United States
Environmental Protection
Agency
Office of Research
and Development
Gulf Breeze. FL
EPA/600/9-91/046a
October 1991
Research and Development
*>EPA
C 0)
<*> *
«£
2 £
O V
c£ .
) "73
CO > 0)
Alaska Oil Spill
Bioremediation Project
Science Advisory Board
Draft Report
Sections 1 through 6
-------
EPA/600/9-91/046a
October 1991
ALASKA OIL SPILL BIOREMEDIATION PROJECT
SCIENCE ADVISORY BOARD DRAFT REPORT
SECTIONS 1 THROUGH 6
P.H. Pritchard, EPA/ERL-Gulf Breeze, Scientific Coordinator
C.F. Costa, EPA/EMSL-Las Vegas, Program Manager
L. Suit, TRI-Rockville, Technical Editor
Contributors:
R. Araujo, EPA/ERL-Athens; D. Chaloud, Lockheed; L. Cifuentes, Texas A&M University;
J. Clark, EPA/ERL-Gulf Breeze; L. Claxton, EPA/RTP; R. Coffin, EPA/ERL-Gulf Breeze;
R. Cripe, EPA/ERL-Gulf Breeze; D. Dalton, TRI; R. Gerlach, Lockheed;
J. Glaser, EPA/RREL-Cincinnati; J. Haines, EPA/RREL-Cincinnati;
D. Heggem, EPA/EMSL-Las Vegas; F. Kremer, EPA/CERI-Cincinnati; J. Mueller, SBP;
A. Neale, Lockheed; J. Rogers, EPA/ERL-Athens;
S. Safferman, EPA/RREL-Cincinnati; M. Shelton, TRI;
A. Venosa, EPA/RREL-Cincinnati
U.S. Environmental Protection Agency
Region 5, Library (PL-12J)
P.vnn rnn»rJhutA«. 77 West Jackson Boulevard, 12th Floor
Exxon Contributors. Chicag0i )L 60604-3590
J. Bragg, R. Chianelli, and S. Hinton, Annandale, N.J.;
S. McMillen, Houston, Texas; R. Prince, Annandale, N.J.
Prepared by:
ENVIRONMENTAL RESEARCH LABORATORY
OFFICE OF RESEARCH AND DEVELOPMENT
U.S. ENVIRONMENTAL PROTECTION AGENCY
SABINE ISLAND
GULF BREEZE, FLORIDA 32561
-------
DISCLAIMER
"Mention of trade names or commercial products does not constitute
endorsement or recommendation for use."
NOTICE
This document is a preliminary draft intended for review by the
Science Advisory Board regarding its technical merit. It has not
been formally released by the U.S. Environmental Protection
Agency and should not at this stage be construed to represent
Agency policy. Following scientific and policy reviews, a final
draft of the document will be published and made available to the
public.
11
-------
ABSTRACT
The Alaska Oil Spill Bioremediation Project was initiated to demonstrate the feasibility
of oil bioremediation as a secondary cleanup tool on selected beaches in Prince William
Sound, and to further the understanding of the microbial ecology of oil biodegradation
on shorelines. It was shown that the addition of oleophilic slow- release/granular and
nutrient solution fertilizers to oil-contaminated beaches in Prince William Sound
increased oil biodegradation rates greater than four-fold over removal rates on untreated
oiled beaches. The application of fertilizer solutions proved to be the most efficient
system for exposing oil-degrading microorganisms to nutrients. Data on the rate and
extent of microbial degradation of oil was crucial to the acceptance of bioremediation
as a cleanup technology. This enhanced biodegradation was evidenced by changes in
several constituent hydrocarbon groups resulting in the disappearance of oil residues.
Supporting studies demonstrated that bioremediation of oil is a reasonable and
environmentally sound secondary cleanup procedure. It appears to work in both surface
and subsurface beach material. Although there was an overall lack of general oil
biodegradation at Disk Island, studies during the summer of 1990 at Elrington Island
showed that a pulse application of nutrients provides sustained accelerated
biodegradation of oil over a three to four week period. This pulse application
phenomenon has significant potential for addressing future oil spills since it is as
effective as a continuous long-term application. In addition, the use of sampling baskets
containing homogenized beach material was a reliable method to supplant direct
sampling of beach material.
111
-------
TABLE OF CONTENTS
ACKNOWLEDGEMENTS xliii
EXECUTIVE SUMMARY xlv
SECTION 1 - INTRODUCTION 1
BACKGROUND 1
DESCRIPTION OF THE AREA IMPACTED BY THE SPILL 2
BIOREMEDIATION APPROACH 6
FIELD OPERATIONS - SUMMER 1989 12
Test Beach Selection 12
Fertilizer Selection 13
Snug Harbor Demonstration 13
Passage Cove Demonstration 14
Supporting Laboratory Studies 14
Follow-Up Research - Winter 1989/1990 15
FIELD OPERATIONS - SUMMER 1990 16
Monitoring Program 16
Disk Island Demonstration 16
Elrington Island Demonstration 17
SECTION 2 - FERTILIZER SELECTION AND CHARACTERISTICS 18
BACKGROUND 18
DESCRIPTION OF FERTILIZER FORMULATIONS 18
WOODACE Briquettes 19
PAR EX Granules 19
OSMOCOTE Briquettes 19
MAGAMP 20
SIERRA CHEMICAL Granules 20
INIPOL Oleophilic Fertilizer 20
METHODS 21
Static Tests 22
Intermittent Submersion Tests 22
Field Tests 23
RESULTS 23
WOODACE Briquettes 23
PAR EX Granules 28
iv
-------
TABLE OF CONTENTS (Continued)
OSMOCOTE Briquettes 29
MAGAMP Briquettes 29
SIERRA CHEMICAL Granules 31
INIPOL Oleophilic Fertilizer 35
SUMMARY AND CONCLUSIONS 35
SECTION 3 - TEST PLOT DESIGN AND SAMPLING 39
STUDY AREA DESCRIPTIONS 39
Snug Harbor 39
Passage Cove 43
Disk Island 43
Elrington Island 44
SAMPLING METHODS IN 1989 44
Snug Harbor , 44
Passage Cove 50
SAMPLING METHODS IN 1990 50
Disk Island 52
Elrington Island 54
Basket Removal and Sampling 59
METHOD OF FERTILIZER APPLICATION 61
Slow-Release Fertilizers 61
Oleophilic Fertilizer 67
Fertilizer Solution 67
SECTION 4 - CHEMICAL AND BIOLOGICAL ANALYTICAL PROCEDURES 79
NUTRIENT ANALYSIS 79
Nitrite and Nitrate 82
Ammonia 82
Phosphate 82
OIL CHEMISTRY 83
Oil Residue Weight 83
Oil Composition 85
Respirometric Studies 85
-------
TABLE OF CONTENTS (Continued)
MICROBIOLOGY 87
Numbers of Oil-Degrading Microorganisms 87
Mineralization of Radiolabeled Hydrocarbons 88
ECOLOGICAL MONflTORING 89
Chlorophyll 90
Primary Productivity 90
Bacterial Abundance 90
Bacterial Productivity 90
Caged Mussels 91
Field Toxicity Tests 91
LABORATORY FLASK STUDIES 92
Shake Flasks (Exxon) 92
Respirometric Flasks 93
Biometer Flasks 95
Measurement of Carbon Dioxide 98
MICROCOSM STUDIES 98
Jar Microcosms 99
Tank Microcosms 99
Column Microcosms 101
CHEMICAL EFFECT OF OLEOPHILIC FERTILIZER 105
TOXICITY STUDIES 107
Acute Tests 107
Chronic Estimator Toxicity Tests 107
Toxicity of INIPOL and its Constituents to Mammalian and Avian Wildlife 108
MUTAGENICITY TESTS 109
STABLE ISOTOPES 109
Stable Istope Microcosm Study 112
Bacterial Bioassays and Nucleic Acid Concentration 112
Ammonium and Nitrate Distillations 113
Isotopic Analysis 113
Other Analyses 114
VI
-------
TABLE OF CONTENTS (Continued)
SECTION 5 - QUALITY ASSURANCE/QUALITY CONTROL 115
BACKGROUND 115
QUALITY ASSURANCE/QUALITY CONTROL COMPONENTS 116
QA Plan 116
Design Characteristics 117
Quality Assurance Samples 119
Quality Control Samples 119
Replicate Samples - Disk Island and Elrington Island 119
Analytical Duplicate/Triplicate 121
Field Audit Blank Sample 121
Reagent Blank 123
Biometer Test Blanks - Passage Cove, Disk Island, Elrington Island 123
Quality Control Check Sample 123
Detection Limit QCCS 125
Matrix Spike 125
Calibrations 125
Instrument Detection Limits 126
Assistance Audits 126
SECTION 6 - SNUG HARBOR FIELD RESULTS 134
VISUAL OBSERVATIONS 134
NUTRIENT CONCENTRATIONS 137
OIL CHEMISTRY 146
Oil Residue Weight 146
Oil Composition 161
MICROBIOLOGY 216
Numbers of Oil-Degrading Bacteria 216
ECOLOGICAL MONITORING 221
Nutrients from Nearshore Waters 223
Chlorophyll Analyses 223
Phytoplankton Primary Productivity 223
Bacterial Abundance 231
Bacterial Production 231
Caged Mussels 236
MICROCOSM STUDIES 237
Tank Microcosms 237
vu
-------
TABLE OF CONTENTS (Continued)
Jar Microcosms 242
MUTAGENICITY TESTS 245
SUMMARY AND CONCLUSIONS 247
SECTION 7 - PASSAGE COVE FIELD RESULTS 251
VISUAL OBSERVATIONS 251
NUTRIENT CONCENTRATIONS 253
OIL CHEMISTRY 253
Oil Residue Weight 253
Oil Composition 263
WINTER 1989/1990 SAMPLING 300
SPRING 1990 SAMPLING 309
MICROBIOLOGY 311
Numbers of Oil-Degrading Bacteria 311
Microbial Activity 313
ECOLOGICAL MONITORING 321
Nutrients 321
Chlorophyll Analysis 321
Phytoplankton Primary Productivity 321
Bacterial Abundance 325
Bacterial Productivity 325
Caged Mussels 325
Field Toxicity Tests 330
SUMMARY AND CONCLUSIONS 333
SECTION 8 - DISK ISLAND FIELD RESULTS 336
NUTRIENT RELEASE TEST 336
NUTRIENT CONCENTRATIONS 344
OIL CHEMISTRY 344
MICROBIOLOGY 369
SUMMARY AND CONCLUSIONS 376
Vlll
-------
TABLE OF CONTENTS (Continued)
SECTION 9 - ELRINGTON ISLAND FIELD RESULTS 378
VISUAL OBSERVATIONS 378
NUTRIENT CONCENTRATIONS 379
OIL CHEMISTRY 379
Oil Residue Weight 379
Oil Composition • 387
MICROBIOLOGY 400
Numbers of Oil-Degrading Bacteria 400
Microbial Activity 401
Dissolved Oxygen and Nutrient Uptake 410
MICROCOSM STUDIES 416
Effect of Oil Concentration 417
WINTER SAMPLING 421
SUMMARY AND CONCLUSIONS 423
SECTION 10 - SUPPLEMENTAL LABORATORY STUDIES 427
SHAKE FLASKS 427
RESPIROMETRY 435
TOXICITY OF OLEOPHILIC FERTILIZER 443
Laboratory Bioassays 443
Wildlife Toxicity Issues 447
Atmospheric Ammonia 448
Aqueous Ammonia 454
Urea 455
2-butoxy-ethanol 455
Lauryl sulfate (surrogate for laureth phosphate) 456
TOXCITY OF CUSTOMBLEN PELLETS TO BIRDS 457
BIOMETER STUDIES 457
Fertilizer Specific Activity 457
Agitation Rates 464
Fertilizer Application Strategies 464
Bioaugmentation Studies 475
IX
-------
TABLE OF CONTENTS (Continued)
Oleophilic Fertilizer Mode of Action 475
SUMMARY AND CONCLUSIONS 483
SECTION 11 - STABLE ISOTOPES 484
FOOD CHAIN STUDIES 484
Food Chain Studies at Snug Harbor and Passage Cove, Summer 1989 484
Fertilizer Nitrogen Assimilation by Food Chains on Beaches, Summer 1989 487
Food Chain Studies at Elrington and Disk Island, Summer 1990 492
Fertilizer Nitrogen Assimilation by Food Chains on Beaches, Summer 1990 495
ANCILLARY FIELD DATA, SUMMER 1990 498
MICROCOSM EXPERIMENT 498
SECTION 12 - MODELING OIL BIOREMEDIATION 506
APPROACH AND ANALYSIS 506
Modeling Procedures for Biological Waste Treatment Systems 507
Modeling Approach for Bioremediation 508
Review of Field and Laboratory Data Collected the Summer of 1990 511
Methodology to Calibrate Bioremediation Model 512
Modeling Assumptions 513
Bench-Scale Data Analysis 516
Field Data Analysis 517
SUMMARY AND CONCLUSIONS 521
SECTION 13 - COMMERCIAL PRODUCTS TESTING 523
LABORATORY STUDIES 523
METHODS 524
Electrolytic Respirometry 524
Experimental Design 525
Shake Flasks 527
Sampling 529
Nutrient Analysis 529
Oil Chemistry 529
Microbiology 529
RESULTS 530
Respirometry 530
Nutrient Concentrations 530
-------
TABLE OF CONTENTS (Continued)
Statistical Analysis of Alkane Degradation Data 530
Total Alkane Reduction 533
Typical Chromatographic Profiles 533
Total PAH Reduction 536
Typical Mass Spectral Profiles 536
Microbiology 539
SUMMARY AND CONCLUSIONS 542
FIELD STUDIES 543
METHODS 544
Plot Description 544
Sampling 547
Nutrients 547
Nutrient Application 547
Schedule 548
Sediment Chemistry 548
Microbiology 548
Data Analysis 548
RESULTS 549
Persistence of Nutrients 549
Numbers of Oil-Degrading Microorganisms 549
Oil Residue Weight 554
Total Resolvable Alkanes 557
Total Resolvable Alkanes as a Percent of the Residue Weight 557
SUMMARY AND CONCLUSIONS 562
LITERATURE CITED 564
REFERENCES 566
XI
-------
LIST OF FIGURES
Figure l.la. Diagram of the Oil and Its Impact 3
Figure l.lb. Oil Impacted Areas in Prince William Sound 4
Figure 1.2. Exxon Cleaning a Beach Using Water Under High Pressure and
Temperature 5
Figure 1.3. Unfractionated Prudhoe Bay Crude Oil 7
Figure 1.4. Prudhoe Bay Crude Oil, Aliphatic Fraction 8
Figure 1.5. Prudhoe Bay Crude Oil, Aromatic Fraction 9
Figure 2.1. Diagram of Fertilizer Testing for Nutrient Release Characteristics 24
Figure 2.2. Cumulative Release of Ammonia, Total Kjeldahl Nitrogen (TKN), and
Total Phosphorus from WOODACE Briquettes in Static Flask
Experiments 25
Figure 2.3. Daily Nutrient Release Rate of Ammonia (NH4), Total Phosphorus (TP),
and Total Kjeldahl Nitrogen (TKN) from WOODACE Briquettes 25
Figure 2.4. . Ammonia Release From IBDU Briquettes at 9 and 21 Degrees Centigrade
in 3 Different Water Sources 27
Figure 2.5. Cumulative Release of Ammonia and Total Kjeldahl Nitrogen (TKN)
from IBDU Fertilizer Granules Contained in Bags in Static Flask
Experiments 28
Figure 2.6. Cumulative Release of Ammonia, Total Kjeldahl Nitrogen (TKN), and
Total Phosphorus from OSMOCOTE Briquettes in Static Flask
Experiments 30
Figure 2.7. Cumulative Release of Ammonia, Total Kjeldahl Nitrogen (TKN), and
Total Phosphorus from MAGAMP Briquettes in Static Flask
Experiments 30
Figure 2.8. Sampling Point Locations for Magnesium Ammonium Phosphate Fertilizer
Field Test 32
Figure 2.9. Magnesium Ammonium Phosphate Fertilizer Test Ammonium
Concentration in Beach Pore Water at 12, 24, and 96 Hours After
Placement of Fertilizer 33
Figure 2.10. Cumulative Release of Ammonia and Nitrate from SIERRA CHEMICAL
Granules in Static Flask Experiments 34
Figure 2.11. Cumulative Release of Ammonia and Total Kjeldahl Nitrogen (TKN)
from INIPOL EAP 22 in Static Flask Experiments 36
xn
-------
LIST OF FIGURES (Continued)
Figure 3.1.
Figure 3.2.
Figure 3.3.
Figure 3.4.
Figure 3.5.
Figure 3.6.
Figure 3.7.
Figure 3.8.
Figure 3.9.
Figure 3.10.
Figure 3.11.
Figure 3.12.
Figure 3.13.
Figure 3.14.
Figure 3.15.
Figure 3.16a.
Figure 3.16b.
Figure 3.17.
Figure 3.18.
Figure 3.19.
Location of Field Sites
Sampling Locations at Snug Harbor, Knight Island, in Prince William
Sound, Alaska
Sampling Locations at Passage Cove, Knight Island, in Prince William
Sound, Alaska
Sampling Design for Snug Harbor and Passage Cove
A 1m by 1m Frame Was Placed on the Beach in the Designated Grid
Cell and Samples Were Collected from the Center of the Frame
Sampling Basket
Disk Island Schematic of the Beach for the Fertilizer Specific Activity
Experiment
Disk Island Fertilizer Specific Activity Plot Map and Rate of
CUSTOMBLEN Granule Application
Scaling Experiment Plots on an Uncontaminated Cobble Beach
Elrington Island Beach Diagram ,
Elrington Island Nutrient Solution Experiment Beach Areas and Rate
of Nutrient Solution Application
Nutrient and Dissolved Oxygen Monitoring Basket
Timeline Depicting Dates of Fertilizer Application and Basket Removal
from Disk and Elrington Island
Bags Filled with WOODACE Briquettes at Snug Harbor
Timelines Depicting Dates of Fertilizer Application for Snug Harbor and
Passage Cove
Placement of the Bags of Fertilizer Briquettes on Otter and Seal
Beaches
Repositioning of the Bags of Fertilizer Briquettes on Otter and Seal
Beaches
CUSTOMBLEN Granules Adhered to Cobble at Passage Cove
Application of Oleophilic Fertilizer Using a Backpack Sprayer
Sprinkler System in Operation at Passage Cove
40
45
46
47
49
51
53
53
55
56
57
58
60
63
64
65
65
66
68
70
Xlll
-------
LIST OF FIGURES (Continued)
Figure 4.1. Nutrient Sample Filtration Apparatus 81
Figure 4.2. Oil Chemistry Sample Extraction 84
Figure 4.3. Biometer Flask - "High Tide" 97
Figure 4.4. Schematic Diagram of the Tank Microcosms 100
Figure 4.5. Schematic of Flow-Through Column Microcosm - Use of Pump 1 to
Fill Columns 102
Figure 4.6. Schematic of Flow-Through Column Microcosm - Use of Pump 2 to
Drain the Columns 103
Figure 4.7. Schematic of Flow-Through Column Microcosm - Use of Pump 3 to
Purge Air from the Columns 104
Figure 4.8. Potential Fate of Oil Carbon in Flow-Through Microcosms 106
Figure 5.1. Control Chart for nC18 for Disk Island and Elrington Island 120
Figure 5.2. Control Chart for the nC18/phytane Ratio for Disk Island and
Elrington Island 120
Figure 5.3. Control Chart of Nutrient Analysis Precision for Phosphate During the
Summer of 1990 122
Figure 5.4. Field Blank Analysis for nC 18 Over Time for Snug Harbor 122
Figure 5.5. Quality Control Check Sample Control Chart for the nC17/pristane
Ratio for the Summer of 1990 124
Figure 5.6. Quality Control Check Sample Control Chart for the nC18/phytane
Ratio for the Summer of 1990 124
Figure 6.1. Approximately 8 to 10 Days Following Application of INIPOL to the
Cobble Beach Plot at Snug Harbor, Reductions in the Amount of Surface
Oil (As Compared to Surrounding Untreated Oiled Areas) Was Evidenced
by a Clean Rectangle on the Beach Surface 135
Figure 6.2. At Ground Level, the Reduction in Oil on the INIPOL-Treated Plot was
also Strikingly Apparent 135
Figure 6.3a. Ammonia Concentrations in Interstitial Water Samples Prior to Fertilizer
Application (T-0) 138
Figure 6.3b. Ammonia Concentrations in Interstitial Water Samples 1 -2 Days Post
Fertilizer Application (T-l) 139
xiv
-------
LIST OF FIGURES (Continued)
Figure 6.3c. Ammonia Concentrations in Interstitial Water Samples 8-10 Days Post
Fertilizer Application (T-2) 140
Figure 6.3d. Ammonia Concentrations in Interstitial Water Samples 30 Days Post
Fertilizer Application (T-3) 141
Figure 6.3e. Ammonia Concentrations in Interstitial Water Samples 6 Weeks Post
Fertilizer Application (T-4) 142
Figure 6.4a. Nitrate/Nitrite Concentrations in Interstitial Water Samples 1 -2 Days
Post Fertilizer Application (T-l) 143
Figure 6.4b. Nitrate/Nitrite Concentrations in Interstitial Water Samples 8-10 Days
Post Fertilizer Application (T-2) 144
Figure 6.4c. Nitrate/Nitrite Concentrations in Interstitial Water Samples 30 Days
Post Fertilizer Application (T«3) 145
Figure 6.5. Change in Oil Residue Weight Through Time for Seal Beach (Untreated
Control) at Snug Harbor (Cobble Surface) 147
Figure 6.6. Change in Oil Residue Weight Through Time for Seal Beach
(WOODACE Briquettes) at Snug Harbor (Cobble Surface) 147
Figure 6.7. Change in Oil Residue Weight Through Time for Seal Beach (INIPOL)
at Snug Harbor (Cobble Surface) 148
Figure 6.8. Change in Oil Residue Weight Through Time for Seal Beach (Untreated
Control) at Snug Harbor (Mixed Sand and Gravel Under Cobble) 148
Figure 6.9. Change in Oil Residue Weight Through Time for Seal Beach
(WOODACE Briquettes) for Snug Harbor (Mixed Sand and Gravel Under
Cobble) 149
Figure 6.10. Change in Oil Residue Weight Through Time for Seal Beach (INIPOL)
at Snug Harbor (Mixed Sand and Gravel Under Cobble) 149
Figure 6.11. Change in the Median Residue Weight, Expressed as Percent of the 6/9/89
Median Over Time for the Briquette, INIPOL, and Untreated Control Beaches
at Snug Harbor (Cobble Surface) 152
Figure 6.12. Change in the Median Residue Weight, Expressed as Percent of the 6/9/89
Median Over Time for the Briquette, INIPOL, and Untreated Control Beaches
at Snug Harbor (Mixed Sand and Gravel Under Cobble) 155
Figure 6.13. Change in Oil Residue Weight Through Time for Otter Beach (INIPOL)
at Snug Harbor (Mixed Sand and Gravel) 157
xv
-------
LIST OF FIGURES (Continued)
Figure 6.14. Change in Oil Residue Weight Through Time for Eagle Beach
(Untreated Control) at Snug Harbor (Mixed Sand and Gravel) 157
Figure 6.15. Change in Oil Residue Weight Through Time for Otter Beach
(WOODACE Briquettes) at Snug Harbor (Mixed Sand and Gravel) 158
Figure 6.16. Change in the Median Residue Weight, Expressed as Percent of the 6/9/89
Median Over Time for the Briquette, INIPOL, and Untreated Control Beaches
at Snug Harbor (Mixed Sand and Gravel Only) 159
Figure 6.17. Change in nC18 Concentration Through Time for Seal Beach
(Untreated Control) at Snug Harbor (Cobble Surface) 163
Figure 6.18. Change in nC22 Concentration Through Time for Seal Beach (Untreated
Control) at Snug Harbor (Cobble Surface) 163
Figure 6.19. Change in nC27 Concentration Through Time for Seal Beach
(Untreated Control) at Snug Harbor (Cobble Surface) 164
Figure 6.20. Change in Sum of Alkane Concentration nC18 to nC27 Through Time for
Seal Beach (Untreated Control) at Snug Harbor (Cobble Surface) 164
Figure 6.21. Change in Pristane Concentration Through Time for Seal Beach
(Untreated Control) at Snug Harbor (Cobble Surface) 165
Figure 6.22. Change in Phytane Concentration Through Time for Seal Beach
(Untreated Control) at Snug Harbor (Cobble Surface) 165
Figure 6.23. Change in nC18 Concentration Through Time for Seal Beach (WOODACE
Briquettes) at Snug Harbor (Cobble Surface) 166
Figure 6.24. Change in nC22 Concentration Through Time for Seal Beach (WOODACE
Briquettes) at Snug Harbor (Cobble Surface) 166
Figure 6.25. Change in nC27 Concentration Through Time for Seal Beach (WOODACE
Briquettes) at Snug Harbor (Cobble Surface) 167
Figure 6.26. Change in Sum of Alkane Concentration nC18 to nC27 Through Time for
Seal Beach (WOODACE Briquettes) at Snug Harbor (Cobble Surface) .. 167
Figure 6.27. Change in Pristane Concentration Through Time for Seal Beach
(WOODACE Briquettes) at Snug Harbor (Cobble Surface) 168
Figure 6.28. Change in Phytane Concentration Through Time for Seal Beach
(WOODACE Briquettes) at Snug Harbor (Cobble Surface) 168
Figure 6.29. Change in nC18 Concentration Through Time for Seal Beach (INIPOL)
at Snug Harbor (Cobble Surface) , 169
xvi
-------
LIST OF FIGURES (Continued)
Figure 6.30. Change in nC22 Concentration Through Time for Seal Beach (INIPOL)
at Snug Harbor (Cobble Surface) 169
Figure 6.31. Change in nC27 Concentration Through Time for Seal Beach (INIPOL)
at Snug Harbor (Cobble Surface) 170
Figure 6.32. Change in Sum of Alkane Concentration nC18 to nC27 Through Time for
Seal Beach (INIPOL) at Snug Harbor (Cobble Surface) 170
Figure 6.33. Change in Pristane Concentration Through Time for Seal Beach (INIPOL)
at Snug Harbor (Cobble Surface) 171
Figure 6.34. Change in Phytane Concentration Through Time for Seal Beach (INIPOL)
at Snug Harbor (Cobble Surface) 171
Figure 6.35. Change in nC18/phytane Ratio Through Time for Seal Beach
(Untreated Control) at Snug Harbor (Cobble Surface) 172
Figure 6.36. Change in nCl8/phytane Ratio Through Time for Seal Beach
(WOODACE Briquettes) at Snug Harbor (Cobble Surface) 172
Figure 6.37. Change in nC18/phytane Ratio Through Time for Seal Beach (INIPOL)
at Snug Harbor (Cobble Surface) 173
Figure 6.38. Change in the Median Residue Weight for Several Hydrocarbons
Expressed as Percent of the 6/9/89 Median Over Time for the
Briquette, INIPOL, and Untreated Control Beaches at Snug Harbor
(Cobble Surface) 174
Figure 6.39. Change in nC18 Concentration Through Time for Seal Beach (Untreated
Control) at Snug Harbor (Mixed Sand and Gravel Under Cobble) 184
Figure 6.40. Change in nC22 Concentration Through Time for Seal Beach (Untreated
Control) at Snug Harbor (Mixed Sand and Gravel Under Cobble) 184
Figure 6.41. Change in nC27 Concentration Through Time for Seal Beach (Untreated
Control) at Snug Harbor (Mixed Sand and Gravel Under Cobble) 185
Figure 6.42. Change in the Sum of Alkane Concentration nC18 to nC27 Through Time
for Seal Beach (Untreated Control) at Snug Harbor (Mixed Sand and
Gravel Under Cobble) 185
Figure 6.43. Change in Pristane Concentration Through Time for Seal Beach
(Untreated Control) at Snug Harbor (Mixed Sand and Gravel Under
Cobble) 186
Figure 6.44. Change in Phytane Concentration Through Time for Seal Beach
(Untreated Control) at Snug Harbor (Mixed Sand and Gravel U-Jcr
Cobble) 186
xvii
-------
LIST OF FIGURES (Continued)
Change in nC18 Concentration Through Time for Seal Beach (WOODACE
Briquettes) at Snug Harbor (Mixed Sand and Gravel Under Cobble) ... 187
Change in nC22 Concentration Through Time for Seal Beach (WOODACE
Briquettes) at Snug Harbor (Mixed Sand and Gravel Under Cobble) ... 187
Change in nC27 Concentration Through Time for Seal Beach (WOODACE
Briquettes) at Snug Harbor (Mixed Sand and Gravel Under Cobble) .... 188
Figure 6.45.
Figure 6.46.
Figure 6.47.
Figure 6.48. Change in the Sum of Alkane Concentration nC18 to nC27 Through Time
for Seal Beach (WOODACE Briquettes) at Snug Harbor (Mixed Sand and
Gravel Under Cobble) 188
Figure 6.49. Change in Pristane Concentration Through Time for Seal Beach
(WOODACE Briquettes) at Snug Harbor (Mixed Sand and Gravel
Under Cobble) 189
Figure 6.50. Change in Phytane Concentration Through Time for Seal Beach (WOODACE
Briquettes) at Snug Harbor (Mixed Sand and Gravel Under Cobble) .... 189
Figure 6.S1. Change in nC18 Concentration Through Time for Seal Beach (INIPOL)
at Snug Harbor (Mixed Sand and Gravel Under Cobble) 190
Figure 6.52. Change in nC22 Concentration Through Time for Seal Beach (INIPOL)
at Snug Harbor (Mixed Sand and Gravel Under Cobble) 190
Figure 6.53. Change in nC27 Concentration Through Time for Seal Beach (INIPOL)
at Snug Harbor (Mixed Sand and Gravel Under Cobble) 191
Figure 6.54. Change in the Sum of Alkane Concentration nC18 to nC27 Through
Time for Seal Beach (INIPOL) at Snug Harbor (Mixed Sand and Gravel
Under Cobble) 191
Figure 6.55. Change in Pristane Concentration Through Time for Seal Beach (INIPOL)
at Snug Harbor (Mixed Sand and Gravel Under Cobble) 192
Figure 6.56. Change in Phytane Concentration Through Time for Seal Beach (INIPOL)
at Snug Harbor (Mixed Sand and Gravel Under Cobble) 192
Figure 6.57. Change in nC18/Phytane Ratio Through Time for Seal Beach (Untreated
Control) at Snug Harbor (Mixed Sand and Gravel Under Cobble) 193
Figure 6.58. Change in nC18/Phytane Ratio Through Time for Seal Beach (WOODACE
Briquettes) at Snug Harbor (Mixed Sand and Gravel Under Cobble) .... 193
Figure 6.59. Change in nC18/Phytane Ratio Through Time for Seal Beach (INIPOL)
at Snug Harbor (Mixed Sand and Gravel Under Cobble) 194
XVlll
-------
LIST OF FIGURES (Continued)
Figure 6.60. Change in the Median Residue Weight for Several Hydrocarbons
Expressed as Percent of the 6/9/89 Median Over Time for the
Briquette, INIPOL, and Untreated Control Beaches at Snug Harbor (Mixed
Sand and Gravel Under Cobble) 196
Figure 6.61. Change in nC18 Concentration Through Time for Eagle Beach (Untreated
Control) at Snug Harbor (Mixed Sand and Gravel) 201
Figure 6.62. Change in nC22 Concentration Through Time for Eagle Beach (Untreated
Control) at Snug Harbor (Mixed Sand and Gravel) 201
Figure 6.63. Change in nC27 Concentration Through Time for Eagle Beach (Untreated
Control) at Snug Harbor (Mixed Sand and Gravel) 202
Figure 6.64. Change in Sum of the Alkane Concentration nC18 to nC27 Through Time
for Eagle Beach (Untreated Control) at Snug Harbor (Mixed Sand and
Gravel) 202
Figure 6.65. Change in Pristane Concentration Through Time for Eagle Beach
(Untreated Control) at Snug Harbor (Mixed Sand and Gravel) 203
Figure 6.66. Change in Phytane Concentration Through Time for Eagle Beach
(Untreated Control) at Snug Harbor (Mixed Sand and Gravel) 203
Figure 6.67. Change in nC18 Concentration Through Time for Otter Beach (WOODACE
Briquettes) at Snug Harbor (Mixed Sand and Gravel) 204
Figure 6.68. Change in nC22 Concentration Through Time for Otter Beach (WOODACE
Briquettes) at Snug Harbor (Mixed Sand and Gravel) 204
Figure 6.69. Change in nC27 Concentration Through Time for Otter Beach (WOODACE
Briquettes) at Snug Harbor (Mixed Sand and Gravel) 205
Figure 6.70. Change in Sum of Alkane Concentration nC18 to nC27 Through Time
for Otter Beach (WOODACE Briquettes) at Snug Harbor (Mixed Sand
and Gravel) 205
Figure 6.71. Change in Pristane Concentration Through Time for Otter Beach (WOODACE
Briquettes) at Snug Harbor (Mixed Sand and Gravel) 206
Figure 6.72. Change in Phytane Concentration Through Time for Otter Beach (WOODACE
Briquettes) at Snug Harbor (Mixed Sand and Gravel) 206
Figure 6.73. Change in nC18 Concentration Through Time for Otter Beach (INIPOL)
at Snug Harbor (Mixed Sand and Gravel) 207
Figure 6.74. Change in nC22 Concentration Through Time for Otter Beach (INIPOL)
at Snug Harbor (Mixed Sand and Gravel) 207
xix
-------
LIST OF FIGURES (Continued)
Figure 6.75. Change in nC27 Concentration Through Time for Otter Beach (INIPOL)
at Snug Harbor (Mixed Sand and Gravel) 208
Figure 6.76. Change in Sum of the Alkane Concentration nC18 to nC27 Concentration
Through Time for Otter Beach (INIPOL) at Snug Harbor (Mixed Sand and
Gravel) 208
Figure 6.77. Change in Pristane Concentration Through Time for Otter Beach
(INIPOL) at Snug Harbor (Mixed Sand and Gravel) 209
Figure 6.78. Change in Phytane Concentration Through Time for Otter Beach
(INIPOL) at Snug Harbor (Mixed Sand and Gravel) 209
Figure 6.79. Change in nC18/phytane Ratio Through Time for Eagle Beach (Untreated
Control) at Snug Harbor (Mixed Sand and Gravel) 210
Figure 6.80. Change in nC18/Phytane Ratio Through Time for Otter Beach (WOODACE
Briquettes) at Snug Harbor (Mixed Sand and Gravel) 210
Figure 6.81. Change in nC18/phytane Ratio Through Time for Otter Beach (INIPOL)
at Snug Harbor (Mixed Sand and Gravel) 211
Figure 6.82. Change in the Median Residue Weight for Several Hydrocarbons
Expressed as Percent of the 6/9/89 Median Over Time for the
Briquette, INIPOL, and Untreated Control Beaches at Snug Harbor
(Mixed Sand and Gravel Only) 213
Figure 6.83. nC17/pristane Ratio versus Log10 Residue Weight Two Weeks Before
Fertilizer Application (5/28/89) 217
Figure 6.84. nC18/phytane Ratio versus Log10 Residue Weight Two Weeks Before
Fertilizer Application (5/28/89) 218
Figure 6.85. nC17/pristane Ratio Versus Log10 Residue Weight at Time Zero of
Fertilizer Application (6/8/89) 219
Figure 6.86. nC18/phytane Ratio Versus Log10 Residue Weight at Time Zero of
Fertilizer Application (6/8/89) 220
Figure 6.87. Phytoplankton Chlorophyll Data (mg Chlorophyll a/L) from Water
Samples Collected Along Cobble and Gravel Shorelines at Snug Harbor
Following June 7 and 8, 1989, Fertilizer Additions to Gravel Shorelines . 230
Figure 6.88. Primary Productivity Estimates (From 14C Uptake; mg C/ms/hour) For
Phytoplankton Samples From Snug Harbor at Various Sample Dates
Following the June 7 and 8, 1989, Fertilizer Additions Along Cobble
and Gravel Shorelines 232
xx
-------
LIST OF FIGURES (Continued)
Figure 6.89. Abundance of Bacteria (cells x 109/L) From Water Samples Taken Along
Cobble and Gravel Shorelines on Various Sample Dates Following the
June 7 and 8, 1989, Fertilizer Additions to Snug Harbor
Shorelines 234
Figure 6.90. Bacterial Productivity, as Measured by Tritiated Thymidine Uptake
(mM Thymidine/L/day), for Bacterial Samples Collected on Various
Sample Dates Adjacent to Cobble and Gravel Shorelines at
Snug Harbor 235
Figure 6.91. Effect of INIPOL on the Relative Numbers of Oleic Acid-Degrading
Microorganisms in Jars Containing Oiled Rocks and Artificial Seawater,
Seawater, or Saline Solution 243
Figure 6.92. Effect of INIPOL on the Relative Numbers of Oil-Degrading
Microorganisms in Jars Containing Oiled Rocks and Artifical Seawater,
Seawater, or Saline Solution 244
Figure 6.93. Mutagenicity of Soil Extracts Using the Spiral Salmonella typhimurium
Assay with Strain TA98 with an Aroclor 1254-Induced CD-I Rat Liver
Homogenate Exogenous Activation System 246
Figure 7.1. Kittiwake Beach at Passage Cove, Treated with Fertilizer Solution from
the Sprinkler System, Showed Extensive Disappearance of Oil Compared to
Untreated Control Plots 252
Figure 7.2. Oil Untreated Control Plots 252
Figure 7.3. Change in Oil Residue Weight Through Time for Raven Beach
(Untreated Control) at Passage Cove (Cobble Surface) 254
Figure 7.4. Change in Oil Residue Weight Through Time for Tern Beach (INIPOL +
CUSTOMBLEN) at Passage Cove (Cobble Surface) 254
Figure 7.5. Change in Oil Residue Weight Through Time for Kittiwake Beach
(Fertilizer Solution) at Passage Cove (Cobble Surface) 255
Figure 7.6. Change in the Median Residue Weight, Expressed as Percent of the
7/22 Median Over Time for Kittiwake, Raven, and Tern Beaches at
Passage Cove 255
Figure 7.7. Change in Oil Residue Weight Through Time for Raven Beach
(Untreated Control) at Passage Cove (Mixed Sand and Gravel) 261
Figure 7.8. Change in Oil Residue Weight Through Time for Tern Beach (INIPOL +
CUSTOMBLEN) at Passage Cove (Mixed Sand and Gravel) 261
Figure 7.9. Change in Oil Residue Weight Through Time for Kittiwake Beach
(Fertilizer Solution) at Passage Cove (Mixed Sand and Gravel) 262
xxi
-------
LIST OF FIGURES (Continued)
Figure 7.10. Change in nC18 Alkane Concentration Through Time for Raven Beach
(Untreated Control) at Passage Cove (Cobble Surface) 264
Figure 7.11. Change in the Sum of Alkane Concentration nC18 to nC27 Through Time
for Raven Beach (Untreated Control) at Passage Cove (Cobble Surface) . 265
Figure 7.12. Change in Phytane Concentration Through Time for Raven Beach
(Untreated Control) at Passage Cove (Cobble Surface) 265
Figure 7.13. Change in nC18 Alkane Concentration Through Time for Tern Beach
(INIPOL + CUSTOMBLEN) at Passage Cove (Cobble Surface) 266
Figure 7.14. Change in the Sum of Alkane Concentration nC18 to nC27 Through
Time for Tern Beach (INIPOL + CUSTOMBLEN) at Passage Cove
(Cobble Surface) 266
Figure 7.15. Change in Phytane Concentration Through Time for Tern Beach
(INIPOL + CUSTOMBLEN) at Passage Cove (Cobble Surface) 267
Figure 7.16. Change in nC18 Alkane Concentration Through Time for Kittiwake
Beach (Fertilizer Solution) at Passage Cove (Cobble Surface) 267
Figure 7.17. Change in the Sum of Alkane Concentration nC18 to nC27 Through Time
for Kittiwake Beach (Fertilizer Solution) at Passage Cove (Cobble
Surface) 268
Figure 7.18. Change in Phytane Concentration Through Time for Kittiwake Beach
(Fertilizer Solution) at Passage Cove (Cobble Surface) 268
Figure 7.19. Change in Composition, Expressed as Percent of the Median
Concentration of Individual Hydrocarbons on the 7/22/89 Sampling
for Raven, Tern, and Kittiwake Beaches at Passage Cove 271
Figure 7.20 Hydrocarbon Composition on August 6, 1989, Expressed as Percent of
the Median Concentration of Individual Hydrocarbons on the 7/22/89
Sampling for Raven, Tern, and Kittiwake Beaches at Passage Cove .... 273
Figure 7.21. Hydrocarbon Composition on August 20, 1989, Expressed as Percent of
the Median Concentration of Individual Hydrocarbons on the 7/22/89
Sampling for Raven, Tern, and Kittiwake Beaches at Passage Cove .... 273
Figure 7.22. Change in the nC18/phytane Ratio Through Time for Raven Beach
(Untreated Control) at Passage Cove (Cobble Surface) 274
Figure 7.23. Change in the nC18/phytane Ratio Through Time for Tern Beach
(INIPOL + CUSTOMBLEN) at Passage Cove (Cobble Surface) 274
Figure 7.24. Change in the nC18/phytane Ratio Through Time for Kittiwake Beach
(Fertilizer Solution) at Passage Cove (Cobble Surface) 275
xxii
-------
LIST OF FIGURES (Continued)
Figure 7.25. Relationship Between the nC18/phytane Ratio and Oil Residue Weight
for All Beaches at Passage Cove (Cobble Surface) at Time Zero 277
Figure 7.26. Relationship Between the nC18/phytane Ratio and Oil Residue Weight
for All Beaches at Passage Cove (Cobble Surface) at Time Zero +
Time One Sampling 278
Figure 7.27. Change in Phytane, Expressed as Percent of the 7/22/89 Median Over Time
for Kittiwake, Raven, and Tern Beaches at Passage Cove 280
Figure 7.28. Relationship Between Oil Residue Weight and the Sum of Alkane
Concentration nC18 to nC27 for Raven Beach (Untreated Control)
at Passage Cove (Cobble Surface) 281
Figure 7.29. Relationship Between Oil Residue Weight and the Sum of Alkane
Concentration nC18 to nC27 for Tern Beach (INIPOL + CUSTOMBLEN)
at Passage Cove (Cobble Surface) 282
Figure 7.30. Relationship Between Oil Residue Weight and the Sum of Alkane
Concentration nC18 to nC27 for Kittiwake Beach (Fertilizer Solution)
at Passage Cove (Cobble Surface) 283
Figure 7.31. Percent Remaining (Hopane Normalized) of Different Hydrocarbons at
Kittiwake Beach at Passage Cove on 8/20/89 284
Figure 7.32a. Aromatic Homologs Relative to Hopane (Phenanthracene/Anthracene
Group) 287
Figure 7.32b. Aromatic Homologs Relative to Hopane (Dibenzothiophene Group) 287
Figure 7.33. Change in the Median Concentration of Summed Alkanes nC18 to nC27
(Arithmetic Scale) Over Time, for Kittiwake, Raven, and Tern
Beaches at Passage Cove 288
Figure 7.34. Change in the Median Concentration of Summed Alkanes nC18 to nC27
(Log Scale) Over Time, for Kittiwake, Raven, and Tern Beaches at
Passage Cove 288
;
Figure 7.35. Change in nC18 Alkane Concentration Through Time for Raven Beach
(Untreated Control) at Passage Cove (Mixed Sand and Gravel) 291
Figure 7.36. Change in the Sum of Alkane Concentration nC18 to nC27 Through Time
for Raven Beach (Untreated Control) at Passage Cove (Mixed Sand and
Gravel) 291
Figure 7.37. Change in Phytane Concentration Through Time for Raven Beach
(Untreated Control) at Passage Cove (Mixed Sand and Gravel) 292
XXlll
-------
LIST OF FIGURES (Continued)
Figure 7.38. Change in nC18 Alkane Concentration Through Time for Tern Beach
(INIPOL + CUSTOMBLEN) at Passage Cove (Mixed Sand and Gravel) .. 292
Figure 7.39. Change in the Sum of Alkane Concentration nC18 to nC27 Through Time
for Tern Beach (INIPOL + CUSTOMBLEN) at Passage Cove (Mixed Sand
and Gravel) : 293
Figure 7.40. Change in Phytane Concentration Through Time for Tern Beach
(INIPOL + CUSTOMBLEN) at Passage Cove (Mixed Sand and Gravel) .. 293
Figure 7.41. Change in nC18 Alkane Concentration Through Time for Kittiwake
Beach (Fertilizer Solution) at Passage Cove (Mixed Sand and Gravel) ... 294
Figure 7.42. Change in the Sum of Alkane Concentration nC18 to nC27 Through Time
for Kittiwake Beach (Fertilizer Solution) at Passage Cove (Mixed Sand
and Gravel) 294
Figure 7.43. Change in Phytane Concentration Through Time for Kittiwake Beach
(Fertilizer Solution) at Passage Cove (Mixed Sand and Gravel) 295
Figure 7.44. Change in the nC18/phytane Ratio Through Time for Raven Beach
(Untreated Control) at Passage Cove (Mixed Sand and Gravel) 295
Figure 7.45. Change in the nC18/phytane Ratio Over Time for Tern Beach
(INIPOL + CUSTOMBLEN) at Passage Cove (Mixed Sand and Gravel) .. 296
Figure 7.46. Change in the nC18/phytane Ratio Over Time for Kittiwake Beach
(Fertilizer Solution) at Passage Cove (Mixed Sand and Gravel) 296
Figure 7.47. Relationship Between the Sum of Alkane Concentration nC18 to nC27
and Oil Residue Weight for Raven Beach (Untreated Control) at
Passage Cove (Mixed Sand and Gravel) 297
Figure 7.48. Relationship Between the Sum of Alkane Concentration nC18 to nC27
and Oil Residue Weight for Tern Beach (INIPOL + CUSTOMBLEN) at
Passage Cove (Mixed Sand and Gravel) 298
Figure 7.49. Relationship Between the Sum of Alkane Concentration nC18 to nC27
and Oil Residue Weight for Kittiwake Beach (Fertilizer Solution) at
Passage Cove (Mixed Sand and Gravel) 299
Figure 7.50. Oil Residue Weights (Log Scale) for the Low Tide Zone at Passage Cove
During the Winter of 1989 302
Figure 7.51. nC18 Concentration (Log Scale) for the Low Tide Zone at Passage Cove
During the Winter of 1989 302
Figure 7.52. nC27 Concentration (Log Scale) for the Low Tide Zone at Passage Cove
During the Winter of 1989 303
xxiv
-------
LIST OF FIGURES (Continued)
Figure 7.53. Sum of Alkane Concentration nC18 to nC27 (Log Scale) for the Low Tide
Zone at Passage Cove During the Winter of 1989 303
Figure 7.54. Phytane Concentration (Log Scale) for the Low Tide Zone at Passage
Cove During the Winter of 1989 304
Figure 7.55. nC18/phytane Ratio (Log Scale) for the Low Tide Zone at Passage Cove
During the Winter of 1989 304
Figure 7.56. Oil Residue Weights (Log Scale) for the High Tide Zone at Passage Cove
During the Winter of 1989 305
Figure 7.57. nC18 Concentration (Log Scale) for the High Tide Zone at Passage Cove
During the Winter of 1989 305
Figure 7.58. nC27 Concentration (Log Scale) for the High Tide Zone at Passage Cove
During the Winter of 1989 306
Figure 7.59. Sum of Alkane Concentration nC18 to nC27 (Log Scale) for the High
Tide Zone at Passage Cove During the Winter of 1989 306
Figure 7.60. Phytane Concentration (Log Scale) for the High Tide Zone at Passage
Cove During the Winter of 1989 307
Figure 7.61. nC18/phytane Ratio (Log Scale) for the High Tide Zone at Passage Cove
During the Winter of 1989 307
Figure 7.62. Oil Residue Weight in the Subsurface for the Three Treatments in the
Spring of 1990 310
Figure 7.63. Mineralization of Radiolabeled Phenanthrene (»,O), Naphthalene (D,l),
and Hexadecane (A,*), in Samples From Oiled (Open Symbols) and
Unoiled (Closed Symbols) Beaches 314
Figure 7.64. Mineralization of Phenanthrene in Samples From (A) Water-Soluble
Fertilizer-Treated; (B) Untreated Control; and (C) Oleophilic Beaches
at Passage Cove 316
Figure 7.65. Mineralization of Phenanthrene in Oleophilic-Treated Microcosm with
Oiled Layer Over Clean Layer, Showing Lack of Bacterial Activity in
Upper Layer 320
Figure 7.66. Mean Chlorophyll a/L Measurements (+ SD) from 4 Replicate Plankton
Samples Taken at Passage Cove Study Site Following July 25, 1989,
Fertilizer Applications to Shorelines 322
xxv
-------
LIST OF FIGURES (Continued)
Figure 7.67. Mean Primary Productivity Measurements (+ SD) From 14C-Uptake
(mg C/ms/hour) From 4 Replicate Plankton Samples Taken at Passage
Cove Study Site Following July 25, 1989, Fertilizer Applications
to Shorelines 324
Figure 7.68. Abundance of Bacteria (cells x!0°/L) From Water Samples Collected at
the Passage Cove Study Site. Fertilizer was Applied on
July 25, 1989 326
Figure 7.69. Bacterial Productivity Measurements From Tritiated Thymidine Uptake
(mM Thymidine/L/day) in Water Samples Collected at Various Sites in
Passage Cove Following Nutrient Application to Shorelines on
July 25, 1989 328
Figure 8.1. Ammonia Concentrations in Wells in Each Plot on the Incoming Tide for the
Scaling Experiment on 6/19/90 , 337
Figure 8.2. Ammonia Concentrations in Wells in Each Plot Using Root Feeders on the
Incoming Tide for the Scaling Experiment on 6/19/90 337
Figure 8.3. Ammonia Concentrations in Wells in Each Plot on the Outgoing Tide for the
Scaling Experiment on 6/19/90 338
Figure 8.4. Ammonia Concentrations in Wells in Each Plot on the Incoming Tide for the
Scaling Experiment on 6/20/90 338
Figure 8.5. Ammonia Concentrations in Wells in Each Plot on the Outgoing Tide for the
Scaling Experiment on 6/20/90 339
Figure 8.6. Ammonia Concentrations in Wells in Each Plot on the Incoming Tide for the
Scaling Experiment on 6/21/90 339
Figure 8.7. Ammonia Concentrations in Wells in Each Plot on the Outgoing Tide for the
Scaling Experiment on 6/21/90 340
Figure 8.8. Phosphate Concentrations in Wells in Each Plot on the Incoming Tide for the
Scaling Experiment on 6/19/90 340
Figure 8.9. Phosphate Concentrations in Wells in Each Plot Using Root Feeders on the
Incoming Tide for the Scaling Experiment on 6/19/90 341
Figure 8.10. Phosphate Concentrations in Wells in Each Plot on the Outgoing Tide for the
Scaling Experiment on 6/19/90 341
Figure 8.11. Phosphate Concentrations in Wells in Each Plot on the Incoming Tide for the
Scaling Experiment on 6/20/90 342
Figure 8.12. Phosphate Concentrations in Wells in Each Plot on the Outgoing Tide for the
Scaling Experiment on 6/20/90 342
xxvi
-------
LIST OF FIGURES (Continued)
Figure 8.13. Phosphate Concentrations in Wells in Each Plot on the Incoming Tide for the
Scaling Experiment on 6/21/90 343
Figure 8.14. Phosphate Concentrations in Wells in Each Plot on the Outgoing Tide for the
Scaling Experiment on 6/21/90 343
Figure 8.15. Change in Ammonia Concentration Over Time for the Incoming Tide
for All plots for the Disk Island Fertilizer Application Rate Study 345
Figure 8.16. Change in Ammonia Concentration Over Time for the Outgoing Tide
for All plots for the Disk Island Fertilizer Application Rate Study 345
Figure 8.17. Change in Phosphate Concentration Over Time for the Incoming Tide
for All plots for the Disk Island Fertilizer Application Rate Study 346
Figure 8.18. Change in Phosphate Concentration Over Time for the Outgoing Tide
for All plots for the Disk Island Fertilizer Application Rate Study 346
Figure 8.19. Change in Nitrate Concentration Over Time for the Incoming Tide for
All plots for the Disk Island Fertilizer Application Rate Study 347
Figure 8.20. Change in Oil Residue Weight Over Time for Untreated Control Plot #1
for the Disk Island Application Rate Study 349
Figure 8.21. Change in Oil Residue Weight Over Time for Untreated Control Plot #2
for the Disk Island Application Rate Study 350
Figure 8.22. Change in Oil Residue Weight Over Time for the 50 g/m2 Fertilizer
Application for the Disk Island Application Rate Study 351
Figure 8.23. Change in Oil Residue Weight Over Time for the 100 g/m2 Fertilizer
Application for the Disk Island Application Rate Study 351
Figure 8.24. Change in Oil Residue Weight Over Time for the 500 g/m3 Fertilizer
Application for the Disk Island Application Rate Study 352
Figure 8.25. Change in Oil Residue Weight Over Time for the 1,000 g/m2 Fertilizer
Application for the Disk Island Application Rate Study 352
Figure 8.26. Change in nC18 Concentration Over Time for Untreated Control Plot #1
for the Disk Island Application Rate Study 353
Figure 8.27. Change in nC18 Concentration Over Time for Untreated Control Plot #2
for the Disk Island Application Rate Study 353
Figure 8.28. Change in nC18 Concentration Over Time for the 50 g/m2 Fertilizer
Application for the Disk Island Application Rate Study 354
xxvn
-------
LIST OF FIGURES (Continued)
Figure 8.29. Change in nC18 Concentration Over Time for the 100 g/m2 Fertilizer
Application for the Disk Island Application Rate Study 354
Figure 8.30. Change in nC18 Concentration Over Time for the 500 g/m2 Fertilizer
Application for the Disk Island Application Rate Study 355
Figure 8.31. Change in nC18 Concentration Over Time for the 1,000 g/m3 Fertilizer
Application for the Disk Island Application Rate Study 355
Figure 8.32. Change in nC27 Concentration Over Time for Untreated Control Plot #1
for the Disk Island Application Rate Study 356
Figure 8.33. Change in nC27 Concentration Over Time for Untreated Control Plot #2
for the Disk Island Application Rate Study 356
Figure 8.34. Change in nC27 Concentration Over Time for the 50 g/m2 Fertilizer
Application for the Disk Island Application Rate Study 357
Figure 8.35. Change in nC27 Concentration Over Time for the 100 g/m2 Fertilizer
Application for the Disk Island Application Rate Study 357
Figure 8.36. Change in nC27 Concentration Over Time for the 500 g/m2 Fertilizer
Application for the Disk Island Application Rate Study 358
Figure 8.37. Change in nC27 Concentration Over Time for the 1,000 g/m2 Fertilizer
