initial stratification. Red spots are areas where bottom stress values exceed the lower limit for the resuspension threshold τ0/ρw = 1 × 10−4 m2 s−2.

555 556 557 558 559

after a relatively long period will trace the pathways of deep layer circulation and, therefore, can outline the above suggested branches of near-bottom currents (if they do exist).

Figures 5 and 6 show distribution 561 of the matter in the bottom, mid562 dle, and surface layers simulated by a 563 27-day period of easterly and northerly 564 winds, respectively. Pattern of the 560

FIGURE 4 Same as in Figure 3, but for summer stratification and northerly (top left), southerly (top right), easterly (bottom left), and westerly wind (bottom right).

tracer concentration in the bottom layer clearly shows a bifurcation of the matter pathway in the Bornholm Basin, which is undoubtedly caused by bifurcation of the near bottom current just in accordance with the hypothesis by Emelyanov et al. (1995). Note that the bottom layer in the central part of the Bornholm Basin including the location of the Bornholm munitions dump site is totally free of tracer, while the above lying middle and even surface layers contain some remnants of tracer with low concentration of the matter.

565

Particle Release in the Bornholm Deep BBL

580

A number of numerical experiments were performed with particle release in the bottom layer of the Bornholm Deep just in the center of an area where CW had been dumped after World War II (see Figure 1). So such experiments might have a practical importance in view of hypothetical possibility of a leakage from the munitions. At the northerly and easterly wind conditions, the particles initially move to the west toward the Bornholm Island and then get involved in either northern or southern detour around the Bornholm Deep (Figure 7). In contrast to the northerly and easterly wind conditions, at the westerly and southerly wind conditions particles with settling velocity of wS = 2 m/day initially move to the northeast for some 20 km, then get involved in cyclonic rotation and remain forever trapped within an area of 30 km in diameter (see Figure 8, top panels, where zoom-in particle tracks are presented). Simulation experiments showed that particles with w S = 2 m/day being released in the Bornholm Deep BBL remain trapped within the basin at

582


Number 1

566 567 568 569 570 571 572 573 574 575 576 577 578 579

581

7

583 584 585 586 587 588 589 590 591 592 593 594 595 596 597 598 599 600 601 602 603 604 605 606 607 608 609 610 611

612 613 614 615 616 617 618 619 620 621 622 623 624 625 626

FIGURE 5

FIGURE 6

Distribution of a passive matter in the bottom, middle, and surface layers simulated by 27-day period of an easterly wind of 0.2 N m−2 wind stress. The instant source of the matter was placed in the bottom layer of the Bornholm Strait at (14°30′E, 55°15′N).

Same as in Figure 5, but for the northerly wind.

the SSE, S, SSW, SW, SWW, W, NWW, and NW wind conditions. If the settling velocity is small enough (w S < 1.3 m/day and w S < 1.6 m/day for the southerly and westerly wind conditions, respectively), the particles are no longer trapped within the Bornholm Deep (Figure 8, middle and bottom panels). Most of them make an anticyclonic detour of the Bornholm Basin and flow in the Bornholm Strait to enter the Arkona Basin. At the westerly wind conditions, some of the particles with w S
1.3 m/day and wS > 1.6 m/day at southerly and westerly winds, respectively). A summary plot of particle releases in the center of the Bornholm dumpsite at different wind conditions is given in Figure 9. One hundred particles with settling velocity of 2 m/day were released at the bottom in the center of a circle—primary dumpsite

(15.62 E, 55.35 N) at 16 different wind conditions. Particle tracks for 669 the period of 3 or 6 months are plot670 ted. This summary plot has been 671 used to choose sampling transects dur672 ing the last cruise under MERCW 673 project (winter, 2008). It is seen that 674 at SSE, S, SSW, SW, SWW, W, 675 NWW, and NW winds, particles 676 move NE for a distance of 20-30 km, 667 668

rotate cyclonically and remain trapped within the Bornholm Deep. Note that westerly winds are most typical for the region. For this reason, sampling at a transect of 30-km long from the dumpsite center towards the NE seems to be most preferable. At N, NNE, NE, NEE, E, SEE, and SE winds, particles initially move to the west for a distance of 15-20 km and then turn to the north or south. Therefore, a 20-km-long transect from the dumpsite center towards the west seems to be the second preferable transect. Similar numerical experiments were performed with particle release in the bottom layer of the Bornholm Strait and in the Słupsk Furrow and presented by Zhurbas et al. (2010). Such experiments have an importance for getting a wider understanding of dispersion of a passive matter in the Baltic deep waters.