Application for the Disk Island Application Rate Study 358
Figure 8.38. Change in the Sum of the Alkane Concentration nC!8 to nC27 Over
Time for Untreated Control Plot #1 for the Disk Island Application
Rate Study 359
Figure 8.39. Change in the Sum of the Alkane Concentration nC18 to nC27 Over
Time for Untreated Control Plot #2 for the Disk Island Application
Rate Study 359
Figure 8.40. Change in the Sum of the Alkane Concentration nC18 to nC27 Over
Time for the 50 g/m2 Fertilizer Application for the Disk Island
Application Rate Study 360
Figure 8.41. Change in the Sum of the Alkane Concentration nC18 to nC27 Over
Time for the 100 g/m2 Fertilizer Application for the Disk Island
Application Rate Study 360
Figure 8.42. Change in the Sum of the Alkane Concentration nC18 to nC27 Over
Time for the 500 g/m2 Fertilizer Application for the Disk Island
Application Rate Study 361
XXVlll
-------
LIST OF FIGURES (Continued)
Figure 8.43. Change in the Sum of the Alkane Concentration nC18 to nC27 Over
Time for the 1,000 g/m2 Fertilizer Application for the Disk
Island Application Rate Study 361
Figure 8.44. Change in Phytane Concentration Over Time for Untreated Control
Plot #1 for the Disk Island Application Rate Study 362
Figure 8.45. Change in Phytane Concentration Over Time for Untreated Control
Plot #2 for Untreated Disk Island Application Rate Study 362
Figure 8.46. Change in Phytane Concentration Over Time for the 50 g/m2
Fertilizer Application for the Disk Island Application Rate Study 363
Figure 8.47. Change in Phytane Concentration Over Time for the 100 g/m3
Fertilizer Application for the Disk Island Application Rate Study 363
Figure 8.48. Change in Phytane Concentration Over Time for the 500 g/m2
Fertilizer Application for the Disk Island Application Rate Study 364
Figure 8.49. Change in Phytane Concentration Over Time for the 1,000 g/m2
Fertilizer Application for the Disk Island Application Rate Study 364
Figure 8.50. Change in the nC18/phytane Ratio Over Time for Untreated Control
Plot #1 for the Disk Island Application Rate Study 365
Figure 8.51. Change in the nC18/phytane Ratio Over Time for Untreated Control
Plot #2 for the Disk Island Application Rate Study 365
Figure 8.52. Change in the nC18/phytane Ratio Over Time for the 50 g/m2
Fertilizer Application for the Disk Island Application Rate Study 366
Figure 8.53. Change in the nC18/phytane Ratio Over Time for the 100 g/m2
Fertilizer Application for the Disk Island Application Rate Study 366
Figure 8.54. Change in the nC18/phytane Ratio Over Time for the 500 g/m2
Fertilizer Application for the Disk Island Application Rate Study 367
Figure 8.55. Change in the nC18/phytane Ratio Over Time for the 1,000 g/m2 Fertilizer
Application for the Disk Island Application Rate Study 367
Figure 8.56. Plot of Rate of CO, Production Versus Rate of Fertilizer
Concentration 370
Figure 8.57. Radiolabeled Hexadecane Mineralization Over Time (Standard Deviation)
for Both Untreated Control Plots at Disk Island 372
Figure 8.58. Radiolabeled Hexadecane Mineralization Over Time (Standard Deviation)
for the Four Treated Plots at Disk Island 373
xxix
-------
LIST OF FIGURES (Continued)
Figure 8.59. Summary of Radiolabeled Hexadecane Mineralization Over Time for All
Plots at Disk Island 374
Figure 8.60. Measurements of Total Heterotrophic and Oil-Degrading Bacterial
Populations Over Time at All Plots at Disk Island 375
Figure 9.1. Change in Ammonia Concentration Through Time for the Incoming Tide at the
Untreated Control Beach and Sprinkler Beach at Elrington Island 380
Figure 9.2. Change in Nitrate Concentration Through Time for the Incoming Tide at the
Untreated Control Beach and Sprinkler Beach at Elrington Island 380
Figure 9.3. Change in Phosphate Concentration Through Time for the Incoming Tide at the
Untreated Control Beach and Sprinkler Beach at Elrington Island 381
Figure 9.4a. Change in Oil Residue Weight Through Time for the Sprinkler Beach Oiled
Subsurface Layer at Elrington Island 382
Figure 9.4b. Change in Oil Residue Weight Through Time for the Untreated Control Beach
Oiled Subsurface Layer at Elrington Island 382
Figure 9.4c. Change in Oil Residue Weight Through Time for the Bath Beach Oiled
Subsurface Layer at Elrington Island 383
Figure 9.5. Change in Oil Residue Weight Through Time for the Bottom Unoiled Layer for
All Beaches at Elrington Island 386
Figure 9.6. Change in Oil Residue Weight Through Time for the Top Unoiled Layer for All
Beaches at Elrington Island 386
Figure 9.7. Change in the nC18 Concentration Over Time for the Oiled Subsurface
Layer for All Beaches at Elrington Island 388
Figure 9.8. Change in the nC19 Concentration Over Time for the Oiled Subsurface
Layer for All Beaches at Elrington Island 388
Figure 9.9. Change in the nC20 Concentration Over Time for the Oiled Subsurface
Layer for All Beaches at Elrington Island 389
Figure 9.10. Change in the nC21 Concentration Over Time for the Oiled Subsurface
Layer for All Beaches at Elrington Island 389
Figure 9.11. Change in the nC22 Concentration Over Time for the Oiled Subsurface
Layer for All Beaches at Elrington Island 390
Figure 9.12. Change in the nC23 Concentration Over Time for the Oiled Subsurface
Layer for All Beaches at Elrington Island 390
xxx
-------
LIST OF FIGURES (Continued)
Figure 9.13. Change in the nC24 Concentration Over Time for the Oiled Subsurface
Layer for All Beaches at Elrington Island 391
Figure 9.14. Change in the nC25 Concentration Over Time for the Oiled Subsurface
Layer for All Beaches at Elrington Island 391
Figure 9.15. Change in the nC26 Concentration Over Time for the Oiled Subsurface
Layer for All Beaches at Elrington Island 392
Figure 9.16. Change in the nC27 Concentration Over Time for the Oiled Subsurface
Layer for All Beaches at Elrington Island 392
Figure 9.17. Change in the Sum of the Alkane Concentration for nC18 to nC27 Over
Time for the Oiled Subsurface Layer for All Beaches at Elrington Island 393
Figure 9.18. Change in the Phytane Concentration Over Time for the Oiled Subsurface
Layer for All Beaches at Elrington Island 393
Figure 9.19. Change in the nC18/Phytane Ratio Over Time for the Oiled Subsurface
Layer for All Beaches at Elrington Island 394
Figure 9.20a. Change in the CO, Concentration Over Time for all Beaches at Elrington
Island on July 11, 1990 402
Figure 9.20b. Change in the CO, Concentration Over Time for all Beaches at Elrington
Island on July 21, 1990 403
Figure 9.20c. Change in the CO2 Concentration Over Time for all Beaches at Elrington
Island on July 30, 1990 404
Figure 9.20d. Change in the CO, Concentration Over Time for all Beaches at Elrington
Island on August 7, 1990 404
Figure 9.2la. Production of Radiolabeled CO, from Phenanthrene Through Time in Oiled
Beach Samples Taken From the Sampling Baskets (Layer 3) on 7/11/90, for the
Sprinkler, Untreated Control, and Bath Beaches at Elrington Island .... 407
Figure 9.2 Ib. Production of Radiolabeled CO, from Phenanthrene Through Time in Oiled
Beach Samples Taken From the Sampling Baskets (Layer 3) on 7/21/90, for the
Sprinkler, Untreated Control, and Bath Beaches at Elrington Island 408
Figure 9.2 Ic. Production of Radiolabeled CO, from Phenanthrene Through Time in Oiled
Beach Samples Taken From the Sampling Baskets (Layer 3) on 7/30/90, for the
Sprinkler, Untreated Control, and Bath Beaches at Elrington Island .... 409
Figure 9.22. Median Percent Total CO, Production (Triplicate Samples) Over Time for
the Sprinkler, Untreated Control, and Bath Beaches at Elrington Island . 411
xxxi
-------
LIST OF FIGURES (Continued)
Figure 9.23. Change in Dissolved Oxygen Concentration Over Time for the
Untreated Control Beach at Elrington Island on July 6, 1990 413
Figure 9.24. Change in Dissolved Oxygen Concentration Over Time for the
Sprinkler Beach at Elrington Island on July 3, 1990 413
Figure 9.2S. Change in Dissolved Oxygen Concentration Over Time for the Bath Beach
at Elrington Island on July 7, 1990 414
Figure 9.26. Change in Dissolved Oxygen Concentration Over Time for the Untreated
Control Beach at Elrington Island on August 3, 1990 414
Figure 9.27. Change in Dissolved Oxygen Concentration Over Time for the Sprinkler
Beach at Elrington Island on August 5, 1990 415
Figure 9.28. Change in Dissolved Oxygen Concentration Over Time for the Bath
Beach at Elrington Island on August 3, 1990 415
Figure 9.29. Cumulative Mineralization of Oil Carbon from Flow-Through Column
Microcosms 418
Figure 9.30. Average Daily Oil Carbon Mineralization During High and Low Tidal
Cycles of Replicate Column Microcosms 419
Figure 9.31. Average Daily Oil Carbon Mineralization in Column Microcosms for Each
Nutrient Treatment 420
Figure 9.32. Change in CO2 Concentration Over Time for Variable Oil Concentrations
at Elrington Island 422
Figure 9.33. Number of Oil Degraders at the Bath, Sprinkler, and Untreated Control
Beaches at Elrington Island 424
Figure 9.34. Hexadecane Mineralization Activity for the Untreated Control,
Sprinkler, and Bath Beaches at Elrington Island 425
Figure 10.1. Gas Chromatographic Profiles Showing the Effect of Different Inocula
on Degradation of Artificially Weathered Prudhoe Bay Crude Oil 428
Figure 10.2. Gas Chromatographic Profiles Showing the Effect of Temperature on the
Degradation of Artificially Weathered Prudhoe Bay Crude Oil 429
Figure 10.3. Gas Chromatographic Profiles Showing the Effect of Temperature on the
Degradation of Artificially Weathered Prudhoe Bay Crude Oil Treated with
INIPOL 430
Figure 10.4. Gas Chromatographic Profiles Showing the Effect of Different
Concentrations of INIPOL (% of Oil Concentration) on the Degradation
of Artificially Weathered Prudhoe Bay Crude Oil 431
xxxii
-------
Figure 10.5.
LIST OF FIGURES (Continued)
Gas Chromatographic Profile Showing the Effect of Different
Fertilizers, Under Poisoned and Unpoisoned Conditions, on the
Figure 10.6.
Figure 10.7.
Figure 10.8.
Figure 10.9.
Figure 10.10.
Figure 10.11.
Figure 10.12.
Figure 10.13.
Figure 10.14.
Figure 10.15.
Figure 10.16.
Figure 10.17.
Figure 10.18.
Figure 10.19.
Figure 10.20.
Figure 10.21.
Figure 10.22.
Gas Chromatographic Profiles Showing the Effect of INIPOL, Under
Poisoned and Unpoisoned Conditions, on the Degradation of Oil on
Beach Material Taken from Prince William Sound
Cumulative Oxygen Uptake on Weathered Prudhoe Bay Crude Oil
Gas Chromatographic Profiles of Alkanes at 0 and 6 Weeks After
Initiation of Flask Studies
Gas Chromatographic Profiles of Aromatics at 0 and 6 Weeks After
Initiation of Flask Studies
Fertilizer Specific Activity (O3 Consumption, CO2 Production) for Six
Treatments Over Time
Mineralization of Radiolabeled Phenanthrene Over Time As Influenced
by Fertilizer Application
Stimulation of Oil-Degrading Microflora with Inorganic Nutrient
Supplementation Over Time
Specific Relationship Between N Concentration and the Activity of
Oil-Degrading Microflora
Cumulative Percent Mineralization of 14C Phenanthrene Over Time ....
Tidalnates" Counts per Sample Interval
Tidalnates" Counts per Sample Interval
Effect of Various Treatments on the Activity of Indigenous, Oil-
Degrading Microorganisms
Effect of Various Treatments on the Activity of Indigenous, Oil-
Degrading Microorganisms
Effects of INIPOL and Soluble Nutrients on 14C Oleic Acid and 14C
Phenanthrene Mineralization Over Time
Distribution of 14C from Radiolabeled Phenanthrene
Accelerated Biodegradation of 14C-Phenanthrene and 14C-Oleic Acid . .
Cumulative Total CO2 Production Over Time
^*r**
434
436
437
439
459
460
461
463
465
466
466
468
469
470
473
474
477
XXXlll
-------
LIST OF FIGURES (Continued)
Figure 10.23. Cumulative Percent Mineralization of 14C-Phenanthrene Over Time .... 478
Figure 10.24. Effect of INIPOL and Inorganic Nutrients on the Activity of Oil-
Degrading Microflora Over Time 479
Figure 10.25. Effect of INIPOL and its Constituents on Microbial
Activities Over Time 480
Figure 10.26. Oil Chemistry Data 481
Figure 11.1. The 61SC and 51BN of Plants, Animals, and Seston Collected from
Snug Harbor (6/13/89, 7/16/89, 8/10/89) and from Passage Cove
(8/13/89, 8/20/89, 8/22/89) 488
Figure 11.2. Mean «1SC and «1BN of Plants, Animals, and Seston Collected Directly
from Untreated Control and Fertilized Plots 489
Figure 11.3. Limpet, Periwinkle, M. edilus, Balanus spp, and Whelk Data Showing
«15N Plotted as a Function of «1SC 491
Figures 11.4A, 46. Stable Carbon Isotope Values for Ecological Samples Collected at Disk
Island and Elrington Island 493
Figures 11.5A, 5B. Stable Nitrogen Isotope Values for Ecological Samples Collected
at Disk Island and Elrington Island 494
Figures 11.6A, 6B. Stable Nitrogen Isotope Values Over Time for Ecological Samples
Collected at Elrington Island 496
Figures 11.7A, 7B. Stable Nitrogen Isotope Values Over Time for Ecological Samples
Collected at Disk Island 497
Figure 11.8. Acridine Orange Direct Counts (AODC) of Bacteria for the Four
Microcosm Treatments Over Time 501
Figure 11.9. Organic Carbon Content (PC) for the Four Microcosm
Treatments Over Time 501
Figure 11.10. Elemental Carbon to Nitrogen Ratio (C:N)wt for the Four Microcosm
Treatments Over Time 502
Figure 11.11. 515N for the Four Microcosm Treatments Over Time 502
Figure 11.12. «1SC for the Four Microcosm Treatments Over Time 503
Figure 12.1. Chemostat Reactor 508
Figure 12.2. Schematic for Modeling the Beach Bioremediation System 510
xxxiv
-------
LIST OF FIGURES (Continued)
Figure 13.1.
Figure 13.2.
Figure 13.3.
Figure 13.4.
Figure 13.5.
Figure 13.6.
Figure 13.7.
Figure 13.8.
Figure 13.9.
Figure 13.10.
Figure 13.11.
Figure 13.12.
Figure 13.13.
Figure 13.14.
Figure 13.15.
Figure 13.16.
Figure 13.17.
Figure 13.18.
Figure 13.19.
Net Oxygen Uptake Curves for Products and Mineral Nutrients:
a) Products E, G, B, A, and D; (b) Products C, J, H, F, and I 532
Total Alkane Reduction in the Product Flasks: (a) day 11, (b) day 20 ... 534
Chromatographic Profile of the Alkane Data at Day 11 for
Products E, A, and I 535
Total Aromatic Reduction in the Product Flasks at Days 11 and 20 537
GC/MS Profile of the Aromatic Data at Day 11 for Products E, A, I ... 538
Yield of Oil-Degraders for All Products at Day 11 540
Log Increase in Oil-Degraders for All Products in 11 Days 541
Schematic Diagram of the Experimental Plot Layout on Disk Island 545
Schematic Diagram of a Typical Beach Plot Showing Dimensions
and Location of Sampling Bags 546
Average Changes in Ammonia-N Levels in the Four Days Between
Fertilizer Applications 550
Average Changes in Nitrate-N Levels in the Four Days Between
Fertilizer Applications 551
Average Changes in Phosphorus Levels in the Four Days Between
Fertilizer Applications 552
Oil-Degrader Counts in All Plots of Blocks 2 and 3
as a Function of Time 553
Changes in Oil Residue Weight Averaged Over All Four Blocks as
a Function of Time 555
Changes in Oil Residue Weight in Each Block as a Function of Time ... 556
Changes in Total Resolvable Alkanes Averaged Over All Four Blocks
as a Function of Time 558
Changes in Total Resolvable Alkanes in Each Block as a Function of
Time
559
Changes in Total Resolvable Alkanes as a Percent of Oil Residue
Weight Averaged Over All Four Blocks as a Function of Time 560
Changes in Total Resolvable Alkanes as a Percent of Oil Residue
Weight in Each Block as a Function of Time 561
XXXV
-------
LIST OF TABLES
Table 1.1. Calculated Ratios of nC17/Pristane and nC18/Phytane 10
Table 2.1. INIPOL EAP 22 Chemical Composition 21
Table 2.2. Total Kjeldahl Nitrogen (TKN) Released under Static Conditions from I8DU
Granular Fertilizer in Bags with Different Surface to Volume Ratios 29
Table 2.3. Release of Ammonia, Total Kjeldahl Nitrogen (TKN), and Total Phosphorus
(TP) from INIPOL EAP 22 during Intermittent Submersion Experiment 37
Table 3.1. Physical Description of Treatment Areas at Snug Harbor 41
Table 3.2. Analysis of Oil Extracted from Mixed Sand and Gravel Samples Taken from
Otter Beach in Snug Harbor on May 28, 1989, Two Weeks Prior to Fertilizer
Application 42
Table 3.3. Physical Description of Treatment Areas at Passage Cove 43
Table 3.4. Fertilizer Treatments at Snug Harbor, Passage Cove, Disk Island, and Elrington
Island 62
Table 3.5. Summary of Field Tests 71
Table 3.6. Summary of Supporting Field and Laboratory Tests 73
Table 4.1. Nutrient Sampling Schedule for Snug Harbor, Passage Cove, Disk Island, and
Elrington Island 80
Table 4.2. Experimental Design for Respirometric Studies 94
Table 4.3. Experimental Design of Flask Studies 95
Table 5.1. Analytical Laboratory Within-Batch Measurement Quality Objectives for
Beach Substrate 118
Table 5.2. Analytical Laboratory Within-Batch Measurement Quality Objectives for
QA/QC Samples for Nutrient Additions 118
Table 6.1. Median Values (mg/g) and Statistical Comparisons of Oil Residue
Weights in Cobble Surface Samples from Different Beach
Treatments at Snug Harbor 151
Table 6.2. Rate Analysis of Natural Log-Transformed Oil Residue Weights in
mg/g in Cobble Surface Samples Versus Time for Test Beaches at
Snug Harbor 153
Table 6.3. Median Values (mg/g) and Statistical Comparisons of Oil Residue
Weights in Mixed Sand and Gravel Samples Under Cobble from
Different Beach Treatments at Snug Harbor 154
xxxvi
-------
LIST OF TABLES (Continued)
Table 6.4. Rate Analysis of Natural Log-Transformed Oil Residue Weights in
mg/g in Mixed Sand and Gravel Samples Under Cobble Versus
Time for Test Beaches at Snug Harbor 156
Table 6.5. Median Values (mg/g) and Statistical Comparisons of Oil Residue
Weights in Mixed Sand and Gravel Only Samples from Different Beach
Treatments at Snug Harbor 160
Table 6.6. Rate Analysis of Natural Log-Transformed Oil Residue Weights in
mg/g in Mixed Sand and Gravel Only Samples Versus Time for Test
Beaches at Snug Harbor 161
Table 6.7. Change in Hydrocarbon Composition Through Time at Snug
Harbor, Expressed in Percent of the Median Concentration of
Individual Hydrocarbons on the 6/9 Sampling (Cobble Surface) 175
Table 6.8. Number of Samples, out of Approximately 21 Samples, Taken at
Each Sampling Time at Snug Harbor, with Alkane Concentration
Below Detection Limit (Cobble Surface) 176
Table 6.9. Median Values and Statistical Comparisons of Oil Residue Weights for
Summed Alkanes in Cobble Surface Samples from Different Beach Treatments
at Snug Harbor 177
Table 6.10. Rate Analysis of Natural Log-Transformed Oil Residue Weights for
Summed Alkanes in Cobble Surface Samples Versus Time for Test Beaches at Snug
Harbor 178
Table 6.11. Median Values and Statistical Comparisons of the nC18/Phytane
Ratio in Cobble Surface Samples from Different Beach Treatments
at Snug Harbor 180
Table 6.12. Rate Analysis of nC18/Phytane Ratios in Cobble Surfaces Versus
Time for Test Beaches at Snug Harbor 181
Table 6.13. Change in Hydrocarbon Composition Through Time at Snug
Harbor, Expressed in Percent of the Median Concentration of
Individual Hydrocarbons on the 6/9 Sampling (Mixed Sand and
Gravel Under Cobble) 182
Table 6.14. Number of Samples, out of Approximately 21 Samples, Taken at Each Sampling
Time at Snug Harbor, with Alkane Concentration Below Detection Limit (Mixed
Sand and Gravel Under Cobble) 183
Table 6.15. Median Values and Statistical Comparisons of Oil Residue Weights for
Summed Alkanes in Mixed Sand and Gravel from Different Beach Treatments
at Snug Harbor 197
xxxvn
-------
LIST OF TABLES (Continued)
Table 6.16. Rate Analysis of Natural Log-Transformed Oil Residue Weights for
Summed Alkanes in Mixed Sand and Gravel Under Cobble Versus Time for
Test Beaches at Snug Harbor 198
Table 6.17. Median Values and Statistical Comparisons of the nC18/Phytane Ratio in
Mixed Sand and Gravel from Different Beach Treatments at Snug Harbor 199
Table 6.18. Rate Analysis of the nC18/Phytane Ratio in Mixed Sand and Gravel
Samples Versus Time for Test Beaches at Snug Harbor 200
Table 6.19. Change in Hydrocarbon Composition Through Time at Snug Harbor, Expressed in
Percent of the Median Concentration of Individual Hydrocarbons on the 6/9
Sampling 215
Table 6.20. Relative Levels of Oil-Degrading Microorganisms in Snug Harbor Mixed
Sand and Gravel Test Plots 222
Table 6.21. Ammonia Nitrogen (/i N/L) from Nearshore Water over Gravel Beaches at Snug
Harbor 224
Table 6.22. Ammonia Nitrogen (n N/L) from Nearshore Water over Cobble Beaches at Snug
Harbor 225
Table 6.23. Phosphate (n P/L) from Nearshore Water over Gravel Beaches at Snug Harbor. 226
Table 6.24. Phosphate (n P/L) from Nearshore Water Over Cobble Beaches at Snug Harbor. 227
Table 6.25. Tidal Variation in Measurements of Bacterial Abundance and Plankton Chlorophyll
a for Sampling Stations at Snug Harbor on 7/26-27/89 228
Table 6.26. Tidal Variation in Measurements of Bacterial Abundance and Plankton chlorophyll
a for Sampling Stations at Passage Cove on 8/7/89 229
Table 6.27. Total PAH's (Mg/g) in Caged Mussels at Snug Harbor at 6 Stations Over Time . 236
Table 6.28. Chemical Analysis of Mixed Sand and Gravel Microcosms Sampled 17 Days After
Initiation of Fertilizer Application 238
Table 6.29. Residue Weight of Oil in Cobble Microcosms Analyzed 26 Days After Fertilizer
Application 240
Table 6.30. Ratios of Hydrocarbons in Oil from Cobble Microcosms Analyzed 26 Days After
Fertilizer Application 240
Table 6.31. Comparison of nC17/Pristane Ratios and nC17/Norhopane Ratios as Measures of
Oil Degradation in Samples Taken from Cobble Microcosm 42 Days After
Initiation of Fertilizer Application 241
XXXVlll
-------
LIST OF TABLES (Continued)
Table 6.32. Use of Dibenzothiophene Peaks/Norhopane Ratios as Relative Measures of the
Degradation of Aromatic Components in Oil Sampled from Cobble Microcosms 42
Days after Initiation of Fertilizer Application 241
Table 7.1. Median Values and Statistical Comparisons of Oil Residue Weights
from Different Beach Treatments 256
Table 7.2. Rate Analysis of Natural Log-Transformed Oil Residue Weights on
Cobble Surfaces for Test Beaches in Passage Cove 258
Table 7.3. Rate Analysis of Nontransformed Oil Residue Weights (Zero-Order) on
Cobble Surfaces in Passage Cove 259
Table 7.4. Change in Hydrocarbon Composition Through Time Expressed in
Percent of the Median Concentration of Individual Hydrocarbons
on the 7/22 Sampling 269
Table 7.5. Of Approximately 21 Samples, Number of Samples Taken at Each
Sampling Time with Alkane Concentration Below Detection Limit 270
Table 7.6. Comparison of Zero-Order and First-Order Rates (Linear
Regression Fit to Medians) of Change in Hydrocarbon Composition,
Based on Decreases Through Time in Hydrocarbon Concentrations
of the Summed Alkanes, nC18 to nC27 290
Table 7.7. Relative Concentration (Log10 of the Cell Number/g of Beach
Material) of Oil-Degrading Microorganisms in Passage Cove 312
Table 7.8. Relative Concentration (Log10 of the Cell Numbers/g of Beach
Material and Standard Deviation) of Oil-Degrading Microorganisms
in Samples from Beaches That Were Not Impacted by Oil 312
Table 7.9. Mineralization of 10 pg 14C-Phenanthrene Per g Passage Cove
Beach Material Prior to Application of Fertilizer (7/22/89) 315
Table 7.10. Mineralization of 0.32 pg of 14C-Phenanthrene and 0.44 jig 14C-
Hexadecane Per g Passage Cove Beach Material 317
Table 7.11. Total PAH's (pg/g) in Caged Mussels at Passage Cove at 5 Stations
Over Time 330
Table 7.12. 48-Hour Survival and Development of Larvae Tested with 100% Site Water
(Undiluted Except for Salinity Adjustment) 331
Table 9.1. Residue Decay Rates for Elrington Island 384
Table 9.2. Analysis of Beach Plot Differences by Date for Elrington Island
Based on nC18/Phytane Ratio 395
xxxix
-------
LIST OF TABLES (Continued)
Table 9.3. Analysis of Beach Plot Differences by Date for Elrington Island
Based on nC18 Linear Alkane 397
Table 9.4. Analysis of Beach Plot Differences by Date for Elrington Island
Based on the Sum of the Linear Alkanes from nC18 to nC27 398
Table 9.5. Estimated Decay Rates in mg/kg/day for Selected Hydrocarbons
Based on Concentration Changes Between t • 0 and t - 7 Days 399
Table 9.6. Number of Oil Degraders (MPN's x 104) in Samples from
Elrington Island 401
Table 9.7. Mineralization Rates from Total CO2 Production
(n moles/100 m/hour) and Calculated Oil Degradation
(mg/kg/day) in Oiled Beach Material Taken from
Sampling Baskets on Elrington Island 405
Table 9.8. Ratio of Nutrient Concentrations Before and After Exposure to
Oiled Beach Material in Special Sampling Baskets from Elrington
Island Beaches 416
Table 9.9. Relationship Between Oil Mineralization Rates and Oil
Concentration in Beach Material from Elrington Island 421
Table 10.1. Shaker Flask Studies: Biodegradation of Aromatics (PAHs) 440
Table 10.2. Results of Laboratory Toxicity Tests with the Oleophilic Fertilizer,
INIPOL, and Various Marine Species 444
Table 10.3. Summary of Toxicity Data for Ammonia 449
Table 10.4. Summary of Toxicity Data for Aqueous Ammonia 450
Table 10.5. Summary of Toxicity Data for Urea 451
Table 10.6. Summary of Toxicity Data for 2-Butoxy-ethanol 452
Table 10.7. Summary of Toxicity Data for Sodium Lauryl Sulfate 453
Table 10.8. Changes in Oil Concentration and Distribution During
Biodegradation with Various Nutrient Amendments 462
Table 10.9. Total Peak Areas: GC Analysis of Oil Residue
from Shaker Experiment 467
Table 10.10. Changes in Oil Concentration and Oil Distribution After 6 Days
Incubation and Treatment with INIPOL and/or Inorganic Nutrient
Solutions 472
xl
-------
LIST OF TABLES (Continued)
Table 10.11. Organisms Isolated from Oiled Beach Material in Prince William
Sound, Alaska 476
Table 10.12. Comparison of GC Profile Total Peak Areas from Microbial
Inoculation Biometer Study 482
Table 10.13. Mineralization of 9-14C-Phenanthrene (24 hour Incubation)
Following 7 Days of Treatment Exposure 482
Table 10.14. Changes in Oil Concentration and Oil Distribution After 7 Days
Incubation 483
Table 11.1. Isotopic Data For Samples Taken from Selected Beaches of Prince
William Sound, AK between June and August, 1989 485
Table 11.2. Stable Carbon and Nitrogen Isotope Data From Disk Island and
Related Bioassay Experiments 499
Table 11.3. Stable Carbon and Nitrogen Isotope Data of Carbon and Nitrogen
Sources in the Microcosm Experiment 499
Table 11.4. Stable Carbon Isotope Data of Material Collected on GF/F Filters,
Nominally Considered To be Bacteria, and Nucleic Acid Extracts
from the Microcosm Experiment 504
Table 12.1. Growth Rate Calculations for 350 ppm Nitrogen Flask 518
Table 12.2. Growth Rate Calculations for 0.35 ppm Nitrogen Flask 518
Table 12.3. Growth Rate Calculations for Elrington Island CO3 Evolution Data
Fertilizer-Treated Beach Material 519
Table 12.4. Growth Rate Calculations for Elrington Island CO2 Evolution Data,
Untreated Control Beach Material 519
Table 12.5. Predicted S (Residual Oil) Values, mg Oil/kg for Elrington Island
Untreated Control for Different n and MPN Values 520
Table 12.6. Predicted S (Residual Oil) Values, mg Oil/kg for Elrington Island
Fertilizer Treated Beach Material for Different n and MPN Values 520
Table 13.1. Experimental Design for Respirometric Studies 526
Table 13.2. Experimental Design for the Shaker Flask Studies 528
Table 13.3. NHS-N levels in Each Product Flask at the Start of the Experiment 531
Table 13.4. Tukey's Studentized Range Test for Detecting Differences in Mean
Percent Removal of Alkanes by Products in 11 Days 531
xli
-------
LIST OF APPENDICES
APPENDIX A PROJECT ORGANIZATION AND RESPONSIBILITIES A-l
APPENDIX B CHRONOLOGY OF EVENTS B-1
xlii
-------
ACKNOWLEDGEMENTS
The success of this project depended on the contributions of numerous individuals. Sincere
gratitude is extended to the outstanding field teams, comprised of W. Kinney (U.S. Environmental
Protection Agency, Ada, Oklahoma), K. Cabbie, M. Dillon, R. Dushek, E. Eschner, N. Halsell, G.
Merritt, J. Baker (Lockheed Engineering & Sciences Co., Las Vegas, Nevada); and R. Wright (Science
Applications International Corporation, San Diego, California). Chemical analyses were provided by
dedicated staffs, under the leadership of J. Clayton, W. Horn, and J. Payne (Science Applications
International Corporation, San Diego, California). Microbiological analyses and laboratory
experiments were conducted by personnel from Technical Resources, Inc., (Gulf Breeze, Florida), and
by teams headed by E. Brown (University of Alaska, Fairbanks, Alaska) and G. Winter (Alaska
Department of Environmental Conservation, Valdez, Alaska).
Outstanding logistics support was critical to project success. Personnel from the U.S.
Environmental Protection Agency-Region X (Anchorage, Alaska) were instrumental in coordinating
logistics in the time frame immediately following the spill. K. Schmidt and R. Shokes (Science
Applications International Corporation, San Diego, California) accomplished the monumental task of
logistics support in 1989; in 1990, D. Peres (Lockheed Engineering & Sciences Co., Las Vegas,
Nevada) performed in an equally outstanding manner. Gratitude is extended to J. Harvey and B.
Wireman of Technical Resources, Inc. (Gulf Breeze, Florida) for their many contributions to the
project. Capable and efficient onsite administrative support was provided by the U.S. Environmental
Protection Agency (various locations) and E. Sullivan, E. Gray, K. Brown, D. Rosenblatt, L. Suit, and
S. Doherty of Technical Resources, Inc. (Rockville, Maryland and Gulf Breeze, Florida).
Appreciation is also extended to T. Baugh (U.S. Environmental Protection Agency, Washington, D.C.)
for his onsite and offsite support and dedication to administrative tasks.
Rigorous quality assurance provided by the U.S. Environmental Protection Agency (Las Vegas,
Nevada) and Lockheed Engineering & Sciences Co. (Las Vegas, Nevada) helped assure the validity
of the results presented in this report. Database management was provided by Science Applications
International Corporation (San Diego, California) in 1989 and by Lockheed Engineering & Sciences
Co. (Las Vegas, Nevada) in 1990. Additionally, experimental design and data analysis statistics were
provided by the U.S. Environmental Protection Agency (Las Vegas, Nevada) and Lockheed
Engineering & Sciences Co. (Las Vegas, Nevada). The authors extend their gratitude for the superior
work performed in these areas.
xliii
-------
Deepest appreciation is expressed to the crews of the F/V AUGUSTINE (G. Tyler, P. Tyler,
J. Tyler, T. Morgan, and V. Boyd); the F/V CARMEN ROSE (B. King, D. Kuntz, S. Carter, and
G. Covel); the F/V INSPECTOR (S. Illigs, B. Costello, and L. Corbell); the F/V JOLLY ROGER
(L. Gray, N. Stewart, and M. Jochen); the Aircraft N756AF (D. Thacker and K. Lobe); and the ERA
helicopters (D. Fehrenkamp, B. Nelms, and D. Spaltey) for their support and safe transportation of
personnel, equipment, and samples. D. Carlson and K. Broker (Norcon, Inc., Valdez, Alaska)
demonstrated superior craftsmanship in the construction of the two field laboratories. Prince William
Sound Community College (Valdez, Alaska) generously provided facilities for the chemical laboratory.
Special acknowledgement is owed to B. Wyatt for providing space for housing and the field
laboratories as well as for construction of the office space known as the Wyatt Federal Center.
Appreciation is extended to individuals in the U.S. Environmental Protection Agency's
Headquarters offices and the various research laboratories for their onsite and offsite support and
guidance. Recognition is also extended to R. Bare, D. Elemendorf, and M. Grossman (Exxon
Research and Engineering Co., Annandale, New Jersey); F. Kaiser and J. O'Bara (Exxon Research
and Engineering Co., Florham Park, New Jersey); R. Requejo (Exxon Production Research Co.,
Houston, Texas); and J. Wilkinson (Exxon Alaska Operations Office, Anchorage, Alaska). Finally,
gratitude is extended to the people and businesses of Valdez, Alaska, for their support of the project
and onsite scientists.
xliv
-------
EXECUTIVE SUMMARY
The feasibility of using bioremediation as cleanup tool for oil contaminated beaches in Prince
William Sound, Alaska, has been verified by a series of field and laboratory studies conducted during
1989 and 1990. Initial field studies at Snug Harbor during the first summer involved the application
of two different types of fertilizer formulations (oleophilic and slow-release) directly to the beach
surface. The results from this study provided sufficient information to allow Exxon to consider
bioremediation on a large scale during the month of August 1989 to assist in the overall oil cleanup
program. Approximately 70 miles of the Prince William Sound beaches were subsequently treated
with a combination of oleophilic and slow-release fertilizers. A second field study at Passage Cove
during the first summer was used to verify the effectiveness and safety of this large-scale application.
Following laboratory research during the winter of 1989/1990, activities were initiated for a
second summer to optimize fertilizer application. Field studies were conducted at sites on Elrington
Island and Disk Island. The research results demonstrated the effectiveness of pulsed fertilizer
application and established new methods for assessing the effectiveness of bioremediation.
Enhancement of oil biodegradation was greater during the summer of 1990 than the previous summer.
Specific summaries and conclusions from the studies at each of the field sites is given below.
Snug Harbor
Two fertilizer formulations, an oleophilic fertilizer and fertilizer briquettes, were tested
separately at Snug Harbor. Two types of contaminated beach material were also evaluated:
cobblestone overlying mixed sand and gravel, and mixed sand and gravel alone.
a) Visual inspection of beaches treated with oleophilic fertilizer showed that oil was removed
from the beach surface approximately 2 to 3 weeks after fertilizer application. The effect
was most apparent on cobble beaches, where initially much of the surface oil was removed.
No visible decreases in the oil occurred, however, on the beaches treated with the slow-
release fertilizer briquettes or on the untreated control beaches. Disappearance of oil on
oleophilic-treated plots continued over time, eventually leading to the disappearance of oil
from most of the beach material surfaces.
xlv
-------
b) No oil slicks or oily materials were observed in the seawater following application of the
fertilizers, and no oil or petroleum hydrocarbons were detected in mussels contained in cages
just offshore from the fertilizer-treated beaches. This suggested that removal of oil from the
beaches did not appear to be a result of dispersing phenomena.
c) Analysis of oil extracted from all beach plots showed that the oleophilic fertilizer caused the
greatest initial reduction in oil residues on the cobble surfaces, accompanied by substantial
changes in oil composition. These results were significantly different when compared to the
untreated control and the briquette fertilizer-treated plot. However, it is believed that the
full extent of this change was initially masked through interferences in the gas
chromatographic analyses by components in the oleophilic fertilizer.
d) The greatest changes in oil composition were observed in samples from the briquette
fertilizer-treated plots. However, changes were primarily observed during the first two
weeks of the test in cobble plot samples, suggesting that the fertilizer-enhanced changes were
short-lived. A more sustained but less extensive effect was seen in the mixed sand and
gravel plots. These results lead to the conclusion that fertilizer briquettes, or a similar
formulation that releases inorganic nitrogen and phosphorus, would likely affect changes in
oil composition on both the cobble surface and within the mixed sand and gravel matrix.
e) All changes in oil composition were accompanied by large decreases in the nC18/phytane
ratio. This represents a differential change in chemically similar hydrocarbons, and can only
be attributed to biodegradation processes. Thus, fertilizer application appeared to enhance
oil biodegradation.
f) Numbers of oil-degrading microorganisms did not appear to increase as a result of fertilizer
application. However, large heterogeneity in the microbial population precluded observing
statistical differences. This was further complicated by a high number of oil-degrading
bacteria in the oiled beach material prior to fertilizer exposure (averaging 1 to 10% of the
total bacterial population). The high numbers represented an enrichment of oil-degrading
microorganisms of approximately 10s to 105 compared to beaches not exposed to oil. These
results demonstrate that the beaches were well primed for bioremediation.
g) Extensive ecological monitoring studies indicated that the addition of fertilizer to oiled
shorelines did not cause ecologically relevant increases in planktonic algae or bacteria, or any
xlvi
-------
measurable nutrient enrichment in adjacent embayments. These studies were supported by
stable nitrogen isotope analyses of intertidal algae and heterotrophic organisms. Stable
nitrogen isotope ratios demonstrated that when fertilizer was assimilated by algae on the
beach, trophic structures were not disrupted. Finally, mutagenicity studies showed that
mutagenic materials associated with Prudhoe Bay crude oil were lost over time from both
treated and untreated control plots. In conjunction with chemical analysis, these studies
demonstrated that decreases in mutagenicity were due to both fertilizer-enhanced
biodegradation and other natural processes.
Passage Cove
Two fertilizer applications were also tested at Passage Cove. A combination of the oleophilic
fertilizer and fertilizer granules (instead of briquettes) was applied to one beach, and a fertilizer
solution was applied (via a sprinkler system each day at low tide) to another beach.
a) The visual reduction in oil due to application of the oleophilic/granular fertilizer
combination was similar to Snug Harbor results. This visual reduction became apparent
approximately two to three weeks following application of the fertilizers; the untreated
control beach, on the other hand, essentially did not change visually. The effect was perhaps
more dramatic in Passage Cove since oil from both the cobble surface and the subsurface
mixed sand and gravel visually disappeared in a shorter timeframe. It is possible that when
the beaches in Passage Cove were physically washed, oil was distributed over a large surface
area, subsequently creating improved conditions for biodegradation of oil.
b) Application of fertilizer solutions from a sprinkler system also caused oil to visually
disappear in approximately the same general timeframe as Snug Harbor results (3 to 4 weeks).
This observation provided definitive proof that biodegradation (and not chemical washing)
was likely responsible for the oil removal, since there is no other reasonable mechanism to
explain this effect of nutrient addition to the oil. The application of fertilizer solutions,
therefore, proved to be the most efficient system for exposing oil-degrading microorganisms
to nutrients in a controlled and reproducible manner.
c) Application of the fertilizer solution produced a statistically significant enhancement of oil
biodegradation relative to the untreated control beach. Rates of total oil residue loss were
greater than four-fold faster than rates of removal on the untreated control beach. The loss
xlvii
-------
of oil residues was accompanied by extensive changes in oil composition. This included large
decreases in the nC18/phytane ratio. Thus, enhanced biodegradation was probably
responsible for changes in oil residue and composition. Results from the fertilizer solution
treatment further support that oil biodegradation in Prince William Sound was limited by the
availability of nutrients and not by the availability of the oil itself. In addition, reapplication
of nutrients (the extreme in the case of the fertilizer solution) is probably important for
sustaining enhanced biodegradation.
d) Application of the oleophilic/slow-release granular fertilizer combination also substantially
enhanced oil biodegradation. At a slightly lower degree of statistical confidence (90%
confidence level instead of 95%), this fertilizer combination produced a significant two-to
three-fold enhancement in the removal of total oil residues relative to the untreated control
beach. This was accompanied by an extensive change in the composition of the oil as well.
e) Mechanistically, there is no evidence to suggest that the application of the oleophilic/slow-
release granular fertilizer combination worked differently than the application of the
fertilizer solution; each process provided enough nutrients to the oil-degrading microbial
populations to enhance biodegradation. Results from changes in oil composition during the
initial two weeks following fertilizer application suggest that the oleophilic fertilizer uniquely
caused simultaneous degradation of the higher and lower molecular weight hydrocarbons.
Results from the untreated control beach and the fertilizer solution-treated beach showed
that during the same time period, a more typical response was observed; that is, the lower
molecular weight hydrocarbons degraded faster than the higher molecular weight
hydrocarbons. It is also believed that the eventual greater response from the fertilizer
solution application was due to higher nutrient concentrations sustained over a longer period.
Reapplication of the oleophilic/slow-release granular fertilizer combination every three to
four weeks might produce the same effect observed with the fertilizer solution application.
f) Further monitoring of the fertilizer-treated beaches through early summer 1990 revealed that
even subsurface oil (to a depth of approximately 0.3 m) was virtually completely removed
within approximately 10 months. However, significant, but patchy amounts of oil remained
on the untreated control beach after this time period. This suggests that bioremediation
greatly reduced beach cleanup time.
xlviii
-------
g) Due to high variability in the numbers of oil-degrading bacteria in each sample, it was not
possible to show statistically significant increases in the oil-degrading microbial populations
as a result of the fertilizer addition.
h) No widespread or persistent adverse ecological effects were observed from the monitoring
program that was designed to measure toxic responses, eutrophication, and bioaccumulation
of oil residues. Ammonia, the only component in the oleophilic fertilizer that was potentially
toxic to indigenous species, never reached toxic concentrations outside the immediate zone
of application (as inferred from the toxicity test results). Measurements of chlorophyll,
primary productivity, and bacterial production indicated eutrophication did not occur. The
absence of oil residues in caged mussels, held just offshore of the fertilizer-treated areas,
supported the tenet that oil was not released from the beaches into the water column as a
result of the fertilizer treatment.
Disk Island
A field study at Disk Island was conducted to estimate fertilizer dose response. Plots at Disk
Island were dosed with different concentrations of slow-release granular fertilizer, to evaluate the
effectiveness of concentrations above and below the recommended application rate. Effects on
enhanced oil biogradation were measured. Special sampling baskets were developed for field sampling
to ease the analytical burden.
a) Analysis of ammonia, nitrate and phosphate in interstitial beach water showed nutrient
release was generally proportional to fertilizer application rate. However, nutrient release
represented more of a pulsed high concentration during the first 2 to 3 days following
application rather than a slow release over time. Stable nitrogen isotope analysis confirmed
that the fertilizer was the source of NH4+ and NOS" in pore waters.
b) Analytical chemistry results showed that the addition of fertilizer failed to enhance oil
biodegradation, regardless of the application rate. Changes in oil composition occurred,
although slowly, but were similar for both treated and untreated plots. It is not known what
conditions at the Disk Island site could have precluded enhanced oil biodegradation. Disk
Island may represent a type of beach (low energy, low slope profile, less porous beach
material) that is not amenable to oil bioremediation because of insufficient oxygen
availability and/or high natural organic matter content (peat deposits). However, stable
xlix
-------
carbon isotope analysis in microcosm experiments demonstrated that oil carbon was a
substantial proportion of the total carbon supply to the bacterial assemblage.
c) Measurements of oil mineralization, based on total CO2 production or 14CO2 released from
radiolabeled hydrocarbons (phenanthracene and hexadecane), did show effects due to
fertilizer application rate. It was quite evident that considerable oil biodegradation was
occurring in the beach samples. Production of large amounts of radiolabeled CO2 from
phenanthracene and hexadecane strongly suggested that this was due to biodegradation of oil
and not other types of organic matter (humic materials). Adding fertilizer generally caused
significant increases in mineralization rates relative to the untreated control plots, with the
greatest stimulation occurring with the 500 g/m2 application (approximately 2 to 5 times
greater than the untreated controls). The largest application (1,000 g/m2) actually seemed
to inhibit mineralization to some extent. A calculated dose response indicated that a six-fold
increase in fertilizer granule application produced a two-fold increase in oil degradation
rates. In addition, oil-degrading bacteria increased following fertilizer application.
Concentrations were almost 10-fold greater than the untreated controls on the plots receiving
the two highest fertilizer concentrations.
d) Stable carbon and nitrogen isotopes were used to trace nitrogen from bioremediation
treatments into intertidal beach food chains and to examine trophic structures. This
assimilation of fertilizer nitrogen was species-specific and related to the proximity of the
organism to the treated plot. Stable nitrogen isotope analysis of heterotrophic organisms
revealed that the fertilizer was assimilated into the food chains, and assimilation depended
on the feeding strategy of the organism. While nitrogen from fertilizer was assimilated into
beach food chains, no adverse effect to the food chain structure was observed.
Elrington Island
This field study focused on optimizing the fertilizer solution application used in Passage Cove
the previous summer. The effectiveness of multiple pulse doses of fertilizer solution was evaluated
against a single pulse dose, and bioremediation of subsurface oil was also examined. '
a) It was concluded that bioremediation of subsurface oil was reasonable, if sufficient quantities
of nitrogen and phosphorus nutrients can be supplied. This was accomplished by using
fertilizer solution.
1
-------
b) A single pulse application of fertilizer (4 hours, once at low tide) enhanced oil biodegradation
for as long as 3 to 4 weeks. This application was as effective, if not more so, than a multiple
dose application. These results raise the question of whether the effects of fertilizer
application during the summer of 1989, which also consisted of large initial pulses followed
by a gradual release over time of nutrients at lower concentrations, may have been related
more to the extent of nutrient exposure to the microbial communities at one time than the
length of exposure.
c) Fertilizer-enhanced oil biodegradation rates on Elrington Island were the highest recorded
in the field demonstrations. These rates were approximately 100 mg/kg of beach
material/day, almost six-fold higher than the untreated Control beach. This occurred despite
the fact that rates on the untreated Control beach were approximately three-fold higher than
rates reported at Passage Cove the previous summer. This increase in effectiveness of
fertilizer application could be due to extensive colonization of the subsurface oil by oil-
degrading microorganisms, and/or increased availability of oil to the bacteria by
impregnation with glacial till (greater exposed surface area).
d) Measurement of oil mineralization rates in the field and the laboratory using total CO2
production, oxygen uptake, and nutrient assimilation generally coincided with changes in oil
concentration and composition. Relative to oil chemistry analysis, mineralization
measurements could therefore provide simpler procedures for assessing the effect of
fertilizer-enhanced biodegradation in future field studies.
e) The Elrington Island study clearly validated the use of laboratory oil degradation information
from flask and microcosm studies to predict bioremediation events in the field.
f) The use of sampling baskets containing homogenized beach material proved to be a reliable
method to supplant direct sampling of beach material. It considerably reduced sampling
variability and provided information that was representative of the beach it modeled.
g) Stable isotope studies were also conducted on Elrington Island and compared to results on
Disk Island. On Elrington only one alga, Fucus distichus, was observed to assimilate fertilizer
nitrogen. Differences between beaches are attributed to a steeper beach slope on Elrington
Island, resulting in stronger definition of ecological niches.
li
-------
Other Studies
a) To substantiate and interpret results obtained in field studies, a series of laboratory research
projects were completed during the winter of 1989/1990. These projects included evaluation
of application strategies for fertilizers, investigation of the mode of fertilizer action, and
bioaugmentation studies. Results from shake flasks and microcosm studies showed that a
single pulse of fertilizer nutrients was as effective as multiple pulses for enhancing oil
degradation over a three to four week period. This meant that fertilizer application strategies
could ultimately be greatly simplified as was demonstrated in the Elrington Island field study.
Studies with oleophilic fertilizer verified that the product affected oil degradation primarily
through the provision of nitrogen and phosphorus, and not through some alternative
supplement. In other studies, several pure cultures with demonstrated ability to degrade
Prudhoe Bay crude oil were isolated from oiled Prince William Sound beach material. When
high concentrations of these cultures were reinjected into the oiled beach material, great* r
initial oil biodegradation occurred compared to oiled beach material receiving only nutrients.
Thus, bioaugmentation, based on laboratory tests, merited further study.
b) Due to the complicated ecology involved in the biodegradation of oil by natural microbial
communities, it was important to develop an initial predictive model to relate responses of
the microbial communities to fertilizer application. Laboratory and field data were used to
test a simple deterministic model. The results showed that much more information is needed
on the production and fate of the microbial biomass to make this modeling approach
generally applicable to oil bioremediation.
c) Field evaluation of two commercial bioremediation products yielded inconclusive results.
Most of the readily biodegradable compounds in the oil had likely disappeared during the
sixteen months that had elapsed between the oil spill and the initiation of the field test.
Lacking sufficient substrate, it was not possible to measure significant differences among the
treatments.
lii
-------
SECTION 1
INTRODUCTION
BACKGROUND
Major oil spills have galvanized public attention to alternative cleanup technologies. Oil
biodegradation in aquatic (marine and freshwater), terrestrial, and groundwater environments has
been extensively studied in laboratory systems over the past 20-30 years (Atlas, 1981 and 1984;
National Academy of Sciences, 1985; Leahy and Colwell, 1990; and Bartha, 1986), but it is only
recently that this information has been considered for large-scale bioremediation efforts in aquatic
environments (Nelson et al., 1987; Bartha, 1986; Morgan and Watkinson, 1989; and Lee and Levy,
1989). Definitive success in the restoration of gasoline-contaminated aquifers (Raymond et al., 1976
and 1978; Minugh et al., 1983; Yaninga et al., 1985; and Brown et al., 1985) and oil-contaminated soils
(Bartha, 1986; Rittmann and Johnson, 1989; and Dibble and Barth, 1979) has, however, shown the
usefulness of bioremediation and has indicated the importance of a basic research database.
Enhancing biodegradation processes to assist in the cleanup of oil spills in marine environments
has been suggested several times, with much emphasis on the treatment of oil on open waters (Lee and
Levy, 1989; Halmo, 1985). Several approaches have been discussed and debated, including
accelerating oil biodegradation rates by increasing the availability of oil to bacteria through the use
of dispersants and seeding oil-contaminated areas with hydrocarbon-degrading bacteria. Both
approaches have had mixed results, and the complex logistics of open-water monitoring has made
assessments of success ambiguous and inconclusive.
The simplest approach for enhancing oil biodegradation is the addition of nitrogen and
phosphorus nutrients in a well-oxygenated environment. It is well known that enrichments of oil-
degrading microorganisms occur rapidly following oil spills in most environments. But with a large
amount of degradable oil carbon present, biodegradation quickly becomes limited by nutrient and
oxygen availability (Lee et al., 1988; Atlas, 1981 and 1984; National Academy of Sciences, 1985;
Leahy and Colwell, 1990; Bartha, 1986; and Morgan and Watkinson, 1989). Numerous laboratory and
field studies have shown that attempts to overcome these limitations generally lead to successful
optimization of oil biodegradation rates and extents (Nelson et al., 1987; Atlas, 1984; Leahy and
Colwell, 1990; Morgan and Watkinson, 1989; and Lee and Levy, 1989). This approach has never been
used before on a large scale to directly assist in cleanup operations following a major oil spill.
-------
INTRODUCTION
On March 24, 1989, the Exxon Valdez ran aground on Bligh Reef in Prince William Sound,
Alaska, releasing approximately 11 million gallons of Prudhoe Bay crude oil. The oil spread onto an
estimated 1,000 miles of shoreline (350 miles in Prince William Sound). The oil spill provided a
unique opportunity to test the feasibility of bioremediation on a large scale (EPA ORD, 1989), sirce
Prudhoe Bay crude oil had been the focal point of several previous biodegradation studies in cold
water environments (Atlas and Busdosh, 1976; Fedorak and Westlake, 1981; Horowitz and Atlas, 1977;
Atlas et al., 1978; and Cook and Westlake, 1974).
DESCRIPTION OF THE AREA IMPACTED BY THE SPILL
The site of the Exxon Valdez oil spill is a harsh and diverse environment with poor accessibility.
The shoreline is geologically young, composed largely of metamorphic rock, and ranges from verticil
cliffs to boulder and pebble beaches. High-energy beaches are common, with tides ranging from H 4
to -1 m. In some areas, glacial and snow melt introduce large amounts of fresh water to nearshoi e
waters of Prince William Sound. The Sound has a considerable population of seals and sea otter;,
extensive herring and salmon spawning areas, and significant numbers of seabirds and shorebhd:.
There is a substantial migration of birds that feed at beaches and intertidal areas.
Major contaminated shoreline areas included Knight Island, Eleanor Island, Smith Island, Green
Island, and Naked Island (Figure 1.1 A and B). Knight Island, the largest and one of the most heavily
polluted of these islands, has restricted tidal flushing action in some bays and coves. The oil settled
into the beach gravel, on rock surfaces, and on the faces of vertical cliffs. Contamination occurred
primarily in the intertidal zone.
Initial weathering resulted in a loss of approximately 15% to 20% of the oil mass by
volatilization. Volatilized components included normal aliphatic hydrocarbons of less than 12 carbon
atoms and aromatic hydrocarbons such as benzene, toluene, xylene, and some methyl-substituted
naphthalenes. The resulting residue consisted of alkanes, branched alkanes, heterocyclic chemicals,
multi-ring aromatic compounds, high-molecular-weight waxes, and asphaltenes. On most beaches
in Prince William Sound the weathered oil was black and viscid and not brown and mousse-like.
Beaches were physically cleaned by Exxon using a combination of flooding and application of
water under high pressure and high temperature (140*F) (Figure 1.2). The extent of physical washing
was dependent upon the degree of contamination. Vacuum extraction and physical skimming were
-------
Indicates
C-x-) observed
distribution
Kenai
Peninsula
Cook
Inlet
Composite overview of
oil spill tracking from
March 24, 1989 to
June 20. 1989. All
categories of oil
are represented
Approximate area:
28,500 sq. km.
90
Prince
William
Sound
"Exxon Valdez"
grounding site
Alaska
Peninsula
State of Alaska
Dept. of Environmental
Conservation
Kilometers
Figure 1.1 A. Diagram of the Oil and Its Impact.
-------
State of Alaska
Dept of Environmental
Conservation
Key
Heavy/Medium OH
Prince
anuij. William
Sound
Hlnchlnbrook
Montague Island
Island
Gulf of Alaska
0 5 10 15 20
Scale In Miles
Figure 1.1 B. OH Impacted Areas In Prince William Sound.
-------
Figure 1.2. Exxon Cleaning a Beach Using Water Under High Pressure and
Temperature.
-------
INTRODUCTION
used to remove the released oil from the water surface. The cleaning process partially removed oil
from the surface of rocks and beaches, particularly pools of oil, but did not effectively remove the
oil trapped in and below the matrix of gravel and cobble. The washing process also spread a thin
layer of oil over a greater surface area of rock and gravel.
The composition of oil found on the beaches following this washing (weathered oil) is commonly
measured by extracting the oil from beach material and analyzing it with gas chromatogsraphy. A
typical gas chromatogram of fresh and weathered Prudhoe Bay crude oil is shown in Figure 1.3. The
weathered crude oil was taken from a Prince William Sound beach (Northwest Bay) in late spring
1989. The major peaks represent detector responses for the normal alkanes; the annotated numbers
are the carbon lengths of the appropriate alkane. Normal aliphatic hydrocarbons of 12 carbons or less
are absent in the weathered oil, while large quantities of biodegradable hydrocarbons (nC13-nC28)
remain. Gas chromatograms for oil samples fractionated into the aliphatic and aromatic components
are shown in Figures 1.4 and 1.5. These fractionated samples of oil showed the presence of smal!
quantities of aromatic hydrocarbons in the weathered oil, but hydrocarbons up to the methyl
naphthalene were absent.
Pristane and phytane, branched alkanes, are generally recognized to slowly biodegrade relative
to straight chain alkanes (Atlas, 1981) and have thus been used as conserved internal standards to
measure biodegradation. Changes in the ratios of hydrocarbon concentration for the linear alkanes
relative to the branched alkanes can be used to indicate biodegradation since changes brought about
by other physical and chemical processes will not differentially affect the fate of these two
hydrocarbon types (Pritchard and Costa, 1991). Table 1.1 gives the calculated ratios of nC17 linear
alkane to pristane and nCI8 linear alkane to phytane for samples taken from Prince William Sound
from April 4 to May 2, 1989. Relative to fresh Prudhoe Bay crude oil, biodegradation of the oil at
most beaches had not occurred. The sample from Disk Island (gravel) was the only one with a
significant difference in these ratios relative to fresh Prudhoe Bay crude oil. This suggested natural
biodegradation was probably occurring at this beach.
BIOREMEDIATION APPROACH
After learning of the magnitude of the spill, the EPA Assistant Administrator for the Office of
Research and Development (ORD) convened a meeting of nationally and internationally recognized
scientists in the field of oil biodegradation in April, 1989, to evaluate the feasibility of using
-------
10 11 12 13 14 15 18
Fresh
16
17
18
19
m»+t
^yN^WWW
20
21
22
Weathered
23
Figure 1.3. Unfractionated Prudhoe Bay Crude Oil (Number Indicates Carbon
Atoms of Alkane).
-------
10 11 12 13 14 15 16
Fresh
15
14
13
12
10 11
Lx^r
^
u
^
Irf
«
16 17
Jtf
*
^
18
A
19
*
20
a
21
kjjA
22
itak
23 Weathered
24 25
25 26
1 272829
1 1
JLa «u_ . ^. 1*1 L. -«*. — ! • J, 1
T» - ^ TT I1 1 flF^TW *"f T •iT-rr^!H<
Rgure 1.4 Prudhoe Bay Crude OH, Aliphatic Fraction (Number Indicates
Carbon Atoms of Alkane).
-------
Fresh
Weathered
Figure 1.5. Prudhoe Bay Crude Oil, Aromatic Fraction.
-------
INTRODUCTION
TABLE 1.1. CALCULATED RATIOS OF NC17/PRISTANE AND NC18/PHYTANE
Sample
nC17/
Pristane
nC18/
Phytane
Fresh Prudhoe Bay Crude Oil
Eleanor Island
Northwest Bay
1.7
2.0
Surface
Surface Control*
6" Depth
6" Depth Control*
Seal Island
Smith Island
Disk Island
Gravel
Fresh Oiled Rock
Weathered Oiled Rock
1.5
<0.47
1.4
<0.45
1.6
1.5
0.8
1.4
1.8
1.9
—
1.7
__
2.1
1.9
1.0
1.7
2.0
* Sample taken from an uncontaminated beach area.
bioremediation to assist cleanup operations. The scientific group recommended that ORD plan and
conduct a field demonstration project to evaluate the use of fertilizers for accelerating natural
biodegradation of the spilled oil in Prince William Sound. This recommendation was based on the
following conclusions:
• The presence of readily degradable hydrocarbons from the spilled oil would enrich naturally
occurring oil-degrading bacteria.
10
-------
INTRODUCTION
• Oil biodegradation in Prince William Sound waters was probably limited by the availability
of nitrogen and phosphorus; therefore, fertilizing the beaches with these nutrients would
enhance natural degradation of the oil.
• Past studies have convincingly shown that enhancement of oil biodegradation readily occurs
through nutrient addition. Further verification of these studies by laboratory experiments
was unnecessary.
• Successful bioremediation would require consideration of the logistics and mechanics of
long-term nutrient application and the physical agitation of oil.
• An oleophilic fertilizer, such as produced by Elf Aquitaine Chemical Company, may be the
only way to assure extended contact of the nutrients with the oil-contaminated beach
material.
• Bioremediation should be used as a finishing step for any cleanup program. Once the bulk
oil was removed (regardless of the method), bioremediation would further reduce the
amount of residual oil.
• Treatment of the beaches with fertilizer would not necessarily remove all residues (i.e., little
visual improvement) but it would considerably reduce, if not eliminate, ecological
availability of the oil.
• Inoculation of oil-contaminated beaches with hydrocarbon-degrading microorganisms
enriched from Prince William Sound waters was considered inappropriate as an initial
approach but should be considered, in an experimental context, for future spills.
EPA, having a large research program in bioremediation (EPA ORD, 1990-600/9-90/041),
including the necessary technical personnel and in-house contractors, consequently responded rapidly
to the workshop recommendations.
A bioremediation research project was thus initiated. The goal was to perform an initial field
demonstration of bioremediation as a cleanup tool, and, if successful, make recommendations for
wider-scale use to Exxon. In addition, EPA would provide a follow-up field study to large-scale
application as definitive indication of bioremediation success. A research plan was developed
containing the following objectives:
• To examine the rate and extent of natural biodegradation on oil-contaminated beaches.
• To determine if oil biodegradation rates on oil-contaminated beaches could be enhanced by
the addition of nutrients in the field to merit the use of bioremediation as a cleanup tool.
• To develop methods for long-term application of nutrients to oil-contaminated beaches.
11
-------
INTRODUCTION
• To establish methods for monitoring potential ecological effects resulting from nutrient
addition.
• To develop information on the movement of nutrients in beach substrata (beach mechanics).
• To examine the possibility of enhancing oil biodegradation through microbial inoculation.
The plan was reviewed by a special committee of EPA's Scientific Advisory Board, and they
recommended implementation with minor modifications.
Following development of the research plan, a cooperative effort was proposed to Exxon under
the Federal Technology Transfer Act of 1986. On June 2, 1989, the two parties reached an
agreement, and the project was formally initiated. Exxon agreed to provide all logistical support
(transportation from Valdez to test sites, field laboratory facilities, and subsistence) and $1.6 million
for direct support of the field demonstration project. EPA provided $1.6 million for management
personnel, scientific expertise, quality assurance, and operations technical support.
A team of experts from the different research laboratories within the ORD (see Appendix A for
listing) was assembled to implement the field demonstration project. A brief overview of the major
events is given below. A more detailed chronology of events is given in Appendix B.
FIELD OPERATIONS - SUMMER 1989
Field operations began in early May 1989, using the mobilization capability of ORD laboratories
at Las Vegas, NV; Gulf Breeze, FL; Cincinnati, OH; Athens, GA; Research Triangle Park, NC; and
Ada, OK (see Appendix A for support personnel). It was imperative to initiate the field
demonstration as quickly as possible to provide enough time during the summer for large-scale
application if the results were favorable.
Test Beach Selection
Test beaches at Snug Harbor and Passage Cove on Knight Island were selected for testing during
the summer of 1989. These beaches were mainly comprised of large cobblestone overlying a mixed
sand and gravel base. The Snug Harbor beaches had a moderate degree of oil contamination confined
to a broad band within the intertidal zone. At the initiation of the project, oiled beaches that had
been physically washed by the Exxon process were not available. Thus, the Snug Harbor study site
12
-------
INTRODUCTION
was chosen to approximate beach conditions following physical washing. Passage Cove, a more
heavily oiled beach, was selected later in the summer after it was physically washed by Exxon.
Physical washing resulted in the removal of the bulk oil and spread the remaining oil over the beach
surfaces. Both beaches had a thin layer of oil covering the surface of the cobblestone, as well as oil
mixed into the sand and gravel under the cobblestone to varying depths.
Fertilizer Selection
The spill situation necessitated rapid evaluation and selection of fertilizers, and was based on
considerations of application strategies, logistical problems for large-scale application, commercial
availability (particularly if large-scale application became reasonable), and the ability to deliver
nitrogen and phosphorus nutrients to the microbial communities on the surface and subsurface beach
material for sustained periods of time. Three fertilizer application strategies were adopted for testing:
commercially available slow-release formulations, an oleophilic fertilizer, and fertilizer solution
(Rogers et al., 1990).
Snug Harbor Demonstration
Initial field fertilizer application was conducted at Snug Harbor in June using oleophilic and
slow-release fertilizers. Effects of the fertilizer applications on the oil-degrading microbial
communities in the beach material were examined. Oil biodegradation was tracked through time
using analytical chemistry (to determine changes in oil residue weight and composition) and
microbiological techniques. In addition, visual observations were also recorded.
A monitoring program was established to investigate potential adverse ecological effects and
verify the safety of bioremediation as a cleanup tool. The potential for eutrophication was
investigated through measurements of ammonia, phosphate, chlorophyll, bacterial abundance and
productivity, and phytoplankton primary productivity. To address the question of physical removal
of the oil (as opposed to biodegradation), caged mussels were placed offshore and sacrificed for oil
concentration measurements.
In addition to the field testing, laboratory and microcosm experiments were designed to
supplement the field activities. Potential mutagenicity activity associated with the biodegradation of
oil was investigated, and incorporation of applied fertilizers into the food web was assessed using
stable isotopes.
13
-------
INTRODUCTION
By the middle of July 1989, results from the field demonstration project convinced Exxon to
use large-scale bioremediation as part of their cleanup effort. Exxon began fertilizer application in
early August to approximately 70 miles of physically washed beach in Prince William Sound.
Increasing the biodegradation rate of oil at this point was very important in order to achieve maximal
degradation before winter conditions curtailed cleanup operations.
Passage Cove Demonstration
The Passage Cove study was initiated as the definitive technical support site for the large-scale
application of fertilizers. Nutrient solution and oleophilic plus slow-release fertilizer were applied
to Passage Cove in late July 1989. Oil degradation was again assessed using analytical chemistry
(changes in oil residue weight and composition) and microbiological techniques (most probable
number counts and respiration). Visual observations were again recorded.
The same ecological monitoring program for direct toxicity and eutrophication was implemented,
and possible adverse ecological effects resulting from direct toxicity of the fertilizer were also
addressed. Caged mussels were again placed offshore of the field site to evaluate physical removal
of the oil.
Laboratory tests conducted to supplement the field testing at Passage Cove included: 1) use of
microcosms to test the effectiveness of subsurface oil bioremediation under more controlled
conditions than the field; 2) assessment of the incorporation of fertilizers into the food web using
stable isotopes; and 3) measurements of microbial activity at Passage Cove.
Supporting Laboratory Studies
In addition to the laboratory studies conducted to specifically supplement the activities at the
two field sites, experiments were also conducted to investigate more general parameters. These
included: 1) laboratory biodegradation screening evaluations to determine if degradation in Prince
William Sound was limited by nutrient availability; 2) respirometric analyses to obtain additional
information on the effect of oleophilic fertilizer for enhancing the degradation of different
concentrations of artificially weathered oil; 3) experiments to determine the mechanism by which
oleophilic fertilizer enhanced oil degradation; 4) tests to evaluate the rock-washing characteristic of
14
-------
INTRODUCTION
oleophilic fertilizer under conditions that precluded biological activity; and 5) acute toxicity tests of
oleophilic fertilizer to fish, invertebrates, and algae.
Follow-Up Research - Winter 1989/1990
The bioremediation research conducted during the summer of 1989 showed enhancement of
biodegradation through specific introduction of fertilizers. Several major problems, however, needed
to be addressed to assist in interpreting these results and assessing future fertilizer applications.
Laboratory studies were therefore conducted during the winter of 1989/1990 to complement the
summer of 1989 research and provide insight for the summer of 1990 field studies. Three study areas
were examined during the winter of 1989/1990:
• Mechanism of action of oleophilic fertilizers and optimization of fertilizer application for
maximal rate of oil degradation;
• Toxicity of fertilizers to marine biota; and
• Eutrophication modeling in select bays and areas of Prince William Sound
Studies on the mechanism of action of oleophilic fertilizers were conducted using microcosms.
Changes in oil composition were used as the indicator of oil degradation. Experiments for optimizing
fertilizer application included fertilizer-specific activity studies and alternative treatment application
scenarios.
Additional sampling was conducted in Passage Cove in November 1989, sediment chemistry data
were obtained and oil degradation rates were analyzed. Tests were conducted on compounds such as
oleic acid or laureth phosphate, and initial studies with a soybean oil product for enhanced
degradation rates.
To determine the toxicity of fertilizers to marine biota, the relative toxicity of urea/ammonia
and lauryl phosphate was examined in standard laboratory tests. Acute and chronic toxicity tests were
conducted on several INIPOL exposure regimes, and a literature review was conducted on ammonia
toxicity to marine invertebrates.
To investigate the possibility of eutrophication, a eutrophication model, EUTRO4, was instituted
for Passage Cove and Snug Harbor. Residence times for nutrients in both bays were calculated, and
15
-------
INTRODUCTION
eutrophication potential was determined. A number of nutrient loading scenarios were tested to
determine the potential for eutrophication under worst case conditions.
FIELD OPERATIONS - SUMMER 1990
Monitoring Program
As a result of summer 1989 and winter 1989/1990 studies, the 1990 cleanup plan for the
remaining oil-contaminated shorelines in Prince William Sound that was agreed upon by the U.S.
Coast Guard, Exxon, EPA, the Alaskan Department of Environmental Conservation (ADEC), and
other resource agencies and landowners involved bioremediation as an integral part, both as a primary
and secondary cleanup method. However, further substantiation of bioremediation effectiveness was
required.
To provide this information, a joint bioremediation monitoring program was conceived and
implemented by scientists from Exxon, EPA, ADEC, and the University of Alaska using Exxon
logistical and resource support. Three beaches that were part of the cleanup plan were selected for
monitoring. Changes in oil chemistry, dissolved oxygen, and nutrient concentrations were monitored
in surface and subsurface samples, as well as increases in the number of oil degraders, hydrocarbon
mineralization activity, and toxicity to bioassay test species. The results from this study are reported
under separate cover.
To complement this monitoring program, additional experimental research in the field was also
planned. The intent of the research was to strengthen the success of bioremediation and optimize its
effectiveness under different field conditions. The beaches selected for experimentation were Disk
Island and Elrington Island. Additional laboratory studies were conducted to complement and verify
the field studies.
Disk Island Demonstration
The Disk Island study was designed to determine the specific activity of fertilizer-enhanced
biodegradation, or the extent of rate enhancement per quantity of nutrients released. To evaluate the
effect of applying different concentrations of fertilizer on biodegradation, samples were taken for
16
-------
INTRODUCTION
oil chemistry (changes in oil residue weight and composition), increases in oil-degrading microbial
activity and biomass, and resulting nutrient concentrations.
A scaling experiment was also conducted at an uncontaminated cobble beach in association with
the Disk Island study, to assure that the application of fertilizer granules would release nutrients into
a defined plot size. Subsurface wells were used to measure nutrient release following fertilizer
application.
Elrington Island Demonstration
The Elrington Island study was initiated to investigate the effects of different applications of
nutrient solution on stimulating oil biodegradation. To assess potential increases in biodegradation,
analytical chemistry analyses (changes in oil residue weight and composition), increases in oil-
degrading microbial activity and oil-degrader biomass, dissolved oxygen uptake, and uptake of
inorganic nutrients were evaluated. The ability of bioremediation to work effectively on oil in the
beach subsurface (0.3 to 1.0 m depths) was emphasized.
17
-------
SECTION 2
FERTILIZER SELECTION AND CHARACTERISTICS
BACKGROUND
An important aspect of this project was the selection of fertilizers for the field tests. The goal
was to find fertilizer formulations which would either slowly release nitrogen and phosphorus
nutrients or keep the nutrients in contact with surface microbial communities for extended time
periods. In addition, consideration was given to formulations amenable to practical and inexpensive
application to contaminated shorelines, in light of the possibility of future large-scale applications.
Three general types of fertilizer were selected for testing:
1) Solid, slow-release fertilizer, in which nutrients would be slowly released from a point source
and distributed over the beach surface by tidal action. The product had to deliver sufficient
quantities of nutrients for several weeks.
2) Liquid oleophilic fertilizer, in which nutrients would "dissolve" into the oil covering the rock
and gravel surfaces. This sequesters nutrients in the oil phase, facilitating bacterial growth on
the surface over sustained periods. Nutrient distribution over the beach material would be
accomplished in the original fertilizer application.
3) Fertilizer solutions, in which inorganic nitrogen and phosphorus would be dissolved in seawater
and applied through spray irrigation (a fixed sprinkler system). This type of application would
introduce nutrients into the oiled beach material, particularly oil below the surface, in a defined,
controlled, and reproducible way.
Several commercially available fertilizer formulations that satisfied these requirements were
selected and their nutrient-release characteristics determined. A small study was also conducted to
see how specific solid fertilizer formulations physically behaved under field conditions.
DESCRIPTION OF FERTILIZER FORMULATIONS
The fertilizer formulations selected for initial testing prior to use in the field are described
below.
18
-------
FERTILIZER SELECTION
WOODACE Briquettes
This fertilizer formulation contains isobutyraldehydediurea (IBDU),1 a chemical that
spontaneously hydrolyzes into isobutylaldehyde and urea. This process is responsible for the
characteristic slow release of nitrogen. Hydrolysis is temperature dependent, and while slower at
lower temperatures is still significant. The source of phosphorus is Linstar, a citric acid soluble
phosphatic fertilizer developed by Mitcubichi Chemical Corp. Each briquette weighs approximately
17 grams, on average, and has a specific gravity of 1.5 to 1.8. This fertilizer has an N:P:K ratio of
14:3:3.
PAR EX Granules
This fertilizer is granular, with an N:P:K ratio of 24:4:12, formulated to give an immediate and
sustained release of nutrients. The product used was PAR EX, produced by Estech, Inc. All of the
nitrogen is derived from ammonium phosphate, urea, and IBDU. A minimum of 45% of the nitrogen
is derived from the IBDU. The available phosphorus is derived from potassium magnesium phosphate
and iron is also present as ferrous sulfate. The granules have a specific gravity of approximately 1.3.
OSMOCOTE Briquettes
Manufactured by SIERRA CHEMICAL, this fertilizer contains urea formaldehyde as the
nitrogen source in a slow-release formulation created by thermoplastic resin (natural plant product
binder) encapsulation. The urea formaldehyde released from the briquettes must be biologically
hydrolyzed to produce ammonia. Phosphorus is present as calcium phosphate and iron is also present
as ferrous sulfate. Each briquette weighs 21 grams, on average, and the fertilizer N:P:K ratio is
20:10:5.
1Isobutyraldehyde Diurea (IBDU) is a condensation product of urea. The reaction can be carried
out both in aqueous solution and between solid urea and liquid aldehyde as follows:
CHS CHS NHCONH2
> CHCHO + 2 NH2CONH2 - > CHCH < + H2O
CH3 CH3 NHCONH2
19
-------
FERTILIZER SELECTION
MAGAMP
This fertilizer formulation contains magnesium ammonium phosphate (MAGAMP), which is
sparingly soluble in water. It is made by Martin Marietta Magnesia Specialties using specialized
manufacturing procedures. The fertilizer can be cast into different shapes (granules, briquettes,
bricks) of varying densities, and will slowly release ammonia and phosphate when submersed. This
fertilizer has an N:P:K ratio of 7:40:0 and is available in briquettes that weigh 209 g each, on average,
and bricks of 8 or 40 Ibs.
SIERRA CHEMICAL Granules
This fertilizer formulation (also referred to as CUSTOMBLEN) consists of inorganic nutrient
sources (ammonium nitrate, calcium phosphate, and ammonium phosphate) contained in a vegetable
oil coating (polymerized by reaction with a diene). The coating gives the fertilizer its slow-release
characteristic. The N:P:K ratio is 28:8:0 and the granules have a specific gravity of approximately 1.8.
INIPOL Oleophilic Fertilizer
The only oleophilic fertilizer available in sufficient amounts to use in a scaled-up operation was
the Elf Aquitaine (France) product, INIPOL EAP 22. This is a mixture of nutrients encapsulated by
oleic acid (the external phase). It is theorized that oleic acid and surfactants in the fertilizer
formulation cause the nutrients to become sequestered in the oil phase, preventing rapid release of
the nutrients into the aqueous phase and subsequent washout. INIPOL is a yellow liquid with a
specific gravity of 0.996, a viscosity of 250 cSt, a pour point of 1 PC, and a flash point of >100°C.
The N:RK ratio is 7.3:2.8:0.
The main ingredients in INIPOL are oleic acid and urea, plus chemicals to maintain them in a
microemulsion. The chemical composition of INIPOL is given in Table 2.1. The product is designed
to initially stimulate oleic acid-degrading bacteria. The quantity of nitrogen and phosphorus present
is sufficient to allow the natural oleic acid degraders found in the receiving environment to consume
all of the oleic acid carbon present in the INIPOL. Once the added oleic acid is consumed, and the
numbers of oleic acid-degraders have increased substantially, oil biodegradation is thought to
commence. It is not exactly clear why this pattern occurs, but many oleic acid-degrading bacteria
are known to degrade petroleum hydrocarbons. Elf Aquitaine representatives have suggested that the
20
-------
FERTILIZER SELECTION
oleic acid-degrading microorganisms may die once they reach a certain density, creating a natural
recycling of nutrients through mineralization of this dead biomass.
TABLE 2.1. INIPOL EAP 22 CHEMICAL COMPOSITION
Chemical Purpose
Oleic acid Oleophilic Phase (Continuous)
Primary Carbon Nutrient
Tri[laureth-4]phosphate Phosphate Nutrient Surfactant
2-Butoxy-l Ethanol Co-Surfactant
Emulsion Stabilizer
Urea Nitrogen Nutrient
Water Hydrophilic Phase
All of these fertilizers were tested in the laboratory, field, or both in the summer of 1989 prior
to field application, in order to determine their nutrient release characteristics and subsequent
suitability for field testing. The methods and results from the tests are presented below.
METHODS
To simulate the effect of tidal activity on the nutrient release characteristics of the selected
fertilizers, static and intermittent submersion laboratory tests were developed. These tests were
conducted to determine which fertilizers would satisfy the minimum requirements for field testing.
The static test represented high tide, submerged conditions when turbulence was at a minimum. The
intermittent submersion test represented the turbulent condition, simulating water moving from low
to high tide and back. For the INIPOL fertilizer, varying amounts of fertilizer were applied to
determine the best application rate and nutrient retention characteristics. Excess INIPOL fertilizer,
when not adsorbed to the oiled surfaces, loses its solution properties on contact with water and
releases urea very rapidly.
21
-------
FERTILIZER SELECTION
Static Tests
Tests for granular fertilizers were conducted by sealing a specific weight of the granules in
porous, 100% cotton cloth bags, or placing a specific weight of the granules at the bottom of the
beaker, and submerging in a beaker of artificial seawater (Instant Ocean, supplied by Aquarium
Systems, Inc.). The briquettes were tested by placing them in the beaker and submerging with the
seawater. The beakers were maintained at 1S°C without mixing. According to an established
schedule, water was decanted out of the beakers and replaced with new seawater. Consequently, the
effect of the water exchange on the quantity of nutrients released could be assessed. The amount of
total Kjeldahl nitrogen (TKN) (EPA method 365.4), ammonia (EPA method 350.1), nitrate (EPA
method 353.1), and total phosphorus (EPA method 365.4), were monitored, depending on the fertilizer
tested.
To determine the effect of microbial activities on the release of ammonia from the TKN leached
out of the IBDU briquettes, flask studies were conducted in which briquettes were covered with three
types of water, deionized, sterile seawater (filter sterilized - 0.22/j), and non-sterile seawater. The
experiments were run at two different temperatures (9°C and 21°C) and the amount of ammonia
released over time determined. To verify that using Prince William Sound beach material would
produce similar results, a test was also conducted with the beach material.
The oleophilic fertilizer was tested by applying the fertilizer with an air paint sprayer to the
surface of oil-contaminated rocks obtained from Prince William Sound. The rocks were then covered
with seawater within 5 minutes and accumulation of TKN, NH4, NO3, and NO2 in the water over
time was monitored.
Intermittent Submersion Tests
A rocker table equipped with a 8 cm long, 10 cm wide pipette tray was used for the intermittent
submersion tests. A bag containing fertilizer granules or the fertilizer briquettes alone was buried in
clean Alaskan beach material at one end of the tray. The air paint sprayer was again used to apply
the oleophilic fertilizer directly on the beach material. The rocker table was maintained in a cold
room at 15°C, generally operated for 120 minutes, and was sampled for nutrients on a set schedule.
The table was then stopped, the water drained, and the beach material allowed to remain undisturbed
for 4 hours. The beach material was covered again with water, the table was operated for one hour,
22
-------
FERTILIZER SELECTION
and sampling was repeated. The process was repeated several times. Other operational variations
were tested, depending on the intent of the specific experiment.
Field Tests
To test nutrient release characteristics of magnesium ammonium phosphate (MAGAMP) under
field conditions, 8 Ib and 40 Ib MAGAMP bricks were placed on a sand and gravel beach at Snug
Harbor. Sampling stations were placed downgradient of the fertilizer briquettes. The weight of the
bricks was expected to minimize their movement by tidal and wave action. Nutrient samples were
collected 12, 24, and 96 hours after placing the bricks on the beach.
RESULTS
The results of the field and laboratory tests for the three general types of fertilizers chosen for
testing are detailed below. Figure 2.1 diagrams the overall testing scheme for fertilizer selection.
WOODACE Briquettes
The cumulative nutrient release pattern for ammonia, total phosphorus, and TKN from static
tests with the IBDU briquettes is shown in Figure 2.2. Release rates per day for each nutrient are
shown in Figure 2.3. The release of ammonia from urea will only occur in the presence of a
biological agent. Bacteria were not purposely introduced, and consequently ammonia release should
be minimal. The release rate was relatively constant after the initial surge, probably caused by the
powder on the briquettes. Release of TKN accounted for 17% of the total available nitrogen after
17 days, 31% after 60 days, and 45% after 140 days. Release rates for total phosphorus, although
somewhat variable, were constant over time. Total phosphorus continued to be released over the test
period, with values averaging approximately 3 mg/L and a cumulative release of 36%. The variability
was probably associated with analytical error. The schedule of water exchanges appeared to have a
negligible effect on release rates for this fertilizer (data not shown).
It is apparent that small amounts of ammonia and phosphorus are released with each 24-hour
soaking of the briquettes. The average amount of ammonia released per day is approximately 100-
fold higher than background levels in Snug Harbor waters (see Section 6: Nutrients). However,
considering the rapid dilution of ammonia that will occur in the field following release from
23
-------
Fertilizers Tested
WOOOACE Briquettes
PAR EX Granules
OSMOCOTE Briquettes
MAGAMP
SIERRA CHEMICAL Granules (CUSTOMBLEN)
Oleophilic (INIPOL EAP 22)
Static Tests
Intermittent
Submersion Tests
WOODACE Briquettes
PAR EX Granules
OSMOCOTE Briquettes
MAGAMP
SIERRA CHEMICAL Granules
(CUSTOMBLEN)
Oleophilic (INIPOL EAP 22)
I
Field Tests
WOOOACE Briquettes
MAGAMP
Oleophilic (INIPOL EAP 22)
WOODACE Briquettes
MAGAMP
Fertilizers Recommended for Field Testing
Initial 1989 Fieid Demonstration (Snug Harbor): WOODACE Briquettes; INIPOL
EAP 22
> Second 1989 Field Test (Passage Cove): SIERRA CHEMICAL Granules
(CUSTOMBLEN) and INSPOL EAP 22 together
• initial 1990 Field Test (Disk Island): SIERRA CHEMICAL Granules (CUSTOMBLEN)
»Second 1990 Fieid Test (Eirington island): Inorganic Nitrogen and Phosphorus
(granular)* Mixed with Saawater
• Product Not Tested in 1989 Fertiliser Test Studies
Figure 2.1. Diagram of Fertilizer Testing for Nutrient Release Characteristics.
24
-------
100
90 -
80 -
8 70-1
C
- 60 H
= 50-
•
1 40 -
30 -
Each point represents one water exchange
TKN .. .
—•••••••*
—""
40
60
80
100
120
140
Time (Days)
Figure 2.2. Cumulative Release of Ammonia, Total Kjeldahl Nitrogen (TKN),
and Total Phosphorus from WOODACE Briquettes in Static
Flask Experiments.
160
Tim* (Days)
Figure 2.3. Daily Nutrient Release Rate of Ammonia (NH4), Total
Phosphorus (TP), and Total Kjeldahl Nitrogen (TKN) from
WOODACE Briquettes.
25
-------
FERTILIZER SELECTION
briquettes, it would be unlikely to measure any increased concentrations of ammonia (discarding
contributions from the TKN) in the field as a result of fertilizer application. The total amount of
ammonia released is only a small fraction of the total nitrogen available in the formulation.
To determine if the TKN was a source of ammonia to bacteria under natural conditions,
briquettes were soaked for 3 succesive 1 -hour periods (water was changed for each period) using 2
different temperatures and 3 different sources of water. The results are shown in Figure 2.4.
Significant amounts of ammonium were released under all conditions. The lowest amount of ammonia
was released with filtered seawater (containing no microorganisms). Since this was less than that
released in the presence of deionized water, it suggests that there was an ionic strength effect on the
ammonia release. Temperature had little effect on release rates in these media. However, with
unfiltered seawater, considerably more ammonia was released, particularly at the higher temperatures.
This indicated a possible biological effect on ammonia release, presumably through the microbial
breakdown of the TKN fraction.
Nutrient release from the IBDU briquettes was also tested using the intermittent submersion test.
In general, concentrations of the nutrients and TKN released were minimal.
Under all conditions, the physical integrity of the IBDU briquettes was excellent, with very little
change in shape and consistency after being submersed in water for four months. A simple freeze/
thaw experiment was also conducted on the WOODACE briquettes. The experiment consisted of
alternately freezing and thawing submerged and non-submerged briquettes, then weighing and
visually observing changes. The results indicated good durability. The briquettes appeared to be a
good choice for field application.
Studies were also performed on the movement of briquettes broadcast on the beach. Results
showed that unconfined briquettes will not retain their original position and were redistributed after
several tidal cycles. Greater redistribution occurred on sand and gravel beaches as compared to cobble
beaches. Due to the redistribution of briquettes, this form of fertilizer application is best applied in
containers which hold the briquettes in place. Unconfined briquettes may be of some limited use on
sheltered cobble beaches, where wave action may have less influence on the beach.
26
-------
350 r
300 •
•? 250 '
ot
*•*
o
01
u
o
O
50 '
Deionized Water
Filtered Seawater
Unfiltered Seawater
1 2
Soaking Time (Hours)
a
350
300
250
£ 200
|
2 150
o
U
100
50
21°C
1 2
Soaking Time (Hours)
Figure 2.4. Ammonia Release From IBDU Briquettes at 9 and 21 Degrees
Centigrade in 3 Different Water Sources.
27
-------
FERTILIZER SELECTION
PAR EX Granules
A typical cumulative nutrient release pattern for ammonia, phosphate, and TKN from PAR EX
granules in bags from one of the static tests is shown in Figure 2.5. High amounts of nutrients were
released initially, followed by a very slow release. If the experiment was repeated with the granules
layered on the bottom of the beaker (i.e., no bag to contain the granules), a steeper release pattern was
observed. The number of water exchanges had a much greater effect on these experiments than when
the fertilizers were placed in bags (data not shown).
100
90 -
? 80 -
]„-,
c
. 60 -
c
I 50-1
I
>
e
40 -
30 -
20 -
10 -
0
Each point represents one water exchange
TKN
H
Ammonia
....... .». . 1 .A AA..A
20
40
Tim* (Days)
60
Figure 2.5. Cumulative Release of Ammonia and Total Kjeldahl Nitrogen (TKN) from
IBDU Fertilizer Granules Contained In Bags In Static Flask Experiments.
Additional studies also showed that if the granule bag volume is reduced relative to the bag
surface area, slightly more nutrients were released (Table 2.2). Thus, the more water passing over the
granules, the higher the release rate. As expected, the results of the test using Prince William Sound
beach material were similar to the test without rocks.
28
-------
FERTILIZER SELECTION
TABLE 2.2. TOTAL KJELDAHL NITROGEN (TKN) RELEASED UNDER STATIC
CONDITIONS FROM IBDU GRANULAR FERTILIZER IN BAGS WITH DIFFERENT
SURFACE TO VOLUME RATIOS
% of Cumulative
Available Nitrogen
Bag Volume/Surface Area
(cubic cm/square cm)
1.8
1.3
0.6
Fertilizer
Weight (g)
893
256
32
Seawater
Volume (ml)
4,800
1,400
450
(TKN) Released in:
24 Hours
34
36
41
45 Days
38
42
49
OSMOCOTE Briquettes
The cumulative nutrient release pattern for ammonia, phosphate, and TKN for static tests is
shown in Figure 2.6. After 2 months of testing, approximately 25% of the available nitrogen was
released, primarily as TKN, whereas almost 60% of the phosphate was released over this time period.
The physical form of the briquette was unstable, flaking soon after initial submersion and further
decomposing over time. A dye within the briquette turned the water green with each water exchange.
These briquettes, despite their good nutrient release characteristics, appeared unsuitable for long-term
use in the field.
MAGAMP Briquettes
When high-density and low-density MAGAMP briquettes were tested, the low-density briquette
(about half the weight of the high density) disintegrated almost immediately upon submersion in the
seawater, and consequently was not tested. Release of ammonia, TKN, and phosphate from the high-
density briquettes was slow and constant (Figure 2.7). After 10 days, only 1.5% of the available
nitrogen was released. At 75 days, approximately 5% to 6% had been released. The release rate
appeared to be independent of the number and volume of water exchanges. The high-density
briquettes appeared to be very durable. When MAGAMP powder was tested, it congealed to a putty-
like consistency soon after the experiment was started. Accumulative release of ammonia was about
the same as the briquettes.
29
-------
100
90 -
Each point represents one water exchange
20
Total Phosphorus
40
60
80
100
120
Time (Days)
Rgure 2.6. Cumulative Release of Ammonia, Total Kjeldahl Nitrogen (TKN),
and Total Phosphorus from OSMOCOTE Briquettes in Static
Flask Experiments.
100
90-
~ 80-
60-
1 50H
40-
Each point represents one water exchange
TKN
Ammonia
Total Phosphorus
100
Time (Days)
Figure 2.7. Cumulative Release of Ammonia, Total Kjeldahl Nitrogen (TKN),
and Total Phosphorus from MAGAMP Briquettes in Static Flask
Experiments. 30
-------
FERTILIZER SELECTION
Results from the intermittent submersion tests showed the same low release rates. A high burst
of ammonia seen within the first half hour of the test was explained by initial flaking of the
briquettes. Flaking at later times did not occur.
MAGAMP can be formulated into dense bricks. Bricks weighing 8 and 40 Ibs were field tested
as an alternative physical form for fertilizer application. These bricks are useful in the field because
of their positional stability on the beach without an anchoring device. However, these bricks could
not be produced in large quantity and were, therefore, unavailable for use in any of the fertilization
studies. However, because of their potential promise as an alternative physical form of fertilizer,
separate beach mechanics studies were conducted to evaluate the nutrient release and distribution
characteristics of these bricks. The very slow release of ammonia from MAGAMP made it necessary
to determine, under controlled field conditions, if nutrient release could be detected in the field.
Beach pore water sampling stations were placed down-gradient from MAGAMP bricks as shown
in Figure 2.8. Samples for ammonia analysis were collected 12, 24, and 96 hours after initial
placement of the bricks. The data are shown in Figure 2.9. The 40 Ib brick released up to 138 n of
nitrogen as ammonia, with the highest concentrations directly down-gradient from the block.
Significant quantities of ammonia were observed up to 2 m from the bricks at low tide. Ammonia
also appears to be well distributed around the brick. The data suggest that this type of point source
for fertilizer application could be quite useful in the future.
SIERRA CHEMICAL Granules
Figure 2.10 shows the cumulative nutrient release pattern with a variable exchange rate (5
exchanges on the first day, 2 exchanges per day thereafter through the 10th day, daily thereafter
through the 40th day, and then every other day. The amount of nitrogen (ammonia and nitrate)
released after 80 days was 77% of the total available nitrogen. When the frequency of water
exchanges was doubled, 95% of the total nitrogen (approximately half ammonia and half nitrate) was
released after 80 days. The shape of the release curves were similar. This effect of water exchanges
was probably due to the mechanical agitation of the system prior to each exchange.
31
-------
Top Transect
Middle Transect
Left
Position
Fertilizer
Brick
(TM)
0.5m
(M\\* — 1.0m — *^MM^-« — 1.0m — *^Mm
i^
1.0m
|
Middle
Position
Mean High Tide
Mean Low Tide
Right
Position
Figure 2.8. Sampling Point Locations for Magnesium Ammonium
Phosphate Fertilizer Field Test.
32
-------
12 hours alter placemen!
24 hours alter placement
96 hours alter placement
u>
u>
140 —1 4 ° — I
5:f
f z
11 70 —
!*
h
—
n
-
2.0 —
-
.
4.0 —
2.0 —
FIRM n
° TM
140 —
ff
1 | 70 —
?l
s|
4.0 —
^—
•H
S
3 2.0 —
;
| _v_
!
N
k
5 _
Ml* MM Mn
140 —
ff
I E 70 —
£
2
S *
4.0 —
«
*
5
>
S
«
<
!
!
"
V
S
s
s
s
«.
•t
a —
|
i 2.0 —
5
:
: ~~
s
,
s ^
_
nm
TM TM
4.0 —
~
2.0 —
,B | ,
r-
1
il H_
E« H» 1
ffW^^
HM_
ML MM MR ML MM MR
4.0 —
_..
2.0 —
1^1 r-ilS3«l ml n
BL BM BR
MAP 1
40- Ib. Brick
il
ri^l n^ •
§
1
BL BM BR BL BM BR
MAP 2 MAR s
8- Ib. Brick g. ,b Brlck
Figure 2.9. Magnesium Ammonium Phosphate Fertilizer Test: Ammonium Concentration in Beach Pore Water at 12,24 and 96 Hours After
Placement of Fertilizer (Sampling Locations Given in Figure 2.8.).