FIGURE 8 Tracks of suspended particles with different value of the settling velocity wS released in the bottom layer of the Bornholm Deep under the southerly and westerly wind conditions.

FIGURE 9 Summary plot of particle releases in the center of the Bornholm dumpsite at different wind conditions. Settling velocity wS = 2 m/day.


Number 1

9

685 686 687 688 689 690 691 692 693 694 695 696 697 698 699 700 701 702 703 704 705 706 707 708

709 710 711 712 713 714 715 716 717 718 719 720 721 722 723 724 725

Calculation of Predicted Environmental Concentrations of CWA The last simulations, the distribution of passive matter, were aimed to - perform simulations of 3D fields of CWA Predicted Environmental Concentration (PEC) in the sea for the continuous release scenario under different wind conditions; - calculate synthesized climatic 3D fields and 2D maps of stationary CWA PEC in the water based on circulation and matter transport models presented in Results of Numerical Experiments under different wind conditions and wind stress statistics; and - develop an approach to extrapolate measured CWA concentration to whole water domain on the basis of simulation results. Since the boundary forcing condition for the circulation model is formulated in terms of the wind stress rather than wind velocity, it seems worthwhile to produce the wind stress statistics. The wind stress is roughly a quadratic function of the wind speed, and its components (τx , τy) can be parameterized as

726 727 728

τ x ; τ y ¼ ρa cD Wx ; Wy W ; W ¼ Wx2 þ Wy2

0:5

;

wherein Wx, Wy are components of wind velocity vector at 10-m level, ρa ≈ 1.28 kg m−3 is the air density, and CD is the resistance coefficient calculated from an empirical formula (Large and Pond, 1981). cD ⋅103 ¼

1:14 0:49 þ 0:065W

1

for W ≤ 10m s for 10 m s1 ≤ W ≤ 25m s1

FIGURE 10 Distribution of cumulative wind stress for the period 1961–1998.

10


To analyze wind stress statistics, we use data of the Christiansø Weather Station (55 19.26 N, 15 11.16 E), the closest station to the Bornholm dump site (the center 55 20.77 N, 15 37.28 E). For the wind stress statistics, we calculate the shares of cumulative wind stress (the product of the wind stress and wind duration) distributed between 16 wind direction sectors of 22.5° angle range each. Distribution of cumulative wind stress between 16 wind directions calculated for Christiansø is shown in Figure 10. There are two preferential wind directions—the westerly and easterly wind petals comprising 47.3% and 28.6% of the wind stress total, respectively.

729

Mapping of CWA PEC for the Bornholm Dumpsite

748

Following Sanderson et al. (2008), let us consider a continuous release of a CWA from a primary dumpsite and calculate maps of CWA PEC using the circulation and matter transport models presented in Results of Numerical Experiments. An approach aimed to map CWA PEC is as follows. First, we run the circulation model applying wind stress of 0.2 N/m2 (approximately 10 m/s of wind speed) for a 3-day period and for each of the 16 wind directions. Then we run the matter transport model with frozen current velocity and diffusivity fields taken from the 16 circulation model runs for a period much greater than the CWA degradation half-life in order to obtain 3D asymptotic (stationary) field of CWA PEC for fixed wind directions. After 16 stationary fields of CWA PEC are obtained, we calculate a synthesized climatic 3D CWA PEC field by averaging the fixed wind CWA PEC fields with weights equal to the shares of the

750

730 731 732 733 734 735 736 737 738 739 740 741 742 743 744 745 746 747

749

751 752 753 754 755 756 757 758 759 760 761 762 763 764 765 766 767 768 769 770 771 772 773 774 775

776 777 778 779 780 781 782 783 784 785 786 787 788 789 790 791 792 793 794 795 796 797 798 799 800 801 802 803 804 805 806 807 808 809 810 811 812 813 814 815 816 817 818 819 820 821 822 823