-------
Each point represents one water exchange
20
40 60
Tim* (Days)
80
100
Rgure 2.10. Cumulative Release of Ammonia and Nitrate from SIERRA
CHEMICAL Granules In Static Rask Experiments.
34
-------
FERTILIZER SELECTION
INIPOL Oleophilic Fertilizer
The results from static tests with this fertilizer are shown in Figure 2.11. All of the nitrogen
(>100%) was released within the first few water exchanges. The release of more nitrogen than was
theoretically thought to be in the INIPOL formulation suggests that manufacturer's specifications for
this batch of INIPOL were incorrect or TKN was present on the oiled beach material.
An intermittent submersion test was conducted on the oleophilic fertilizer applied to oil-covered
Prince William Sound beach material. The data are shown in Table 2.3. Within 5 minutes after
INIPOL-treated oiled rocks were covered with seawater, over 60% of the available nitrogen was
released as TKN. However, following this initial burst, TKN appeared to be released more slowly
(i.e., very little increase in TKN occurred over the next 115 minutes). After decanting the water off
the beach material, allowing it to sit unsubmerged for 6 hours and recovering the rocks with water,
only 8.3% of the available nitrogen was further released as TKN. Concentrations of ammonia and
phosphate released were quite low, but generally followed the same pattern as the TKN.
Allowing the fertilizer to remain in contact with the oil for 6 hours prior to the addition of water
did not change the nutrient release patterns. This suggests that the amounts of nutrient which
sequester with the oil (i.e., not washed off) are incorporated very soon after fertilizer application.
It is unclear, however, why the nutrient did not sequester with the oil in the static experiments.
In addition, mixing the beach material as the INIPOL was applied, or warming the INIPOL to
25°C before application, did not significantly change the amount of nitrogen released in the first few
minutes. It is unclear why the static tests did not show similar results.
SUMMARY AND CONCLUSIONS
From these studies, three fertilizer formulations were selected for field testing: the WOODACE
slow-release fertilizer briquette formulation, the CUSTOMBLEN fertilizer granule formulation, and
the INIPOL oleophilic fertilizer formulation.
It was concluded that bags of WOODACE fertilizer briquettes would be used in the initial field
demonstration for slow-release fertilizer. This fertilizer had good nutrient release characteristics,
excellent durability in the field, and ready availability. Given the time constraints of the
bioremediation field demonstration project, this fertilizer was a reasonable first choice.
35
-------
120
I
TKN
Total Phosphorus
———••
Ammonia
Each point represents one water exchange
Time (Days)
I
8
10
Rgure 2.11. Cumulative Release of Ammonia and Total KJeldahi Nitrogen
(TKN) from INIPOL EAP 22 In Static Flask Experiments.
36
-------
FERTILIZER SELECTION
TABLE 2.3. RELEASE OF AMMONIA, TOTAL KJELDAHL NITROGEN (TKN), AND
TOTAL PHOSPHORUS (TP) FROM INIPOL EAP 22 DURING INTERMITTENT
SUBMERSION EXPERIMENT
Minutes from 5 Minute 6 Hour
Start of Experiment Contact Time' Contact Time*
Ammonia Released (mg N/L)b
5 1.1 0.5
15 1.1 0.4
30 1.4 0.5
60 1.3 0.7
120 1.4 0.7
510C 0.2
540 0.1
600 0.0
TP Released (mg P/L)
5 1.3 1.4
15 1.2 1.2
30 1.0 1.1
60 1.5 1.3
120 1.0 1.0
510C 1.1
540 0.9
600 0.9
TKN Released (mg N/L)b
5
15
30
60
120
510C
540
600
24.6
26.1
27.2
32.5
29.4
4.6
4.6
4.3
29.8
34.8
35.5
34.3
32.3
* Time between fertilizer application and initial submersion
b Initial concentration of nitrogen = 57 mg/L
0 Water drained; beach material remained unsubmerged for 6 hours; seawater replaced
37
-------
FERTILIZER SELECTION
Recognizing that bagged briquettes could not be produced in sufficient quantities for large-scale
application, slow-release fertilizer granules (SIERRA CHEMICAL, CUSTOMBLEN) were selected
for the second field test, since they could be easily broadcast over the beach surface in a large-scale
operation. The granules had good nutrient release characteristics but were not as long lasting or as
durable as the briquettes.
Tests with the oleophilic fertilizer, INIPOL EAP 22, showed that it did not retain nutrients on
the surface of oil, losing approximately 90% to 100% of the available nitrogen in the first minutes
following application. However, more elaborate microcosm studies using more realistic environmental
conditions (Roger Prince, Jim Bragg, Exxon) have shown that only approximately 50% of the nitrogen
is lost in the first 24 hours and small amounts are released thereafter for 2 to 3 weeks. Other
published studies on this oleophilic fertilizer have shown it to work well on sandy beaches, but similar
testing has not been done with cobble beaches. Therefore, its use on the cobble beaches found in
Prince William Sound represented a new application.
38
-------
SECTION 3
TEST PLOT DESIGN AND SAMPLING
STUDY AREA DESCRIPTIONS
Beaches in Snug Harbor and Passage Cove were selected in 1989 as test sites for the application
of the two slow-release fertilizers (WOODACE briquettes and CUSTOMBLEN granules), and the
oleophilic fertilizer (INIPOL). In addition, the application of a fertilizer solution was also tested at
the Passage Cove site.
Selection of these test sites was based on the following criteria:
• Typical shoreline of Prince William Sound; i.e., cobblestone overlain on mixed sand and
gravel beaches with a gradual vertical rise
• Sufficient area for testing with fairly uniform distribution of beach material
• Protected embayment with adequate staging areas
• Uniform oil contamination
• Minimal impact from freshwater inputs
In the summer of 1990, beach areas at Disk Island were selected as test sites for application of
CUSTOMBLEN granules, and beach areas at Elrington Island were selected for the application of
nutrient solution.
Snug Harbor
Snug Harbor is located on the southeastern side of Knight Island. The shoreline utilized for the
demonstration is located on the western side of this harbor (Figure 3.1). This shoreline represented
a beach with oil contamination that approximated the degree of contamination remaining after a
moderately oiled beach had been physically washed by the Exxon process. This approximation was
required because at this time there were no beaches available for testing in Prince William Sound that
had been physically washed. The area is surrounded by mountains, reaching an elevation of
approximately 2,000 feet, with steep vertical ascents. Major sources of freshwater runoff are from
precipitation and snowmelt, which is typical of islands in Prince William Sound. Although other
39
-------
PASSAGE /
COVE /
Prince William Sound
3.1. Location of Field Sites. Alphanumeric Codes Were Those
Designated by Exxon and the Prince William Sound Shoreline
Cleanup Committee.
40
-------
PLOT DESIGN AND SAMPLING
shorelines in Snug Harbor were heavily contaminated with oil, it appeared that little oil was being
released to the water, thus minimizing the prospect of reoiling the beaches chosen as control and
treatment and control. Table 3.1 identifies the beach types and dimensions.
TABLE 3.1. PHYSICAL DESCRIPTION OF TREATMENT AREAS AT SNUG HARBOR
Beach
Eagle
Otter
Otter
Seal
Seal
Seal
Beach Type
Sand, gravel
Sand, gravel
Sand, gravel
Cobble
Cobble
Cobble
Length (m)
21
21
35
28
28
21
Depth (m)
12
12
12
12
12
8
Oil contamination in the test area was present as a continuous band along the length of the
beach. This band was approximately 1S to 20 meters wide and corresponded roughly with the average
boundaries of the high and low tides observed in Snug Harbor. To determine the approximate
distribution of oil on the beach, samples of beach material from one of the designated mixed sand and
gravel plots were taken on May 25, 1989. The samples were extracted, and the oil weight and
chemical composition were determined. Methods for the sampling and analysis are given in Section
4. The oil residue weights and ratios of nC17/pristane and nC18/phytane at two different depths are
shown in Table 3.2. It is clear that oil concentrations varied considerably, ranging from a high of
67,200 mg/kg of beach material to a low of 8 mg/kg of beach material. In general, higher
concentrations were found in the top 10 cm of the beach. Changes in the ratios, relative to fresh
Prudhoe Bay crude oil, were also apparent in some samples, indicating biodegradation of the oil.
Changes were quite variable, but it does appear that biodegradation may have been occurring at the
lower depths.
41
-------
PLOT DESIGN AND SAMPLING
TABLE 3.2. ANALYSIS OF OIL EXTRACTED FROM MIXED SAND AND GRAVEL
SAMPLES TAKEN FROM OTTER BEACH IN SNUG HARBOR ON MAY 28, 1989, TWO WEEKS
PRIOR TO FERTILIZER APPLICATION*
TOD (0- 10 cm^
Block
No.
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
Mean
StdDev
Residue
wt. (mg/kg)
100
29,000
30,100
6,070
2,030
6,600
1,440
1,030
7,600
9,820
658
67,200
1,560
8.190
12,242
+/- 18,556
nC17/
Pristaneb
0.8
1.6
1.5
1.5
1.2
1.2
1.1
0.8
1.4
1.5
1.5
1.6
1.0
L6
1.3
+/-0.3
nC18/
Phytaneb
1.0
1.9
1.7
1.8
1.5
1.7
1.4
1.1
1.7
1.8
1.9
1.8
1.3
L2
1.6
+/-0.3
Bottom HO-20
Residue
wt. (mg/kg)
253
18,300
296
2,600
37
97
365
469
412
512
8
9,280
9,620
8,100
45
538
80
622
125
1.790
2,169
+/-4,842 -t
nC17/
Pristane
0.9
1.6
1.0
1.5
0.8
1.1
0.8
0.9
0.9
1.2
1.0
1.6
1.5
1.6
0.9
1.3
0.9
1.1
0.9
L4
1.1
-/-0.3
cm)
nC18/
Phytane
0.8
2.0
1.2
1.9
1.0
1.3
1.1
1.1
1.1
1.5
1.2
1.8
1.8
1.9
0.9
1.6
1.3
1.4
1.3
L6
1.4
+/-0.3
* Otter Beach was divided into three equal zones lengthwise across the beach to represent high, mid,
and low tide areas. Each zone was divided into 7 equal blocks, and blocks were numbered from left
to right consecutively, starting with the high tide zone.
b nC17/pristane and nC18/phytane ratios in fresh Prudhoe Bay crude oil are approximately 1.7 and
2.0, respectively.
42
-------
PLOT DESIGN AND SAMPLING
Passage Cove
Passage Cove is located on the northwestern side of Knight Island (Figure 3.1). This site was
originally heavily contaminated with oil and was subjected to physical washing by Exxon. Even after
physical washing, considerable amounts of oil remained at this site, mostly spread uniformly over the
surface of rocks and in the beach subsurface. Pools of oil and mousse-like material were minimal on
the surface. Contamination was apparent to approximately 50 cm below the beach surface. All beach
areas tested were cobblestone set on a mixed sand and gravel base. Table 3.3 lists Passage Cove beach
designations and plot dimensions. This site served as the main reference beach for the large-scale
application of fertilizers and was used to evaluate the application of fertilizer solutions.
TABLE 3.3. PHYSICAL DESCRIPTION OF TREATMENT AREAS AT PASSAGE COVE
Beach
Raven
Tern
Kittiwake
Beach Type
Cobble over mixed
sand and gravel
Cobble over mixed
sand and gravel
Cobble over mixed
sand and gravel
Length (m)
28
35
28
Depth (m)
21
21
21
Disk Island
Disk Island is a small island located between Ingot and Knight Islands. Study site DI-067a is
located on the upper northwest corner of Disk Island (Figure 3.1). The site was moderately oiled and
not subjected to physical washing by Exxon. The beach was a fairly wide, low energy, cobble beach
with both surface and subsurface contamination. The test plots within the beach were located near
a pond, and large trees lined the back of the beach. In addition to the experiments conducted by
EPA's Bioremediation Project at Disk Island, experiments were also conducted at the same site by
EPA/ Cincinnati utilizing commercial microbial products. This study is discussed in Section 13.
43
-------
PLOT DESIGN AND SAMPLING
Elrington Island
Elrington Island was the southernmost test area utilized by the oil spill project, and is located
between Latouche and Evans Islands. Test site section ER-20 on the north end of Elrington Island
represented a narrow, cobble, high-energy beach with typical subsurface oil contamination (Figure
3.1). A berm separated the beach from the treeline. The surface of the beach was relatively free of
oil but extensive amounts of oil were found approximately 15 to 30.5 cm below the surface. This
section was therefore an appropriate beach to determine the effectiveness of fertilizer solutions in
enhancing biodegradation of the oil in the subsurface.
SAMPLING METHODS IN 1989
Snug Harbor
The beach sampling design was formulated to generate scientifically defensible conclusions
relative to the success of bioremediation (Figures 3.2 and 3.3). Unless otherwise specified, each test
site was divided into a series of plots within the beach. The plots were generally 21 to 35 m long and
12 m wide running the length of the beach. Plot size was controlled by the available beach (i.e.,
sections of relative uniformity), the extent of beach covered by the oil, and the prominence of certain
topographical features. Buffer zones of at least 5 m separated the plots. Larger buffer zones (>20 m)
were established between treated and reference plots to minimize cross-contamination. Cross-
contamination of nutrients between plots was not expected because of extensive dilution.
Zonal sampling (low, mid, and high tide) was used to uncover any effect due to seawater
coverage, rainfall, and freshwater runoff or temperature (exposure to sun, air, ocean, etc.). Sampling
intensity was gauged to minimize biological and physical effects. Sampling was designed to permit
collection of a sufficient number of samples to establish active biodegradation in each zone. If
degradation occurred in all three zones, comparisons could be made to establish trends from high to
low tide zones.
Blocks were derived from each intertidal zone by dividing the beach plot length into seven equal
segments (Figure 3.4). For three zones that created a total of 21 blocks. It was recognized that certain
sampling points on the beach were not representative of the entire beach. For instance, stream runoff
over one section of the beach might have been caused by an underlying solid rock outcrop near the
44
-------
Ecological Monitoring Stations
D 10.0m from mean tow tide
A 1 .Om from mean tow tide
O caged mussels
0 control site
• • • • " "• «*"•••••
Pnnce
William
Sound
Cobble Water-Soluble
srna//
Gravel Water-Soluble
Gravel Oleophil
Gravel Control
Figure 3.2. Sampling Locations at Snug Harbor, Knight Island, in
Prince William Sound, Alaska.
45
-------
Prince William Sound
Ecological Monitoring Stations
o 0.5m, 5.0m from mean low tide
A 0.5m from mean tow tide
o caged mussels
0 control site
I beach plots
(control)
bluff
Tern
(oleophilic fertilizer
+ fertilizer granules)
Kittiwake
(fertilizer solution)
Figure 3.3 Sampling Locations at Passage Cove, Knight Island, in Prince
William Sound, Alaska.
-------
Sampling Grid Cell
Figure 3.4. Sampling Design for Snug Harbor and Passage Cove
* Each group of seven blocks represents an intertidal zone (high, mid, and low)
-------
PLOT DESIGN AND SAMPLING
surface of the beach. Collecting seven samples for each beach stratum accounted for collection of a
nonrepresentative sample, and for the possibility of obvious gross error in a sample due to incorrect
sampling or analysis. In addition, seven samples were needed to ensure adequate power of statistical
tests.
Each block was sampled within 1 m x 1 m sampling grid cells. Therefore, although the number
of blocks within plots did not vary with beach size, the number of sampling grid cells within a block
for a particular plot did. The boundaries of each plot were established using rope secured to rebar
stakes. Squares of PVC pipe (1 m x 1 m) were used to delineate sampling grid cells.
For each designated sampling time, a sample was taken from one grid cell within each block for
all anticipated analyses. The sampling grid cell selection procedure used the following steps:
• The sampling crew began at the upper-left-hand corner of each block and picked two
numbers from a random number table that fell within the confines of the block. For
example, if the block size for the particular plot was 5 m in length and 3 m in width, the
table was used to choose a number from 1 to 5 to designate the distance along the block
boundary where the grid cell would be established from the starting point. The intersection
of the two randomly selected points was the upper-left-hand corner of the selected sampling
grid cell. The same sampling grid cell location was used for all blocks in a single plot during
a single sampling event.
• A 1 m x 1 m frame was placed on the beach in the designated grid cell and samples were
collected from the center of the frame (Figure 3.5)
• If a sample could not be taken at the center of the grid cell, a random number between 1 and
12 was chosen. These numbers represented positions on the face of a clock, in which 12
pointed to high tide line. The sampler then moved away from the center of the frame
toward the indicated clock position until an appropriate site was found within the sampling
frame. The sampling crew used judgment in many situations; for example, if a large boulder
was encountered, the site was discarded and the selection procedure was repeated.
This procedure was continued for each block until sampling was completed. Except for
nutrients, sampling site selection was the same for all sample analyses. All sampling was performed
at low tide. One to two days were required to sample all plots at each test site depending on the
height of low tide. When two days were required for sampling, only one-half of the untreated control
plots were sampled each day.
48
-------
PLOT DESIGN AND SAMPLING
Figure 3.5. A 1m by 1m Frame Was Placed on the Beach in the Designated
Grid Cell and Samples Were Collected from the Center of the
Frame.
The overall design of beach sampling efforts was statistically non-optimal. The major limitation
arose from the lack of duplicate beaches for each treatment and reference site. Measured effects were
attributable to both nutrient treatment effects and beach effects. It could not be determined
statistically whether an increased bioremediation rate at a site was due to either the treatment or to
a fortuitously good location, since these two variables were confounded. When only one treated beach
was successful, low confidence should be assigned to the result; however, because two types of
beaches and two types of treatments were used, when one or both treatments were successful on both
types of beaches, confidence in the results may be high.
Samples were taken on mixed sand and gravel beaches by placing a bottomless metal pail onto
the beach surface and working the bucket into the substratum. As small rocks were encountered that
prevented the pail from going further into the beach material, the material around the pail was
49
-------
PLOT DESIGN AND SAMPLING
excavated and the rock removed. If 50% of the rock was inside the perimeter of the pail, it was added
to the pail and included in the sample. If 50% or more was on the outside, it was excluded from the
sample. All large rocks (approximately 4 cm or larger in any dimension) were discarded from the
sample, since the amount of oil covering their surface was insignificant relative to oil in the entire
sample, and exclusion of these rocks reduced variability in substrate characteristics of the sample.
Once the pail was inserted approximately 13 to 14 cm into the beach material (using marks on
the inside of the pail), all beach material to 10 cm (including small rocks that protruded more than
50% above that mark), were included in the sample. Rocks that did not protrude more than 50%
above the mark were not included. All beach material removed from the sampler was placed in
washed and rinsed new paint cans. The contents of the paint can were then thoroughly mixed with
a steel spoon. A subsample of material sufficient to fill a 400 ml wide-mouth jar was taken from the
mixed sample. The jar and its contents were also subsampled for microbiology analysis and then
frozen.
Cobblestone beaches were sampled in a similar manner. The bottomless pail was worked down
over the cobble to the surface of the mixed sand and gravel layer under the cobble. Approximately
5 to 10 cobblestones were then sampled at random without regard to the extent of their oiling, and
the cobblestones were placed on large squares of aluminum foil. The stones were then double-
wrapped with the foil and placed in a ziploc plastic bag. The package was then frozen. Cobblestones
remaining in the bottomless pail were removed and the mixed sand and gravel sampled as above.
Passage Cove
The methods for test plot design and sampling for Passage Cove were identical to Snug Harbor,
but different plot sizes were delineated.
SAMPLING METHODS IN 1990
Sampling baskets (16 cm x 16 cm x 16 cm wire mesh containers with lids) containing
homogenized oiled beach material were prepared in the field for Disk Island and Elrington Island
(Figure 3.6). Oiled rock material from Prince William Sound beaches were sieved for material within
12.5 mm, and then manually thoroughly homogenized. The homogenized material was then placed
into each basket. Two types of baskets were created: one filled completely with homogenized oiled
50
-------
Beach Surface
Wire Mesh Top
Un-Oiled
Beach
Material
Oiled
Beach
Material
Subsurface Oil
Subsurface Oil
Un-Oiled
Beach
Material
Wire Mesh Basket
Figure 3.6. Sampling Basket.
-------
PLOT DESIGN AND SAMPLING
beach material (Disk Island) and the other with three layers of material: a 4 cm layer of oiled
homogenate material sandwiched between two 8 cm layers of homogenized clean beach material
(Elrington Island). The baskets were then implanted into fixed plots on a beach. The specific
methods used to prepare sampling baskets in the field are described below:
• Oiled beach material was sieved through a 12.5 mm sieve, and the material that passed
through the sieve was retained. The retained material was thoroughly mixed using hoes in
a large plywood box for at least 20 minutes.
• Enough sample was randomly removed by using a hand trowel to fill each sampling I-chem
jar. Six 1-kg samples of homogenized oiled beach material were collected from the
homogenization box. Five samples of the oiled homogenate and three samples of the clean
material from each plot were analyzed for oil chemistry; one of the oiled samples from each
homogenate was analyzed in triplicate. The material was removed from a different area
within the box each time. Samples were placed in I-chem jars, tightly capped, and stored
in a cooler with frozen gel pacs. If microbiological analyses were not performed on the
samples, dry ice was substituted for gel pacs to freeze the samples for oil chemistry analysis.
Samples were returned as quickly as possible to Valdez for subsampling (if necessary) and
analysis.
• If the mesh size of sampling basket exceeded the minimum size of the material, the basket
was lined with fiberglass screening. The basket was then filled with the homogenized
material and a lid was wired onto each basket.
• At Disk Island baskets were implanted in the beach plots with the top of the basket flush
with the beach surface. At Elrington Island baskets were buried approximately 15 to 20 cm
below the beach surface.
• For the subsurface oil layer baskets, the procedure outlined above was repeated, but clean
beach material was substituted for oiled material. At least 3 subsamples of the homogenized
clean beach material were collected. These baskets were filled as follows: a layer of clean
material; a layer of oiled material; and a layer of clean material. The oil layer was about 8
cm thick and the top and bottom layers about 5 cm. Baskets designed for monitoring
dissolved oxygen and nutrient uptake consisted of a well of unslotted PVC pipe placed
approximately 2 cm into the bottom clean layer, and another unslotted PVC well placed at
the interface of the oil layer and top clean layer. The wells were approximately 2 cm from
the sides of the basket.
Disk Island
A diagram of the Disk Island beach used for this experiment is shown in Figure 3.7. Six 3 m
x 3 m plots were marked with rebar and nylon rope (Figure 3.8). Four wells, each cased with 51 mm
diameter slotted stainless steel, were installed in the center of the four treatment plots. Each well was
0.8 m from the center of the plot. The two outer untreated control plots each contained two wells.
52
-------
trees
Figure 3.7. Disk Island Schematic of the Beach for the Fertilizer Specific
Activity Experiment.
tre£ line
Rgure 3.8. Disk island Fertilizer Specific Activity Plot Map and Rate of
CUSTOMBLEN Granule Application.
53
-------
PLOT DESIGN AND SAMPLING
Small, square sampling baskets containing homogenized oiled beach material were clustered around
the wells on each of the plots. Material excavated for basket placement flush with the beach surface
was removed from the test plot area. The four treatment plots received CUSTOMBLEN fertilizer at
different application rates. Nutrient concentrations were monitored in the wells.
For the scaling experiment conducted in association with the Disk Island study, three plots on
an uncontaminated cobble beach were marked with rebar and nylon: 3 m x 3 m, 6 m x 6 m, and 9
m x 9 m (Figure 3.9). Four wells, each cased with 1m lengths of 44 mm diameter slotted stainless
steel, were installed in each plot, 0.6 m from the center of the plot so only the top unslotted 102 mm
section remained above ground. The wells were capped to prevent tidal water from entering. Wells
were numbered 1 though 4, beginning with the one farthest from the water line at low tide and
proceeding clockwise. Each plot was treated with water-soluble fertilizer granules (CUSTOMBLEN).
Nutrient concentrations were monitored in each well through time.
Elrington Island
Instead of marking plots, three areas of Elrington Island beach were delineated: the Bath beach,
Sprinkler beach, and untreated Control beach (Figure 3.10). Beach experiment dimensions and rate
of nutrient solution application are given in Figure 3.11. The Bath beach received one application
of nutrient solution, the Sprinkler beach received multiple applications of nutrient solution, and the
untreated Control beach received no treatment. A rock outcrop separated the Bath and Sprinkler
areas. The Sprinkler and untreated Control beaches were separated by a 12 m buffer zone. Ten
sampling baskets identical to those described for Disk Island above were arranged in a row on each
beach. Larger cylindrical stainless steel baskets were used on the Sprinkler and untreated Control
beaches; smaller square baskets were used on the Bath beach. Separate homogenates were prepared
for each beach due to logistical constraints.
Nine of the baskets were used for microbiology and oil chemistry analyses. The tenth basket
on each plot contained two wells cased with 51 mm diameter unslotted polyvinylchloride (PVC) pipe.
These wells were used to monitor nutrient concentrations and dissolved oxygen at the upper and lower
interface of the subsurface oil layer (Figure 3.12). Four wells were installed in both the Sprinkler and
untreated Control beaches. The wells were located 0.6 m on either side of the baskets; wells on the
same side were approximately 2 m apart.
54
-------
9m
3m
6m
9m
A10 6m
B20
C30
WATER
WELL POINTS
Rgure 3.9. Scaling Experiment Plots on an Uncontaminated Cobble
Beach.
55
-------
Stream
Sprinkler Head
Oiled Berm
Oiled Area
Low Tide
BATH
Oiled Area
OV Oiled Area
Qoo
Low Tide
SPRINKLER
Low Tide
CONTROL
Area Covered By
Feterlizer Solution
Figure 3.10. EIrington Island Beach Diagram.
-------
Ol
Sprinkler Head
X
DC unc unnnn
DOnC UDDD
nnnnc unnnn
UNTREATED CONTROL
SPRINKLER
8m
• 0.75 inches/hr
•13.6g/m2 N
• 2.7 g/m2 P
• 4 hour spray irrigation
• one application
• 0.4 inches/hr
• 6.88 g/m2 N
•1.37 g/m2 P
• 4 hour spray irrigation
• six applications
0.4 inches/hr
seawater only
4 hour spray irrigation
six applications
Figure 3.11. EIrington Island Nutrient Solution Experiment Beach Areas and Rate
of Nutrient Solution Application.
-------
Cn
00
5.08 cm PVI
DO Pro
4cm
i
8cm
i
i
4cm
]
l
r
L
1
be
1
•
's':?//|'1!^^;f
-' ,»~*\ • <'- .*-
/* ' ' w/J-. ""v * f ...f.^.. t f t V
1
DO Probe
1
Beach Surface
*\ % - ' *
t ',-.••'
%«•
f f t !
f
S
Un-Oiled
Material
Oiled
Material
Un-Oiled
-^* „ Rnnf*h
Material
Figure 3.12. Nutrient and Dissolved Oxygen Monitoring Basket.
-------
PLOT DESIGN AND SAMPLING
Basket Removal and Sampling
Baskets were removed at set intervals from Disk Island and Elrington Island and subsampled for
microbiology and oil chemistry analyses. Prior to application of the fertilizer, one basket was
removed from each plot or beach on Disk Island to provide zero-time data for oil chemistry and
microorganism activity. A timeline depicting basket collection from the test areas at Disk Island and
Elrington Island is given in Figure 3.13. The schedule for basket collection depended on the
experimental design. Baskets were removed from the beach, wrapped in aluminum foil, and placed
in a small plastic garbage bag. The wrapped baskets were placed in a cooler with frozen gel packs.
Dry ice was not used unless the basket was only used for oil chemistry analysis. Otherwise the baskets
were returned to the lab immediately after collection and fully processed within 12 hours of the time
of collection. Clean material was used to fill the hole left by the basket removal.
Subsampling of the baskets was performed in the laboratory in Valdez. Baskets which contained
only oiled beach material (no layers of unoiled beach material) were subsampled as follows:
The top 1 to 2 cm of material was scraped off with a metal spoon and discarded. The beach
material below was sampled in the middle of the basket with the spoon, avoiding material at the
basket edges. If fertilizer granules were visible in the material, the first subsample was taken for oil
chemistry to avoid fertilizer grains in samples for microbiological analysis. Oil chemistry samples
were frozen in a glass jar at -20 °C. Samples for microbiology were contained in scalable bags or glass
jars and were stored at 4°C. Any unusual observations (odor, visible color changes) were noted.
Baskets containing a subsurface oiled layer were sampled by first inserting a plastic ruler on one
side of the basket to provide accurate depth measurements for removal of each layer. The top 25 mm
of material was removed with a metal spoon, transferred to a sheet of aluminum foil, thoroughly
homogenized with a spoon, and scooped into the sample containers. The subsampling procedure was
repeated for each layer. Layers were defined as follows: a surface layer consisting of the top one
inch of clean material; an upper oil interface located 13 mm on either side of interface; an oil layer
located 25 mm from the middle of the layer; a lower oil interface located 13 mm on either side of
interface; and a bottom layer located 25 mm from the bottom of the clean material layer. Any
unusual observations (odor, visible color changes) were noted. Oil chemistry was conducted for all
five basket layers from all baskets collected. Microbial respiration was measured in biometers for the
upper oil interface, oil layer, and lower oil interface for two baskets from each plot within the
beaches. Microbial activity was also determined for all five basket layers from all baskets collected.
59
-------
DISK ISLAND
B-O
10 B B
I |
1
\ [ \ l
I I I I I I I I I I I I I I I I I I
20 300 10 20
June uuiy
B
I I
M | | I ! I I | |
300 10
ELRINGTON ISLAND
Bath i
Vi'
B
1
B
i
B
r i i
20 300
I lnnA 1 L
I I I I | M I I
10 20
B
I
June
Sprinkler i
•July-
III T|
300 10
-J '-August-'
IF) CF)
E© © B © B
U M i i
I M
20 300
'——' '
I I II I I \ M
10 20
June
Untreated
contro'
July
300
' '-
August
10
-'
s s as
i i i
B S SB SB
U U i
20 300
I luna 1 L
I I I I I I I I I I
10 20
June
July-
300 10
I— A linnet—I
August-
Legend
I = Date Baskets Were Installed in the Beach
F = Date of Fertilizer Application
B = Date of Basket Removal
B-O = Time 0 Basket Removal
S s Date of Seawater Application
Figure 3.13. Timeline Depicting Dates of Fertilizer Application and Basket
Removal From Disk and Elrington Islands.
60
-------
PLOT DESIGN AND SAMPLING
METHOD OF FERTILIZER APPLICATION
Slow-Release Fertilizers
Herring-seine net bags filled with slow-release fertilizer briquettes (WOODACE) were
positioned on Otter and Seal beaches at Snug Harbor to provide complete exposure of the beach
material to nutrients leached from the bags. Bags filled with WOODACE briquettes at Snug Harbor
are shown in Figure 3.14. Each bag contained approximately 33 pounds of briquettes. Application
of the briquette bags occurred on June 11,1989. A timeline depicting fertilizer application to the test
beach plots at Snug Harbor is given in Figure 3.15. The total quantity of briquettes applied to Otter
Beach (35 m x 12 m test area) was 800 pounds, representing approximately 100 pounds nitrogen and
24 pounds phosphorus (as P2O6). The bags were tethered to 0.9 m sections of 2.9 cm diameter steel
rods buried 15 cm below the surface of the beach. Figure 3.16a indicates the positioning of the
24 bags in the experimental area. Three rows of eight bags were placed at 2 m, 6 m, and 10 m from
the top of the plot.
*
On June 20 and 21, 1989, the bags were repositioned according to the diagram in Figure 3.16b,
as the bags located at the 2 m row were not consistently submerged by the high tide. In addition,
preliminary data indicated that the nutrients were being channelled vertically down the beach. Four
more bags were added to the previous 24 bags for a total of 28 bags, resulting in 920 pounds of fer-
tilizer at Otter beach, or 130 pounds N and 30 pounds P (Table 3.4).
The same arrangement and repositioning was used for the briquette bags on Seal beach. This
beach was smaller (28 m wide versus 35 m), so the weight of briquettes applied per bag was 26 pounds
(versus 33 pounds) for a total of 620 pounds. This figure increased to 730 pounds when the four new
bags were added, resulting in 103 pounds N and 22 pounds P (Table 3.4).
Slow-release fertilizer granules (CUSTOMBLEN) were applied to Tern beach in Passage Cove
on July 25, 1989. Figure 3.17 shows a picture of the granules adhered to cobble at Passage Cove. A
timeline depicting fertilizer application to the test beach plots at Passage Cove is given in Figure 3.15.
The granules were applied using a commercial broadcast fertilizer spreader, at a rate of approximately
0.0033 lbs/ft2. The total application of nitrogen and phosphorus was approximately 400 Ibs and 40
Ibs, respectively. The granules adhered to the oil on the rock surfaces and were therefore not easily
displaced from the beach or redistributed by the tidal action. CUSTOMBLEN granules were also
61
-------
PLOT DESIGN AND SAMPLING
TABLE 3.4. FERTILIZER TREATMENTS AT SNUG HARBOR, PASSAGE COVE, DISK
ISLAND, AND ELRINGTON ISLAND
Beach or Plot
within the beach
Treatment
Snug Harbor
Otter Beach
Seal Beach
Eagle Beach
Passage Cove
Tern Beachb
Kittiwake Beach
Raven Beach
Disk Island (Fertilizer
Specific Activity)
Control 2
RA4
RA3
RA2
RA1
Control 1
Elrington Island
Control Beach
Sprinkler Beach
Bath Beach
130 Ibs N, 30 Ibs P per plot (WOODACE briquettes)
Two applications* of INIPOL - 10 gallons (10.9 g/m2 N and
4.4 g/m2 P) and 10.5 gallons (11.5 g/m2 N and 4.2 g/m2 P)
103 Ibs N, 22 Ibs P per plot (WOODACE briquettes)
Two applications* of INIPOL - 13 gallons (10.7 g/m2 N and
4.1 g/m2 P) and 14 gallons (11.5 g/m2 N and 4.4 g/m2 P)
None- Untreated control
400 Ibs N, 40 Ibs P entire beach area (CUSTOMBLEN
granules) plus INIPOL - 57 gallons (21.5 g/m2 N and
8.2 g/m2 P)
Approximately 1,100 gal applied daily with
7 mg/L N and 3 mg/L P (Nutrient solution)
None- Untreated control
None- Untreated control
1000 g/m2 (CUSTOMBLEN granules)
500 g/m2 (CUSTOMBLEN granules)
100 g/m2 (CUSTOMBLEN granules)
50 g/m2 (CUSTOMBLEN granules)
None- Untreated control
Six applications of seawater only
Six applications of 6.88 g/m2 N, 1.37 g/m2 P (nutrient
solution)
One application of 13.6 g/m2 N, 2.7 g/m2 P (nutrient
solution)
*INIPOL was reapplied because of the conditions under which it was originally applied (stormy and
rainy).
bINIPOL values are best estimates.
62
-------
Figure 3.14. Bags Filled with WOODACE Briquettes at Snug Harbor.
63
-------
SNUG HARBOR
Oner
Beach
Seal
Beach
c
o w
i I
c
O* R
c c
11 i 11'"';
CMS«9t.8)
llll|mi|nTir |
0 10 20 30 0
M I I II M I M I
10 20 30 0 10 20 30 0 10 20 31
—June 1 '—July 1 >—August—'
c
o w
co* R c
U U i i
I M | I M
10
c c c
-------
, ,
7 \
1
7 5
7 \
7 \
, ,
r T
• i
, j
7 \
f 1
7 T
7 \
7 T
^ ^
7 \
7 \
7 \
7 \
f 1
Rgure 3.16A. Placement of the Bags of Fertilizer Briquettes on Otter and
Seal Beaches (See Figure 3.2 for Beach Locations).
Rgure 3.16B. Repositioning of the Bags of Fertilizer Briquettes on Otter
and Seal Beaches (See Figure 3.2 for Beach Locations).
65
-------
Figure 3.17. CUSTOMBLEN Granules Adhered to Cobble at Passage Cove.
66
-------
PLOT DESIGN AND SAMPLING
applied to plots on Disk Island on July 1,1990, July 5, 1990, July 9, 1990, and July 13, 1990. A
timeline depicting fertilizer application to these test beaches is given in Figure 3.10.
Oleophilic Fertilizer
The oleophilic fertilizer was applied using a backpack sprayer with a capacity of four gallons
(Figure 3.18). The fertilizer was applied to both beaches as the tide was going out in the evening on
the first application, and was initiated at the top of the beach an hour after the tide was past the
lowest zone in the plot. A second application to both beaches occurred in the morning. The fertilizer
was initially warmed, to ensure uniform application and prevent clogging of the spray nozzle.
Oleophilic fertilizer (INIPOL) was first applied to Otter beach in Snug Harbor on June 8,1989 and
to Seal beach in Snug Harbor on June 9,1989 (Figure 3.IS). The beaches received 13 and 10 gallons,
respectively. The applications represented approximately 5% of the estimated weight of the oil on
the beach. INIPOL was applied in combination with slow-release CUSTOMBLEN fertilizer granules
to Tern beach in Passage Cove on July 25,1989 (Figure 3.15). The following computations were made
to determine the application rate:
For a plot 20 m x 12 m there are 240 m2 or 2,600 ft2. Assuming an oil depth of 6 inches, there
is a total volume of oiled beach material of 1300 ft*. From an estimated void volume of 20%, and a
specific gravity of rock equal to 160 lbs/fts, this total volume contains approximately 160,000 pounds
of rock. Based on a manufacturer's recommended 5% loading rate of the INIPOL relative to an
estimated oil weight of 1% (1600 Ibs of oil), 83 pounds or 10 gallons of INIPOL should be applied.
A second application of 10.5 gallons of INIPOL was performed on June 17,1989, to Otter beach
based on recommendations from Elf Aquitaine representatives. The second application to Seal beach
occurred on June 18, 1989, at a rate of 14 gallons of INIPOL.
Fertilizer Solution
Kittiwake beach in Passage Cove was used to evaluate the effectiveness of daily nitrogen and
phosphorus application via spray irrigation. A timeline depicting fertilizer application to Kittiwake
beach is given in Figure 3.15. The sprinkler system began operating on August 2, 1989, using
sprinkler heads typical of lawn sprinklers. The fertilizer was dissolved in seav/ater, and the solution
t
67
-------
Figure 3.18. Application of Oleophilic Fertilizer Using a Backpack Sprayer.
68
-------
PLOT DESIGN AND SAMPLING
was pumped by a gasoline-driven well pump to four sprinkler heads set on rebar stakes placed at
approximately the midpoint on each side of the plot. Each sprinkler swept a 180° arc across the beach
repeatedly during application (Figure 3.19). Typical applications consisted of 0.4 inches of water per
day. A total of 17 pounds of ammonium nitrate fertilizer and 7 pounds triple-superphosphate
fertilizer were applied. Application rates were established to supply 7 mg/L of nitrogen and 3 mg/L
of phosphorus to pore water in the saturated beach material to a depth of 2 m. The pump delivered
approximately 55 psi while operating, and pumped approximately 1100 gallons per hour.
At Elrington Island, fertilizer solution was also applied by a sprinkler system (see Figure 3.10
for application schedule). The system drew liquid fertilizer from a tank, mixed it with seawater, and
applied the mixture at a predetermined rate over a 4-hour period at low tide. The treatment was
reapplied approximately every 4 days on the Sprinkler beach, but only once on the Bath beach. The
application rate information is given in Table 3.4. The sprinkler heads were commercially available
heads used in agricultural irrigation systems. The same system was used to sprinkle unamended
seawater over the untreated Control beach.
Over the 2-year period in which research was conducted, more than 5 field and 20 laboratory
tests were implemented at several different sites in Prince William Sound. Table 3.5 summarizes all
field tests conducted, including the beaches subjected to the various fertilizer treatments, and
laboratory analyses performed. Table 3.6 summarizes the laboratory tests conducted, including the
experimental design and laboratory analyses performed. The detailed descriptions of field and
laboratory test designs and methods are described in Section 4.
69
-------
Figure 3.19. Sprinkler System in Operation at Passage Cove.
70
-------
TABLE 3.5. SUMMARY OF FIELD TESTS
PART A - SUMMER OF 1989
Beach
Location
Beach/Type Oiling
Fertilizer Treatment
Analyses
Eagle
Snug Harbor
Sand, gravel/moderate
Otter
Otter
Seal
- Seal
Seal
Raven
Tern
Kittiwake Beach
Snug Harbor
Snug Harbor
Snug Harbor
Snug Harbor
Snug Harbor
Passage Cove
Passage Cove
Passage Cove
Sand, gravel/moderate
Sand, gravel/moderate
Cobble over mixed sand
and gravel/moderate
Cobble over mixed sand
and gravel/moderate
Cobble over mixed sand
and gravel/moderate
Cobble over mixed sand
and gravel/heavily oiled-
physically washed
Cobble over mixed sand
and gravel/heavily oiled-
physically washed
Cobble over mixed sand
and gravel/heavily oiled-
physically washed
Untreated control-
none
INIPOL
WOODACE briquettes
WOODACE briquettes
INIPOL
Untreated control-
none
Untreated control-
none
INIPOL +
CUSTOMBLEN
granules
Nutrient solution
(sprinkler)
Oil chemistry (oil residue weight,
composition); nutrients;
microbiology (MPN); ecological
(chlorophyll; phytoplankton
primary prod; bact. abundance
and produc; caged mussels)
Same as above
Same as above
Same as above
Same as above
Same as above
Oil chemistry (oil residue weight,
composition); nutrients;
microbiology (MPN); ecological
(chlorophyll; phytoplankton
primary production; bact.
abundance and produc; caged
mussels; field toxicity tests)
Same as above
Same as above
-------
TABLE 3.5. (CONTINUED)
PART B - SUMMER OF 1990
Beach
Location
Beach/Type Oiling
Fertilizer Treatment
Analyses
f\3
Disk Island
Bath
Sprinkler Beach
Untreated Control
Beach
Seal Beach
Disk Island
Elrington Island
EIrington Island
Elrington Island
Snug Harbor
Small cobble mixed into
mixed sand and gravel
Cobble overlying mixed
sand and gravel/
Significant subsurface oil
layer
Cobble overlying mixed
sand and gravel/
Significant subsurface oil
layer
Cobble overlying mixed
sand and gravel/
Significant subsurface oil
layer
Oil totally gone
CUSTOMBLEN
granules
One application of
nutrient solution
(sprinkler)
Multiple applications
of nutrient solution
(sprinkler)
Untreated control-
none
CUSTOMBLEN
granules
Baskets: Oil chemistry (oil residue
weight, composition); microbial
activity (CO2 produc., MPN);
Wells: Nutrients; Ecological
(intertidal food webs)
Baskets: Oil chemistry (oil residue
weight, composition); microbial
activity (O, consumption, CO2
produc., MPN); dissolved O2
uptake;
Wells: Nutrients; Ecological
(intertidal food webs)
Same as above
Same as above
Wells: Nutrients
-------
TABLE 3.6. SUMMARY OF SUPPORTING FIELD AND LABORATORY TESTS
PART A - SUMMER OF 1989
Experiment Type
Test Materials
Experimental Design
Analyses
Tank Microcosm Studies
Column Microcosm Studies
Jar Microcosm Studies
Shake Flask Studies (Exxon)
Snug Harbor (mixed sand &
gravel and cobble).
Mixed sand and gravel from
Hell's Hole Beach
(uncontaminated) and Raven
Beach at Passage Cove.
Snug Harbor.
Bushnell Haas Medium; Prince
William Sound water; Alyeska
ballast water; Artificially
weathered Prudhoe Bay crude
oil.
Artificially weathered Prudhoe
Bay crude oil plus INIPOL
and/or WOODACE Briquettes.
Artificially weathered Prudhoe
Bay crude oil plus INIPOL and
Alyeska ballast water.
Effect of Oleophilic fert. (INIPOL)
and soluble fert. in bags (not slow-
release) on oil degradation.
Effect of INIPOL on oil movement
and bacterial activity.
Different nutrient mediums with and
without INIPOL.
Effect of inocula in the presence of
nutrients.
Effect of combined INIPOL and
water-soluble nutrients.
Effect of different concentrations of
INIPOL; 3, 10, 20, and 50% of oil
concentration.
Oil chem. (oil residue weight,
composition); GC/MS.
Numbers of oil-degrading
microorganisms; Mineralization of
HC labeled phenanthrene &
hexadecane.
Numbers of oleic acid-degrading
microorganisms; oil chem (oil
composition).
GC/FID.
-------
TABLE 3.6. (CONTINUED)
PART A - SUMMER OF 1989
Experiment Type
Test Materials
Experimental Design
Analyses
Shake Flask Studies (Exxon)
(cont'd.)
Respirometric Flask Studies
Chemical Effect of Oleophilic
Fertilizer (Exxon)
Toxicity Tests
Mutagenicity Tests
Artificially weathered Prudhoe
Bay crude oil plus INIPOL or
WOODACE Briquettes.
Prince William Sound oiled
beach material plus INIPOL in
poisoned and non-poisoned
conditions.
Artificially weathered Prudhoe
Bay crude oil with
uncontaminated beach material
and Alyeska ballast water; or
Snug Harbor seawater.
Oiled gravel.
Laboratory-reared test
organisms.
Snug Harbor mixed sand and
gravel.
Effect of incubation temperatures
(20 "C, 15 'C and 5 *C) in presence
of INIPOL and soluble nutrients.
Effect of INIPOL on oil degradation
from oiled beach material.
Relative comparison of the effects of
INIPOL and soluble fertilizer and
effects of inoculation.
INIPOL added to oiled rocks and oil
release measured.
Toxicity of oleophilic fertilizer; Fish,
invertebrates, or alga + a) INIPOL,
seawater; b) INIPOL, artificially
weathered Prudhoe Bay crude oil.
Mutagenic effect of oil beach
material exposed to different
fertilizers; INIPOL, briquettes.
Analytical respirometry; GC/FID and
GC/MS.
96-hour LC50.
Spiral Salmonella assay.
-------
TABLE 3.6. (CONTINUED)
PART A - SUMMER OF 1989
Experiment Type
Test Materials
Experimental Design
Analyses
Food Chain Bioaccumulation of
Carbon and Nitrogen
Biological samples from:
Tatitalek Island, Snug Harbor,
Passage Cove;
Seston samples from : Snug
Harbor, Passage Cove;
Particulate matter from root
feeders.
Movement of fertilizer nitrogen and
oil carbon into food chain in
sampling from fertilizer-treated and
untreated beaches.
Biological: Stable isotope analyses of
primary producers, consumers, and
seston;
Chemical: Stable isotope analyses of
dissolved ammonium.
CJl
-------
TABLE 3.6 (CONTINUED)
PART B - WINTER OF 1989/1990
Experiment Type
Test Materials
Experimental Design
Analyses
Biometer Flask Studies
Oiled beach material from Bay Effect of nutrient concentration and
of Isles, artificial seawater. timing of INIPOL application.
Biometer Flask Studies Coupled
to Micro-Oxymax Respirometer
Biometer Flask Studies
Biometer Flask Studies
ibid.
ibid.
ibid.
Biometer Flask Studies
ibid.
Effect of various concentrations of
soluble nutrients day 1; N only; and P
only.
Effect of different agitation rates;
SO, 75, 100 and 125 rpm.
Different fertilizer application
scenarios 1) INIPOL day 1 + soluble
nutrients daily; 2) soluble nutrients
daily; 3) soluble nutrients day 1 only;
INIPOL day 1 only; control.
Inoculation studies 1) Oil-enriched
mixed culture + soluble nutrients; 2)
Oil-enriched strain El2V + soluble
nutrients; 3) Soluble nutrients only;
4) Prince William Sound water only.
Cumulative CO2 production;
mineralization of UC labeled
phenanthrene and oleic acid; oil
chemistry (CH2Cl2/hexane extraction;
gravimetric analysis; GC analysis).
Cumulative O2 consump., cumulative
CO2 production; mineralization of
UC phenanthrene; oil chemistry
(CH?Cl2/hexane extraction;
gravimetric analysis; GC analysis).
Mineralization of UC phenanthrene;
oil chemistry (CH2Cl2/hexane
extraction of tidal-nate and rocks for
gravimetric and GC analysis).
Cumulative CO2 production;
mineralization of UC phenanthrene &
oleic acid.
Cumulative CO2 production;
mineralization of UC phenanthrene;
CH2Cl2/hexane extraction of tidal-
nate and rocks (gravimetric analysis;
GC analysis).
-------
TABLE 3.6 (CONTINUED)
PART B - WINTER OF 1989/1990
Experiment Type Test Materials Experimental Design Analyses
Toxicity Tests Not Applicable. 1) INIPOL toxicity to wildlife 1) Acute LC 50- to CUSTOMBLEN
(Acute LC50- quail); granules;
2) INIPOL toxicity to fish; 2) 7 day LC50; survival & growth.
3) Literature review: toxicity of
INIPOL and constituents to
mammalian and avian wildlife.
-------
TABLE 3.6 (CONTINUED)
PART C - SUMMER OF 1990
Experiment Type
Test Materials
Experimental Design
Analyses
Biometer Flask Studies
Column Microcosm studies
Microcosm Studies (Stable
Isotopes)
Stable Isotopes
Elrington Island, sampling
baskets.
Elrington Island.
Oiled gravel.
Disk and Elrington Islands.
Mineralization of oil in samples from
Bath Beach, Sprinkler Beach,
Untreated Control Beach.
Control with oil; Control without oil
pulse doses 20 ml nutrient solution;
2-hr doses 40 ml nutrient solution.
1) Fertilizer + seagrass detritus; 2)
Fertilizer only; 3) Seagrass detritus
only; 4) Control.
Fertilizer granules and fertilizer
solution.
Microbial activity (CO2 produc.)
mineralization of 1*C labeled
phenanthrene & hexadecane.
Nutrients; Cumulative CO, produc.,
TOC; oil chemistry (oil residue
weight).
Nitrogen & carbon isotope ratios.
Biological: Stable isotope analyses of
seston, bacteria;
Chemical: Stable isotope analyses of
ammonium, algae, consumers,
predators.
-------
SECTION 4
CHEMICAL AND BIOLOGICAL ANALYTICAL PROCEDURES
Detailed information on the standard operating procedures are given in the Quality Assurance
plans, prepared under the direction of the program Quality Assurance Officer, Dan Heggem (Papp
et al., 1989; Chaloud et al., 1990). Only brief accounts of the analytical procedures have been
included here.
NUTRIENT ANALYSIS
For the summer of 1989, water samples taken for nutrient analysis were filtered (Whatman glass
fiber filter) and placed in ISO mL plastic screw capped bottles. The bottles were immediately frozen
with a dry ice-antifreeze solution. Water samples taken offshore were collected with a clean bucket
and subsamples were taken for nutrient analysis. Water samples from the beach were collected behind
or in front of an ebbing or flooding tide, using a commercial root feeder. The root feeder was
outfitted with rubber tubing and a peristaltic pump to allow interstitial pore water to be drawn into
the feeder tube and sampled at the top of the feeder tube. The feeder was inserted approximately 20
cm into the mixed sand and gravel. Pore water was flushed through the feeder for one minute prior
to sampling.
For the summer of 1990, interstitial water samples were collected from monitoring wells using
a peristaltic pump. Samples were collected as the incoming tidal water filled each well and/or after
the outgoing tide had exposed the wells. A brief lag was required following outgoing tide well
exposure in order to allow excess tidal water to percolate through the beach material. The nutrient
sampling schedule was experiment-specific and is summarized in Table 4.1.
To collect water samples, tubing was inserted below the water line in the well. The pump was
turned on and allowed to run for a full, slow count of 10 (10 to IS seconds). Pumped water was
discarded away from the test area. After this rinse, the sample container was rinsed with
approximately 25 to SO mL of pumped water, capped, and vigorously shaken, ensuring that all interior
surfaces were contacted. The rinse was discarded. Sample bottles were filled to within 1.3 cm of the
top, tightly capped, and placed in Ziploc bag. The bottles were immediately stored in cooler with
frozen gel packs. Water collection was repeated for each well.
79
-------
ANALYTICAL PROCEDURES
TABLE 4.1. NUTRIENT SAMPLING SCHEDULE FOR SNUG HARBOR, PASSAGE COVE,
DISK ISLAND, AND ELRINGTON ISLAND
Snug Harbor (Root Feeder Interstitial Water Samples)
Ammonia - Samples taken on both the incoming and outgoing tide
1) Prior to fertilizer application (T«0);
2) One to two days post application (T-l);
3) Eight to ten days post application (T=2);
4) Thirty days post application (T=3); and
5) Six weeks post application (T=4)
Nitrate/Nitrite - Samples taken on both the incoming and outgoing tide
1) One to two days post application (T-l);
2) Eight to ten days post application (T=2); and
3) Thirty days post application (T-3)
Passage Cove
Data lost.
Disk Island (Well Samples)
1) Pretreatment (zero-time data);
2) Incoming tide following treatment;
3) Next outgoing tide;
4) Every other outgoing/incoming tidel thereafter, through Day 4 (fourth day following fertilizer
treatment); and
5) An incoming and an outgoing tide on days 8, 12, and 16.
Elrineton Island (Well and Basket Samples)
1) Samples were taken from the wells on each of the 3 days between applications on the Sprinkler
plot;
2) Samples were taken from the monitor baskets on the day preceding nutrient application.
Within a few hours after collection, each sample was filtered through a 0.7 p GFF filter to
remove aquatic organisms and particulate matter into one 250-mL and one 125-mL prewashed
polyethylene or Nalgene wide-mouth sample bottles. The filter apparatus is shown in Figure 4.1.
After filtration, all samples for nutrient analysis were stored at -20°C. If samples were not filtered
within 12 hours after collection, they were frozen in the field using dry ice.
80
-------
FUNNEL
00
BASE
CHAMBER
SCREEN
D-RINGS
RING
HOLDER
STOPPER
HOSE
Figure 4.1. Nutrient Sample Filtration Apparatus.
-------
ANALYTICAL PROCEDURES
Bulk water samples used in ecological monitoring and microbiology analyses for summers 1989
and 1990 were filtered using a 0.2 /j filter. Filtration was performed immediately after collection
using a standard filtration set-up with either a peristaltic or vacuum pump.
Phosphate, ammonia, nitrate, and nitrite were measured in all field samples; only phosphate and
ammonia were measured in laboratory-generated samples. The standard methods outlined below were
designed to quantitate the total amount of phosphorus in the form of phosphate and the total amount
of nitrogen in the form of nitrate, nitrite, and ammonia.
Nitrite and Nitrate
Nitrite was determined by the Griess reaction in which sulfanilamide and N-(l-Naphthyl)
ethylenediamine dihydrochloride (NNED) is reacted with nitrite in an aqueous acidic solution to form
an intense pink diazo dye with an absorption maximum of 540 to 543 nm. This method was also used
for nitrate, following initial reduction to nitrite by passing through a column containing copperized
cadmium fillings (Parsons et al., 1984). Detection limits for nitrate and nitrite were expected to be
0.05 n and 0.01 p, respectively.
Ammonia
Ammonia was determined by the Colorimetric Phenate method or Phenol-Hypochlorite method,
in which hypochlorite and phenol react with ammonium in an aqueous alkaline solution to form
indophenol blue, an intensely blue chromophore with an absorption maximum at approximately 637
to 640 nm (Parsons et al., 1984). The detection limit for ammonia was expected to be approximately
0.1 n.
Phosphate
Phosphate (i.e., orthophosphate) was determined as phosphomolybdic acid, which has an
absorption maximum at 880 to 885 nm in its reduced form in the presence of antimony (Parsons et
al., 1984). The detection limit for phosphate was expected to be 0.03 /*.
82
-------
ANALYTICAL PROCEDURES
OIL CHEMISTRY
Beach samples for chemical analysis consisted of either mixed sand and gravel contained in 400
mL I-Chem jars or cobblestones wrapped in aluminum foil and frozen prior to analysis. The mixed
sand and gravel was thawed immediately prior to the initiation of oil analysis, and the contents were
mixed thoroughly.
The following procedures were intended to assess the total amount of oil degradation and the
change in hydrocarbon composition as a result of biodegradation.
Oil Residue Weight
Figure 4.2 shows the detailed steps of this extraction scheme.
A 100 g subsample of mixed sand and gravel was removed from the I-chem jars and mixed
thoroughly with methanol in an Erlenmeyer flask. The slurry was shaken for five minutes, and the
methanol decanted out of the Erlenmeyer flask. The samples were similarly reextracted two times
with HPLC grade methylene chloride. The weight of extracted mixed sand and gravel was
determined by drying. The organic fractions were combined and backextracted with 3% aqueous
sodium chloride. The phases were separated and the aqueous portion was extracted with fresh
methylene chloride. All methylene chloride extracts were combined.
Several boiling chips were added to the methylene chloride and the volume of solvent was
reduced using a three-ball Snyder column attached to the round-bottom flask heated on a steam bath.
Volume was reduced until the color was approximately the color of dilute weathered oil (ca 15 mg/2
mL methylene chloride). The final volume of the extract was measured and an aliquot was
transferred to a GC autosampler vial.
All cobblestones were extracted using the same procedure (methanol, followed by methylene
chloride), except that shaking was replaced by gentle swirling to remove oil from the rock surfaces.
A measured aliquot of methylene chloride extract was allowed to dry for residue oil weight
analysis. For selected samples, a 2-mL aliquot of the methylene chloride extract was extracted with
hexane. The hexane extract was then brought to dryness and the residue weighed. These analyses
83
-------
Sand / Gravel (ca. 100 g)
Extract With 75 mL MeOH/5 mln
Extract With 75 mL DCM/5 mln _
Extract With 75 mL DCM/5 mln
I
Sediment Dried and Weighed
DCM/MeOH
I
Extract With 75 mL 3% NaCI
DCM NaCI / MeOH
Extract With 25 mL DCM
DCM
Discard
NaCI /MeOH
Condense DCM to > 3.00 mL
(Measure Volume)
Remove Measured Volume (2.00 mL) for Residue Weight Analysis
N2 Slowdown of Remaining DCM To Dryness
Add 10 mL Hexane, Vortex Mix, Centrifuge
Add 10 mL Hexane, Vortex Mix, Centrifuge
Add 10 mL Hexane, Vortex Mix, Centrifuge
Hexane Supernate
I
Hexane Supernate
Discard
Pellet
^- Hexane Supernate
Condense Combined Hexane Supernates (23.00 mL)
(Measure Volume)
-^•Remove
Measured Volume (2.00 mL)
For Hexane Extractable Weight
Analysis
Analyze By FID-GC
Figure 4.2. Oil Chemistry Sample Extraction.
84
-------
ANALYTICAL PROCEDURES
provided hexane extractable residue weight, and the residual hexane-unextractable component. The
hexane unextractable residue weight was assumed to be asphaltenes, which are not degradable by
bacteria.
Oil Composition
Hydrocarbon analysis of methylene chloride extracts or the hexane extractable fraction was
determined by GC/FID. The conditions for the GC during the summer of 1989 were as follows:
Column: DBS, 30 m X 0.25 mm, film thickness 0.25 n
Initial Temperature: 4S°C, 5 minute hold
Temperature Rate: 3.5°C/minute
Final Temperature: 280°C, 60 minute analysis
Injector. Splitless, 1 minute valve closure
Injector Temperature: 285°C
Injection: 2.0 /jL
Detector FID, 350°C
The conditions for the GC during the summer of 1990 were similar, but had a 30 m X 0.32 mm
column, a final temperature of 280 °C with a 20 minute hold, and a 1.0 pL injection.
Respirometric Studies
Methylene chloride was added to sample seawater in a volume ratio of 1:10 methylene
chloridetseawater. The sample was extracted using EPA SW 846 Method 3510 (separatory funnel
method): the seawater sample was transferred into a 250 mL separatory funnel; 1 mL of a 50 ppm
HC surrogate standard and 1 mL of 1 ppm PAH surrogate standards were added. The sample bottle
was rinsed with 30 mL methylene chloride and the extract added to the separatory funnel. The funnel
was sealed and shaken for 1 to 2 minutes; the organic layer was allowed to separate from the water
phase and the methylene chloride extract was collected. The extraction was repeated twice using 30
mL of methylene chloride, passed through an anhydrous sodium sulfate column, combined in an
evaporation concentrator, and condensed to a final volume of 1 mL. The extract was then passed
through a column of silica gel and again concentrated to 1 mL. Aliphatic hydrocarbons were analyzed
using GC/FID under the following GC conditions:
85
-------
ANALYTICAL PROCEDURES
Column: DB-5, 0.75 mm ID X 30 m
Initial Temperature: SO°C, 5 minute hold
Temperature Rate: 7°C/minute
Final Temperature: 300°C, 75 minute or less analysis
Injector Splitless
Injector Temperature: 250°C
Injection: 2.0 /*L
Detector. FID, 350°C
Aromatic hydrocarbons were analyzed using CD/MS under the following GC conditions:
Column: DB-5, 0.25 mm ID X 30 m
Initial Temperature: 50°C, 5 minute hold
Temperature Rate: 8'C/minute
Final Temperature: 300°C, 50 minute analysis
Injector Splitless, 0.8 minute valve closure
Injector Temperature: 270eC
Injection: 2.0 jiL
Detector MSD, 350°C
Selected Ion Monitoring (SIM) mode with 100 msec dwell time
A slightly different analytical procedure was used for laboratory oil biodegradation studiei
conducted during the winter of 1989/1990. One hundred mLs of methylene chloride were added to
biometer flasks at the end of an experiment, shaken (200 rpm) for 1 min, and the organic phase
transferred to a clean flask. This extraction procedure was repeated two more times with 50 mL
volumes of methylene chloride. Combined organic phases were passed through a layer of anhydrous
sodium sulfate (ca. 25 g) to remove residual water and suspended solids. An aliquot of known volume
was transferred to a clean (methylene chloride-rinsed) tared 25 mL test tube, and methylene chloride
was removed under a stream of dry nitrogen at 25 to 30°C. Residual solvent was removed by placing
tubes in a desiccator for 48 hours and the tubes were weighed to determine the amount of methylene
chloride-extractable residue.
Methylene chloride-extractions were subsequently extracted 3 times with 10 mL volumes of
hexane. Separation of hexane-soluble/methylene chloride-soluble materials (hexane insoluble
fraction) was facilitated by centrifugation (5000 rpm, 10 min). Hexane-soluble fractions were
transferred to clean, tared test tubes and hexane was removed under a stream of dry nitrogen at 30°C.
Residual solvent was removed from both the hexane-soluble and hexane-insoluble fractions under
desiccation. Hexane-soluble fractions were weighed, and changes in the chemical profile of this
fraction were determined by gas chromatographic analysis.
86
-------
ANALYTICAL PROCEDURES
"Tidal waters" from flask experiments were also extracted with methylene chloride as described
above.
MICROBIOLOGY
Numbers of Oil-Degrading Microorganisms
The Most Probable Number (MPN) technique was used to determine the number of total
heterotrophic and hydrocarbon-degrading microorganisms. Numbers of oil-degrading
microorganisms were measured at Snug Harbor and Passage Cove (summer 1989) by an extinction to
dilution procedure using oil as the carbon source.
The defined nutrient medium used in these tests contained (per liter of distilled water): NaCL,
24 g; MgS04.7H20, 1.0 g; KC1, 0.7 g; KH2PO4, 2.0 g; Na2HPO4, 3.0 g; and NH4NO3, 1.0 g. The pH
of the medium was adjusted to 7.4 with 1.0 N NaOH following autoclaving. The medium was
distributed in 4.5 mL portions to sterile dilution tubes. Initial dilutions were prepared by adding 5.0
g wet weight of sand and gravel subsample to the prepared dilution bottles containing 50 mL
autoclaved defined.nutrient medium. Following vigorous mixing by hand for 15 seconds, a 0.5 mL
sample of the initial dilution was used to prepare a dilution series from 102 to 1010. Each tube was
then amended with 20 /iL of sterile weathered Prudhoe Bay crude oil collected from an oil-
contaminated beach in Prince William Sound. Tubes were incubated at approximately 15°C for 21
days with 15 second shaking every three days. The tubes were scored independently by two
individuals at 21 days of incubation. Tubes that showed visible microbial turbidity or changes in the
physical form of the oil (oily droplets converted to stringy and flaky particulate material) were
considered positive. Numbers of oleic acid-degrading bacteria were determined using standard plate
counting procedures on defined nutrient agar medium supplemented with 1% oleic acid.
The standard "5-tube" MPN (APHA 1985), as modified for hydrocarbon-degrading
microorganisms and field considerations, was employed during the summer of 1990. Hydrocarbon-
degrading microorganisms were defined as those capable of emulsifying a Prudhoe Bay oil sheen
layered on Bushnell-Haas marine mineral salts broth. Total heterotrophs were defined as those
capable of growth (turbidity) in marine broth.
87
-------
ANALYTICAL PROCEDURES
The MPN technique required inoculation of five 100 ftL aliquots of each serially diluted sample
into sterile 24-well microtiter plates containing approximately 1.75 mL of sterile broth. Following
inoculation, a sheen of sterile Prudhoe Bay crude oil was applied to each well of the Bushnell-Haas
plates. Each microtiter plate was incubated at 16 ± 2°C for three weeks following inoculation. Wells
were scored as positive when oil emulsification was clearly indicated by disruption of the sheen.
Based on the results of a number of replicate inoculations (typically either three or five), the
statistically significant MPN of microbes (selected for or by the medium) per unit volume was
calculated. If the numbers fell below or above the dilution series selected, then the final numbers
were reported as either less than or greater than the table value.
Mineralization of Radiolabeled Hydrocarbons
Evolution of 14CO2 from phenanthrene-9-14C, a polynuclear aromatic compound, hexadecane-
1-14C, a straight chain aliphatic, and naphthalene-1-14C was used to measure the activity of
indigenous petroleum-degrading microorganisms as influenced by the addition of oleophilic (INIPOL)
and water-soluble fertilizers. Duplicate 5.0 g samples of beach material (1 to 5 mm diameter)
obtained from oiled beaches with and without fertilizer treatments were added to 10 mL artificial
salt-water medium (ASWM) in clean, sterile 100 mL Wheaton bottles. Each bottle was spiked with
0.1 /
-------
ANALYTICAL PROCEDURES
lined septa. Each vial was injected with 5 or SO pL of a 20 g/L solution of a radiolabeled hexadecane
(in acetone). The resulting initial concentration of added hydrocarbon was 10 or 100 /jg/vial (wet
sediment). One mL of 4N HC1 was injected into one vial of each series at time zero to determine the
amount of radiolabel added. The remaining vials in each series were incubated without shaking for
1,2, and 5 days with 1 or 10 ppm radiolabeled substrates. All incubations were conducted in the dark
at 16 ± 2°C.
The extent of hydrocarbon transformation was measured by recovering the 14CO2 produced from
the 14C-labeled hexadecane, calculating the rate of 14CO2 production (r), and converting this rate to
a hydrocarbon transformation rate (R) using the following equation:
R - r(Sn + A)
where Sn was the ambient hydrocarbon concentration and A was the added hydrocarbon concentration
in /ig/g dry weight of sediment. The rate of 14CO2 production (r) was calculated based on zero-order
or first-order kinetics, depending on the fraction of the added label that appeared as CO2 during the
incubation period. The equation was based on the assumption that added 14C-hydrocarbons were
completely mixed and equilibrated with ambient hydrocarbons in the slurries.
To recover 14CO2, acidified samples were purged for 15 minutes with N2 gas (30 mL/minute)
through a Harvey trap containing 15 mL of acidified toluene. The trap effectively scavenged
unoxidized or partially oxidized volatile hydrocarbons purged from the sample along with the CO2.
The gaseous stream was then bubbled through a standard liquid scintillation vial containing 10 mL
of CO2-sorbing phenethylamine cocktail. The radioactivity in the vial was counted in a Beckman
Model LSC 1800 liquid scintillation counter with automatic quench correction.
All dry weight determinations were obtained for each sediment sample by removing
approximately 50 g of sediment and weighing in a tared container. The samples were dried at 90°C
for 24 hours, cooled, and reweighed. Sediment dry weights were used to standardize all of the data.
ECOLOGICAL MONITORING
Water samples collected offshore in Cubitainers were transported to the laboratory in Valdez and
analyzed for several parameters that might be affected by bioremediation research efforts. Analysis
89
-------
ANALYTICAL PROCEDURES
included measurements reflecting possible eutrophication, release of oil from the beaches, toxic
effects from the fertilizers, and the presence of mutagenic oil residues. Procedures for these
measurements are as follows:
Chlorophyll
One-liter water samples were filtered through glass fiber filters, and the filters were extracted
with a solution of 90% acetone and 1 N NaOH. After overnight incubation in the refrigerator,
samples were centrifuged and the optical density of the supernatant was determined at 750 nm (total
absorbance) and 665 nm (chlorophyll a). Phaeophytin was determined by rereading the optical
densities after the addition of 10% HC1.
Primary Productivity
Photosynthetic productivity by phytoplankton was estimated by incorporation of 14C-
bicarbonate. Plankton samples collected in the field were transported to the Valdez laboratory,
incubated in BOD bottles in an outside waterbath, filtered, and frozen. Prior to July 5,1989, samples
were then sent to the U.S. EPA Environmental Research Laboratory (ERL)/Gulf Breeze for analysis
using a liquid scintillation counter. Once the liquid scintillation counter was operational at the Valdez
laboratory (July 5) primary productivity samples were counted there.
Bacterial Abundance
Estimates of the numbers of bacteria per mL of water in the water column were determined
using acridine orange direct counting with fluorescent microscopy (Hobbie et al., 1977). Water
samples were filtered through black Nucleopore 0.2 n pore size filters and stained with buffered
acridine orange solution (Fisher Chemical). A minimum of 200 bacterial cells were counted in 5 to
10 grid fields in the microscope.
Bacterial Productivity
The thymidine incorporation method of Fuhrman and Azam (1982) was used to measure
bacterial productivity. Triplicate water samples were spiked with 5 nL of 5H-methyl thymidine (1.1
i; 2.86 nM final concentration), incubated for 20 minutes and then extracted with 5 mL of cold
90
-------
ANALYTICAL PROCEDURES
10% trichloroacetic acid (TCA). Samples were filtered through 0.22 pm Millipore filters, washed with
cold TCA, and the radioactivity on the filter was measured in a liquid scintillation counter.
Caged Mussels
Mussels (Mytilus edulis) collected from Tatitlek Narrows, an area not affected by the oil spill,
were kept in cages at various sites at Snug Harbor (for ca. 10 weeks) and Passage Cove (for ca. 6
weeks). At each site four cages filled with 25 mussels each were deployed to measure the uptake of
petroleum hydrocarbons that might be released into the water column following fertilizer application
on the beaches. Mussels were removed at approximately weekly intervals from the cages and sent to
EPA/Gulf Breeze for analysis of polynuclear aromatic hydrocarbons (PAHs).
At each sampling, 3 mussels from each cage were sacrificed and the tissues were removed from
the shell and frozen. The frozen tissues were returned to the laboratory, where the tissues from all
3 mussels from a single cage were extracted by homogenizing and spiking approximately 20 g of tissue
with appropriate surrogates, digested with 6 N KOH at 35°C for 18 hours. The sample was then
serially extracted with 3 X 30 mL portions of ethyl ether (Warner, 1976). Extracts were then cleaned
up by elution from silica gel columns (SOP #EV89-5,1989). Concentrated, cleaned-up extracts were
then analyzed by capillary column gas chromatography (SOP #EV89-2,1989). Samples were analyzed
for 16 PAHs and total identified and unidentified PAHs. Percent moisture (EPA method 3550) was
determined for each sample so results could be expressed on a wet weight or dry weight basis.
Field Toxicity Tests
To characterize the extent to which toxic concentrations might develop during or immediately
after fertilizer application to oiled shorelines, a series of toxicity tests were conducted using field
water samples and a testing scheme similar to that used to test acute toxicity of industrial effluents.
Water samples were collected at specified intervals before and after application of INIPOL to
shorelines in Passage Cove and were sent to the consulting laboratory MEC (Marine Environmental
Consultants in Tiberon, CA.) for 48-hour toxicity tests with oyster larvae. Crassostrea gigas. Tests
followed the ASTM (1980) practice for conducting static acute toxicity tests with larval molluscs. The
MEC final report to this project specifies test methods and daily observations (MEC, 1989). A variety
91
-------
ANALYTICAL PROCEDURES
of laboratory and field controls were utilized during testing. Endpoints monitored for these tests were
larval survival and percentage of larvae that exhibited abnormal development.
One water sample (field control) was collected at the field control site, immediately outside of
the test area, just before the initiation of fertilizer application. At the beach where fertilizer was
applied (a 100 m stretch of shoreline), water was collected at 0.5 m depth (just above the bottom)
immediately offshore. Water samples were collected immediately before fertilizer application (pre-
application, which was 2 hours before low tide), following the completion of application (2 hours
after low tide), and again after 1 hour, 3 hour, 6 hour, 12 hour, and 18 hour intervals. Sampling
stopped at this time in order to return samples for shipping. All water samples were maintained at
4°C until toxicity tests began.
Oyster larvae toxicity tests were conducted with a standard dilution series (100%, 56%, 32%,
18%, and 10%) prepared for each water sample collected after application. Because the salinity of site
water was 26 ppt, field samples were adjusted to 28 ppt by addition of 90 ppt brine solution before
test dilutions were prepared. The salinity adjustment accounted for approximately 3% dilution and
was selected as the minimum change necessary to ensure that salinity was sufficient to sustain normal
development of oyster larvae (this dilution was not accounted for in the subsequent reporting of
sample concentrations). The same brine was diluted to 28 ppt and tested as a "hypersaline control"
to characterize the adequacy of the brine mixture as a test solution. Laboratory seawater was diluted
from 32 ppt to 28 ppt and tested as a seawater control.
LABORATORY FLASK STUDIES
Shake Flasks (Exxon)
Flask studies used samples of Prince William Sound water and/or oiled beach material. All flasks
were incubated with slow shaking at constant temperature. At each sampling, flask contents were
sacrificed and extracted with methylene chloride. Extracts were analyzed as described above for
hydrocarbon composition analysis.
For experiments on the effects of different inocula, samples of artificially weathered Prudhoe
Bay crude oil (volatiles removed, 30% weight loss, by distillation) (1% by weight) were placed in
sterile Bushnell-Haas medium, to provide nitrogen and phosphorus equal to 3.5% and 4.1% by weight
92
-------
ANALYTICAL PROCEDURES
of oil, respectively. This medium was used uninoculated (control) or inoculated with either a 10%
inoculum of water from the Alyeska ballast treatment facility or seawater from Prince William Sound.
All flasks were incubated at 1S°C for 16 days prior to analysis of oil composition. To determine the
effects of incubation temperature, flasks were incubated for 38 days at 15* and 5'C prior to analysis
of oil composition.
To determine the relative effectiveness of oleophilic fertilizer, INIPOL, at 3, 10, 20, and 50%
of the oil concentration, was added to a poisoned (50 mg/L HgCl2) and a nonsterile flask. Extents
of oil degradation were compared to flasks receiving water-soluble fertilizer (WOODACE briquettes)
in nonsterile flasks at a rate sufficient to produce a mixture of fertilizer and oil with 0.4% added N
and 0.09% added P.
Measurements of INIPOL-enhanced oil degradation on rock surfaces was studied using oiled
beach material from Prince William Sound that was covered with INIPOL at concentrations
approximating 10% of the oil concentration. Untreated oiled beach material and poisoned controls
(50 mg/L HgCl3) served as untreated controls.
Respirometric Flasks
Laboratory flask studies were also conducted using analytical respirometry to determine oil
biodegradation rates, and GC/FID chromatography to determine changes in oil composition.
Two nutrient formulations, INIPOL and a defined minimal-salts medium (OECD), were
compared. Microbial inocula consisted of seawater from Snug Harbor, beach material collected from
an uncontaminated beach in Valdez, weathered crude oil from the spill, and indigenous biota from
the Alyeska ballast water treatment plant.
The oil was fractionated into the aliphatic, aromatic, and polar fractions using standard silica
gel column chromatography. Composition of the aliphatic and aromatic fraction was measured by gas
chromatography using flame ionization detection (GC/FID) and gas chromatography/mass
spectrometry (GC/MS), respectively. Samples were collected at 0 weeks, 6 weeks, and 26 weeks (see
methods described under oil chemistry).
93
-------
ANALYTICAL PROCEDURES
Respirometry experiments were performed in a Voith Sapromat B-12 respirometer, consisting
of a temperature controlled water bath with 12 measuring units, and a recorder for direct plotting of
oxygen uptake curves. Each measuring unit comprised a reaction vessel with a CO2 adsorber, an
oxygen generator, and a pressure indicator. Microbial activity created a vacuum in the reaction
vessel, and it was recorded by the pressure indicator. Pressure was balanced by electrolytic oxygen
generation from the dissociation of copper sulfate and sulfuric acid. The recorder/plotter constructed
an oxygen uptake graph automatically.
Design of the respirometry experiments is summarized in Table 4.2. All vessels contained 2
grams of uncontaminated beach material from Valdez and 1000 mL of seawater collected offshore at
Snug Harbor. The vessels containing beach material, oil, and INIPOL were charged by first adding
the beach material, pouring a measured amount of oil onto the sand, adding the INIPOL to the oiled
rocks, and finally filling the vessel with the Snug Harbor seawater. All reaction vessels were mixed
with stirring turbines and incubated at 15'C in the dark.
TABLE 4.2. EXPERIMENTAL DESIGN FOR RESPIROMETRIC STUDIES
Reaction
Vessel*
V1,V1R
V2.V2R
V3.V3R
V4.V4R
C5
C6
C7
C8
Oil
Concentration
(mg/L)
1000
300
100
1000
_
-
1000
—
INIPOL
Concentration
(mg/L)
50
15
5
50
50
-
50
50
Alyeska
Ballast Water
(mL)
.
-
-
10
_
-
-
—
* V - Vessel
R = Replicate
C » Control
94
-------
ANALYTICAL PROCEDURES
Flask microcosm experiments were also conducted to provide further support for the
respirometric studies. Each flask contained 20 g of uncontaminated beach material but was prepared
in the same manner as the respirometric flasks. The experimental design for these experiments is
summarized in Table 4.3. Flasks were incubated on a shaker at 15'C.
TABLE 4.3. EXPERIMENTAL DESIGN OF FLASK STUDIES
Flask'
F1.F1R
F2,F2R
F3,F3R
F4,F4R
Cl
C2
Oil
Concentration
(mg/L)
10,000
10,000
10,000
10,000
10,000
10,000
INIPOL OECDb
Concentration
(mg/L)
500
500
+
+
_
— •••
Alyeska
Ballast
(mL)
.
10
-
10
-
10
" F - Flask
R - Replicate
C - Control
b OECD, a defined minimal-salts medium was composed of the following constituents added to
provide the specified final concentration (mg/L) in the test solution: KH2PO4 (170), K2HPO4 (435),
Na2HPO4 (668), NH4C1 (50), MgSO4.6H2O (45), CaCl2 (55), FeCls.6H2O (2.5). It included the
following trace elements added to provide final concentrations (Mg/L) in the test solution: MnSO4
(60.4), H3BOS (114.4), ZnSO4.7H2O (85.6), (NH4)6MO7O24 (69.4), and FeCls EDTA (200). To
prevent trace nutrient limitation, either 1 mL/L of a stock yeast extract solution (15 mg/100 mL), or
the following vitamins, biotin (0.4), nicotinic acid (4.0), thiamine (4.0), p-aminobenzoic acid (2.0),
pantothenic acid (2.0), pyridoxamine (10.0), cyanocobalamine (4.0), and folic acid (10.0).
Biometer Flasks
Aerobic biodegradation of organic substrates results in the consumption of oxygen and the
production of carbon dioxide (respiration). Biometer flasks were used to measure stimulatory effects
of various fertilizer treatments on the activity of indigenous, aerobic, oil-degrading microbes by
measuring accelerated levels of microbial respiration.
95
-------
ANALYTICAL PROCEDURES
Biometer flasks contained 100 g of homogenized oiled beach material from Prince William Sound
sieved (<12.5 mm, >4.75 mm diameter, summer 1990; >2.75 diameter, winter 1989/1990), and mixed
well to generate homogenized test substrate of uniform oiling.
To simulate conditions similar to Prince William Sound, oiled material was exposed to an
artificial tidal cycle consisting of a 12 hour high tide followed by a 12 hour low tide. Rocks were
incubated at 15°C in the dark and tidal action was simulated by gentle mixing (75 rpm). To facilitate
tidal cycling while maintaining the integrity of the closed system, the apparatus was modified by
placing a Teflon tube, fitted with a swivel lock and sealed with a 10 mL hypodermic syringe, through
the rubber stopper holding the ascarite trap (Figure 4.3). A 12 hour "high-tide" period was simulated
by adding SO mL of an aqueous solution of sufficient volume to submerge all rocks through the
Teflon tube. For "low-tide", a 60 mL hypodermic syringe was connected to the Teflon tube, aqueous
solutions were withdrawn, and rocks were incubated for 12 hours. Each high-tide solution was
retained and tested for oil residues physically removed from the test systems.
At each tidal change NaOH trapping solutions were sampled to measure microbial respiration
rates. Evolved CO2 was trapped in 10.0 mL of a 0.5 N NaOH solution (prepared with CO2-free
water) located in the side-arm of the biometer flask. For each 12-hour interval, NaOH was removed
from the biometer flask and replaced with 10.0 mL of fresh trapping solution. The amount of trapped
CO2 was determined by acidifying NaOH samples (pH<2.5 with 8.5% phosphoric acid) and analyzing
headspace gases by gas chromatography. Production of CO2 was measured for 5 to 7 days, and
background CO2 concentrations were determined for each sampling point. This procedure was
repeated to simulate the requisite number of tidal cycles. For certain experiments, a Micro-Oxymax
autorespirometer was used to measure both carbon dioxide production and oxygen consumption
simultaneously.
Radiolabeled phenanthrene-9-14C (s.a.=13.1 mCi/mmol) of [U]-14C-oleic acid (s.a.=907
mCi/mmol) were used to measure the activity of "INIPOL-degrading" and oil-degrading microflora.
Radiolabeled substrates were introduced with the final aqueous high-tide solutions (2.5x106 dpm/50
mL) and added directly to the oiled beach material in the biometer flasks. Release of 14CO2 was
determined by liquid scintillation analysis on duplicate, 1.0 mL samples of the NaOH solutions
recovered at 12 hour intervals over a 3 to 5 day incubation period.
96
-------
Rubber Stopper
Cotton
Ascarite Trap 8 To 20 Mesh
Stopcock
Rubber Stopper
250mL Erlenmeyer Rask
50ml_ Sterile Seawater
Oiled Beach Material
Incubation Conditions
• Daily Tidal Exchange
• 12 Hours High Tide /12 Hours Low Tide
-15° C, Dark
• Agitation - 50 rpm
• Sterile Control
Rubber Stoppers
15 Gauge Needle
SOmL Tube
NaOH
Polyethylene Tubing
Figure 4.3. Biometer Flask - "High Tide".
-------
ANALYTICAL PROCEDURES
Measurement of Carbon Dioxide
The procedure for measurement of CO2 in aqueous solutions used a Dohrmann DC-80 carbon
analyzer, which was installed, plumbed, and operated according to the low temperature system set-up
described in the operator's manual with three exceptions. Since only CO2 was measured, the reactor
reagent was 2% phosphoric acid and not the 2% potassium persulf ate/phosphoric acid mixture called
for by the manual. Also, the UV lamp was not lit. Finally, the sample was not acidified and sparged
prior to injection.
An appropriate amount of sample (approximately 20 pL) was hand injected into the DC-80. The
operator's manual provides the information to determine amount of sample to inject. The DC-80
possesses a self-contained pumping system which cycles (at > 2.5 mL/minute) reactor reagent past the
injection port. When a sample is injected the pH shift (to < pH 2) results in the immediate release
of CO2 from the NaOH. The CO3 is cycled through the reactor and carried to the Non-dispersive
Infrared Detector (NDIR) on a stream of O3 (zero grade, 200 cc/min). The detector measures the
peak as an electrical output and the processor in the DC-80 converts the peak to a parts per million
carbon (ppm C) value.
In some tests samples were placed in a biometer flask coupled to the Micro-Oxymax auto
respirometer to detect changes in the oxygen and carbon dioxide concentration in the headspace gases
of each biometer flask. This was accomplished by cycling the gases through an electrochemical fuel
cell to detect oxygen and an infrared spectrophotometer to detect CO2 at 3 hour intervals. Results
were reported as rate of O2 and CO2 consumption/production. Following 3 days of respirometric
analysis with the Micro-Oxymax, a fourth high-tide solution with radiolabeled phenanthrene was
added to each biometer to monitor specific microbial activities.
MICROCOSM STUDIES
Microcosm studies permit the testing of bioremediation concepts under idealized conditions to
provide complementary data and information to field demonstration projects. Three types of
microcosms were tested: jar, tank, and column.
98
-------
ANALYTICAL PROCEDURES
Jar Microcosms
Experiments to determine the numbers of oil-degrading and oleic acid-degrading
microorganisms resulting from the application of INIPOL fertilizer were conducted in chemically
clean (I-Chem) jars, each containing approximately 200 g of oiled beach material plus either seawater,
artificial seawater, or sodium chloride solution (20%). The artificial seawater was composed of (per
liter of distilled water): NaCL (24 g) MgSO4.7H20 (1.0 g) KCL (0.7 g) KH2PO4 (2.0 g). Na2HPO4 (3.0
g), and NH4NOS (1.0 g). The medium pH was adjusted to 7.4 with 1.0 N NaOH following
autoclaving. For sterile systems, the oil-contaminated beach material was autoclaved in I-Chem jars.
This process removed the water from the oil but did not remove the oil from the beach material.
INIPOL application consisted of dripping 3 mL of INIPOL on the beach material surface and allowing
the treated material to incubate for 3 hours before filling the jars with the appropriate aqueous phase
(about 100 mL). Except for the jar containing unautoclaved seawater, sterile medium (seawater,
artificial seawater, or NaCl solution) was used in each microcosm. Subsamples (1.0 mL) were removed
at 24-hour intervals for bacterial enumeration. Oleic acid-degrading bacteria were enumerated on
oleic acid-containing agar plates supplemented with nitrogen and phosphorus. Oil-degrading bacteria
were enumerated by the dilution to extinction technique described under Microbiology in Section 4.
After collecting bacterial enumeration samples, the aqueous phase from one set of jars was decanted
off and replaced with fresh sterile medium (fresh seawater was added to the nonsterile seawater jar).
The decanted solution was frozen for analysis of residual oil components.
Tank Microcosms
Microcosms were constructed onboard the motor vessel AUGUSTINE to simulate field
demonstration treatment and untreated control plots within the beaches. Six troughs were used to
hold nine 2-gallon polyethylene tanks. A schematic of the microcosm system is shown in Figure 4.4.
Twenty-seven of the containers were filled with homogenized oiled mixed sand and gravel (mixed
in a large plywood box on the beach) obtained from the beaches in Snug Harbor. The remaining 27
containers were first filled about one-fourth full with the homogenized mixed sand and gravel, and
then filled with oiled cobble. The microcosm containers had four 2.5 cm holes in the bottom to allow
percolation of the water through the beach material as the troughs filled. Seawater from the harbor
was pumped into the troughs, held for 6 hours (high tide) and withdrawn to simulate tidal cycles. The
tanks, therefore, remained dry for 6 hours. Within each trough with nine tanks, three replicate tanks
were sacrificed at three intervals. These were analyzed to characterize weight and composition of the
remaining oil. Intermittent samples were taken for nutrient analyses.
99
-------
Pump ^ j
Seawater <
Mixed Sand and Gravel
Control
Water-Soluble
Soluble
Nutrients
Oleophilic
PUMPS
Cobblestone
©©©
©©©
©©©
Pump C j
Control
©©©
Water-Soluble
Soluble
'Nutrients
©©©
Oleophilic
7
Beach .
Material
Drain
•* T — —-
Seawater
Figure 4.4. Schematic Diagram of the Tank Microcosm.
100
-------
ANALYTICAL PROCEDURES
Two types of fertilizer were tested in the microcosms. Oleophilic fertilizer (INIPOL) was
applied to the microcosms on June 16,1989 with portable backpack sprayers, in sufficient quantities
to coat the exposed surface of the beach material. Eighty IBDU briquettes were placed in a container
so that water entering the microcosm flushed over the briquettes. However, because ammonia in the
microcosms was never above background concentrations during the first week of operation, the
briquettes were replaced with small bags filled with commercial granular fertilizer (N:P:K ratio of
16:5:5, not slow-release), to ensure adequate levels of nutrients were maintained. This approach
continuously produced ammonia concentrations around 400 to 700 mg/L at each filling of the micro-
cosms.
Column Microcosms
Microcosms consisted of jacketed chromatographic columns, 450 mm long and 25 mm in
diameter (Ace Glassware, Inc., Cat. No. 5821-26). A refrigerated bath/circulator was used to
circulate water through the column jackets to maintain a temperature of 15 ± 1.5 °C. A small plug
of glass wool was inserted into the bottom of each column and 300 g of oiled beach material from
Elrington Island was added. This material contained approximately 6 g of oil per kg and was sieved
to between 4.75 and 12.5 mm.
Peristaltic pumps and silicone tubing were used for moving air and water to and from the
microcosms. Peristaltic pumps, controlled by a 10-program, microprocessor-based timer/controller,
were used to simulate two tidal cycles per 24-hour period. A tidal cycle consisted of: 1) pumping
water into the bottom of the microcosm for approximately 2 hours (Figure 4.5); 2) maintaining a "high
tide" for an additional 4 hours; 3) draining the column by flushing carbon dioxide-scrubbed air
through the top of the column and continuing to purge air from the headspace for 1 hour (Figure 4.6);
4) maintaining "low tide" for 6 hours; and 5) purging air from the column to remove all carbon
dioxide formed during low tide (Figure 4.7).
Water from Prince William Sound was passed through a glass fiber filter and air was pumped
through Ascarite to remove carbon dioxide. Effluent water from the systems was acidified and the
carbon dioxide purged into sodium hydroxide traps for determination of oil mineralization during the
high tide phase. Air exiting the microcosms during low tide was passed through other sodium
hydroxide traps for assessment of carbon dioxide produced during the low tide phase. Carbon dioxide
was assayed using a Dohrmann Total Carbon Analyzer with a Dual Sparging Unit. Effluent water was
101
-------
AIR OUT
PWSW
PUMP
I
TIDE IN
EFFLUENT TRAP
J
'v
A
NtOH ACIDIFIED
rRAP EFFLUENT
some
PWSW« Prince William Sound Water
Figure 4.5. Schematic of Flow-Through Column Microcosm - Use of
Pump 1 to Fill Columns.
102
-------
PWSW
PUMP
I
TIDE IN
EFFLUENT TRAP
NtOH ACIDIFIED
TRftP EFFLUENT
BOTTLE
AIR IN
PUMP
2
TIDE OUT
TIDE OUT
COLUMN AIR PURGE
rteOH EXCESS
PUMP
3
DO
CAP PUMP
C02-FREE
AIR INPUT
PWSW- Prince William Sound Water
Figure 4.6. Schematic of Flow-Through Column Microcosm - Use of
Pump 2 to Drain the Columns.
103
-------
1
PWSW
PUMP
I
TIDE IN
EFF
«—
LUENT TRAP
k
A
"V
NiOH ACIDIFIED
TBAP CFFLUEHT
BOTTLE
AIR IN
PUMP
2
(X)
TIDE OUT
AIR OUT
<
COLUMN AIR "PURGE"
<
N.OH "CESS
TRAP H2°
1K" TRAP
PUMP
3
CAP PUMP
C02-FREE
AIR INPUT
PWSW. Prince William Sound Water
Figure 4.7. Schematic of Flow-Through Column Microcosm - Use of
Pump 3 to Purge Air from the Columns.
104
-------
ANALYTICAL PROCEDURES
also analyzed for total organic carbon. Three 300-g samples of oiled beach material were analyzed
for oil residue weight at the beginning of the test, and contents of all columns were analyzed for oil
residue weight at the end of the test, as a gross index of oil removal.
Carbon from petroleum hydrocarbons may be transformed within the microcosm in one or more
of the following manners (Figure 4.8): 1) oil may be mineralized to carbon dioxide and then trapped;
2) organic products may exit the column with the effluent, either as a result of incomplete oil
degradation (i.e., organic acids) or bioemulsification and suspension of oil into the water column; or
3) oil may be converted into microbial biomass and stay associated with the oiled beach material. Oil
carbon exiting the system as a result of the first two processes should be quantified by the selected
procedures (carbon dioxide by sparging into the TOC analyzer and incomplete degradation products
and emulsified oil by TOC analysis). No attempt was made to quantify oil converted to biomass,
which suggests that information provided by this system may be conservative relative to the overall
effects of bioremediation.
Each treatment was tested in duplicate. Two columns containing combusted oiled beach material
provided data for normalizing carbon dioxide in the exiting water and air of all other treatments.
Two columns containing oiled rock without nutrients provided a control for natural effects without
bioremediation. Two columns were treated with a nutrient "bath", consisting of 20 mL of a 35 ppm
N, 7 ppm P solution, added with high tide every 4 days (similar to one of the shake-flask tests). Two
columns were also treated with a nutrient "sprinkle", consisting of 40 mL of a 175 ppm N, 35 ppm P,
solution applied for 2 hours every 4 days. The test was run for a 2-week period.
CHEMICAL EFFECT OF OLEOPHILIC FERTILIZER
The test system designed to address the effectiveness of INIPOL as a potential "rock-washer"
consisted of a separatory funnel containing a small amount of oiled beach material. INIPOL was
added to the oiled beach material at 5% of the oil concentration and the material was refrigerated at
5'C for 1 hour. Artificial sea water at 5'C was added to cover the beach material, and was again
refrigerated at 5'C for 6 hours. The beach material was then drained, and the amount of oil in the
water was estimated. A typical test used four washing cycles.
105
-------
OIL CARBON
BIOMASS
(effluent, air)
PARTIAL OR
NO DEGRADATION
(exit system in effluent)
Figure 4.8. Potential Fate of Oil Carbon in Flow-Through Microcosms.
106
-------
ANALYTICAL PROCEDURES
TOXICITY STUDIES
Acute Tests
Using acute tests to account for worst-case conditions, the oleophilic fertilizer, INIPOL, was
tested in seawater alone. Since the fertilizer is very likely to become bound to oil after application
to oil-contaminated shoreline, and data generated by the manufacturer show that toxicity of the
fertilizer is appreciably decreased in the presence of oil, a second test treatment was selected. This
test treatment involved layering of oil on seawater, spraying on fertilizer, mixing, and testing. The
oil was also tested alone to provide data for comparative purposes.
Battelle Northwest laboratories conducted acute toxicity tests with silver salmon smolts
(Onchorhynchus kisutch), Pacific herring (Clupea harengus passasii), and bay mussel larvae (Mytilus
edulis). The oil used in these tests was weathered ANS crude oil collected May 5,1989, during Exxon
Valdez oil spill clean-up operations, from skimmer #81 working west of Disk Island in Prince William
Sound. This oil sample and the INIPOL used for testing was supplied by scientists at Exxon Research
and Engineering, Annandale, New Jersey. Toxicity testing procedures were adapted from EPA
protocols published for testing crude oils and dispersants (EPA, 1984), substituting INIPOL as the
dispersant chemical. Tests with mussel larvae followed ASTM standard methods (ASTM, 1988) with
minor modifications based on dispersant test methods (EPA, 1984). A detailed methods description
can be found in the Battelle data report to the EPA (Antrim and Word, 1989).
EVS Consultants tested threespine sticklebacks (Gasterosteus aculeaius). Pacific oyster larvae
(Crassostrea gig as), mysids (Mysidopsis bahia), and pandalid shrimp (Pandalus danae) commonly
known as rock shrimp. Exxon supplied EVS with samples of the same oil and INIPOL that was tested
by Battelle. Stocks of test solutions prepared from oil, INIPOL, and mixtures of oil plus INIPOL were
treated as effluent samples and acute toxicity tests were conducted following the EPA-recommended
methods of Peltier and Weber (1985). Tests with oyster larvae followed ASTM methods for molluscs
(ASTM, 1986). Complete test details are in the EVS final report (EVS, 1990).
Chronic Estimator Toxicity Tests
Chronic estimator toxicity tests with INIPOL and two standard estuarine test species of fish, the
inland silverside, Menidia beryllina, and the sheepshead minnow, Cyprinodon variegatus, were
107
-------
ANALYTICAL PROCEDURES
conducted at the EPA Laboratory in Gulf Breeze. A contract laboratory conducted a parallel test with
Menidia. All tests followed Methods for Estimating Chronic Toxicity of Contaminants to Marine
Organisms (EPA, 1987). During the studies, INIPOL test solutions were replaced daily in static
chambers containing larval fish that were 7 to 10 days old at test initiation. Survival and growth of
the fish were monitored as test endpoints for the seven-day test. Results are expressed as the 7-day
LC50 (concentration lethal to 50% of the test population), maximum test concentration resulting in
no significant lethal effects, and maximum test concentration yielding no significant reduction of fish
growth.
Toxicity of INIPOL and its Constituents to Mammalian and Avian Wildlife
Computerized data bases were searched employing a combination of key words that cover data for
routine laboratory test species and also incorporate information that might exist for wildlife. Searches
were conducted on three databases maintained by the National Library of Medicine's Toxicology
Information Program: 1) the Registry of Toxic Effects of Chemical Substances (January, 1990
version); 2) Hazardous Substances Data Bank; and 3) Toxline. A search was also conducted in the
TERRE-TOX database maintained at the EPA Environmental Research Laboratory in Corvallis,
Oregon. These computerized data bases summarize current and historical information from scientific
journals, government reports, industrial reports, books on industrial and chemical safety and hygiene,
and published regulations and standards. The information stored in these databases includes acute
toxicity, chronic effects, metabolism, and sublethal effects. In addition, two EPA documents on
ammonia were reviewed.
The search strategy used the following key words (in both the singular and plural form) to probe
for animal toxicity data: mammal, rat, mouse, guinea pig, rabbit, cat, dog, wildlife, bird, avian,
shorebird, duck, eagle, and raptor. INIPOL and its chemical constituents were searched by name and
their Chemical Abstract System (CAS) number, if assigned. INIPOL was reviewed (no CAS #), as was
ammonia (CAS 7664-41-7), aqueous ammonia (CAS 1336-21-6), urea (CAS 57-13-6), 2-butoxy-
ethanol (CAS 111-76-2), lauryl phosphate (no CAS # found), tri-laureth phosphate (no CAS #
found), sodium lauryl phosphate (CAS 7423-32-7), dodecyl phosphate (no CAS # found). Since there
was no toxicity information available for the lauryl phosphate component of INIPOL, lauryl sulfate
(CAS 151-41-7, synonyms include dodecyl sulfate) and sodium lauryl sulfate (CAS 151-21-3) were
also searched.
108
-------
ANALYTICAL PROCEDURES
MUTAGENICITY TESTS
Soil samples for the mutagenicity studies were collected from Snug Harbor approximately 2
months apart in early June, late July, and early September, 1989. Due to the characteristics of some
complex mixtures (e.g., insolubility), the standard assay can sometimes be impractical. The Alaskan
oil samples were mixtures that were difficult to test in the standard assay. Therefore, the spiral
Salmonella assay, a modification of the standard Salmonella plate incorporation assay, was chosen
for the monitoring of these samples (Houk et al., 1989 and Houk et al., 1991). This assay was chosen
because it required less total sample material, did not require solvent exchanged into another solvent
such as dimethylsulfoxide, eliminated potential artifacts, and saved labor (Maron and Ames, 1983;
Claxton, 1987; and Williams et al., 1988).
To prepare the samples for the assay, the samples were extracted using sonication and
dichloromethane. The extracts were filtered through silanized glass wool and concentrated to < 100
mL using roto-evaporation. After drying with anhydrous NaSO4, all samples were concentrated or
diluted to a concentration of 10 mg/mL (a reference point derived from preliminary testing) and
stored in a freezer at -30'C until used for the bioassay.