cumulative wind stress related to the 16 wind directions (Figure 10). First of all, let us consider yperite (or sulfur mustard, CWA2), the CWA dumped in the greatest quantity. The amount of yperite estimated to have been dumped is M = 7027 tons and the compound degradation halflife is T1/2 = 56 days (Sanderson et al., 2008). Taking Texp = 60 years for the exposure period and A1 = 97 km2 (a circle with radius of 3 nautical miles, its center in 55 20.77 N, 15 37.28 E) for the primary dumpsite area we obtain for the rate of CWA supply to the near-bottom sea layer E 1 = M/A1Texp = 3.83 × 1011 kg/m2. Some part of the CWA supply forms the CWA flux to sediment which is taken into account too (see Sanderson et al., 2008, for details). The matter transport equation was integrated numerically for the period of 256 days, which was found to be enough to arrive at the asymptotic stationary CWA PEC values with reasonable accuracy. Figure 11 presents the maps of stationary yperite PEC in the water just above the bottom calculated for the two most probable wind directions (W and SEE winds). The maximum CWA PECs are located at the outer edge of the primary dumpsite area (a circle on the top and middle panels of Figure 4), just in accordance with Sanderson et al. (2008). The tail of enhanced yperite PEC is directed to NEE and W for W and SEE wind directions, respectively, and the tail directions are well correlated with the suspended particle pathways calculated for the same wind directions. Figure 12 (left panels) presents the map of synthesized climatic 2D stationary yperite PEC in the water just above the bottom calculated using the wind stress statistics.

FIGURE 11 Maps of the stationary yperite PEC (μg/L) in the water just above the bottom calculated for W (left) and SEE (right) winds. The primary dump site is marked by a circle. The bottom panels present suspended particle tracks for the same wind directions.

It is not surprising that the climatic 825 2D Yperite PEC map looks like a 826 superposition of CWA patterns related 827 to most probable wind stress direction 828 petals (cf. Figures 11 and 12). It is great829 ly encouraging that the maximum yper830 ite PEC from our model (4.6, 3.4, and 831 3.3 μg/L; see color scales for Figures 11 832 and 12) fits well the respective value of 824

3.98 μg/L obtained by Sanderson et al. (2008) from their simplified 2D advection-diffusion model. Using the above-described approach, we developed a synthesized climatic PEC map for one of the short living CWA (monochlorobenzene, CWA9) (T1/2 = 4 days; see Figure 12, right panels). The shape of PEC

FIGURE 12 Synthesized climatic 2D stationary PEC maps in the water just above the bottom calculated using the wind stress statistics for yperite (CWA2, left panels) and monochlorobenzene (CWA9, right panels). The concentration unit for the color scale is μg/L.

January/February 2012

Volume 46

Number 1

11

833 834 835 836 837 838 839 840 841

842 843 844 845 846 847 848 849 850

distribution for CWA2 and CWA9 is about identical; the size of high concentration plume for CWA2 seems a little bit larger than that of CWA9 due to the longer half-life (56 days vs. 4 days). Similar to CWA2, the maximum PEC for CWA9 from our model (0.58 μg/L) fits the respective value 0.705 μg/L obtained by Sanderson et al. (2008).

851

Some Methodical Calculations

852

The shapes of CWA PEC maps for CWA2 and CWA9 were about identical despite the huge difference in the degradation half-life. To demonstrate clearly effect of half-life on stationary distribution of CWA concentration, we repeated CWA2 simulations with the half-life time 10 times larger and smaller (i.e., T1/2 = 560 and 5.6 days). The maps of PEC at SEE wind for these two cases are shown in Figure 13). The 100 times decrease of the degradation half-life time results in minor decrease

853 854 855 856 857 858 859 860 861 862 863 864

of the maximum PEC (7%) and considerable shrinking of a peripheral 867 region of relatively high concentrations. 865 866

Application of Simulated CWA PEC Maps to Extrapolate the 870 Observed CWA Concentrations 868 869

Applying the circulation and mat872 ter transport models with real wind 873 stress statistics, we have simulated 3D 874 stationary concentration of decaying 875 CWAs for continuous release scenario. 876 This approach is expected to forecast 877 plausibly (reasonably well) the shape 878 of spatial distribution of CWA, while 879 the absolute values of the simulated 880 concentration may be far from real881 ity because the exposure period 882 (60 years) and therefore the release 883 rate of CWA was arbitrary. However, 884 having VERIFIN’s (Finnish Institute 885 for Verification of the Chemical Weap886 ons Convention, University of Helsinki, 887 MERCW participant) measured data

888 889 890 891 892 893

871

FIGURE 13 Maps of the stationary yperite PEC in the water just above the bottom calculated for SEE wind and 10 times increased (top) and decreased (bottom) degradation half period.

12

for some sites (Missiaen et al., 2010; Sanderson et al., 2010), we can introduce a scaling factor to fit the simulated concentrations to the measured ones and therefore extrapolate the measured concentrations to the whole domain.