STABLE ISOTOPES
The samples were collected between three and five months after the oil spill and usually within
weeks after fertilizer application. Thus, the food-web structure was documented shortly after the
accident, and long-term effects were assessed by comparing these results to monitoring efforts
conducted later in the summer.
A variety of biological samples were collected for isotopic analyses from Snug Harbor and
Passage Cove. A more limited sampling was also conducted on June 13, 1989 on an uncontaminated
beach on Tatitalek Island. Samples included both primary producers and consumers. All organisms
were collected from the intertidal zone, and in some cases were collected directly from the treatment
areas. After collection, samples were immediately placed in Ziploc bags, frozen over dry ice, and
stored in a freezer at -20°C. To prepare for isotopic analyses at Texas A&M University, the samples
were thawed, rinsed with copious amounts of distilled water and freeze-dried. The samples were
placed in a 50°C oven for 24 hours, then ground in a mortar and pestle, and stored in vials in a
desiccator.
109
-------
ANALYTICAL PROCEDURES
A few samples of water-column particulate matter (seston) were also collected from Snug Harbor
and Passage Cove. Up to 20 L of water sample were filtered through a precombusted glass fiber
filter. Preparation of the filter residue containing seston was similar to the biological samples
discussed above. Particulate material in samples from root-feeders were used as subsurface samples.
For these samples, a liter of filtrate was collected. Samples for dissolved ammonium were analyzed
according to Velinsky et al. (1989).
All samples were analyzed isotopically by a modified Dumas combustion that converts organic
carbon and organic nitrogen to CO2 and N2 gas for mass spectral analysis (Macko, 1981). Between
3 and 5 g of living tissue were placed in quartz tubes with Cu and CuO, and the tubes were then
evacuated and sealed. The tubes were heated to 900°C at a rate of 450°C h-1, kept at 900CC for 2
hours, and cooled to room temperature at a rate of 60°C h-1. The slow cooling cycle ensured that any
oxides of nitrogen were decomposed to N2. CO2 gas was separated from N2 gas by cryogenic
distillation, and CO2 and N2 were then analyzed.
Stable carbon and nitrogen isotope ratios are reported according to the standard formula:
fiX = [(Rsample/Rstandard) - 1] x 10s %o
where SX is either «1SC or «16N, and R is either 1SC/12C or 16N/14N. The standard for carbon was
PDB Belemnite, and the standard for nitrogen was ultra-pure tank nitrogen that was standardized
against atmospheric nitrogen.
The reproducibility of the measurement for £1SC of particulate matter was ±0.2 %o with a
minimum sample size of 50 ng. For £15N, the precision was ±0.3 %o for particulate samples, and ±0.5
%o for NH4+. The minimum sample size for $16N analysis was 50 ng. Samples were compared from
untreated and treated beaches on several dates throughout the summer.
Biological samples were also taken on a weekly basis during and/or after fertilizer application
on Disk and Elrington Islands. Representative algae and heterotrophic consumers commonly found
on these beaches were chosen for analysis. By choosing a diverse group of organisms from the
intertidal zone, we were able to take advantage of specific spatial orientations and examine the range
of the effect of fertilizer nitrogen on the food chains. Green, brown and red algae species included
Urospora sp., Fucus disticus, and Odonthalia sp., respectively. Consumer organisms were selected to
110
-------
ANALYTICAL PROCEDURES
include a comprehensive selection of feeding strategies: barnacles (Balanus glandula) are suspension
feeders; periwinkles (Littorina sitlcuma) and limpets (Tacetara persona) feed on organic matter
deposited on rocks in the intertidal zone; and whelks (Nucella emegginata) and eel blenny (Anoplarcus
purpurceus) represented consumers from higher trophic levels. Seston, found in beach interstitial and
coastal waters, were used as an estimate of stable carbon and nitrogen isotope values in phytoplankton.
For each sample, four organisms were taken from the beaches and stored in coolers until
transported back to the laboratory and frozen (approximately 2 hours). Samples were taken from the
intertidal zone below the Sprinkler beach at Elrington Island. Samples from Disk Island were taken
directly from a test beach treated with CUSTOMBLEN fertilizer (1 kg/m2). Frozen samples were
shipped to Texas A&M University for isotope analyses. To prepare for analysis, samples were
thawed, rinsed in deionized water, and dried in an oven at 50°C. All four organisms were ground
together using a mortar and pestle, samples were oxidized in quartz tubes with cupric oxide, and N2
and CO2 gases were isolated by cryogenic distillation. Ratios of 15N/14N and 1SC/12C were analyzed
with magnetic-sector mass spectroscopy.
Well samples were collected from Disk Island, from a beach area that had received
approximately 100 g/m2 CUSTOMBLEN fertilizer granules, and from a well on a beach area that was
not fertilized. Samples were also recovered from adjacent cove waters at Disk Island. Some samples
were also collected from wells at Elrington Island, and in the adjacent waters. These latter samples
are not discussed in this report.
Well sampling was accomplished during the incoming tide by removing 50 L of interstitial beach
water from the bottom of wells using a peristaltic pump and filtering through a 1 /i cartridge filter
into a 50 L container. This 50-L sample was processed for collection of bacterial concentrates as
discussed below. An additional 20 L was collected without pre-filtering in a 20-L collapsible
container (Cubitaner), the container was vigorously shaken, and suspended particulate matter (SPM)
was concentrated by pushing the sample through a 4.7 cm glass fiber filter (GF/F; pre-heated at
480°C for 2 hours). In turn, part of the filtrate was collected in 1-L Nalgene bottles and stored at
-20°C. A 500-mL subsample was then filtered through a 2.5 cm GF/F filter for analyses of
particulate organic carbon (POC) and nitrogen (PON). Filters were dried at 50°C in an oven Hushed
with N2 gas and were stored in Petri dishes.
Ill
-------
ANALYTICAL PROCEDURES
Water column samples were collected by passing up to 20 L of water sample (or the volume
collected when the flow of water was reduced to a trickle) directly through a 4.7 cm glass fiber filter
(GF/F; preheated at 480°C for 2 hours). At selected stations, part of the filtrate was collected in 1-L
Nalgene bottles and stored at -20°C. A 500-mL subsample was filtered through a 2.5 cm GF/F filter
for analyses of POC and PON. For collection of bacterial concentrates, 50 L of sample were pushed
through a 1 -/i cartridge filter into a 50-L container and processed as discussed below.
Stable Isotope Microcosm Study
A microcosm experiment was conducted using four treatments of oiled gravel: 1) fertilizer;
2) fertilizer and seagrass detritus; 3) seagrass detritus; and 4) an untreated control. Well-sorted oiled
gravel from Disk Island was placed in 20-L Nalgene tanks, which were located in the laboratory at
ambient temperature. No attempt was made to limit the availability of light in these containers. Each
day 0.2-ji filtered water (did not contain bacteria) was added to the containers to cover the gravel
for 12 hours. The water was drained from the bottom of the container and the gravel was then
exposed to air for the next 12 hours. The water recovered from the microcosms was filtered through
a 1-n cartridge filter, an aliquot of this filtrate was collected for bacterial abundance (AODC)
measurements, and the remaining water was filtered through both 2.5 and 4.7 cm GF/F filters for
elemental (POC.PON) and stable carbon (£1SC) and nitrogen isotope («15N) analyses. The duration
of the experiment was 12 days.
Bacterial Bioassays and Nucleic Acid Concentration
Natural bacteria were incubated in filtered cove or filtered interstitial water samples (Coffin et
al., 1989). These water samples were filtered through 0.2-n cartridge filters, and then incubated with
a 1% inoculum of a 1.0-p filtered fraction from the same water sample. These samples were
incubated for 72 hours in the dark to allow enough bacterial growth. At the end of the incubation,
the sample was pushed through a GF/F filter to concentrate the bacteria.
Details of the bacterial concentration and nucleic acid extraction procedure are found in Coffin
et al., 1990. Briefly, nucleic acids were extracted from concentrates of bacteria following the protocol
described by Marmur (1961) with modifications to increase the recovery of nucleic acids (Maniatus
et al., 1981). Bacteria were lysed chemically with 0.2% SDS at 60"C for 30 minutes. After lysis,
nucleic acids were extracted twice with phenol (distilled and frozen prior to use), twice with a 50/50
112
-------
ANALYTICAL PROCEDURES
mixture phenol/chloroform and twice with chloroform. Nucleic acids were precipitated in ethanol
and 2% NaCl at -70°C for 10 minutes, and centrifuged. The pellet was rinsed with cold 70% ethanol
and dried in a vacuum oven. Finally, nucleic acid was resuspended in distilled water, and the purity
and quantity of the extracted material was estimated by UV absorption at wavelengths of 260 and 280
nm (Maniatus et al., 1981). Nucleic acids were freeze-dried in preparation for the stable isotope
analyses.
Ammonium and Nitrate Distillations
The technique for measuring the 516N of ammonium (NH4+) was reported by Velinsky et al.
(1989). After NH4* was steam-distilled from the 0.8-L Labconco distillation -flask and trapped on
the zeolite sieve, 3 mL of 8 N NaOH and 600 g of Devarda's alloy were added to the distillation flask.
The flask was then reassembled into the steam-distillation system and a new zeolite sieve trap was
added to the distillate side (Velinsky et al., 1989). Heating tape was wrapped around the distillation
flask and the sample was heated to 100°C for 30 minutes to convert the nitrate (NO,") to NH^. After
heating, the NH4+ was steam-distilled and prepared for isotopic analyses as described in Velinsky et
al. (1989). The zeolite with the exchanged NH4+ was analyzed isotopically as described below. The
precision was ±0.5%o for NH4+. The precision of our modified NOS" technique (±0.5) was better than
that reported by previous work (Kreitler, 1975; Cifuentes et al., 1989) and in the range of the NaOBr
method used by Liu and Kaplan (1989).
Isotopic Analysis
Suspended paniculate matter (SPM) samples that were analyzed for carbon isotopes were put into
glass petri dishes and placed in a glass desiccator with concentrated HC1 fumes. After 4 hours the
samples were gently dried to remove the acid without loss of labile nitrogen. All samples (SPM,
bacterial bioassays, and nucleic acid extracts) were analyzed isotopically by a modified Dumas
combustion that converts organic carbon and organic nitrogen to CO2 and N2 gas for mass spectral
analysis (Macko, 1981). CO2 gas was analyzed on a Finnigan MAT 251 (Laboratory of Dr. Ethan
Grossman, Texas A&M University), and N2 gas was analyzed on a Nuclide 3-60-RMS. The
reproducibility of the measurement for 51SC of particulate matter was ±0.2%o with a minimum sample
size of 50 /ig. For S16N, the precision was ±0.3%o for particulate samples. The minimum sample size
for £16N analysis was 50 jig.
113
-------
ANALYTICAL PROCEDURES
Other Analyses
Particulate organic carbon and paniculate organic nitrogen concentrations of bacterial extracts
and biological samples were measured on a Carlo-Erba CNS analyzer. Bacterioplankton were counted
with the acridine orange direct count technique (Hobbie et al., 1977).
114
-------
SECTION 5
QUALITY ASSURANCE/QUALITY CONTROL
BACKGROUND
A comprehensive quality assurance program was applied to the Oil Spill Bioremediation Project.
The environmental measurements obtained were verified correct and validated for use in this project.
The measurement data, as found in the finalized data base, are fully usable for the intended purpose.
The QA philosophy for this project centered on the concept of "quality assistance". Quality
Assistance provides help to the research scientist at critical times during a research project. Scientific
research quality assurance is very different than regulatory quality assurance. Quality assurance
should support a research project, not restrict it. Quality assurance in this case helped the principle
investigator with organization, provided advice on what was logistically possible, kept measurement
systems in control, and controlled or reported the amount of variability in the total measurement
system.
A vital element of the quality assurance program was the preparation of Quality Assurance Plans
(summer and winter 1989, and summer and winter 1990). These QA Plans were presented to the
scientific researcher as an aid to conduct scientifically sound and acceptable experiments. They were
prepared in accordance with the guidelines and specifications provided in 1983 by the Quality
Assurance Management Staff of the U.S. Environmental Protection Agency Office of Research and
Development. They include a project description containing the details of the project design. The
project organization is described and data quality objectives are addressed. Field sampling
techniques, sample handling and preparation, sample analysis, data management, and data analysis
and reporting are described. Detailed standard operating procedures are contained in appendices to
these documents.
Quality assurance objectives are generally defined in terms of detectability, accuracy, precision,
completeness, representativeness, and comparability. These objectives are derived from data quality
objectives (DQOs) which represent the greatest degree of uncertainty allowable in the data; in other
words, the risk associated with making a decision based upon the data. As a research and develop-
ment program, this project initiated several new developmental techniques and procedures for which
discrete DQOs could not be defined prior to operation. Instead, the QA program was designed to
115
-------
QA/QC
allow both control and assessment of measurement uncertainty during the sampling and analysis
phases of the project. Additional information on QA/QC for this project is available upon request
from Dan Heggem at the Environmental Monitoring Systems Laboratory in Las Vegas, NV.
QUALITY ASSURANCE/QUALITY CONTROL COMPONENTS
QA Plan
The data collection criteria provided a balance between time and cost constraints and the quality
of data necessary to achieve the research objectives. The QA Plans were designed to accomplish the
following general objectives:
• Establish the QA and QC criteria to control and assess data collected in the project.
• Document sampling, analytical, and data management methods and procedures.
• Utilize assessment samples and procedures to verify the quality of the data.
• Perform field and laboratory on-site audits to ensure all activities were properly performed
and any discrepancies were identified and resolved.
• Evaluate the data and document the results.
It was necessary to identify both qualitative and quantitative estimates of the quality of data
needed by the data users. Guidelines established by the EPA Quality Assurance Management Staff
(Stanley and Verner, 1985) encourage data users to clearly identify the decisions that will be made
and to specify the calculations, statistical and otherwise, that are applied to the data.
The raw data were collected during the two major operational phases of the project sampling
and analysis. A certain amount of data measurement uncertainty is expected to enter the system at
each phase. Grouping of the data, such as by beach plot configurations, also increases uncertainty.
The sampling population itself is a source of confounded uncertainty that is extremely difficult to
quantify. Generally, DQOs encompass the overall allowable uncertainty from sample measurement
and from the sampling population that the data users are willing to accept in the analytical results
(Taylor, 1987). Due to the many confounded sources of uncertainty, overall DQOs for the project
were not defined.
116
-------
QA/QC
The QA Plan focused on the definition, implementation, and assessment of measurement quality
objectives (MQOs) that were specified for the combined sampling and analysis phases of data. The
MQOs are specific goals defined by the data users that clearly describe the data quality sought for
each of the measurement phases. The MQOs are defined according to the following six attributes:
• Detectability — the lowest concentration of an analyte that a specific analytical procedure
can reliably detect.
• Precision — the level of agreement among multiple measurements of the same
characteristic.
• Accuracy — the difference between an observed value and the true value of the parameter
being estimated.
• Representativeness — the degree to which the data collected accurately represent the
population of interest.
• Completeness — the quantity of data that is successfully collected with respect to the
amount intended to be collected, as specified in the experimental design.
• Comparability — the similarity of data from different sources included within individual
or multiple data sets; the similarity of analytical methods and data from related projects
across regions of concern.
The project MQOs are established on the basis of the selection of appropriate methods to obtain
the data. The MQOs are reviewed by individuals familiar with analytical methods. If the
measurement quality goals cannot be met during the course of the project, the actual level of quality
is used to reassess the intended use of the data. A lower than desired attainment of data quality could
require different approaches in data analysis or modifications to the levels of confidence assigned to
the data. The initial MQOs for oil chemistry and nutrient analysis are presented in Tables 5.1 and 5.2.
Design Characteristics
To control and assess data quality, QA and QC samples are incorporated into the measurement
system. The oil spill project's experiments and analyses incorporated a number of these samples into
their measurement systems. The QA program was not optimal to assess all types of measurement
certainty in this project because it was a research project.
117
-------
QA/QC
TABLE 5.1 ANALYTICAL LABORATORY WITHIN-BATCH MEASUREMENT QUALITY
OBJECTIVES FOR BEACH SUBSTRATE
Parameter
Extractable oil
C8 through C32
C8 through C32
C8 through C32
Aromatic HC
nC18:phytane
nC17:pristane
Method
gravimetric
GC-FID
GC-FID
GC-FID
GC-FID
GC-FID
GC-FID
Reporting
units
mg/kg substrate
%RT
Response Factors
Mg/g
Mg/g
ratio
ratio
IDLb
25 ppm
NA
NA
250 ppmb
5-25 ppbc
NA
NA
Precision
30%RSD
1.0% RSD
25% RSD»
30% RSD
25% RSD
10% RSD
10% RSD
Accuracy
80-120%
NA
80-120%
80-120%
65-135%
80-120%
80-120%
Completeness
80%
80%
80%
80%
80%
80%
80%
* Response factors for nC17, pristane nC18, phytane: 25% RSD initial calibration; 40% for other n-
alkanes
b 250 ppm in total extractable material (/* g/mg extractable material); 100 ppb in beach substrate n g/kg
of beach substrate
c Water 5 p g/L; sediment 25 n g/kg
TABLE 5.2 ANALYTICAL LABORATORY WITHIN-BATCH MEASUREMENT
QUALITY OBJECTIVES FOR QA/QC SAMPLES FOR NUTRIENT ADDITIONS
Parameter
NHS-N
NOj-N
NO2-N
PO4-P
Method
SPEC
SPEC
SPEC
SPEC
Reporting
units
pMO.l
H M0.05
M M0.01
It M0.03
IDLb
15%
15%
15%
15%
Precision
90-110%
90-110%
90-110%
90-110%
Accuracy
80%
80%
80%
80%
Completeness
* Instrument readings that have been converted (where necessary) to calculated reporting units
b Instrument detection limit (method-specific) in reporting units
118
-------
QA/QC
Quality Assurance Samples
QA samples are samples known to the QA staff but are either blind or double blind to the
analytical laboratory. Blind and double blind samples have concentration ranges that are unknown
to the analysts, but a double blind sample cannot be distinguished from a routine sample (Taylor,
1987). These samples provide an independent check on the QC process and can be used to evaluate
whether the MQOs have been met for any given run or batch, or for all batches (e.g., overall
measurement uncertainty). Important characteristics of the QA audit samples include their similarity
to routine samples in matrix type and concentration level, and their homogeneity and stability. Every
QA sample has a specific purpose in the data assessment scheme. Single blind QA samples were
incorporated into all nutrient analyses. Five concentration levels were used to cover the expected
ranges of both field- and laboratory-generated samples.
Quality Control Samples
To consistently produce high quality data, the laboratory was required to analyze certain types
of QC samples known to the laboratory staff that could be used by the analysts to identify and control
analytical measurement uncertainty. Each QC sample has certain specifications that must be met
before data for the parameter or analytical run are accepted. The QC samples were nonblind samples
to assist the laboratory in meeting analytical MQOs. The QC samples were analyzed by the laboratory
and permitted assessment of the accuracy of the physical and chemical analysis.
Replicate Samples - Disk Island and EIrington Island
Two or three sampling baskets were collected from each beach. Initially, triplicate oil chemistry
analyses and triplicate biometer tests were conducted on subsamples from each basket. Thereafter,
one basket was analyzed for oil chemistry in triplicate, and single oil chemistry analyses were
performed on the remaining baskets. One biometer test was conducted on a subsample from each of
the three baskets. Four wells were located on each beach at EIrington Island for collection of nutrient
samples; these four samples represented replicates for nutrient analyses. The standard deviation
among replicates was used to estimate field variability, and 95 percent confidence intervals were used
to estimate the range of confidence in the mean population estimate (accuracy). Examples of control
charts for nC18 and the nC18/phytane ratio are located in Figure 5.1 and 5.2.
119
-------
100
90-
80
60
50
40
30
20
10
0
o o
10 15 20 25
Replicate Sequence Number
30
35
Legend. + + + Otod Homoganate * * * Unoited Homogenate
o o o Disk Island Baskets a a a Elrington - Bath Beach Baskets
o o o Brington - ConW Beach Baskets A A o, Elrington - Sprinkler Beach Baskets
Figure 5.1. Control Chart for nC18 for Disk Island and Elrington Island.
100-f
90
80
| 60
"5 50
40
30
20
10
0
0000
a o o <*
10 15 20 25
Replicate Sequence Number
30
35
* + •»• Orted Homogenate
ooo Disk Island Baskets
o o o Elnngton — Control Qooch
» * * Knotted Homogenate
aon Elnngton - Bath Beach Baskets
A A A Elrington - SpmMer Beach Baskets
Rgure 5.2. Control Chart for the nC18/phytane Ratio for Disk island and
Elrington Island.
120
-------
QA/QC
Analytical Duplicate/Triplicate
A duplicate subsample of a routine sample extract (oil chemistry) was selected as the analytical
duplicate sample and was used to ensure that within-run MQOs were being satisfied. In the summer
of 1989 duplicate oil chemistry was performed on the same extracts. In the summer of 1990,
however, replicate extracts were taken from the same sample. In some cases in 1990 triplicate oil
chemistry analyses were conducted. Analytical duplicates were also used in nutrient, carbon dioxide,
and TDC analyses. One laboratory split was measured in each analytical run. Precision was calculated
as the relative standard deviation (RSD, coefficient of variance):
RSD - std. dev. / mean
Nutrient analysis precision was calculated as the relative percent difference of the duplicate samples
(RPD):
RPD «= | R - D | / X x 100
An example of a control chart of nutrient analysis precision for phosphate conducted during the
summer of 1990 is shown in Figure 5.3.
Field Audit Blank Sample
The field audit blank (FAB) sample was sent to the sampling crews by the QA staff and was
handled using the same procedures as routine samples. The FAB for nutrients consisted of 3 percent
NaCl in distilled water. The FAB was used to identify system contamination stemming from sampling
and laboratory operations. Pooled data from all FAB samples provided an estimate of the system
detection limit (SDL), which was calculated as three times the pooled standard deviation of the FAB
samples. FABs were done at Snug Harbor and Passage Cove for the oil chemistry analysis. These
samples consisted of solvent extracted non-oiled beach material collected at a site not affected by the
Valdez oil spill. An example of the analysis of these blank samples for nC18 is reported in Figure 5.4.
121
-------
2
o
fl
cc
au
40
30
20
10
-10
-20
-Oft
•
•
•
•
•
• . • '
r " • ' . "
" • •» •*"•"•• • " • •* •
^
1 1 1 1 1 1 1
20 40
duplicate sample number
60
Figure 5.3. Control Chart of Nutrient Analysis Precision for Phosphate
During the Summer of 1990.
10.000
1'000
0.100
CD
5
i
c
0.010
%
o
<$>
08 Jun 89 24 Jun 89 10 JU 89 26 Jut 89 11 Aug 89 27 Aug 89 12 Sep
Sampbng Date
Analysts Vedue
Detection Limit Reported
Rgure 5.4. Field Blank Analysis for nC18 Over Time for Snug Harbor.
122
-------
QA/QC
Reagent Blank
A reagent blank consists of all reagents in the same quantities used in preparing a routine sample
for analysis. It is subjected to the same procedures as a routine sample, including digestions or
extractions. The blank verifies the absence of contamination in the reagents and laboratory sample
preparation procedures. The concentration of the reagent blank should be less than or equal to the
instrument detection limit (IDL). Any reagent blank exceeding the IDL should be investigated to
determine and eliminate the source of contamination. One or more reagent blanks were included in
each analytical run for nutrient analyses. For oil chemistry extractions, one or more reagent blanks
were initially prepared and analyzed; if no contamination was observed, reagent blanks were prepared
only when a reagent lot was changed.
Biometer Test Blanks - Passage Cove, Disk Island, Elrington Island
Two types of blanks were included in each biometer test. One flask contained only air to detect
changes in atmospheric oxygen or carbon dioxide. The second blank was a formaldehyde-treated
flask. The formaldehyde killed any microorganisms in the flask, creating a laboratory "reagent" blank.
Quality Control Check Sample
A quality control check sample (QCCS) is a standard of known concentration used to verify the
calibration curve during sample analysis. The QCCS may be prepared by the analyst or obtained
commercially. If purchased, the QCCS should be from a different source than the standards used to
generate the calibration curve. The concentration of the QCCS should be in the midrange of the
calibration curve or in the midrange of the expected concentration of the routine samples. Values
obtained from repeated analyses of the QCCS are plotted on control charts. Examples of control charts
for the nC17/pristane and nC18/phytane ratios are shown in Figures 5.5 and 5.6. Control limits are
initially set at ±10 percent until sufficient data are collected to establish warning limits (95 percent
confidence interval) and control limits (99 percent confidence interval). A single value outside the
control limit or two consecutive values outside the warning limits initiates corrective action (i.e.,
instrument recalibration and reanalysis of all samples analyzed since the last acceptable QCCS value).
The QCCS values are used to assess precision and accuracy of the data.
123
-------
3.0
2.8
2.6
2.4
2.2
2.0
1.8
1.6
1.4
1.2
1.0
O.B
0.6
0.4
0.2
0.0
mean
(W-)3SD
12 16 20 24
QCCS number
28 32 36
40
Figure 5.5. Quality Control Check Sample Control Chart for the
nC17/pristane Ratio for the Summer of 1990.
3.0
2.8 -
2.6 -
2.4 -
2.2 -
2.0 -
1.8 -
1.6 -
1.4
1.2
1.0
0.8
0.6
0.4
0.2
0.0
•mm"
mean
(W-)3SD
12 16 20 24
QCCS number
28 32 36 40
Figure 5.6. Quality Control Check Sample Control Chart for the
nC18/phytane Ratio for the Summer of 1990.
124
-------
QA/QC
Unweathered Prudhoe Bay crude oil was used as the QCCS for oil chemistry analyses. One
QCCS was analyzed with each group of 10 samples, in addition to QCCS analysis at the beginning and
end of each analytical run. For nutrient analysis, a volume of an appropriate concentration of
standard solution for each nutrient was analyzed as the first sample, after every 10 samples, and as
the last sample in each analytical run. For carbon dioxide and TOC measurements, a standard solution
in the midrange of the sample concentrations was analyzed as the first sample, after every 15 samples
or midbatch, and as the last sample in each analytical run.
Detection Limit QCCS
The detection limit QCCS (DL-QCCS) is a sample containing the analyte of interest at a
concentration 2 to 3 times the IDL. The purpose of the DL-QCCS is to eliminate the need for
formally determining the IDL for every analytical run. A DL-QCCS was included in each analytical
run of nutrient samples. Warning and control limits were established at 2 times and 3 times the
standard deviation of the nominal value of the DL-QCCS.
Matrix Spike
A matrix spike consists of the analyte of interest at a concentration approximating 10 times the
detection limit. The spike recovery should be 90 to 110 percent of the known value of the spike. If
spike recovery is not within this range, the problem should be corrected prior to continuing analysis
of routine samples. Two matrix spike samples were included in each analytical run of nutrient
samples.
Calibrations
Calibration is the establishment of a relationship between standards of known values and the
recorded output of a measurement system. The concentrations of the standards should bracket the
expected range of routine sample concentrations. Generally, three to five standards covering the
linear range of the particular measurement system are recommended, including one "zero concentra-
tion" or detection limit standard. Three-point calibrations were performed daily for nutrient, carbon
dioxide, and TOC sample analyses and two-point (zero and 100) calibrations were performed every
9 hours for oxygen and carbon dioxide monitors used in the biometer tests. For the gas
chromatograph an initial three-point calibration was performed; thereafter a single-point check was
125
-------
QA/QC
performed daily. The three-point calibration was repeated if the single-point check fell outside
limits. Additionally, all balances were checked weekly with a minimum of three "S" class weights.
A single weight near the weight of the material to be weighed or in the midrange of the balance was
checked daily.
Instrument Detection Limits
The IDL is generally defined as 3 times the standard deviation of 7 to 10 nonconsecutive blank
analyses. The IDL is formally determined prior to initiation of sample analysis, at periodic intervals
thereafter, and following any major disruption of the instrument (repair, move, etc.). A DL-QCCS
eliminates the need to formally determine the IDL on a daily basis. The IDL was determined for
nutrient analyses and oil chemistry.
Assistance Audits
An important component of the Quality Assurance Program was the use of assistance audits.
An assistance audit places the proper quality management resources in the right place at the right
time. It was a key concept in the implementation of the Quality Assurance Program. Rather than
waiting until the end of a project and assessing the quality of the data, the assistance audit centers
on improving the quality of the data on a "real time" basis. The goal of the assistance audit program
was to provide data which will meet the needs of the program. An Assistance Audit should give
workers a sense of Quality Program "ownership"; give workers a direct way to address corrective
actions; and gives the Quality Manager a quick and easy way to implement QA/QC measures.
A summary of Assistance Audits conducted for the Oil Spill Bioremediation Project is located
below:
Audit Field Sampling Activities
Date: June 8 and June 11, 1989
Auditor Mike Papp
Summary: SOPs, sampling design and QAP were followed. Equipment, supplies, and crew training
were all sufficient for the task.
Corrective Actions: No corrective actions were required.
126
-------
QA/QC
Audit: On-Site Assistance Audit Sediment Oil Chemistry SAIC Laboratory Kasitsna Bay, Alaska
Date: June 10 and 11, 1989
Auditor Werner Beckart
Summary:
• Laboratory personnel were well educated, had adequate experience and were dedicated to
their work.
• Good and adequate instrumentation and equipment were available.
• The QA/QC practiced at the laboratory was adequate.
Corrective Actions: Acquire more freezer space.
Audit Microbiology
Date: June 12, 1989
Auditor Linda Stetzenbach
Summary: Great care must be used in the interpretation of the microbiology data gathered from this
Project.
Corrective Actions:
Review the Most Probable Number determination procedures.
Review the method of detection of positive tubes in the MPN procedure.
Analyze the cobble beach material.
Process the beach material in a timely manner.
Consider limited assay methods.
Review the data analysis procedure.
Results: John Rogers addressed each corrective action in a letter to Dan Heggem dated July 11,1989.
Audit: On-Site Assistance Audit SAIC Nutrient Laboratory Kasitsna Bay, Alaska
Date: July 20, 1989
Auditor Dan Hillman
127
-------
QA/QC
Summary:
The QA/QC in this laboratory exceeded the requirements in the Project QA Plan.
All QC data examined were acceptable.
The use of Control Charts was discussed with Lab personnel.
Field or lab blanks were not included with the routine sample batches.
The use of consistent Instrument Detection Limits and dropping Reporting Limit was
recommended.
The lab personnel were highly motivated and dedicated to achieving QA requirements of the
Project.
Lab performance was outstanding.
Corrective Actions:
Consider the use of Control Charts.
Use consistent Instrument Detection Limits.
Consider dropping the use of Reporting Limits.
Include field and lab audit blanks in routine sample batches.
Investigate automated methods and implement if possible.
Audit- SAIC Oil Analysis Laboratory at Kasitsna Bay, Alaska
Date: June 27 - 30, 1989
Auditor E. Neil Amick
Summary:
Facilities, equipment and personnel were present to provide quality data.
Sample tracking and documentation procedures were reviewed.
Analytical methods were observed.
Data from QC and routine samples were audited.
Data produced were of good quality and met the objectives of the study.
Some deviations from the QAP and SOPs were found.
Corrective Actions:
• Send 10 routine oil extracts and 3 Prudhoe Bay oil audit samples to EMSL-LV for chemical
analysis.
• Follow the recommended procedures in the QAP concerning the use of QCCS samples.
128
-------
QA/QC
Audit Assistance QA Review of the Ecological Monitoring Program
Date: July 21 - August 9, 1989
Auditor Jim Pollard
Summary:
Field sampling reviewed; some early data acquisition problems were resolved.
QA Plan protocols for chlorophyll processing procedures were in error. The QA Plan needed
to be corrected.
Tracking and documentation systems were reviewed and found to be reliable. Corrections
were made in the data base format to ease data analysis and interpretation.
The design for estimation of measurement error needed to be changed to include a different
type of duplicate sample. Field splits vs lab duplicates.
The field replicate procedure was reviewed.
A synthetic performance audit sample for nutrient analysis was suggested.
A flow diagram for ecological monitoring was developed.
A laboratory QC procedure for chlorophyll a measurement was developed.
A post data entry verification procedure for chlorophyll a measurement was developed.
Data entry and bench sheet generation procedures for chlorophyll a, 14C primary production,
and bacterial production were developed.
Audit: Cincinnati Nutrient Lab Assistance Audit
Date: November 2, 1989
Auditor Dan Hillman
Summary:
Personnel were experienced and suitable for the work.
Laboratory facilities were adequate.
Standard methodology was followed with minor modifications necessary due to the sea water
matrix.
All required QA/QC was performed.
Detection limits were not determined.
Corrective Actions: Alaskan lab and Cincinnati lab should not be compared due to detection limit
differences. The Cincinnati lab raw data may be reanalyzed to determine detection limits. A
recalculation of Cincinnati data which was less than 0.2 ppm will result in comparable detection
limits.
129
-------
QA/QC
Audit- Oil-Site Assistance Audit Sediment Oil Chemistry SAIC Laboratory San Diego, CA
Date: November 3, 1989
Auditor Neal Amick
Summary:
• The results from the SAIC San Diego lab should be directly comparable to the Kasitsna Bay
lab.
• Personnel were appropriate for the work.
• The laboratory facility was suitable.
• The quality control and the sample analysis met or exceeded those required in the QAPP.
Corrective Actions: A few deviations from the methods as written in the QAPP were noted. These
were the same changes as noted in the Kasitsna Bay, Alaska audits and were included in the next
revision of the QAPP.
Audit: On-Site Audit of the Nutrient Laboratory Operations at Valdez, Alaska
Date: August 8, 1990
Auditor Mike Papp
Summary:
Facilities were small and cramped. Areas were generally well organized and well maintained to
maximize efficiency of available space. Safe storage of chemicals and inadequate freezer space were
areas of concern. Archived samples were stored in freezer facilities in the microbiology laboratory.
A complete inventory of equipment and materials was not done. In general, equipment and materials
were adequate, both in quantity and condition, for the needs of the project. Analysts record bench
data could have been recorded neater.
The detection limit QCCS was not run throughout the project. Therefore, no documentation of the
instrumental detection limit was available on a batch basis. It was recommended during the audit that
detection limits be formally determined. This was done immediately after the audit.
130
-------
QA/QC
Recommendations and Corrective Actions: Recommendation of the audit included determination of
the instrumental detection limits and implementation of a procedure to guard against transcription
error during computer data entry. Both of these recommendations were implemented. A review of
performance audit data undertaken by the auditors revealed several problems. First, variability of
the results was greater than anticipated. Second, estimates of the audits required that the same audits
be used for both nitrite and nitrate analyses; audits for remaining nitrate analyses were scheduled to
ensure this was done. Finally, ammonia loss was observed in the high concentration audit material.
The remaining aliquots of this audit were removed and destroyed. Following the audit, on-going
review of the data by the QA staff and data base manager revealed at least two incidents in which
dilution factors appeared to contribute to poor results, either because they were incorrectly recorded
or excessive dilution resulted in readings at the low end of the calibration curve. No provisions for
prevention of these types of errors was included in the laboratory procedures or the QA Plan. It was
recommended that a standard procedure be developed or checks implemented to guard against errors
related to sample dilution.
Audit: On-Site Audit of the Microbiology Laboratory Operations at Valdez, Alaska
Date: August 9, 1990
Auditor: Mike Papp
Summary:
Facilities were small and cramped, but the lab was well organized and well maintained to provide
maximum efficiency within the constraints of cramped space. Efforts were made to provide safe
storage of chemicals. Temperatures of the refrigerators and freezers were not recorded daily. It was
recommended that logs be maintained for each refrigerator and freezer unit. Hazardous materials
used in the laboratory included acids, bases, solvents (including methylene chloride) and radioactive
materials (carbon 14). Labeling and handling procedures were generally adequate; however, operation
of the hood was not properly checked. Filters in the hood should have been checked and static
pressure measurements performed.
A complete inventory of equipment was not done. With the exception of the oven, equipment was
generally adequate both in quantity and condition. The oven was potentially dangerous due to a
131
-------
QA/QC
problem with the thermostat, and it was recommended that it be repaired or replaced. It was,
however, clearly labeled, which minimized the risk of overheating.
A recommendation of the audit was to implement a data recording form for the carbon analyzer. This
was done immediately after the audit. Another recommendation was to record date of preparation
and preparer initials on all stock and standard solutions. This, too, was implemented. A standards
logbook was not maintained. It was recommended that such logs be maintained, both to track
chemical usage and to resolve problems related to contaminated chemicals or poof preparation
practices.
Recommendations and Corrective Actions: Specific audit recommendations included the following:
design and implement a data form for the carbon analyzer; implement a procedure for transcription
error checks following computer data entry; improve and standardize logbook layout, including
attachment of printer tapes (carbon analyzer) and identification of the analyst, document Oxymax
calibrations, and make sure stocks and standards were properly labeled, including contents, date of
preparation and identification. All of these specific recommendations were implemented with the
exception of documenting Oxymax calibrations (the Oxymax was nonfunctional since the date of the
audit). Application of QA/QC to microbiology is fairly new; most of the scientists in this laboratory
were not familiar with the concepts of quality assurance. While much remained to be done, the
understanding and application of QA within this laboratory improved over the duration of the
project.
Audit: OB-Site Audit of the Oil Chemistry Laboratory Operations at Valdez, Alaska
date: August 9, 1990
Auditor: Mike Papp
Summary:
Facilities were small and cramped. Areas were generally well organized and well maintained to
maximize efficiency of available space. One inadequacy was safe storage of chemicals—solvents were
tibt property storfed.
132
-------
QA/QC
A complete inventory of equipment and materials was not done. In general, equipment and materials
were adequate, both in quantity and condition, for the needs of the project. The analysts record
bench data could have been recorded neater.
Procedures, including application of QA/QC, were excellent. No problems were found and no
recommendations were necessary.
Recommendations and Corrective Actions: No corrective actions were needed. Procedures appeared
more than adequate to ensure high quality data.
One note—although the chemist wore a respirator while performing methylene chloride extracts, the
extracts were not performed in the hood due to lack of space. Other individuals in the room could
have been exposed.
133
-------
SECTION 6
SNUG HARBOR FIELD RESULTS
VISUAL OBSERVATIONS
Test beaches at Snug Harbor were moderately contaminated. Visually, the cobble plots had a
thin coating of dry, sticky, black oil covering rock surfaces and gravel areas under the cobble. Oil
did not penetrate more than 8 to 10 cm below the gravel surface. In mixed sand and gravel plots, oil
was well distributed over exposed surface areas and commonly found 20 to 30 cm below the surface.
In many areas of the test plots small patches of thick oil and mousse could be found. This material
was very viscous and mixed with extensive amounts of debris.
Approximately 8 to 10 days following oleophilic fertilizer application to the cobble beach plot,
reductions in the amount of oil on rock surfaces were visually apparent. This was particularly evident
from the air where the contrast with oiled areas surrounding the plot was dramatic, etching a clean
rectangle on the beach surface (Figure 6.1). The contrast was also impressive at ground level; there
was a precise demarkation between fertilizer-treated and untreated areas (Figure 6.2).
Close examination of this treated cobble plot showed that much of the oil on the surface of the
rocks was gone. However, considerable amounts of oil were still present under rocks and in the mixed
gravel below these rocks. The remaining oil was not dry and dull like the oil in other areas of the
beach, but appeared softened and more liquid. It was also very sticky, with no tendency to come off
the rocks. At the time of these observations, no oil slicks or oily materials were observed leaving the
beach during tidal flushing.
There also appeared to be a reduction in the amount of oil in the mixed sand and gravel beach
2 to 3 weeks following oleophilic fertilizer application. However, visual differences between treated
and untreated plots were not as dramatic as on the cobble beach. Some loss of subsurface oil in
treated areas was also visually apparent. Reduction of oil contamination was particularly evident
during the sampling process, as noticeably less oil remained on sampling equipment used on this beach
plot. The amount of oil on all other plots at this time appeared unchanged since the initiation of the
study. There were essentially no visual indications of oil removal on plots treated with slow- release
fertilizer briquettes.
134
-------
Figure 6.1. Approximately 8 to 10 Days Following Application of INIPOL to
the Cobble Beach Plot at Snug Harbor, Reductions in the
Amount of Surface OH (As Compared to Surrounding Untreated
Oiled Areas) Was Evidenced by a Clean Rectangle on the
Beach Surface.
Rgure 6.2. At Ground Level, the Reduction In Oil on the INIPOL Treated
Plot was also Strikingly Apparent.
135
-------
SNUG HARBOR
Since several substances in INIPOL are known to act as surfactants or to otherwise change the
consistency of oil on rock surfaces, the question arose as to whether INIPOL acted to alter the
physical characteristics of the oil and could remove oil from rock surfaces without biodegradation.
A study to evaluate the rock-washing characteristics of INIPOL under conditions that precluded
biological activity showed that oleophilic fertilizer removed only 0.84% of the oil in a test where the
fertilizer was added to ensure complete surface coverage. More than 30% of the oil was removed by
some preparations sold specifically for this purpose. Approximately 45% of the oleophilic fertilizer,
by weight, remained with the oiled rock in the testing regime used.
Thus, the oleophilic fertilizer did not wash oil off rocks at typical Prince William Sound water
temperatures. Based on the results of this study, it is reasonable to expect that oil removal associated
with oleophilic fertilizer applications was not the result of a chemical or physical process.
Over the next two to three weeks, the cleaned rectangle on the cobble beach remained clearly
visible. Subsurface oil remained but was increasingly less apparent, and untreated reference plots
appeared relatively unchanged. The oleophilic-treated mixed sand and gravel plot actually showed
a greater loss of oil, appearing increasingly cleaner.
Six to eight weeks after fertilizer application the contrast between the treated and untreated
areas on the cobble beach narrowed. This was due to reoiling from subsurface material concurrent
with the slow removal of oil on the beach material surrounding the plot. However, it was evident that
the total amount of oil on the treated plots had decreased substantially relative to untreated control
plots. The corresponding mixed sand and gravel plot was also reoiled but to a lesser extent. Oil
contamination was still observed at all other plots, but had generally decreased since the initiation of
the study.
Toward the end of the summer season the area used for the bioremediation study became
steadily cleaner, including most of the areas surrounding the test plots. This was attributed to several
storms and more frequent rainfall. An untreated, heavily contaminated area to the south remained
heavily contaminated throughout the summer by all visual criteria.
136
-------
SNUG HARBOR
NUTRIENT CONCENTRATIONS
Figures 6.3a-e show the ammonia concentrations found in interstitial water for the treatment
and reference plots. The initial background ammonia concentrations (T*0, Figure 6.3a) were low,
and uniform throughout the plots. One to two days after application (T*l, Figure 6.3b) of the
fertilizers, an increase in the ammonia concentrations was evident only in the plots treated with the
oleophilic fertilizer. Concentrations within the zones, however, were highly variable. Based on the
literature and laboratory nutrient release experiments described in Section 10, a pulse of ammonia was
expected following application.
In contrast, ammonia concentrations in the plots treated with the slow-release briquettes
remained at background levels. This is not unreasonable, because nutrient release studies with the
briquettes showed nitrogen was released entirely as TKN, probably as urea. The absence of elevated
NH4 concentrations suggests that, on the beaches, hydrolysis of urea by microorganisms leads to
immediate uptake of the resulting ammonia by bacteria or algae.
Eight to 10 days after application of the fertilizers (T«2, Figure 6.3c), ammonia concentrations
were above background only in the sand and gravel plot treated with oleophilic fertilizer. Ammonia
concentrations in plots treated with briquettes were comparable to the reference plot. At
approximately 4 and 6 weeks after the fertilizer application (T«3, Figure 6.3d, and T-4, Figure 6.3e,
respectively), no substantial difference in the ammonia concentrations was apparent between the
treatment and the reference plots.
Figures 6.4a-c show nitrate/nitrite concentrations in interstitial water for the treatment and
reference plots. One to 2 days following application (T-l, Figure 6.4a), notable concentrations of
nitrate were found in samples taken from the briquette-treated beaches. Eight to 10 days after
application (T-2, Figure 6.4b), sand and gravel beaches treated with oleophilic fertilizer showed
substantially higher levels of nitrate/nitrite nutrients than the untreated control plots. Plots treated
with water-soluble fertilizer showed only slightly elevated concentrations. One month after fertilizer
application (T-3, Figure 6.4c), nitrate/nitrite levels in the treated plots were still higher than in the
reference plots, particularly for the cobble beach treated with briquettes. Neither the INIPOL nor
the briquettes contain nitrate or nitrite. Thus, it is possible that the presence of these nutrients
resulted from ammonia conversion to nitrite by nitrification. If this was the case, it is unusual that
it was confined only to the treated beaches.
137
-------
HIGH TIDE
MID TIDE
LOW TIDE
I-
u>
00
I
z
(9
C
|-
•
fti
i
\
I
?** iti
»1 IM
If SM
799 Iff
n
Sample Block
Sample Block
Sample Block
Legend
• ESR
• 050(6-8-89)
• OSW (6-10-89)
0 SCW (6-10-89)
Q SCO (6-9-89)
•SCR
ESR = Control Mixed Sand and Gravel
SCR = Control Cobble
OSO = Oleophilic Fertilizer-Treated Mixed Sand and Gravel
SCO = Oleophilic Fertilizer-Treated Cobble
OSW = Water-Soluble Fertilizer-Treated Mixed Sand and Gravel
SCW = Water-Soluble Fertilizer-Treated Cobble
NS - No Sample Taken
DL = Detection Limit
ND = No Data Available
Figure 6.3a. Ammonia Concentrations In Interstitial Water Samples Prior To
Fertilizer Application (T = 0).
-------
HIGH TIDE
MID TIDE
LOW TIDE
U)
300
Z
=t
co"1"
c
o
E m
4 -
1 HI nl III _I III
1
400
Z MO
=1
CO
c «.
o "•
E
f
-,l a.. .,,,,.
t
fOO
400-
z
v%
CO
c
0 200
E
< ,00
0
t
< .. . . .Jo.
Sample Block
Sample Block
i« 17 it JT
Sample Block
Legend
•ESR
• OSO (6^-89)
• OSW (6-10-89)
0SCW (6-10-89)
n SCO (6-9-89)
• epn
wwfl
ESR =
SCR =
OSO =
SCO =
OSW =
sew =
NS =
DL =
NO =
Control Mixed Sand and Gravel
Control Cobble
Oleophilic Fertilizer-Treated Mixed Sand and Gravel
Oleophilic Fertilizer-Treated Cobble
Water-Soluble Fertilizer-Treated Mixed Sand and Gravel
Water-Soluble Fertilizer-Treated Cobble
No Sample Taken
Detection Umlt
No Data Available
Figure 6.3b. Ammonia Concentrations In Interstitial Water Samples 1-2 Days
Post Fertilizer Application (T = 1).
-------
MID TIDE
LOW TIDE
1-
z
i.
^
"E
o
E
e
888488 888488 888488 888488 118488
40
M
z
II*
•i^M 20'
C
o
E 18
<
M
*
4488 4
488 4
488 4|•_••.
SCR
ESR =
SCR =
OSO =
SCO =
OSW =
sew =
NS =
DL =
ND =
Control Mixed Sand and Gravel
Control Cobble
Oleophilic Fertilizer-Treated Mixed Sand and Gravel
Oleophilic Fertilizer-Treated Cobble
Water-Soluble Fertilizer-Treated Mixed Sand and Gravel
Water-Soluble Fertilizer-Treated Cobble
No Sample Taken
Detection LImtt
No Data Available
Figure 6.3c. Ammonia Concentrations In Interstitial Water Samples 8-10 Days
Post Fertilizer Application (T = 2).
-------
HIGH TIDE
MID TIDE
LOW TIDE
Sample Block
Sample Block
Sample Block
Legend
•ESR
• OSO (6-8-69)
• OSW (6-10-89)
E2 SCW (6-10-89)
D SCO (6-9-89)
• SCR
ESR = Control Mixed Sand and Gravel
SCR = Control Cobble
OSO = Oleophilic Fertilizer-Treated Mixed Sand and Gravel
SCO = Oleophilic Fertilizer-Treated Cobble
OSW = Water-Soluble Fertilizer-Treated Mixed Sand and Gravel
SCW = Water-Soluble Fertilizer-Treated Cobble
NS = No Sample Taken
DL = Detection Limit
ND = No Data Available
Figure 6.3d. Ammonia Concentrations In Interstitial Water Samples 30 Days Post
Fertilizer Application (T = 3).
-------
*
=t
•*
Ammonia
HIGH TIDE
MM
ISIS
HI',
id!
(IIS
MID TIDE
LOW TIDE
CO
'E
o
E
AVMBQ*
Sample Block
Sample Block
Sample Block
Legend
•ESR
HOSO (6-8-89)
• OSW (6-10-89)
E3SCW (6-10-89)
D SCO (6-9-89)
• SCR
ESR = Control Mixed Sand and Gravel
SCR = Control Cobble
OSO = Oleophilic Fertilizer-Treated Mixed Sand and Gravel
SCO = Oleophilic Fertilizer-Treated Cobble
OSW = Water-Soluble Fertilizer-Treated Mixed Sand and Gravel
SCW = Water-Soluble Fertilizer-Treated Cobble
NS = No Sample Taken
DL = Detection Limit
ND = No Data Available
Figure 6.3e. Ammonia Concentrations In Interstitial Water Samples 6 Weeks
Post Fertilizer Application (T = 4).
-------
o
u
c
o
o
5
"o
1
4
in
i
HIGH TIDE
ii tiigpi iiigpi in
i i
it in
An*
81
•^
MID TIDE
LOW TIDE
^^
o01.