Conclusions The simulation results can be summarized as follows: - Scenario simulations made it possible to calculate 3D fields of CWA PEC in the sea for the continuous release scenario under different wind conditions. Using wind stress statistics from the Christiansø weather station and the simulation results, synthesized climatic 3D fields and 2D maps of stationary PEC in the water for yperite and monochlorobenzene were calculated. - A straightforward way is offered to extrapolate measured CWA concentrations to whole water domain by means of scaling of simulated 3D concentrations. In closing, it seems worthwhile to discuss discrepancies between measured and simulated concentrations. Almost all samples taken during the project and analyzed by VERIFIN refer to upper sediments, while modeling results presented here deal with concentrations of dissolved agents in seawater, hence straightforward comparison is hardly justified. To our knowledge, there are no reliable observations that confirm existence of excessive concentrations of dissolved CWAs in the near-bottom water layer of the primary dumpsite in comparison with other regions of the Baltic Sea, which is in conflict with the simulations. There are two potential causes for this discrepancy: (a) fluxes (input) of dissolved CWAs to the near bottom water and/or

894 895 896 897 898 899 900 901 902 903 904 905 906 907 908 909 910 911 912 913 914 915 916 917 918 919 920 921 922 923 924 925 926 927 928 929 930 931 932 933

934 935 936 937 938 939 940 941 942 943 944 945 946 947 948

949 950 951 952 953 954 955 956 957 958

(b) half-life of CWAs decay used in the model are (greatly?) overestimated. Following Sanderson et al. (2008), the fluxes were calculated as total mass of a dumped CWA divided by a 60-year period (the continuous release scenario). However, there are some doubts if substantial quantities of CWAs can reach enter the sea water media in dissolved form. Indeed, being dumped in sediment layer, a CWA leaking from shells and bombs comes contact with pore water and can be hydrolyzed before it reaches the sea water-sediment interface.

Acknowledgments

977 In:

Proc. Int. Symp., Hamburg, Aug. 108, 978 eds. Sunderman, J., Holtz, K.-P., pp. 203-14. 979 Berlin: Springer-Verlag. 980 Blumberg,

A.F., & Mellor, G.L. 1983. 981 Diagnostic and prognostic numerical calcula982 tion studies of the South Atlantic Bight. 983 J Geophys Res. 88:4579-92. http://dx.doi. 984 org/10.1029/JC088iC08p04579. 985 Cappelen,

J., & Jųrgensen, B. 1999. 986 Observed wind direction and speed in 987 Denmark—with Climatological Standard 988 Normals, 1961-1990. DMI Technical 989 Report No. 99-13. E.M., Kravtsov, V.A., Savin, Y.I., 991 Paka, V.T., & Khalikov, I.S. 2010. Influence 992 of chemical weapons and warfare agents on 993 the metal contents in sediments in the 994 Bornholm Basin, the Baltic Sea. Baltica. 995 23(2):77-90.

990 Emelyanov,

This study has been carried out in the framework of the EC-FP6 project “MERCW” (Modeling of Environ- 996 Emelyanov, E.M., Trimonis, E., Slobodyanik, mental Risks related to sea-dumped 997 V., & Nielsen, O.B. 1995. Sediment thick998 ness and accumulation rates. In: Geology of Chemical Weapons) (Contract 999 the Bornholm Basin. Aarhus Geosci. 5:81-4. INCO-CT2005-013408) and supported by the Russian Foundation for 1000 HELCOM. 1996. Third periodic assessment Q2 Basic Research (Grants 11-05-00718 1001 of the state of the marine environment of the 1002 Baltic Sea, 1989–1993, Executive summary and 12-05-00422). 1003 and

background document. Baltic Sea 1004 Environment Proceedings 64A; 64B. 959 960 961 962 963 964 965 966 967

Authors: Victor Zhurbas Shirshov Institute of Oceanology of Russian Academy of Sciences, Moscow/Kaliningrad, Russia and Marine Systems Institute, Tallinn University of Technology, Tallinn, Estonia Email: [email protected]

968 969 970 971 972

973 974 975 976

Q1

Vadim Paka Shirshov Institute of Oceanology of Russian Academy of Sciences Email: [email protected]

References Blumberg, A.F., & Mellor, G.L. 1980. A coastal ocean numerical model. In: Mathematical Modelling of Estuarine Physics.

1005 Hunter,

J.R., Craig, P.D., & Phillips, H.E. 1006 1993. On the use of random walk models 1007 with spatially variable diffusivity. J Comput 1008 Phys. 106(2):366-76. 1009 Jönsson,

A., Danielsson, Å., & Rahm, L. 1010 2005. Bottom type distribution based on 1011 wave friction velocity in the Baltic Sea. 1012 Cont Shelf Res. 25(3):419-35. http://dx.doi. 1013 org/10.1016/j.csr.2004.09.011. 1014 Kuhrts,

C., Fennel, W., & Seifert, T. 2004. 1015 Model studies of transport of sedimentary 1016 material in the western Baltic. J Marine Syst. 1017 52(1-4):167-90. http://dx.doi.org/10.1016/ 1018 j.jmarsys.2004.03.005. 1019 Large,

W.G., & Pond, S. 1981. Open ocean 1020 momentum flux measurements in moderate 1021 to strong winds. J Phys Oceanogr. 11:324-36. 1022 http://dx.doi.org/10.1175/1520-0485(1981) 1023 0112.0.CO;2.