CO
0 ^
0
0
c »•
o
o
"5
0 '
18
n
it 18
i
;?i is
t i
J
m
\ I
J K
rl fi
1 A««
it
»••
Sample Block
Sample Block
Sample Block
Legend
•ESR (6-13-89)
• OSO (6-9-89)
• OSW (6-12-89)
0 SCW (6-12-89)
QSCO (6-10^9)
• SCR (6-13-89)
ESR = Control Mx*d Sand and Gravd
SCR s Control Cobbte
OSO r Otoophlllc FwtiKzw-Traated Mixed Sand and Gravel
SCO = Oteophlllc Fertllt»r-Traated Cobbte
OSW = Water-Solubto Fertilizer-Treated Mixed Sand and Gravel
SCW = Water-Soluble Fertilizer-Treated Cobb*e
NS = No Sample Taken
DL = Detection Umtt
ND = No Data Available
Figure 6.4a.
Nitrate / Nitrite Concentrations In Interstitial Water Samples 1-2 Days Post
Fertilizer Application (T = 1).
-------
HIGH TIDE
MID TIDE
LOW TIDE
t '
eo •
O
Z 4
•5 ,
6
g -
t
CO
r* a
111*89 lit
ii 111
ii 111
!i 111
ii
Sample Block
Sample Block
Sample Block
Legend
• ESR (6-18-89)
• OSO (6-18-89)
•OSW (6-18-89)
E2SCW (6-19-89)
DSCO (6-19-89)
• fM\f%
SCR
ESR =
SCR =
OSO =
SCO =
OSW =
sew =
NS =
DL =
ND =
Control Mixed Sand and Gravel
Control Cobble
Oleophilic Fertilizer-Treated Mixed Sand and Gravel
Oleophilic Fertilizer-Treated Cobble
Water-Soluble Fertilizer-Treated Mixed Sand and Gravel
Water-Soluble Fertilizer-Treated Cobble
No Sample Taken
Detection Limit
No Data Available
Figure 6.4b. Nitrate / Nitrite Concentrations In Interstitial Water Samples
8-10 Days Post Fertilizer Application (T = 2).
-------
HIGH TIDE
MID TIDE
LOW TIDE
Atnngt
Anragt
Sample Block
Sample Block
Sample Block
Legend
•ESR (7-7-89)
• 080(7-7-89)
•OSW (7-7-89)
fg) SCW (7-7-89)
nSCO (7-7-89)
• SCR (7-7-89)
ESR = Control Mixed Sand and Gravel
SCR = Control Cobble
OSO = Oleophilic Fertilizer-Treated Mixed Sand and Gravel
SCO = Oleophilic Fertilizer-Treated Cobble
OSW = Water-Soluble Fertilizer-Treated Mixed Sand and Gravel
SCW = Water-Soluble Fertilizer-Treated Cobble
NS = No Sample Taken
DL = Detection Limit
ND = No Data Available
Figure 6.4c. Nitrate / Nitrite Concentrations In Interstitial Water Samples
30 Days Post Fertilizer Application (T = 3).
-------
SNUG HARBOR
Samples taken in July from streams near Eagle and Otter beaches showed measurable levels of
inorganic nutrients. The stream south of Eagle beach had 5.2 n nitrogen as nitrate. Stream samples
taken adjacent to Otter beach contained an average of 4.8 n nitrogen as nitrate. A sample of snow
collected from a snow pile 300 yards southeast of Eagle Beach (a result of a winter avalanche)
contained 2.8 n of nitrogen as ammonia, 0.54 p of phosphorus as phosphate, and 1.1 n of nitrogen as
nitrate. Although these concentrations were relatively low, they indicate that snow-melt and runoff
may serve as important sources of nutrients for limited sections of the shoreline, particularly in the
spring and early summer. Even though some of the test plots were located near the streams, nutrient
concentrations in the plots were probably unaffected. The stream was an unlikely source of the
nitrate found in the treated beaches, as no elevated nitrate/nitrite was detected in reference beaches
having equal exposure to the freshwater. Also, no nitrate/nitrite was found at T-0 in any of the plots.
On June 19, the briquette bags were repositioned, and all bags were placed in the mid- and low-
tide zones of the plots. The fertilizer was therefore submerged for longer time periods, enhancing
nutrient transport in these zones. In general, this repositioning did not have a detectable impact on
nutrient distribution on the beaches; nutrient concentrations in the zones showed no new trends. It
was still apparent that minimal dispersion of the nutrients was occurring from the briquettes in areas
of the shoreline not subjected to routine tidal washing. Precipitation during the month of June was
probably insufficient to effectively transport nutrients released from the bags of briquettes located
in the high-tide zone.
OIL CHEMISTRY
Oil Residue Weight
Cobble Plots (Seal Beach)
Figures 6.5 to 6.10 show changes (decay curves) in residue weights of oil on the cobble surface
(referred to as cobble samples) and in the beach material below the cobble (referred to as mixed sand
and gravel under cobble samples) for the three test plots on Seal Beach (untreated, treated with
oleophilic fertilizer, and treated with fertilizer briquettes). The residue weights are normalized per
kilogram weight of cobble or mixed sand and gravel. Given the relatively smooth surface of the
cobblestones, it was assumed that a consistent relationship between rock surface area and rock weight
existed. Variability in the residue weights was therefore a function of both the sampling of
146
-------
10
0.1
3 0.01
0.001
0
e
OBJun88 24Jun88 10 Jul 88 26 Jul 86 11 Aug 80 27 Aug 89 12Sep89
Sampling Data
° Analysis Value
• Median
Figure 6.5. Change In OH Residue Weight Through Time for Seal Beach
(Untreated Control) at Snug Harbor (Cobble Surface).
10
!
0.1
3 0.01
0.001
06Jun89 24Jun88 10 Jul 88 26 Jul 88 11 Aug 88 27 Aug 89 12 Sop
Sampling Data
I o Anolvsis Volue •
Figure 6.6. Change In OH Residue Weight Through Time for Seal Beach
(WOODACE Briquettes) at Snug Harbor (Cobble Surface).
147
-------
10 \
1-
0.1
8 0.01
0.001
06Jun89 24Jun89 10Jul89 26 Jul 89 11 Aug 89 27 Aug 89 12 Sep 89
Sampling Date
0 Anglysis Value
Median
Figure 6.7. Change in Oil Residue Weight Through Time for Seal
Beach (INIPOL) at Snug Harbor (Cobble Surface).
10
1
0.1
a01
0.001
08 Jun 89 24 Jun 89 10 Jul 89 26 Jul 89 11Aug89 27 Aug 89 12 Sep
Sampling Date
0 Anolysis Volue
Median
Figure 6.8.
Change in Oil Residue Weight Through Time for Seal Beach
(Untreated Control) at Snug Harbor (Mixed Sand and Gravel
Under Cobble). The Number of Samples Showing
Concentrations Below Detection Limit is Shown Above the
Sampling Date.
148
-------
10f
0.1
0.01
0.001
M
T 1 1 1 1 1 1
OBJunSQ 24 Jun88 10Jul89 26 Jul 8G 11 Aug 88 27Aug88 12 Sop
Sampling Date
0 AnalYiii Value
Figure 6.9. Change In OH Residue Weight Through Time for Seal
Beach (WOODACE Briquettes) for Snug Harbor (Mixed
Sand and Gravel).
10
0.1
0.01
0.001
-, , 1 , i 1 r
OBJuntt 24Jun80 10 Jul 89 26 Jul 89 11 Aug 89 27 Aug 89 12 Sep 89
Sampling Date
0 Anoly»i» Volue
Median
Figure 6.10. Change In Oil Residue Weight Through Time for Seal
Beach (INIPOL) at Snug Harbor (Mixed Sand and Gravel).
149
-------
SNUG HARBOR
heterogenous beach material and the uneven distribution of oil on the beach. Because of this
variability, the median values were used to develop the decay curves. Table 6.1 provides the actual
median values and their percent change with each sampling date. Figure 6.11 graphically represents
the percent change in the medians, and Table 6.2 compares the initial decay rate for each of the
treatments.
Oil residue weight changes over the entire test period were essentially biphasic. On the
untreated control beach, for example, there was relatively little change over the first 29 days of the
test period, but this was then followed by a faster rate of decline. Comparing slopes of the two
different phases (day 0 to day 29; day 29 to day 92) on this control plot showed that the initial rate
is not significantly different from zero at the 95% confidence level (Table 6.2), but the latter slope
was significantly different from zero. The reason for this change in rate between the July 8 and July
29 samplings is not known.
The decay curve for the briquette fertilizer-treated plot appeared to be biphasic as well,
although an anonymously high median value on the July 8 sampling complicates the interpretation.
However, the slopes of the two different phases (day 0 to day 29; day 29 to day 92) were statistically
different from each other (Table 6.2), with the results generally mirroring those seen on the untreated
control.
The decay curve for the IN1POL fertilizer-treated plot also appeared biphasic, but in a different
sense; the initial decay was rapid and the latter decay was slower. During the first 29 days of the test
period there was approximately a 45% decrease in the median oil residue weight on the INIPOL
fertilizer-treated plot, but on the other plots there was essentially little change in the median values
(Table 6.1). The greater weight loss on the INIPOL fertilizer-treated plot corresponds to the visual
observations. Both phases showed slopes significantly different from zero. This response to the
fertilizer application suggests that perhaps the presence of higher nutrient concentrations observed
during the initial 2 to 3 weeks following application caused an enhancement of oil biodegradation
rates, ultimately leading to greater initial loss of oil residues. However, other factors affecting the
loss of oil residue cannot be ruled out (e.g., physical scouring by tide, wave action, chemical effects,
etc.). If it was a nutrient effect, then it apparently was short-lived, as decay rates during the latter
part of the test period were approximately the same as those seen on the untreated control plots.
150
-------
SNUG HARBOR
TABLE 6.1. MEDIAN VALUES (MG/G) AND STATISTICAL COMPARISONS OF OIL
RESIDUE WEIGHTS IN COBBLE SURFACE SAMPLES FROM DIFFERENT BEACH
TREATMENTS AT SNUG HARBOR
Median Values (% of 6/9/89 Median)
Sampling
Date
June 9
June 17
June 25
JulyS
July 29
August 26
September 9
Day
0
8
16
29
50
78
92
Untreated
Control
0.38
0.41 (108)
0.32 (85)
0.32 (85)
0.19 (49)
0.10(27)
0.076 (20)
Briquettes
0.88
0.84 (96)
0.79 (90)
1.09(124)
0.66 (75)
0.25 (28)
0.15(17)
INIPOL
0.59
0.46 (78)
0.37 (63)
0.32 (55)
0.21 (35)
0.11 (18)
0.057(10)
Mann-Whitney Test Results'
Sampling
Date
June 9
June 17
June 25
JulyS
July 29
August 26
September 9
Briquettes vs. Briquettes vs.
INIPOL Untreated Control
B>
B>
B>
B>
B>
B>
B>
B>C
Same
B>C
B>C
B>C
B>C
B>C
INIPOL vs.
Untreated Control
I>C
Same
Same
Same
Same
Same
Same
' 95 Percent Confidence Level
151
-------
UJ
o>
55
§
Briquettes
-e- INIPOL
Untreated Control
DAYS
Figure 6.11. Change in the Median Residue Weight, Expressed as
Percent of the 6/9/89 Median Over Time for the Briquette,
INIPOL, and Untreated Control Beaches at Snug Harbor
(Cobble Surface). All Variability Is not Shown Because the
Actual Data Points are not Presented.
-------
SNUG HARBOR
TABLE 6.2. RATE ANALYSIS OF NATURAL LOG-TRANSFORMED OIL RESIDUE
WEIGHTS IN MG/G IN COBBLE SURFACE SAMPLES VERSUS TIME
(JULY 8. 1989 TO JULY 29. 1989 ONLY) FOR TEST BEACHES AT SNUG HARBOR
Beach
Briquettes
INIPOL
Untreated
Control
Significance of Slope
Slope Greater than Zero Half-Life,
(Std. Dev.) N T-value p* days
-0.006 73 -0.82 0.42 122
(0.007)
-0.016 80 -2.4 0.02 44
(0.007)
-0.006 65 -0.56 0.58 124
(0.010)
Time to Remove
90%, days
404
146
411
" Only the INIPOL rate is significantly different from zero at the 95 percent confidence level
It is unclear why the oil residue weight decay appeared to change suddently around the
beginning of July. If the response is related to nutrient availability, it is possible that the end of the
algal spring bloom may have reduced the competition for available nutrients and provided nutrients
for more oil degradation. However, the field nutrient data do not show much change in the
background nutrient concentrations during this period of the test. It is also possible that the greatest
decay in oil residue weight was due to the warm temperatures experienced in July.
Changes in oil residue weight through time for the mixed sand and gravel under cobble are
shown in Figures 6.8 to 6.10. Decreases in residue weight decay were only apparent on the untreated
control plot but the decreases was not significant (Table 6.3 and Figure 6.12). The oil residue weights
on all plots were quite scattered, with median values randomly increasing or decreasing with time.
In general, very low concentrations of oil were present; this may have been responsible for the scatter.
Correspondingly, none of the decay rates on treated and untreated plots had slopes significantly
different from zero (Table 6.4). Many samples on the untreated control plot had undetectable
concentrations of oil, indicating that the oil on these plots was very heterogeneously distributed. It
is also important to note
153
-------
SNUG HARBOR
TABLE 6.3. MEDIAN VALUES (MG/G) AND STATISTICAL COMPARISONS OF OIL
RESIDUE WEIGHTS IN MIXED SAND AND GRAVEL SAMPLES UNDER COBBLE
FROM DIFFERENT BEACH TREATMENTS AT SNUG HARBOR
Median Values (% of 6/9/89 Median)
Sampling
Date
June 9
June 17
June 25
Julyg
July 29
August 26
September 9
Day
0
8
16
29
50
78
92
Untreated
Control
0.32
0.50(157)
0.31 (95)
0.35(108)
0.22 (69)
0.20 (64)
0.28 (88)
Briquettes
0.57
0.40 (70)
0.57(100)
0.40 (70)
0.56 (98)
0.27 (48)
0.29(51)
INIPOL
0.31
0.47(151)
0.26 (84)
0.31 (100)
0.35(111)
0.25 (80)
0.14(46)
Mann-Whitney Test Results*
Sampling
Date
June 9
June 17
June 25
July8
July 29
August 26
September 9
Briquettes vs.
INIPOL
B> 1
Same
B> I
Same
B> I
Same
B> 1
Briquettes vs.
Untreated Control
Same
Same
B>C
Same
B>C
Same
Same
INIPOL vs.
Untreated Control
Same
Same
Same
Same
Same
Same
I
-------
I
2
Briquettes
INIPOL
Untreated Control
DAYS
Figure 6.12. Change in the Median Residue Weight, Expressed as
Percent of the 6/9/89 Median Over Time for the Briquette,
INIPOL, and Untreated Control Beaches at Snug Harbor
(Mixed Sand and Gravel Under Cobble). All Variability Is not
Shown Because the Actual Data Points are not Presented.
155
-------
SNUG HARBOR
TABLE 6.4. RATE ANALYSIS OF NATURAL LOG-TRANSFORMED OIL RESIDUE
WEIGHTS IN MG/G IN MIXED SAND AND GRAVEL SAMPLES UNDER COBBLE
VERSUS TIME (JULY 8, 1989 TO JULY 29, 1989 ONLY)
FOR TEST BEACHES AT SNUG HARBOR
Beach
Briquettes
INIPOL
Untreated
Control
Slope
(Std. Dev.)
0.008
(0.010)
-0.0008
(0.008)
-0.007
(0.009)
Significance of Slope
Greater than Zero
N T- value p*
81 0.85 0.40
78 -0.11 0.91
77 -0.84 0.40
* None of the rates are statistically different from zero at the 95 percent confidence level
that the INIPOL application to the cobble surface did not displace the oil and cause it to collect in the
mixed sand and gravel below; that is. the apparent large increase in oil residue weight in the mixed
sand and gravel on the INlPOL-treated plot at the first sampling after fertilizer application was also
observed on the untreated control plot.
Mixed Sand and Gravel Plots (Eagle and Otter Beaches)
The decay in oil residue weight on the beaches consisting only of mixed sand and gravel is
shown in Figures 6.13 to 6.15 and Table 6.5. Figure 6.16 showed that the percent change in the
median oil residue weights with respect to the Day 0 sampling. There appears to be a slow steady
decay on both the untreated control and INIPOL fertilizer-treated plots. However, the slopes of these
decay curves (Table 6.6) were not statistically different from a slope of zero. Unfortunately, the
beaches selected for the test were considerably different in terms of oil contamination, with the
untreated control plot containing almost five times more oil (Table 6.5). This difference may have
obscured any significant differences between the untreated control and treated plots. In addition,
I here was a very large drop in oil concentration on both the untreated control plot and the INIPOL
fertilizer-treated plot between the t • 0 and t - 8 day samplings. The cause of these initial decreases
is not clear.
156
-------
100
10
I
0.1
0.01
o
o
e
o
i 1- -i—— 1 1 i r-
OBJunBQ 24Jun89 10JU89 26Jui89 11 Aug 89 27Aug8Q 12Sep80
Sampling Date
° Analysis Value
M«dion
Hgure 6.13. Change in Oil Residue Weight Through Time for Otter Beach
(INIPOL) at Snug Harbor (Mixed Sand and Gravel).
100 4
10
0.1
0.01
8
o
T
OBJunSQ 24Jun88 10Jul8B 26JJ80 11 Aug 89 27Aug89 12Sep89
Sampling Date
0 Anolvsis Volu*
* Mtdlon
Figure 6.14. Change In Oil Residue Weight Through Time for Eagle
Beach (Untreated Control) at Snug Harbor (Mixed Sand and
Gravel).
157
-------
100
10
5 0.1
0.01
8
0
8
e
a
8
8
e
JJunBB
o o 8
o
o o
0 °
o
2 R n 0
"1 t S 1
I 8 ®
o
o
24 Jin 80 10 JJ 89 28 Jul 80
SwnplnoM
I ° Analysis Votut
o
8
— -U
B ~— -^.
0
o
o
tlAugSQ 27Aug89
•
• Mtdian I
0
a
H
w
12 Sept
Rgure 6.15. Change In OH Residue Weight Through Time for Otter Beach
(WOODACE Briquettes) at Snug Harbor (Mixed Sand and
Gravel).
158
-------
UJ
o>
00
O>
«o
Briquettes
INIPOL
Untreated Control
DAYS
Figure 6.16. Change In the Median Residue Weight, Expressed as
Percent of the 6/9/89 Median Over Time for the Briquette,
INIPOL, and Untreated Control Beaches at Snug Harbor
(Mixed Sand and Gravel Only). All Variability Is not Shown
Because the Actual Data Points are not Presented.
159
-------
SNUG HARBOR
TABLE 6.5. MEDIAN VALUES (MG/G) AND STATISTICAL COMPARISONS OF OIL
RESIDUE WEIGHTS IN MIXED SAND AND GRAVEL ONLY SAMPLES
FROM DIFFERENT BEACH TREATMENTS AT SNUG HARBOR
Median Values (% of 6/9/89 Median)
Sampling
Date
June 9
June 17
June 24
JulyS
July 29
August 26
September 9
Day
0
8
16
29
50
78
92
Untreated
Control
2.85
1.92(68)
1.60(56)
1.82(64)
1.23(43)
0.99 (35)
0.34 (12)
Briquettes
1.80
1.73(96)
1.58(88)
1.83(102)
1.98(110)
0.90 (50)
0.63 (35)
INIPOL
0.71
0.37 (53)
0.41 (58)
0.45 (63)
0.21 (30)
0.12(17)
0.11 (16)
Mann-Whitney Test Results*
Sampling
Date
June 9
June 17
June 24
JulyS
July 29
August 26
September 9
Briquettes vs. Briquettes vs. INIPOL vs.
INIPOL Untreated Control Untreated Control
B>
B>
B>
B>
B>
B>
B>
Same
Same
Same
Same
B>C
Same
B>C
-------
SNUG HARBOR
TABLE 6.6. RATE ANALYSIS OF NATURAL LOG-TRANSFORMED OIL RESIDUE
WEIGHTS IN MG/G IN MIXED SAND AND GRAVEL ONLY SAMPLES
VERSUS TIME (JULY 8, 1989 TO JULY 29, 1989 ONLY)
FOR TEST BEACHES AT SNUG HARBOR
Significance of Slope
Beach Slope Greater than Zero
(Std. Dev.) N T-value p'
Briquettes -0.0008 81 -0.07 0.94
(0.012)
INIPOL -0.008 82 -1.25 0.21
(0.007)
Untreated -0.007 80 -0.97 0.33
Control (0.007)
None of the rates are statistically significant from zero at the 95 percent confidence level
Also, oil residue loss on the briquette fertilizer-treated plot did not appear to change
substantially until the August 26 sampling when a very precipitous decrease (>50% change in the
median in the residue weight) occurred. Again, no explanation for this decrease is available.
Oil Composition
Cobble Plots (Seal Beach)
a) Cobble Surface Samples
Changes through time in the concentration of selected normal alkanes (nCI8, nC22, nC27), the
sum of normal alkanes (nC18 to nC27), the branched alkanes pristane and phytane, and the
nC!8/phytane ratios in cobble surface samples taken from the treated and untreated control plots, are
shown in Figures 6.17 to 6.37. All values of hydrocarbon concentration were normalized to the weight
of oil in the extracted sample. The normal alkanes nC18 to nC27 were chosen because correlation
analysis showed that these hydrocarbons tracked each other consistently. In all cases values below
detection limits were treated as zeros.
161
-------
SNUG HARBOR
For comparative purposes, Tables 6.7 and 6.8 summarize the percent change in the medians of
individual hydrocarbons and the number of samples showing a hydrocarbon concentration of zero
(below detection limits) with each sampling period, respectively. Figure 6.38 provides a graphical
representation of the percent change in the medians for several hydrocarbons.
Several general trends can be identified from this data. For the cobble surface samples from all
beach plots, there appeared to be considerable differences between the treated plots and the untreated
control plot during the first 8 to 16 days of the test period. In most cases, median concentrations of
all the individual normal alkanes decreased by at least 40%, and in some cases considerably more
during this initial period. This was particularly true on the INIPOL fertilizer-treated plot (Figures
6.29 to 6.34 and 6.38; Table 6.7) where in some cases decreases were almost an order of magnitude.
Similar decreases were not seen on the untreated control plot (Figures 6.17 to 6.22 and 6.38; Table
6.7). Statistical analysis using the Mann-Whitney test (Table 6.7) showed no statistical difference in
alkane concentration between the first three sampling dates on the untreated control plot; in other
words, a lag had occurred in the hydrocarbon decay. This was not the case on the fertilizer-treated
plots; median alkane concentrations on the 8-day sampling were significantly less than the initial
median concentrations.
Changes following these initial responses, however, were considerably more complex. The
alkane concentrations on the untreated control plot subsequent to the 16-day sampling (June 25;
Figures 6.17 to 6.22 and 6.38) generally decayed steadily and slowly, with a somewhat faster decay
for lower molecular weight alkanes. Alkane concentrations on the briquette fertilizer-treated plots
(Figures 6.23 to 6.28 and 6.38) appeared to change relatively little subsequent to the 16-day sampling
(following the July 29 sampling, alkane decay on all plots dramatically accelerated for reasons that
are not immediately obvious). If decay rates for the summed alkanes covering the initial SO days of
the test are analyzed statistically (Tables 6.9 and 6.10), the slopes are statistically different from zero
for both the untreated control and the briquette fertilizer-treated plots. However, the influence of
the initial drop on the briquette fertilizer-treated plot was significant. That is, if the t-0 sampling
is not considered then only the slope on the untreated control plot was significantly different from
zero. Thus, it appears that the application of fertilizer briquettes had an initial effect on the oil
composition with respect to normal alkanes, but over time less compositional change occurred than
on the untreated control. This could have been related to the considerably higher oil loading on the
cobble surface of the briquette fertilizer-treated plot, requiring more time to see significant changes
in oil composition (see below).
162
-------
5
"6
00
5
i
c
0.1
0.01
» 8
CJ O
9
o
o
8
o
18
08Jun89 24Jun89 10Jul89 26Jul89 11 Aug 88 27 Aug 89 12 Sep 89
Sampling Date
Analysis Vplue
Mqdiqn
Flaure 6.17. Change In nC18 Concentration Through Time for Seal Beach
(Untreated Control) at Snug Harbor (Cobble Surface). The
Number of Samplea Showing Concentrations Below
Detection Limit It Shown Above the Sampling Date.
10
5
is
0.1
0.01
8
8
o
o
3 Jin 89
1
-* —
0
o
8
\J
O
8
24
o
§
88
2 _
H^_J
o g
1
o
o
D
I 1 1 I
Jun89 10JJ89 26Ju)89 11 Aug 89
Sampling Date
I ° Anoivsii Value • Median
o
8
0 0
0
o °
o
o
12 15
27 Aug 89 12 Sep {
I
Figure 6.18. Change in nC22 Concentration Through Time for Seal Beach
(Untreated Control) at Snug Harbor (Cobble Surface). The
Number of Samples Showing Concentrations Below
Detection Limit is Shown Above the Sampling Date.
163
-------
10
1
Q
•5
B
c
0.1
0.01
o
o
10
14
06Jun89 24Jun89 10Jui89 26JJ89 11 Aug 89 27 Aug 89 12 Sep 89
Sampling Data
I ° Anolysis Volue
Medion
Figure 6.19. Change in nC27 Concentration Through Time for Seal Beach
(Untreated Control) at Snug Harbor (Cobble Surface). The
Number of Samples Showing Concentrations Below
Detection Limit is Shown Above the Sampling Date.
§
i
c
§
c
100
10
1
0.1
0.01
o
o
§
o
08Jun89 24Jun89 10Jul89 26JJ89 11Aug89 27Aug89 12Sep89
Sampling Date
0 Anolysis Value
* Mcdion
Figure 6.20. Change in Sum of Alkane Concentration nCl8 to nC27
Through Time for Seal Beach (Untreated Control) at Snug
Harbor (Cobble Surface). The Number of Samples Showing
Concentrations Below Detection Limit is Shown Above the
Sampling Date.
-------
10-1
8
75
0.1
0.01
8
o
o
o
16
06
89 24Jun89 10Jul8G 26JJ8G 11 Aug 89 27Aug89 12Sep88
Sampling Date
Anglyai» Volye
Median
Figure 6.21. Change In Prlstane Concentration Through Time for Seal
Beach (Untreated Control) at Snug Harbor (Cobble Surface).
The Number of Samples Showing Concentrations Below
Detection Limit is Shown Above the Sampling Date.
10
0.1
0.01
11
OBJun89 24Jun8Q 10Jul89 26JJ89 11 Aug 89 27 Aug 89 12Sep
Sampling Date
I ° Anolysia Volua
• Median
Figure 6.22. Change in Phytane Concentration Through Time for Seal
Beach (Untreated Control) at Snug Harbor (Cobble Surface).
The Number of Samples Showing Concentrations Below
Detection Limit is Shown Above the Sampling Date.
165
-------
10 r
CO
5
i
c
0.1
0.01
0 0
§
0 0
0
8
1 2 3
8
o
o o
0
o
o
i
11 16
06 Jun 89 24 Jun 89 10 Jul 89 26 Jul 89 11 Aug 89 27 Aug 89 12 Sep 89
Sampling Date
° Anolysis Volue
* Medion
Figure 6.23. Change in nC18 Concentration Through Time for Seal Beach
(WOODACE Briquettes) at Snug Harbor (Cobble Surface).
The Number of Samples Showing Concentrations Below
Detection Limit is Shown Above the Sampling Date.
10
5
•5
0.1
i
c
0.01
o
8
o
o
1 1
13
08 Jun 89 24 Jun 89 10 Jul 89 26 Jul 89 11 Aug 89 27 Aug 89 12 Sep 89
Sampling Date
° Anolysis Volue
* Medion
Figure 6.24. Change in nC22 Concentration Through Time for Seal Beach
(WOODACE Briquettes) at Snug Harbor (Cobble Surface).
The Number of Samples Showing Concentrations Below
Detection Limit Is Shown Above the Sampling Date.
166
-------
10
5
•8
0.1
i
c
0.01
8
o
o
o o
10
17
06 Jur 89 24Jun8B 10JU88 26JU89 11Aug89 27Aug88 12 Sep
Sampling Dtto
0 Anolviii Volm
• M«diqn
Rgure 6.25. Change In nC27 Concentration Through Time for Seal Beach
(WOOOACE Briquettes) at Snug Harbor (Cobble Surface).
The Number of Sample* Showing Concentrations Below
Detection Limit Is Shown Above the Sampling Date.
100
10
8
I
c
0.1
0.01
1 1
OBJunae 24 Juntt 10Jul89 26Jul89 11 Aug 89 27Aug89 12 Sep 89
Sampling Date
I ° Anolysis Volue
Median
Figure 6.26. Change In Sum of Alkane Concentration nC18 to nC27
Through Time for Seal Beach (WOODACE Briquettes) at
Snug Harbor (Cobble Surface). The Number of Samples
Showing Concentrations Below Detection Limit Is Shown
Above the Sampling Date.
167
-------
S
•8
0.1-
0.01
o
o
06Jun89 24Jun89 10JU89 26JJ89 11Aug89 27 Aug 89 12 Sep 89
Sampling Date
I ° Anolvsis Value
• Median
Figure 6.27. Change in Pristane Concentration Through Time for Seal
Beach (WOODACE Briquettes) at Snug Harbor (Cobble
Surface). The Number of Samples Showing Concentrations
Below Detection Limit is Shown Above the Sampling Date.
10
5
0.1
0.01
o
o
o
o
08Jun89 24Jun8Q 10JJ89 26Jul89 11Aug89 27Aug89 12 Sep 89
SampInQ Date
° Anolvsis Value
Median
Figure 6.28. Change in Phytane Concentration Through Time for Seal
Beach (WOODACE Briquettes) at Snug Harbor (Cobble
Surface). The Number of Samples Showing Concentrations
Below Detection Limit is Shown Above the Sampling Date.
168
-------
10
o
8
o
o
5
•5
00
5
I
c
o
o
0
o
0.1
o
o
o
o
0.01
08Jun89 24Jun89 10Jul89 26Jul89 11 Aug 89 27 Aug 89 12 Sep 89
Sampling Date
° Analysis Volue
Mcdipn
J
Figure 6.29. Change In nC18 Concentration Through Time for Seal Beach
(INIPOL) at Snug Harbor (Cobble Surface). The Number of
Samples Showing Concentrations Below Detection Limit Is
Shown Above the Sampling Date.
10
S
•6
8
0.1
0.01
I o
o
0
11
OBJun89 24Jun89 10JJ89 26Jut8Q tl Aug 89 27 Aug 89 12Sep89
Sampling Date
° Anolvsis Volut
Madian
Rgure 6.30. Change In nC22 Concentration Through Time for Seal Beach
(INIPOL) at Snug Harbor (Cobble Surface). The Number of
Samples Showing Concentrations Below Detection Limit Is
Shown Above the Sampling Date.
169-
-------
10
1
s
•5
0.1
i
c
0.01
o
8
o
o
10
08Jun89 24 Jun89 10JU89 26JJ89 11 Aug 89 27Aug89 12 Sep 89
SampRng Date
0 Anolysis Value
• Median
Figure 6.31. Change in nC27 Concentration Through Time for Seal Beach
(IN1POL) at Snug Harbor (Cobble Surface). The Number of
Samples Showing Concentrations Below Detection Limit is
Shown Above the Sampling Date.
100
10
= 1
oo
5
0.1
0.01
o
o
8
08 Jun89
1 —i — 1 1 1 r
24Jun89 10JJ89 26Jul89 tl Aug 89 27Aug89 12 Sep 89
Sampling Data
I ° Anolvsis Volue
Medion
Figure 6.32. Change in Sum of Alkane Concentration nC18 to nC27
Through Time for Seal Beach (INIPOL) at Snug Harbor
(Cobble Surface). The Number of Samples Showing
Concentrations Below Detection Limit is Shown Above the
Sampling Date.
170
-------
10
1
5
•5
0.1
0.01
8
6
12
08Jun89 24Jun89 10Jul89 26 Jut 89 11 Aug 89 27Aug89 12 Sep 89
Sampling Date
0 Anolvsis Value
Median
Figure 6.33. Change in Pristane Concentration Through Time for Seal
Beach (INIPOL) at Snug Harbor (Cobble Surface). The
Number of Samples Showing Concentrations Below
Detection Limit is Shown Above the Sampling Date.
10-f
5
•5
0.1
0.01
OSJunSO 24 Jun89 10JJ89 26JuJ89 tl Aug 89 27 Aug 89 12 Sep 89
Sampling Date
0 Anolvsis Volue
Median
Rgure 6.34. Change in Phytane Concentration Through Time for Seal
Beach (INIPOL) at Snug Harbor (Cobble Surface). The
Number of Samples Showing Concentrations Below
Detection Limit is Shown Above the Sampling Date.
171
-------
00
5
18 18
06 Jun 89 24 Jun 89 10 Jul 89 26 Jul 89 11 Aug 89 27 Aug 89 12 Sep 89
Sampling Date
° Anolysis Volue
Medion
Figure 6.35. Change In nC18/phytane Ratio Through Time for Seal Beach
(Untreated Control) at Snug Harbor (Cobble Surface). The
Number of Samples Showing Concentrations Below
Detection Limit is Shown Above the Sampling Date.
12 16
08 Jun 89 24 Jun 89 10 Jul 89 26 Jul 89 11 Aug 89 27 Aug 89 12 Sep 89
Sampling Date
° Anolysis Volue
Medion
Figure 6.36. Change in nC18/phytane Ratio Through Time for Seal Beach
(WOODACE Briquettes) at Snug Harbor (Cobble Surface).
The Number of Samples Showing Concentrations Below
Detection Limit is Shown Above the Sampling Date.
172
-------
ib 11
OBJunae 24Jun89 10JUI89 26Jul89 11 Aug 89 27 Aug 89 12 Sep 89
Sampling Date
° Anolvin Vqlue
Medion
Figure 6.37. Change In nC18/phytane Ratio Through Time for Seal Beach
(INIPOL) at Snug Harbor (Cobble Surface). The Number of
Samples Showing Concentrations Below Detection Limit Is
Shown Above the Sampling Date.
173
-------
nC18
14 29 50 79 91
Day
c
CO
'•&
0)
o>
00
(£>
H-
o
+•*
I
o
Q.
nd8tonC27
40
20
0
14 29 50 79 91
Day
Phytane
14 29 50 79 91
Day
c
2
'•5
o
§
nC18/Phytane Ratio
7 14 29 50 79 91
Day
Briquettes
INIPOL
Untreated Control
Figure 6.38. Change In the Median Residue Weight for Several
Hydrocarbons Expressed as Percent of the 6/9/89 Median
Over Time for the Briquette, INIPOL, and Untreated Control
Beaches at Snug Harbor (Cobble Surface). All Variability Is
not Shown Because the Actual Data Points are not
Presented.
174
-------
TABLE 6.7. CHANGE IN HYDROCARBON COMPOSITION THROUGH TIME
AT SNUG HARBOR, EXPRESSED IN PERCENT OF THE MEDIAN CONCENTRATION
OF INDIVIDUAL HYDROCARBONS ON THE 6/9 SAMPLING* (COBBLE SURFACE)
Alkane Beach Code
nC18
nC19
nC20
nC21
nC22
nC23
nC24
nC25
nC26
nC27
nCI8 tonC27
Phytane
B
I
C
B
I
C
B
I
C
B
I
C
B
I
C
B
I
C
B
I
C
B
I
C
B
I
C
B
I
C
B
I
C
B
I
C
6/17
45
46
108
40
50
92
32
35
92
37
49
83
42
46
75
46
42
90
43
35
80
40
30
74
43
31
68
42
28
72
37
42
81
70
50
120
6/25
66
90
76
65
111
72
63
99
65
64
121
76
61
117
73
56
143
89
58
134
90
57
153
84
55
169
74
55
173
102
61
133
76
69
79
102
7/8
28
61
39
36
105
53
42
136
64
40
141
71
42
166
69
38
169
79
40
166
72
41
207
70
39
226
67
43
240
98
38
160
62
61
96
69
7/29
21
28
16
31
55
25
40
55
33
43
83
36
46
111
40
39
103
42
47
102
47
54
116
50
49
135
44
49
150
61
42
87
38
40
52
34
8/26
0
0
0
4
0
0
2
0
0
1
0
0
5
0
0
4
0
7
0
0
0
0
0
0
0
0
2
2
0
4
4
1
4
19
19
6
9/9
0
0
0
4
0
0
0
10
0
0
0
0
0
0
0
0
13
0
0
0
0
0
0
0
0
0
0
0
0
0
0
12
0
30
0
0
* B - Briquette fertilizer-treated plot; I - INIPOL fertilizer-treated plot; C - Untreated Control
175
-------
TABLE 6.8. NUMBER OF SAMPLES, OUT OF APPROXIMATELY 21 SAMPLES,
TAKEN AT EACH SAMPLING TIME AT SNUG HARBOR, WITH ALKANE
CONCENTRATION BELOW DETECTION LIMIT' (COBBLE SURFACE)
Alkanc Beach Code
nC18
nCI9
nC20
nC21
nC22
nC23
nC24
nC25
nC26
nC27
nC18 to nC27
Phytnne
B
I
C
B
I
C
B
I
C
B
I
C
B
I
C
B
I
C
B
I
C
B
I
C
B
I
C
B
1
C
B
I
C
B
1
C
6/9
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
6/17
1
0
0
1
0
0
1
0
0
1
0
0
1
0
0
1
0
0
1
0
1
1
2
3
1
0
0
2
0
0
6/25
2
0
1
1
0
0
1
0
0
1
0
0
1
0
0
1
0
0
1
0
0
1
0
0
1
0
0
2
0
0
1
0
0
1
0
0
7/8
3
0
0
0
0
0
3
0
0
1
0
0
1
0
0
2
0
0
3
0
0
2
0
0
2
0
0
3
0
0
0
0
0
0
0
1
7/29
0
2
3
0
1
1
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
2
2
8/26
11
15
18
5
11
12
8
13
15
10
14
13
4
11
12
7
10
8
12
13
12
12
12
11
13
10
10
10
13
10
1
8
6
2
5
9
9/9
16
11
18
8
10
16
15
5
18
13
11
18
13
8
15
10
5
16
16
11
15
16
9
16
17
9
15
17
10
14
8
4
13
3
8
11
• B - Briquette fertilizer-treated plot; 1 - INIPOL fertilizer-treated plot; C - Untreated Control
176
-------
SNUG HARBOR
TABLE 6.9. MEDIAN VALUES AND STATISTICAL COMPARISONS OF
OIL RESIDUE WEIGHTS FOR SUMMED ALK.ANES IN COBBLE SURFACE SAMPLES
FROM THE DIFFERENT BEACH TREATMENTS AT SNUG HARBOR
Median Values (% of 6/9/89 Median)
Sampling
Date
June 9
June 17
June 25
JulyS
July 29
August 26
September 9
Day
0
8
16
29
50
78
92
Untreated
Control
17.0
13.7(81)
12.9 (76)
10.6 (62)
6.4 (38)
0.71 (4)
0(0)
Briquettes
25.1
9.4 (37)
15.3(61)
9.6 (38)
10.6 (42)
0.95 (4)
0.18(0)
IN1POL
9.9
4.2 (42)
13.2(133)
15.9(160)
8.6 (87)
0.05(1)
1.2(12)
Mann-Whitney Test Results*
Sampling
Date
June 9
June 17
June 25
JulyS
July 29
August 26
September 9
Briquettes vs.
INIPOL
B>I
B>I
B>I
Same
Same
B>I
B<1
Briquettes vs.
Untreated Control
B>C
Same
Same
Same
B>C
B>C
Same
INIPOL vs.
Untreated Control
Same
Same
Same
Same
I>C
Same
I>C
* 95 Percent Confidence Level
177
-------
SNUG HARBOR
TABLE 6.10. RATE ANALYSIS OF NATURAL LOG-TRANSFORMED OIL RESIDUE
WEIGHTS FOR SUMMED ALKANES IN COBBLE SURFACE SAMPLES VERSUS TIME
FOR TEST BEACHES AT SNUG HARBOR
Beach
Briquettes
INIPOL
Untreated
Control
Slope
(Std. Dev.)
-0.017
(0.006)
-0.001
(0.005)
-0.011
(0.005)
Significance of Slope
Greater than Zero
N T- value p
89 -2.9 0.005
100 0.26 0.80
84 -2.4 0.02
It is difficult to explain the results for the plots treated with INIPOL (Figures 6.29 to 6.34 and
6.38). Following a rather dramatic decrease in alkane concentration (especially for the higher
molecular weight alkanes) during the 8 days after fertilizer application, it appears that the
concentrations actually increased significantly. Although it is possible that biodegradation of the
fertilizer components may have produced intermediates that confounded the GC analysis of the
hydrocarbons, it seems very unlikely that these products would be extracted with the methylene
chloride procedure used in the analyses. If the presence of INIPOL caused other less degraded oil to
be mixed in with the original oil, similar increases in residue weight would be expected. This was
not the case. Finally, it is possible that chemical analysts of the samples taken on June 25 was done
improperly. Again, there was nothing to suggest that this occurred, and samples on subsequent dates
also showed elevated concentrations of the normal alkanes. Thus, these results can not be explained
and may have been obscuring any changes that corresponded with the decreases in oil residue weight.
The dramatic increases in hydrocarbon decay rates following the July 29 sampling were also seen on
the INIPOL-treated plot. The extent of alkane removal appeared to be greatest on the INIPOL -
treated plot.
A further comparison of the decay rates between the normal alkanes and the branched alkanes
(pristane and phytane) for all plots (Figures 6.21, 6.22, 6.27, 6.28, 6.33, 6.34 and 6.38) reveals similar
178
-------
SNUG HARBOR
trends, but with several notable exceptions. First, significant decreases in the branched alkane
concentrations occurred in all cobble surface samples. If these decreases were due to biodegradation,
it suggests that the branched alkanes were not as conserved as originally expected, and any ratioing
of the normal alkanes to the branched alkanes will give conservative estimates of biodegradation (see
below).
Second, on the untreated control plot decreases in pristane and phytane concentrations (Figures
6.21 and 6.22) following the initial lag were surprisingly rapid. In fact, the decreases equaled those
for the nC18 alkane and surpassed the decreases observed for the higher molecular weight alkanes.
This represents possibly a unique biodegradation capability on the untreated control plot.
Finally, pristane and phytane concentrations in samples taken from the briquette fertilizer-
treated plot (Figures 6.27 and 6.28) did not show the dramatic change in decay rate following the July
29 sampling observed with many of the other alkanes in the same samples. This implies that the
events responsible for this accelerated decay may have been due to biodegradation.
Decreases in the nC18/phytane ratios for the two treated plots and the untreated control (Figures
6.35-6.37 and 6.38) strongly suggest that biodegradation was affecting the changes in hydrocarbon
composition in all cases. No other physical or chemical process will differentially effect changes in
the concentration of two alkanes that chemically behave very similarly (their boiling points are very
close, causing them to chromatograph very close to one another in a GC column). Statistical analysis
of the decay rates for the ratios is shown in Tables 6.11 and 6.12. All rates were significantly
different from zero at the 95% confidence interval, and the decay rate of the ratios was slightly faster
on the two fertilizer-treated beaches. Thus, changes in alkane composition suggest that the addition
of the fertilizers caused a small but significant enhancement of biodegradation, and this enhanced
biodegradation activity was probably largely responsible for the observed fertilizer-induced changes
in the oil residue weights noted in the previous section.
b) Mixed Sand and Gravel Samples Under Cobble
Changes through time in the concentration of selected normal alkanes (nC18, nC22, and nC27),
the sum of normal alkanes (nC18 to nC27), the branched alkanes pristane and phytane, and the
nC18/phytane ratios in mixed sand and gravel samples under cobble from treated and untreated
control plots, are shown in Figures 6.39 to 6.59. For comparative purposes, Tables 6.13 and 6.14
179
-------
SNUG HARBOR
TABLE 6.11. MEDIAN VALUES AND STATISTICAL COMPARISONS OF THE
nCI8/PHYTANE RATIO IN COBBLE SURFACE SAMPLES FROM DIFFERENT
BEACH TREATMENTS AT SNUG HARBOR
Median Values (Number of above detection limit values for nC18 and phytane)
Sampling
Date
June 9
June 17
June 25
JulyS
July 29
August 26
September 9
Day
0
8
16
29
50
78
92
Untreated
Control (n)
1.06(19)
1.05(19)
0.86(16)
0.68 (9)
0.52(15)
0.32 (3)
0.27(1)
Briquettes (n)
1.46(18)
1.27(17)
1.38(18)
0.92(11)
0.77 (19)
0.11(9)
0.73 (2)
INIPOL (n)
1.24 (20)
1.30(19)
1.40(18)
0.93(19)
0.62(16)
0.18(1)
0.48 (4)
Mann-Whitney Test Results*
Sampling
Date
June 9
June 17
June 25
July8
July 29
August 26
September 9
Briquettes vs.
INIPOL
B> I
Same
Same
Same
B> I
NA
NA
Briquettes vs.
Untreated Control
B>C
Same
B>C
Same
B>C
BC
Same
I > C
Same
I > C
NA
NA
* 95 Percent Confidence Level
NA - Insufficient data to calculate a statistic valid at the 95% confidence level
180
-------
SNUG HARBOR
TABLE 6.12. RATE ANALYSIS OF nC18/PHYTANE RATIOS IN COBBLE SURFACES
VERSUS TIME FOR TEST BEACHES AT SNUG HARBOR
Beach
Briquettes
INIPOL
Untreated
Control
Slope
(Std. Dev.)
-0.012
(0.0015)
-0.011
(0.0016)
-0.009
(0.0013)
Significance of Slope
Greater than Zero
N T- value p*
94 -7.9 0.0001
101 -7.2 0.0001
82 -6.9 0.0001
* All slopes are statistically significant at the 95 percent confidence level
181
-------
TABLE 6.13. CHANGE IN HYDROCARBON COMPOSITION THROUGH TIME
AT SNUG HARBOR, EXPRESSED IN PERCENT OF THE MEDIAN CONCENTRATION
OF INDIVIDUAL HYDROCARBONS ON THE 6/9 SAMPLING"
(MIXED SAND AND GRAVEL UNDER COBBLE)
Alkane Beach Code
nCI8
nC!9
nC20
nC21
nC22
nC23
nC24
nC25
nC26
nC27
nC18 tonC27
Phytane
B
I
C
B
I
C
B
I
C
B
I
C
B
I
C
B
1
C
B
I
C
B
I
C
B
I
C
B
I
C
B
I
C
B
I
C
6/17
58
97
106
54
80
107
50
74
152
56
88
123
62
89
171
68
76
92
69
100
79
87
95
90
80
112
125
89
172
181
67
97
112
86
81
93
6/25
31
56
0
31
48
0
46
40
0
36
57
0
39
57
0
48
46
0
40
55
0
50
61
0
57
66
212
50
67
0
39
55
59
58
60
56
7/8
13
45
44
14
37
43
13
42
72
18
35
58
19
41
111
19
42
64
18
41
65
20
38
45
21
23
60
24
40
71
16
38
65
26
49
55
7/29
14
31
38
15
28
40
19
30
71
15
33
52
25
42
85
22
35
56
27
43
52
29
38
47
26
40
63
31
37
68
23
39
61
27
31
34
8/26
5
20
24
2
17
30
0
11
30
5
17
39
8
17
59
15
26
46
0
17
0
13
13
25
0
17
0
0
16
35
7
18
35
12
21
19
9/9
14
39
55
13
30
54
0
22
44
12
39
56
13
29
62
19
19
66
17
1
69
25
0
74
10
0
0
2
0
36
13
23
59
21
22
42
' B - Briquette fertilizer-treated plot; I - INIPOL fertilizer-treated plot; C - Untreated Control
182
-------
TABLE 6.14. NUMBER OF SAMPLES. OUT OF APPROXIMATELY 21 SAMPLES,
TAKEN AT EACH SAMPLING TIME AT SNUG HARBOR, WITH ALK.ANE
CONCENTRATION BELOW DETECTION LIMIT"
(MIXED SAND AND GRAVEL UNDER COBBLE)
Alkane Beach Code 6/9
nC18
nC19
nC20
nC21
nC22
nC23
nC24
nC25
nC26
nC27
nC18 to nC27
Phytane
B
1
C
B
I
C
B
1
C
B
1
C
B
I
C
B
I
C
B
I
C
B
I
C
B
I
C
B
I
C
B
I
C
B
I
C
0
2
1
0
2
1
0
2
1
0
2
1
0
2
1
0
2
1
0
2
1
0
2
1
0
2
1
0
3
2
0
1
1
0
1
0
6/17
2
2
1
2
2
1
2
2
2
2
3
3
2
3
2
2
3
4
2
3
4
1
3
4
2
4
4
2
3
4
1
2
0
1
2
1
6/25
3
4
12
3
2
12
4
3
15
3
2
15
3
2
15
2
2
14
4
3
17
4
3
15
4
5
6
4
6
16
2
1
2
3
1
6
7/8
0
0
4
0
1
3
0
0
3
0
0
1
0
0
0
0
1
1
0
3
2
0
0
1
2
3
3
1
3
3
0
0
0
0
0
1
7/29
1
0
4
1
0
3
1
0
3
1
0
3
0
0
2
0
0
1
0
0
3
0
0
2
2
0
4
1
0
3
0
0
0
0
0
3
8/26
9
0
8
10
0
8
12
4
8
8
1
8
8
1
8
5
0
5
11
2
13
8
4
9
12
6
13
13
6
8
5
0
3
8
0
9
9/9
5
5
6
5
5
5
8
6
6
5
5
6
6
5
8
3
6
4
6
9
7
4
10
5
6
10
9
7
10
8
1
5
2
2
5
5
' B - Briquette fertilizer-treated plot; I - INIPOL fertilizer-treated plot; C - Untreated Control
183
-------
10
1
s
oo
5
i
c
0.1
0.01
13
08Jun89 24Jun89 10JuiaO 26 Jul 89 11Aug89 27 Aug 89 12 Sep
Sampling Date
0 Anolysis Volue
Medion
Figure 6.39. Change in nC18 Concentration Through Time for Seal Beach
(Untreated Control) at Snug Harbor (Mixed Sand and Gravel
Under Cobble). The Number of Samples Showing
Concentrations Below Detection Limit is Shown Above the
Sampling Date.