Mehta, A.J. 1988. Laboratory studies of cohesive sediments deposition and erosion. In: Physical Processes in Estuaries, eds. Dronkers, J., van Leussen, W., 427-45. New York: Springer-Verlag.

1024

Mellor, G.L. 2004. Users guide for a threedimensional, primitive equation, numerical ocean model. http://www.aos.princeton.edu/ WWWPUBLIC/htdocs.pom.

1029

Missiaen, T., Söderström, M., Popescu, I., & Vanninen, P. 2010. Evaluation of a chemical munition dumpsite in the Baltic Sea based on geophysical and chemical investigations. Sci Total Environ. 408:3536-53. http://dx. doi.org/10.1016/j.scitotenv.2010.04.056.

1033

Risken, H. 1984. The Fokker-Plank Equation: Methods of Solution and Application. Heidelberg: Springer-Verlag.

1039

Sanderson, H., Fauser, P., Thomsen, M., & Sørensen, P. 2008. Screening level fish community risk assessment of chemical warfare agents in the Baltic Sea. J Hazard Mater. 154:846-54. http://dx.doi.org/10.1016/ j.jhazmat.2007.10.117.

1042

Sanderson, H., Fauser, P., Thomsen, M., Vanninen, M., Söderström, M., Savin, Y., … Feller, P. 2010. Environmental hazards of sea-dumped chemical weapons. Environ Scie Technol. 44(12):4389-94. http://dx.doi. org/10.1021/es903472a.

1048

Seifert, T., Fennel, W., & Kuhrts, C. 2009. High resolution model studies of transport of sedimentary material in the south-western Baltic. J Marine Syst. 75(3-4):382-96. http:// dx.doi.org/10.1016/j.jmarsys.2007.01.017.

1054

Seifert, T., Tauber, F., & Kayser, B. 2001. A high resolution spherical grid topography of the Baltic Sea—revised edition. http:// www.io-warnemuende.de/iowtopo.

1059

Thomson, D.J. 1984. Random walk modelling of diffusion in inhomogeneous turbulence. Q J Roy Meteor Soc. 110(466): 1107-20. http://dx.doi.org/10.1002/ qj.49711046620.

1063

Visser, A.W. 1997. Using random walk models to simulate the vertical distribution of particles in a turbulent water column.

1068

January/February 2012

Volume 46

Number 1

13

1025 1026 1027 1028

1030 1031 1032Q3

1034 1035 1036 1037 1038

1040 1041

1043 1044 1045 1046 1047

1049 1050 1051 1052 1053

1055 1056 1057 1058

1060 1061 1062Q4

1064 1065 1066 1067

1069 1070

1071 1072 1073 1074 1075 1076 1077 1078 1079 1080 1081 1082 1083 1084 1085 1086 1087 1088 1089 1090

Mar Ecol-Prog Ser. 158:275-81. http://dx.doi. org/10.3354/meps158275. Zhurbas, V., Elken, J., Väli, G., Kuzmina, N., & Paka, V. 2010. Pathways of suspended particles transport in the bottom layer of the southern Baltic sea depending on the wind forcing (numerical simulation). Okeanologiya. 50(6):890-90. Zhurbas, V., Stipa, T., Mälkki, P., Paka, V., Golenko, N., Hense, I., & Sklyarov, V. 2004. Generation of subsurface cyclonic eddies in the Southeast Baltic Sea: Observations and numerical experiments. J Geophys Res. 109:C05033. doi: 10.1029/2003JC002074. Zhurbas, V.M., Oh, I.S., & Paka, V.T. 2003. Generation of cyclonic eddies in the Eastern Gotland Basin of the Baltic Sea following dense water inflows: numerical experiments. J Marine Syst. 38:323-36. http:// dx.doi.org/10.1016/S0924-7963(02)00251-8.

14


AUTHOR QUERIES AUTHOR PLEASE ANSWER QUERIES Q1: Please insert citation of this reference in the body. Q2: Please provide complete reference details. Q3: Please provide date of access. Q4: Please provide date of access.

END OF AUTHOR QUERIES