10
5
•5
0.1
0.01
16
06 Jun 89 24Jun89 10 Jul 89 26 Jul 69 tl Aug 89 27Aug89 12 Sep 89
Sampling Date
0 Anolysis Volue
Medion
Rgure 6.40. Change in nC22 Concentration Through Time for Seal Beach
(Untreated Control) at Snug Harbor (Mixed Sand and Gravel
Under Cobble). The Number of Samples Showing
Concentrations Below Detection Limit is Shown Above the
Sampling Date.
184
-------
10
B
c
0.01
16
OBJuiW 24JunBB 10JulW 26Ju)8B 11Aug89 27Aug89 12Sep89
SvnpMngMB
0 Analysis Volu>
• Medion
Rgure 6.41. Change in nC27 Concentration Through Time for Seal Beach
(Untreated Control) at Snug Harbor (Mixed Sand and Gravel
Under Cobble). The Number of Samples Showing
Concentrations Below Detection Limit is Shown Above the
Sampling Date.
100
10
«= 1
0.1
I
c
0.01
OBJuntt 24Jun8B 10Jul89 2BJul89 tl Aug 89 27Aug88 12 Sop 89
0 Anolvtii Volm
Medion
Figure 6.42. Change In the Sum of Alkane Concentration nCl8 to nC27
Through Time for Seal Beach (Untreated Control) at Snug
Harbor (Mixed Sand and Gravel Under Cobble). The Number
of Samples Showing Concentrations Below Detection Limit
Is Shown Above the Sampling Date.
185
-------
10
S
T5
0.1
0.01
51 2 J
T— 1 —I 1 1 1 T
OBJwSQ 24Jun8Q 10 Ju) 89 26 Jul 86 tlAug89 27Aug89 12 Sep 89
Sampling Date
0 Anolvii> Volue
Figure 6.43. Change In Pristane Concentration Through Time for Seal
Beach (Untreated Control) at Snug Harbor (Mixed Sand and
Gravel Under Cobble). The Number of Samples Showing
Concentrations Below Detection Limit Is Shown Above the
Sampling Date.
10
8
•8
0.1
0.01
10
i 1 1 1 1 1 r
06Jun89 24Jun89 10 Jul W 26 Jul 88 11 Aug 89 27 Aug 89 12Sep88
SvnpllngMs
AnQlyjij Vglm
Figure 6.44. Change In Phytane Concentration Through Time for Seal
Beach (Untreated Control) at Snug Harbor (Mixed Sand and
Gravel Under Cobble). The Number of Samples Showing
Concentrations Below Detection Limit Is Shown Above the
Sampling Date.
186
-------
10
5
•8
0.1
0.01
10
OBJunSB 24Jun89 10JJ89 26 Jui 89 tl Aug 89 27Aug89 12 Sep 80
Sampling
I ° Anolvin Voim
Figure 6.45. Change in nC18 Concentration Through Time for Seal Beach
(WOODACE Briquettes) at Snug Harbor (Mixed Sand and
Gravel Under Cobble). The Number of Samples Showing
Concentrations Below Detection Limit Is Shown Above the
Sampling Date.
10
8
•8
8
c
0.1
0.01
OBJun89 24Jun89 10 Jui 89 26Jul8Q Tl Aug 89 27 Aug 89 12 Sep 89
Sampling Date
I ° Anqlvsit Value •
Hgure 6.46. Change In nC22 Concentration Through Time for Seal Beach
(WOODACE Briquettes) at Snug Harbor (Mixed Sand and
Gravel Under Cobble). The Number of Samples Showing
Concentrations Below Detection Limit Is Shown Above the
Sampling Date.
187
-------
10
r
5
0.1
8
c
0.01
14
06 Jin 89 24Jun89 10Jul89 26Jul89 tlAugSO 27Aug89 12 Sep 89
ii Value
Median
Figure 6.47. Change In nC27 Concentration Through Time for Seal Beach
(WOODACE Briquettes) at Snug Harbor (Mixed Sand and
Gravel Under Cobble). The Number of Samples Showing
Concentrations Below Detection Limit Is Shown Above the
Sampling Date.
100
10
5
= 1
0.1
1
c
0.01
OBJuiBQ 24Jun8Q 10JJ89 26Jul89 11 Aug 89 27Aug89 12 Sap
Sampling DM
0 AnoiYiii Value
Median
Figure 6.48. Change In the Sum of Alkane Concentration nC18 to nC27
Through Time for Seal Beach (WOODACE Briquettes) at
Snug Harbor (Mixed Sand and Gravel Under Cobble). The
Number of Samples Showing Concentrations Below
Detection Limit Is Shown Above the Sampling Date.
188
-------
10
?
I
m
I 1
8
T5
?
§ 0.1
1
0.01
o
o
IP 8 g
l^lr--^ 8 °
8 r\ 8 8
^\_ 8 °
••"-^ ._ .. , --A.,
1) ^' '" * ^^"--..*.
• ^**^_
o i g ^- • ,
o o §
8
o
_ — . .- — 1 ~ — 1 1 ' ^~™
0
o
0
d
O '
0 /
o
o
»
1
fl
8
o
1
OBJunSe 24Jun8Q 10Jul89 26 Jul 89 Tl Aug 89 27 Aug 89 12 Sep 89
Sampling Date
Analysis Volue
Figure 6.49. Change In Prlstane Concentration Through Time for Seal
Beach (WOODACE Briquettes) at Snug Harbor (Mixed Sand
and Gravel Under Cobble). The Number of Samples
Showing Concentrations Below Detection Limit Is Shown
Above the Sampling Date.
10
5
•8
0.1
0.01
06Jun89 24Jun89 10 Jul 88 26 Jul 89 tl Aug 89 27 Aug 89 12 Sep 89
S*T^«ngD«e
I " Anolytis Vglg« _". •
Figure 6.50. Change In Phytane Concentration Through Time for Seal
Beach (WOODACE Briquettes) at Snug Harbor (Mixed Sand
and Gravel Under Cobble). The Number of Samples
Showing Concentrations Below Detection Limit Is Shown
Above the Sampling Date.
189
-------
10
S
15
§
c
0.1
0.01
J
H
o
T 1 1 1 1 1 r
06 Jin 89 24Jun8Q 10Ju)89 26Jul8Q 11 Aug 89 27Aug89 12 Sap 89
Sampling Date
Msdign
D
Figure 6.51. Change In nC18 Concentration Through Time for Seal Beach
(INIPOL) at Snug Harbor (Mixed Sand and Gravel Under
Cobble). The Number of Samples Showing Concentrations
Below Detection Limit Is Shown Above the Sampling Date.
10
5
TJ
0.1
0.01
o
o
OBJun89 24Jun8Q 10JU89 26 Jul 89 11 Aug 88 27Aug89 12 Sep 88
Figure 6.52. Change In nC22 Concentration Through Time for Seal Beach
(INIPOL) at Snug Harbor (Mixed Sand and Gravel Under
Cobble). The Number of Samples Showing Concentrations
Below Detection Limit Is Shown Above the Sampling Date.
190
-------
10
0.1
8
c
0.01
10
24Jun80 lOJulSQ 26Ju)8Q HAugtt 27Aug88 12Sep89
Sampling Date
0 Anoivtii Volu«
Mtdion
Figure 6.53. Change In nC27 Concentration Through Time for Seal Beach
(INIPOL) at Snug Harbor (Mixed Sand and Gravel Under
Cobble). The Number of Samples Showing Concentrations
Below Detection Limit Is Shown Above the Sampling Date.
100
10
8
= 0.1
0.01
o
0
8 §
06 Jin 89 24Jun8Q 10JU89 26Jul8B 11 Aug 89 27Aug89 12 Sap 89
Sampling Ma
Median
Rgure 6.54. Change in the Sum of Alkane Concentration nC18 to nC27
Through Time for Seal Beach (INIPOL) at Snug Harbor
(Mixed Sand and Gravel Under Cobble). The Number of
Samples Showing Concentrations Below Detection Limit Is
Shown Above the Sampling Date.
191
-------
10
0.1
0.01
08Jui89 24Jun89 10Jd89 26Jul8Q 11 Aug 89 27 Aug 89 12 Sep 89
Sampling Date
0 Angivin Value
Median
Figure 6.55. Change In Prlstane Concentration Through Time for Seal
Beach (INIPOL) at Snug Harbor (Mixed Sand and Gravel
Under Cobble). The Number of Samples Showing
Concentrations Below Detection Limit Is Shown Above the
Sampling Date.
10
5
•5
0.1
0.01
08Jun8B 24Jun89 10Jui89 26Jul89 11 Aug 89 27Aug89 12 Sap 89
Sampling
° Anolviii VQlge
Mtdion
Figure 6.56. Change In Phytane Concentration Through Time for Seal
Beach (INIPOL) at Snug Harbor (Mixed Sand and Gravel
Under Cobble). The Number of Samples Showing
Concentrations Below Detection Limit Is Shown Above the
Sampling Date.
192
-------
I
c
2 13
10
OBJunBQ 24Jun8B 10Jul89 26 Jut 89 11 Aug 89 27 Aug 89 12 Sap 69
Date
0 Anoivinvolue
• MeQipn
Figure 6.57. Change In nC18/Phytane Ratio Through Time for Seal Beach
(Untreated Control) at Snug Harbor (Mixed Sand and Gravel
Under Cobble). The Number of Samples Showing
Concentratlone Below Detection Limit Is Shown Above the
Sampling Date.
I
10
oejunae 24Jun» lOJuiae ajuiae nAugw 2? Aug a
Sampling Data
° Anolyiij Value
Rgure 6.58. Change In nC18/Phytane Ratio Through Time for Seal Beach
(WOODACE Briquettes) at Snug Harbor (Mixed Sand and
Gravel Under Cobble). The Number of Samples Showing
Concentrations Below Detection Limit Is Shown Above the
Sampling Date.
193
-------
I
c
OejunBO 24 Jin 89 10Jul8Q 26Jul89 11 Aug 89 27Aug80 12Sep89
Sampling Data
Mediqn
Figure 6.59. Change In nC18/Phytane Ratio Through Time for Seal Beach
(INIPOL) at Snug Harbor (Mixed Sand and Gravel Under
Cobble). The Number of Samples Showing Concentrations
Below Detection Limit Is Shown Above the Sampling Date.
194
-------
SNUG HARBOR
summarize the percent change in the medians of individual hydrocarbons and the number or samples
with a hydrocarbon concentration of zero (below detection limits) with each sampling period,
respectively. Figure 6.60 provides a graphical representation of the percent change in the mod inns
for several hydrocarbons.
The results for the mixed sand and gravel under cobble represent the second dramatic effect of
the fertilizer application. Changes in concentrations of all alkanes seemed to follow the same general
trend; decay with time was relatively first-order, with the fastest rates seen on the fertilizer
briquette-treated plot and the slowest rates on the untreated control. Rates on the INIPOL fertilizer-
treated plot were generally in-between. The effect of the fertilizers can be shown statistically if the
summed alkanes are used as an exemplary indicator of the trends. As shown in Tables 6.15 and 6.16,
all decay rates were significantly different from zero at the 95% confidence level and the INIPOL and
briquette fertilizers enhanced the rates two- and three-fold, respectively. These rates were also
significantly different from each other. Thus, despite high variability in the samples, differences in
decay rates were large enough to see a fertilizer effect.
Overall, decay in alkane concentrations in mixed sand and gravel samples taken from under
cobble was much less complex than that seen with the cobble surface samples. There were neither
large decreases in concentration during the latter part of the sampling period, nor apparent increases
in concentrations observed in samples from the INIPOL fertilizer-treated plots. Decays for most
alkanes commenced immediately after fertilizer application, but in contrast to the cobble surface
samples, the decay continued and did not appear to level off. A much shorter lag in hydrocarbon
concentration change occurred in the mixed sand and gravel samples compared to the untreated
control plot cobble surface samples. The only peculiar result was a fairly consistent increase in alkane
concentrations the last sampling. This may have been related to other clean-up operations occurring
in Snug Harbor at that time.
An examination of the changes in the nC18/phytane ratios (Figures 6.57,6.58, and 6.59) showed
similar trends, with fastest change on the briquette fertilizer-treated plot, slowest change on the
untreated control, and intermediate change on the INIPOL fertilizer-treated plot. However, the
curves are complicated because of a lack of any significant change in the ratios during the first 2 to
4 samplings. This is reflected in the statistical analysis of the decay curves shown in Tables 6.17 and
6.18, and is due to the same disappearance rate for both the normal alkane and the branched alkane,
a process that could possibly be attributed to fate processes other than biodegradation. However,
195
-------
nC18
nCl8tonC27
14 29 50 79 91
Day
14 29 50 79 91
Day
Phytane
nC18/Phytane Ratio
k. \£.V-
.2 -
"S 100-
Q) i \j\j-
5 :
«r» RO
8
S? 60-
(0 bu
H—
O 40
*-•
C
8 20-
k.
flO-
S B0
rs RO-
{£ OU
O .- -
S 90-
W <-\J -
0)
Q. n
[A
^
V
^
V
V
*•-.
k*^
.-^^^x
y
V
X
>
r
A
^_
"*
14 29 50 79 91 7 14 29 50 79 91
Day Day
-+- Briquettes -•- INIPOL ~m- Untreated Control
Figure 6.60. Change In the Median Residue Weight for Several
Hydrocarbons Expressed as Percent of the 6/9/89 Median
Over Time for the Briquette, INIPOL, and Untreated Control
Beaches at Snug Harbor (Mixed Sand and Gravel Under
Cobble). All Variability is not Shown Because the Actual
Data Points are not Presented.
-------
SNUG HARBOR
TABLE 6.15. MEDIAN VALUES AND STATISTICAL COMPARISONS OF OIL RESIDUE
WEIGHTS FOR SUMMED ALKANES IN MIXED SAND AND GRAVEL FROM
DIFFERENT BEACH TREATMENTS AT SNUG HARBOR
Median Values (% of 6/9/89 Median)
Sampling
Date
June 9
June 17
June 25
July 8
July 29
August 26
September 9
Day
0
8
16
29
50
78
92
Untreated
Control
3.5
3.9(112)
2.1 (60)
2.3 (65)
2.1 (60)
1.2(35)
2.1 (60)
Briquettes
15.9
10.7 (67)
6.2 (39)
2.6(16)
3.7 (23)
1.0(7)
2.0(13)
INIPOL
7.5
7.3 (97)
4.1 (55)
2.8 (38)
2.9 (39)
1.4(18)
1.7(23)
Mann-Whitney Test Results*
Sampling
Date
June 9
June 17
June 25
July8
July 29
August 26
September 9
Briquettes vs.
INIPOL
B> I
Same
B>I
Same
Same
Same
Same
Briquettes vs.
Untreated Control
B>C
B>C
B>C
Same
Same
Same
Same
INIPOL vs.
Untreated Control
I>C
I>C
I>C
Same
Same
Same
Same
* 95 Percent Confidence Level
197
-------
SNUG HARBOR
TABLE 6.16. RATE ANALYSIS OF NATURAL LOG-TRANSFORMED OIL RESIDUE
WEIGHTS FOR SUMMED ALKANES IN MIXED SAND AND GRAVEL
UNDER COBBLE VERSUS TIME FOR TEST BEACHES AT SNUG HARBOR
Significance of Slope
Slope Greater than Zero
Beach (Std. Dev.) N T-value p
Briquettes -0.032 99 -6.8 0.0001
(0.005)
INIPOL -0.023 96 -4.3 0.0001
(0.005)
Untreated -0.013 94 -2.2 0.03
Control (0.006)
198
-------
SNUG HARBOR
TABLE 6.17. MEDIAN VALUES AND STATISTICAL COMPARISONS OF
THE nC18/PHYTANE RATIO IN MIXED SAND AND GRAVEL FROM DIFFERENT
BEACH TREATMENTS' AT SNUG HARBOR
Median Values (Number of above detection limit values for nC18 and phytane)
Sampling
Date
June 9
June 17
June 25
Julyg
July 29
August 26
September 9
Day
0
8
16
29
50
78
92
Untreated
Control (n)
0.53(18)
0.47(18)
0.34 (8)
0.49(15)
0.63(15)
0.53(11)
0.58(11)
Briquettes (n)
1.00(19)
0.73 (19)
0.70(17)
0.56(21)
0.48 (20)
0.79(11)
0.52 (10)
INIPOL (n)
0.69(19)
0.76(18)
0.58(14)
0.52 (20)
0.57(21)
0.64 (20)
0.74(13)
Mann-Whitney Test Results'
Sampling
Date
June 9
June 17
June 25
July8
July 29
August 26
September 9
Briquettes vs.
INIPOL
B>I
Same
Same
Same
BI
BC
Same
B>C
Same
Same
Same
Same
INIPOL vs.
Untreated Control
I>C
I>C
I>C
Same
Same
Same
Same
' 95 Percent Confidence Level
199
-------
SNUG HARBOR
TABLE 6.18. RATE ANALYSIS OF THE nC18/PHYTANE RATIO IN MIXED SAND AND
GRAVEL SAMPLES VERSUS TIME FOR TEST BEACHES AT SNUG HARBOR
Beach
Briquettes
INIPOL
Untreated
Control
Slope
(Std. Dev.)
-0.0014
(0.0012)
-0.0003
(0.0009)
-0.005
(0.006)
Significance of Slope
Greater than Zero
N T-value p*
117 -1.23 0.22
125 -0.37 0.71
96 -0.82 0.42
* None of these slopes are significantly different from zero at the 95 percent confidence level
since the application of nutrients from the fertilizer briquettes did not have a chemical or physical
effect on the oil, it is highly likely that all of the observed changes in oil composition were due to
biodegradation. It is interesting that the large changes in composition were not accompanied by
changes in oil residue weight, as described in the last section. It is possible that the degraded oil is
much more easily removed from the cobble surfaces than the mixed sand and gravel underneath.
Mixed Sand and Gravel Plots (Otter and Eagle Beaches)
Changes through time in the concentration of selected normal alkanes (nC18, nC22, and nC27),
the sum of normal alkanes (nC18 to nC27), the branched alkanes pristane and phytane, and the
nC18/phytane ratios in samples from the mixed sand and gravel plots (no cobble on the surface) are
shown in Figures 6.61 to 6.81. As indicated above, all values of hydrocarbon concentration were
normalized to the weight of oil in the extracted sample. In all cases, values below detection limits
were treated as zero.
The effect of fertilizer application was most apparent on the plot treated with the fertilizer
briquettes. Over the first 29 days there was a greater decrease in the individual hydrocarbon
concentrations on the briquette fertilizer-treated plot relative to the untreated control. Using the
200
-------
100 f
10
8 1
•5
5
i
c
0.01-
08Jun89 24Jun89 10 Jul 89 26 Jul 89 11 Aug 89 27 Aug 89 12 Sep 89
Sampling Date
0 Anolysis Volue
Median
Figure 6.61. Change in nC18 Concentration Through Time for Eagle
Beach (Untreated Control) at Snug Harbor (Mixed Sand and
Gravel). The Number of Samples Showing Concentrations
Below Detection Limit is Shown Above the Sampling Date.
100
10
8
•5
0.01
o
o
§
§
8
08Jun89 24 Jin 89 10JJ89 26Jul89 11 Aug 89 27 Aug 89 12 Sep 89
Sampling Date
I ° Analysis Volue
Median
Figure 6.62. Change in nC22 Concentration Through Time for Eagle
Beach (Untreated Control) at Snug Harbor (Mixed Sand and
Gravel). The Number of Samples Showing Concentrations
Below Detection Limit is Shown Above the Sampling Date.
201
-------
100
10
5 1
•5
0.1
I
c
0.01
08Jun89 24Jun89 10 Jul 89 26Jul89 11Aug89 27Aug89 12Sep89
SampHng Data
I ° Analysis Value
Median
Rgure 6.63. Change in nC27 Concentration Through Time for Eagle
Beach (Untreated Control) at Snug Harbor (Mixed Sand and
Gravel). The Number of Samples Showing Concentrations
Below Detection Limit is Shown Above the Sampling Date.
1,000
100
10
00
c
I
c
0.1
06 Jin 89 24Jun80 10 Jul 89 26 Jul 89 11 Aug 89 27Aug80 12Sep8Q
Sampling Date
I ° Anolysis Volue
Medion
Rgure 6.64. Change in Sum of the Alkane Concentration nC18 to nC27
Through Time for Eagle Beach (Untreated Control) at Snug
Harbor (Mixed Sand and Gravel). The Number of Samples
Showing Concentrations Below Detection Limit is Shown
Above the Sampling Date.
202
-------
100
10
3 1
•8
0.1
0.01
1
,—— 1 1 1 1 I—
06Jun89 24Jun88 10Jul89 26JJ89 11 Aug 89 27Aug89 12Sep
Sampling Date
I ° Anolvais Volue
Median
Figure 6.65. Change in Pristane Concentration Through Time for Eagle
Beach (Untreated Control) at Snug Harbor (Mixed Sand and
Gravel). The Number of Samples Showing Concentrations
Below Detection Limit is Shown Above the Sampling Date.
100
101
0.1
0.01
08 Jin 89 24 Jin 89 10Jul89 26 Jut 80 tl Aug 89 27Aug89 12Sep89
SamplngDate
° Anolvais Volue
• Median
Rgure 6.66. Change in Phytane Concentration Through Time for Eagle
Beach (Untreated Control) at Snug Harbor (Mixed Sand and
Gravel). The Number of Samples Showing Concentrations
Below Detection Limit is Shown Above the Sampling Date.
203
-------
100 f
10 J
S 1
•5
5
i
c
o.oi
08Jun89 24Jun89 10Jul89 26Jul89 11 Aug 89 27 Aug 89 12 Sep 89
Sampling Date
0 Anolysis Value
Median
Figure 6.67. Change in nC18 Concentration Through Time for Otter
Beach (WOODACE Briquettes) at Snug Harbor (Mixed Sand
and Gravel). The Number of Samples Showing
Concentrations Below Detection Limit is Shown Above the
Sampling Date.
100
10
8 1
0.1
I
c
0.01
o
o
@
o
OBJun89 24Jun89 10Jul89 26Jul89 11 Aug 89 27 Aug 89 12 Sep 89
Sampling Date
0 Anolyais Volue
* Medion
Rgure 6.68. Change In nC22 Concentration Through Time for Otter
Beach (WOODACE Briquettes) at Snug Harbor (Mixed Sand
and Gravel). The Number of Samples Showing
Concentrations Below Detection Limit is Shown Above the
Sampling Date.
204
-------
100
a 10
i
8 1
TJ
1
•j
•* 0.1
8
i
c
0.01
9
o
B fi Q
M 1 1 A
0 r • -^
8 o — —
o
0
0
2
08Jun8i 24Jun89 10Jut88 2BJU80 tlAug80
SvnpNng Dri»
I ° Analvili Value • Median
o
S
«.-
o
o
A
i
27Aug6B
1
Q
O
0
Q
1
o
4
12S«pf
Rgure 6.69. Change In nC27 Concentration Through Time for Otter
Beach (WOODACE Briquettes) at Snug Harbor (Mixed Sand
and Gravel). The Number of Samples Showing
Concentrations Below Detection Limit Is Shown Above the
Sampling Date.
1,000
t
§ 100
5
§
1
c 10
?
c ^
i
0.1
o '
o
o
Lli 8 | o
V V Q ^^*S|^*^11^ 0 II
0 0 « ^--^| | 2
9 B I ' — -~-^^^ 5
0 0
o 2
o 8
2
1 I 1 p 1 1 t
OBJun89 24 Jin 88 10Jut8B 2CJU88 11Aug80 27Aug89 12Sepf
SvnplngOrit
I ° Anolvtii Value • Median I
Rgure 6.70. Change In Sum of Alkane Concentration nC18 to nC27
Through Time for Otter Beach (WOODACE Briquettes) at
Snug Harbor (Mixed Sand and Gravel). The Number of
Samples Showing Concentrations Below Detection Limit is
Shown Above the Sampling Date.
205
-------
100
10
0.1
0.01
06 Jin 89 24Jun80 10JJ89 26JJ89 HAugSB 27Au(}89 12Sep89
SonplngDrit
0 Anqtvsia Volue
• Medlon
Figure 6.71. Change In Pristane Concentration Through Time for Otter
Beach (WOODACE Briquettes) at Snug Harbor (Mixed Sand
and Gravel). The Number of Samples Showing
Concentrations Below Detection Limit Is Shown Above the
Sampling Date.
100
10
8 1
•8
0.1
0.01
06 Jin 89 24Jun80 10JJ80 2BJul80 fl Aug 89 27Aug81 12S«p89
SamplngDto
Figure 6.72. Change In Phytane Concentration Through Time for Otter
Beach (WOODACE Briquettes) at Snug Harbor (Mixed Sand
and Gravel). The Number of Samples Showing
Concentrations Below Detection Limit Is Shown Above the
Sampling Date.
206
-------
100 r
10
3 i
•8
•* 0.1
I
c
0.01
0
o
08Jun88 24Jun89 10Jul89 26Jul69 11 Aug 88 27 Aug 89 12 Sep 89
Sampling Date
0 Anolvan Volua
Median
Rgure 6.73. Changt In nC18 Concantratlon Through TImt for Otttr
Btach (INIPOL) at Snug Harbor (Mlxtd Sand and Qravtl).
Tha Numbar of Samplaa Showing Concantratlona Balow
Dtttctlon Limit la Shown Abova tht Sampling Data.
100 {
10
8
"S
8
c
0.1
0.01
I
06Jun86 24Jun89
2BJUI80 11*4)69 27Aug8Q 12 Sap 80
SvnptogDto
0 Anolviii Volm
Mtdlon
Rgurt 6.74. Changt In nC22 Conctntrttlon Through TImt for Otttr
Btach (INIPOL) at Snug Harbor (Mlxtd Sand and Qravtl).
Tha Numbar of Samplaa Showing Concantratlona Balow
Dtttctlon Limit It Shown Abovt tht Sampling Data.
207
-------
100 i
10
5
•8
8
c
0.1
0.01
OBJunBQ 24Jun80 10JU80 26JJ6B 11 Aug 89 27Aug89 12Sep89
SvnpingDM
Anolvili Vo)u«
Mtdlon
Figure 6.75. Change In nC27 Concentration Through Time for Otter
Beach (INIPOL) at Snug Harbor (Mixed Sand and Gravel).
The Number of Samplee Showing Concentrations Below
Detection Limit le Shown Above the Sampling Date.
1,000
100
8
c 10
0.1
OBJunaO 24 Jin 80 10Jd8Q 2BJU8Q HAug8D 27 Aug 89 12Sep8B
SimpHnoDtfi
° Analvili Valut
• Median
Figure 6.76. Change In Sum of the Alkane Concentration nC18 to nC27
Concentration Through Time for Otter Beach (INIPOL) at
Snug Harbor (Mixed Sand and Gravel). The Number of
Samplee Showing Concentrations Below Detection Limit Is
Shown Above the Sampling Date.
208
-------
100
10
3
"8
0.1
0.01
OSJuneO 24Jun80
26Jul8Q TIAugW 27Aug89 12Sep80
0 AoolyiliVolui
Mtdlan
Figure 6.77. Change In Prlstane Concentration Through Time for Otter
Beach (INIPOL) at Snug Harbor (Mixed Sand and Gravel).
The Number of Samplee Showing Concentrations Below
Detection Limit la Shown Above the Sampling Date.
100
10
5 1
•8
0.1
0.01
08 Jun 88 24Jun89 10JJ89 26Jul89 HAugSD 27Aug89 12Sap80
Sampling Date
I ° Anolvais Value •
Rgure 6.78. Change In Phytane Concentration Through Time for Otter
Beach (INIPOL) at Snug Harbor (Mixed Sand and Gravel).
The Number of Samples Showing Concentrations Below
Detection Limit Is Shown Above the Sampling Date.
209
-------
CD
5
i
c
2
1
08Jun89 24Jun88 10Jul89 26Jul89 11 Aug 89 27Aug89 12 Sep 89
Sampling Me
0 Anply»n Value
Meflion
Rgure 6.79. Change in nC18/phytane Ratio Through Time for Eagle
Beach (Untreated Control) at Snug Harbor (Mixed Sand and
Gravel). The Number of Samples Showing Concentrations
Below Detection Limit Is Shown Above the Sampling Date.
8
08Jun89 24Jun89 10Jul89 26Jul89 11 Aug 89 27Aug89 12 Sep 89
Sampling Date
I ° *no!Yiii Vglm
Mtdion
Figure 6.80. Change In nC18/Phytane Ratio Through Time for Otter
Beach (WOODACE Briquettes) at Snug Harbor (Mixed Sand
and Gravel). The Number of Samples Showing
Concentrations Below Detection Limit Is Shown Above the
Sampling Date.
210
-------
8
c
OGJunBO 24Jun80 10Jul80 26Jul80 11 Aug 80 27 Aug 80 12 Sop 80
SvnpHngDito
I ° Anolviii Value •
Figure 6.81. Change In nC18/phytane Ratio Through Time for Otter Beach
(INIPOL) at Snug Harbor (Mixed Sand and Gravel). The
Number of Samples Showing Concentrations Below
Detection Limit Is Shown Above the Sampling Date.
211
-------
SNUG HARBOR
Mann-Whitney test, hydrocarbon concentrations in samples from these two beaches were not
significantly different at the 95% confidence level at the t-0 sample (Table 6.19). Figure 6.82 also
provides a graphic representation of the percent change in the medians. However, hydrocarbon
concentrations were significantly different on the July 8 sampling date; the median values on the
treated plot had decreased by approximately 50-60%, but on the untreated control plot they had
decreased only 20-25%. Therefore, the decay rate for the summed alkanes on the briquette fertilizer-
treated plot was significantly different from zero over this initial 29 day period but the decay rate
on the untreated control plot was not.
By the following sampling date (day 4, July 29), the untreated control appeared to have "caught
up" with the briquette fertilizer-treated plot to some extent; i.e., no differences between the
concentrations of the alkanes. This may have been due to a more accelerated decay on the untreated
control and an apparent slower decay rate on the treated plot. However, over the entire duration of
the study, it is clear that final concentrations of individual hydrocarbon concentrations on the
briquette fertilizer-treated plots were considerably lower than on the untreated control. Decay rates
overall differed by a factor of two for the summed alkanes, suggesting a significant long-term effect
of this fertilizer application.
These changes in hydrocarbon composition can also be attributed to biodegradation since there
was a significant decay in the nC18/phytane ratios on both plots. Differences between the plots were
not significant, but this may be the result of a slightly greater decay rate for phytane on the briquette
fertilizer-treated plot. Again, the ratio method may give a conservative indication of biodegradation
if phytane is being degraded.
Finally, it would appear that the changes in oil composition were not directly coupled to losses
in oil residue weight. For example, on the briquette fertilizer-treated plot, initial decreases in
individual hydrocarbon concentration (Figures 6.67 through 6.72) were not accompanied by decreases
in oil residue weight (Figure 6.15). Just the opposite was true on the untreated control plot.
However, the greater overall decrease in individual hydrocarbon concentrations observed on the
briquette fertilizer-treated plot (Figure 6.69) may have been related to the large decrease in oil residue
weight following the July 29 sampling (Figure 6.14); in other words, biodegradation may have
proceeded to a point where it changed the physical consistency of the oil, causing the degraded oil
residues to be more easily removed from the beach matrix. Correspondingly, changes in oil
composition on the untreated control plot may not have been great enough to elicit changes in oil
residue weight.
212
-------
nC18
nC18tonC27
c
(0
'•5
Q)
o>
GO
200
C
Q)
O
0)
Q.
14 29 50 79 91
Day
14 29 50 79 91
Day
250
Phytane
Briquettes
INIPOL
Untreated Control
91
Figure 6.82. Change In the Median Residue Weight for Several
Hydrocarbons Expressed as Percent of the 6/9/89 Median
Over Time for the Briquette, INIPOL, and Untreated Control
Beaches at Snug Harbor (Mixed Sand and Gravel Only). All
Variability Is not Shown Because the Actual Data Points are
not Presented.
213
-------
SNUG HARBOR
The situation on the INIPOL fertilizer-treated plot is more complicated to interpret. Again,
initial increases in hydrocarbon concentrations were seen relative to the t-0 medians (Table 6.19).
Possible explanations for this response were given above in the discussion of the cobble surface
samples (page 161). Despite these unexplained increases in individual hydrocarbon concentrations,
there was some decrease in the nC18/phytane ratio (Figure 6.81), suggesting that extensive
biodegradation was occurring since it is unlikely that some nonbiological process would affect the
nC)8 and phytane hydrocarbons differentially. Rather dramatic changes occurred in hydrocarbon
concentration following the July 8 (day 29) sampling date, allowing the INIPOL fertilizer-treated plot
to, in essence, "catch up" with the other plots. However, these changes apparently were not large
enough to affect any concomitant decrease in oil residue weights on the INIPOL fertilizer-treated plot
(Tables 6.3 and 6.4).
In summary, for all types of beach material samples (cobble surface, mixed sand and gravel
under cobble, and mixed sand and gravel only), it would appear that fertilizer application did enhance
oil biodegradation. This was most obvious in the initial 29 days of the test on the plots treated with
the fertilizer briquettes. Changes in oil composition, including the nC18/phytane ratios, were most
extensive on these plots. However, this was not generally accompanied by significant changes in oil
residue weight, and thus we would argue that changes in oil composition may not have been sufficient
to cause large changes in oil residue for the fertilizer briquettes.
Significant changes did occur, however, in oil residue in the cobble surface samples with the
INIPOL fertilizer application. Assuming the changes were due to biodegradation, by extrapolation
one would expect large changes in oil composition. This was not the case; concentrations of
individual hydrocarbons appeared to actually increase. It is possible that the INIPOL chemically
caused a loss of oil residues, but laboratory experiments suggest this was not the case. In addition,
no oil residues were detected in mussels suspended in cages just offshore the treatment area; with the
amount of oil released it was expected that some should have bioconcentrated in the mussel tissue.
It is more likely that components in the INIPOL interfered with the gas chromatographic analysis of
the oil, possibly masking any changes in oil composition. If this was the case, it is also possible that
the residual INIPOL materials contributed to the weight of the oil residues, again giving a
conservative estimate of oil biodegradation.
214
-------
TABLE 6.19. CHANGE IN HYDROCARBON COMPOSITION THROUGH TIME AT SNUG
HARBOR, EXPRESSED IN PERCENT OF THE MEDIAN CONCENTRATION OF
INDIVIDUAL HYDROCARBONS ON THE 6/9 SAMPLING'
Alkane Beach Code
nC18
nC19
nC20
nC21
nC22
nC23
nC24
nC25
nC26
nC27
nC18 tonC27
Phytane
B
I
C
B
I
C
B
I
C
B
I
C
B
I
C
B
I
C
B
I
C
B
I
C
B
I
C
B
I
C
B
I
C
B
I
C
6/17
87
73
108
95
85
113
93
89
94
96
89
104
98
93
115
91
97
116
100
110
113
99
114
116
100
133
115
101
117
113
98
94
107
112
148
123
6/25
77
100
91
80
112
95
79
123
99
86
128
100
89
143
109
82
149
113
98
143
110
94
153
115
89
162
110
92
176
119
93
139
106
102
155
124
7/8
37
135
64
42
145
75
47
182
81
45
163
78
46
187
90
43
166
85
46
174
79
44
172
87
41
157
83
40
177
89
43
166
79
79
236
124
7/29
34
30
31
36
33
34
33
38
34
37
40
34
40
39
41
38
49
42
45
44
41
44
39
45
45
47
46
45
55
49
40
42
38
64
62
79
8/26
15
31
16
14
31
21
19
31
19
15
35
22
18
38
21
17
46
27
22
44
26
20
42
36
18
41
31
17
27
25
17
38
24
51
47
58
9/9
16
35
32
20
39
35
22
41
30
21
40
32
22
48
36
23
40
41
21
49
40
24
39
38
25
40
40
18
13
29
21
41
36
44
55
51
' B - Briquette fertilizer-treated plot; I - INIPOL fertilizer-treated plot; C - Untreated Control
215
-------
SNUG HARBOR
Degradation Extent/Oil Residue Weight Relationships
During beach sampling it was obvious that globs of viscous, sticky oil were present in some
areas. Where these globs were encountered, there was concern that spike concentrations of
undegraded oil would mask evidence of degradation. Examination of the data indicated that changes
in the nC 17/pristane and nC 18/phytane ratios were most apparent in the samples containing less total
oil. This is reasonable if one realizes that at low concentrations, the surface area-to-oil residue weight
ratio is large, as it is when oil is dispersed into the beach material as small droplets or films.
Effectiveness of biodegradation will increase as the oil surface area increases. With higher
concentrations of oil, the same degradation rate is probably occurring, but the surface area-to-oil
amount is much less. Because the oil is in bigger globs, the degraded oil on the surface is diluted by
the undegraded oil during sampling and homogenization. If this observation is valid, it should be
possible to normalize the extent of degradation to the amount of oil present. Figures 6.83 through
6.86 show that when the nCl 7/pristane and nCl 8/phytane ratios are plotted against their respective
residue weights, a direct relationship exists. This data is from plots prior to fertilizer treatment.
Regression analysis of the data gave r-values around 0.8 (alpha - 0.0001). By comparing slopes of
this relationship from two different sampling periods, the effect of biodegradation can be seen. The
slopes increased by two- and three-fold over 2 weeks. With more degradation the slope will continue
to steepen to a limit where the data points begin to cluster closer to the origin. This relationship may
have application in further analyzing data from treated and untreated plots. Initial attempts to
normalize the ratios with the oil residue weight to reduce variability of the data have, to date, been
ineffective. The approach, however, seems promising and further work will evaluate its usefulness.
MICROBIOLOGY
Numbers of Oil-Degrading Bacteria
The relative numbers of oil-degrading bacteria present on beach materials were determined from
parallel samples of beach material taken for oil chemistry. Sets of samples from the oleophilic
fertilizer treated beach, the water-soluble fertilizer treated beach, and the untreated control beach
were taken prior to fertilizer treatment, and on several dates after treatment. Numbers of oil-
degrading bacteria were assessed by: serially diluting each sample in a minimal salts medium contain-
ing ammonium and phosphate ion; adding a small quantity of oil to each dilution; incubating the
dilution tubes for 21 days; and then scoring the tubes for the absence or presence of biological growth
and changes in the physical character of the oil.
216
-------
2.1
2.0
1.9
1.8
1.7
1.6
1.5
1.4
1.3
1.2
1 1
1.0
0.9
0.8
0.7
0.6
0.5
0.4
0.3
0.2
0.1
0.0
.Fresh Prudhoe
Bay Crude Oil
r - 0.81
slope • 0.26
Log Residual Weight (mg/kg)
Figure 6.83. nC17/prlstane Ratio versus Log10 Residue Weight Two
Weeks Before Fertilizer Application (5/28/89).
217
-------
2.1
2.0
1.9
1.8
1.7
1.6
1.5
1.4
1.3
1.2
1.1
1.0
0.9
0.8
0.7
0.6
0.5
0.4
0.3
0.2
0.1
0.0
Fresh Prudhoe
Bay Crude Oil
r - 0.79
slope - 0.28
345
Log Residual Weight (mg/kg)
Figure 6.84. nC18/phytane Ratio versus Log10 Residue Weight Two
Weeks Before Fertilizer Application (5/28/89).
218
-------
.8
«
cc
o
Fresh Prudhoe
Bay Crude Oil
3 4
Log Residual Weight (mg/kg)
Figure 6.85. nCl7/prlstane Ratio Versus Log10 Residue Weight at Time
Zero of Fertilizer Application (6/8/89).
219
-------
.2
&
I
i
o
2.1 •
2.0
1.9 k
1.8
1.7
1.6
1.5
1.4
1.3
1.2
1.1
1.0
0.9
0.8
0.7
0.6
0.5
0.4
0.3
0.2
0.1
0.0
.Fresh Prudhoe
Bay Crude OH
r - 0.78
slope - 0.74
Leg Residual Weight (mg/kg)
Figure 6.86. nC18/phytane Ratio Versus Log10 Residue Weight at Time
Zero of Fertilizer Application (6/8/89).
220
-------
SNUG HARBOR
After the dilution tubes were scored, the relative bacterial populations were determined as
follows: for a given dilution series, the highest positive dilution was assigned the last positive tube
below which there was one or no negative tubes. Thus, a negative tube wus ignored only when the
first negative tube was followed by a positive tube. It is believed that this procedure did not
significantly bias the results, especially when focusing on relative comparisons between treated and
untreated control plots. In the first three sampling periods the dilutions were carried to a maximum
factor of 108, and in the last sampling period to a maximum factor of 1010.
The results shown in Table 6.20 are expressed as the Iog10 of the dilution factors used. In
evaluating these results, it is important to recognize that a significant number of the dilution series
were not high enough (i.e., the highest dilution series did not contain all negatives). Whenever
samples have been under-diluted, the normal summary statistics (such as the mean or standard
deviation) are biased. Also, whenever the median is the maximum dilution value, no good estimate
of the average is available. This situation occurred twice, so in these cases the true average is larger
than the median. However, when there are only a few maximum dilution values, the mean and
standard deviation will not be overly biased. In these cases the bias in the mean may be less than the
possible error from using the median. The possible change (error) in the median is 0.5 for the gain
or loss of one data point or for an error in reading any dilution series by one tube.
There was an upward trend in number of organisms from 106 to 107 per unit volume over the
sampling period for the control beach. A similar pattern was observed for the beach treated with
water-soluble fertilizer, except the number increased to 1010 for the last sampling date. The reasons
for this increase are not known. In contrast, a transient increase in bacteria numbers was observed
with samples from the beach treated with oleophilic fertilizer. This increase might be associated with
the transient availability of nutrients released from the oleophilic fertilizer.
ECOLOGICAL MONITORING
The monitoring component of the project was designed to identify ecological effects of nutrients
added to the shore zone on planktonic microorganisms. Sampling stations were established in
nearshore locations next to both treated and untreated control beaches in Snug Harbor (see Sections
3 and 4) and at locations outside Snug Harbor. Samples were collected on 9 occasions; once prior to
the addition of fertilizer, 2 days after addition, and 1, 2, 3, 4, 5, 6, and 8 weeks after addition. After
week 5, the stations 10 m from shore were no longer sampled in order to accommodate the workload
221
-------
SNUG HARBOR
TABLE 6.20. RELATIVE LEVELS OF OIL-DEGRADING MICROORGANISMS
IN SNUG HARBOR MIXED SAND AND GRAVEL TEST PLOTSmb
UNTREATED CONTROL BEACH
Pre-
Treatment
N
n>max
Median
Mean
SD
SD-Mean
22
0
5
5.18
0.91
0.19
6/17/89
21
8
6
6.33
1.53
0.33
WATER-SOLUBLE FERTILIZER-TREATED BEACH
Pre-
Treatmcnt
6/17/89
21
3
6
6.29
1.10
0.24
OLEOPHILIC FERTILIZER-TREATED BEACH
6/17/89
N
n>max
Median
Mean
SD
SD-Mean
20
2
5
5.65
1.04
0.23
Pre-
Treatment
N
n>max
Median
Mean
SD
SD-Mean
20
2
6
5.75
1.29
0.29
20
11
8+
7.00
1.21
0.27
6/24/89
20
1
6
6.05
0.83
0.19
6/24/89
18
0
6
5.78
0.65
0.15
6/24/89
20
2
6
6.05
1.10
0.25
7/8-9/89
19
4
7
7.58
1.61
0.37
7/8-9/89
21
16
10+
9.48
1.12
0.24
7/8-9/89
21
0
6
5.95
0.67
0.15
* The tabulated results show the number of samples run per beach (N), the number of serial dilutions
still showing positive results at the highest dilution (n>max), the median, the mean, the standard
deviation (SD), and the standard deviation of the mean (SD-Mean).
b All results are expressed as the Iog10 of the dilution factors used.
222
-------
SNUG HARBOR
from an additional study site. Data analyzed after week 5 indicated no significant loss in assessment
capability resulted from this decision.
Nutrients from Nearshore Waters
None of the nutrient concentrations increased in waters adjacent to treated shoreline compared
to the control shoreline, as illustrated by the ammonia and phosphorus data in Tables 6.21 through
6.24. These data provide evidence that fertilizers applied to the Snug Harbor shoreline either
remained within the matrix as applied, were taken up by microbial biomass, or were diluted to
background concentrations once they reached the shoreline. In any case, the potential for stimulating
plankton biomass from nutrient enrichment along the shoreline was not evident from these data.
Chlorophyll Analyses
Chlorophyll analyses of phytoplankton samples were used to monitor changes in the abundance
of algae. Nutrient enrichment could stimulate algal growth in Snug Harbor, and increased chlorophyll
concentrations would be evidence that nutrients had washed from the beach and had been
incorporated into algal biomass. None of the chlorophyll data indicated that algal populations within
Snug Harbor were stimulated by fertilizer applications (Figure 6.87). Results demonstrate that
nearshore concentrations were similar to those offshore; differences between samples collected near
treated beaches and reference areas were not ecologically significant. Differences observed between
nearshore (1 m) and offshore (10 m) samples and fertilized and reference shoreline samples were
within the expected range of day-to-day variation (Tables 6.25 and 6.26). All chlorophyll values were
within the expected range for Prince William Sound plankton communities.
Phytoplankton Primary Productivity
Photosynthetic production by phytoplankton is estimated by the incorporation of UC-
bicarbonate, providing a functional measure of the photosynthetic activity of algal cells. It allows an
evaluation of whether the algal population sampled is viable and active, nutrient limited, or enriched.
Comparisons of photosynthetic rates obtained on different sampling dates are not valid because the
light conditions during incubation could have been different enough to significantly affect the daily
productivity estimate. Only treated-versus-control comparisons are valid for each sampling date. On
several dates, primary productivity estimates from stations near fertilized shorelines were significantly
223
-------
SNUG HARBOR
TABLE 6.21. AMMONIA NITROGEN (p. N/L) FROM NEARSHORE WATER
OVER GRAVEL BEACHES AT SNUG HARBOR. MEAN OF FOUR
REPLICATES (STANDARD DEVIATION). (METHOD DETECTION LIMIT - 0.13 n N/L.)
Sample Date
6/10/89
6/14/89
6/21/89
6/28/89
7/5/89
7/12/89
7/23/89
8/9/89
Untreated Control
(Rodney Beach)
1m 10 m
1.5
(0.05)
0.68
(0.10)
0.92
(0.03)
0.21
(0.06)
0.51
(0.11)
0.80
(0.32)
0.13
(0.00)
0.13
(0.00)
1.5
(0.08)
0.65
(0.05)
1.02
(0.06)
0.15
(0.02)
0.52
(0.03)
0.73
(0.19)
__»
--
Oleophilic
(Otter Beach)
1m 10 m
1.6
(0.05)
0.52
(0.09)
0.74
(0.03)
0.13
(0.00)
0.56
(0.09)
0.57
(0.11)
0.13
(0.00)
0.13
(0.00)
1.7
(0.06)
0.58
(0.10)
0.83
(0.05)
0.20
(0.10)
0.57
(0.10)
0.50
(0.05)
--
--
Water-Soluble
(Otter Beach)
1m 10 m
1.5
(0.22)
0.61
(0.08)
0.73
(0.03)
0.13
(0.00)
0.74
(0.16)
0.63
(0.08)
0.13
(0.00)
0.13
(0.00)
1.8
(0.17)
0.58
(0.10)
0.74
(0.06)
0.20
(0.14)
0.53
(0.09)
0.96
(0.57)
—
--
-- « Sample not collected.
224
-------
SNUG HARBOR
TABLE 6.22. AMMONIA NITROGEN (/i N/L) FROM NEARSHORE WATER OVER
COBBLE BEACHES AT SNUG HARBOR. MEAN OF FOUR REPLICATES
(STANDARD DEVIATION). (METHOD DETECTION LIMIT - 0.13 n N/L.)
Sample Date
6/10/89
6/14/89
6/21/89
6/28/89
7/5/89
7/12/89
7/23/89
8/9/89
Untreated Control
(Fred Beach)
1m 10 m
2.1
(0.12)
0.73
(0.03)
0.99
(0.08)
0.24
(0.06)
0.61
(0.12)
0.62
(0.18)
0.13
(0.00)
0.13
(0.00)
1.8
(0.00)
0.70
(0.08)
0.91
(0.06)
0.35
(0.26)
0.65
(0.19)
0.70
(0.20)
__•
--
Oleophilic
(Seal Beach)
1m 10m
1.5
(0.05)
0.45
(0.06)
0.96
(0.04)
0.22
(0.13)
0.52
(0.03)
0.79
(0.16)
0.13
(0.00)
0.13
(0.00)
0.5
(0.10)
0.55
(0.12)
0.82
(0.03)
0.13
(0.00)
0.50
(0.05)
0.75
(0.14)
--
—
Water-Soluble
(Seal Beach)
1m 10m
1.4
(0.27)
0.64
(0.06)
0.87
(0.09)
0.22
(0.11)
0.44
(0.21)
0.86
(0.17)
0.13
0.13
1.4
(0.08)
0.48
(0.06)
0.88
(0.10)
0.18
(0.07)
0.47
(0.05)
0.78
(0.08)
--
--
* -- Sample not collected.
225
-------
SNUG HARBOR
TABLE 6.23. PHOSPHATE (/i P/L) FROM NEARSHORE WATER OVER GRAVEL
BEACHES AT SNUG HARBOR. MEAN OF FOUR REPLICATES (STANDARD DEVIATION).
(METHOD DETECTION LIMIT - 0.20 n P/L FOR SAMPLE DATE 6/10/89,
0.02 n P/L THEREAFTER.)
Sample Date
6/10/89
6/14/89
6/21/89
6/28/89
7/5/89
7/12/89
7/23/89
Untreated Control
(Rodney Beach)
1m 10 m
0.20
(0.00)
0.10
(0.00)
0.44
(0.00)
0.25
(0.00)
0.27
(0.04)
0.23
(0.03)
0.08
(0.00)
0.20
(0.00)
0.13
(0.03)
0.40
(0.03)
0.25
(0.00)
0.27
(0.04)
0.29
(0.03)
a
Oleophilic
(Otter Beach)
1m 10 m
0.34
(0.27)
0.18
(0.04)
0.29
(0.04)
0.15
(0.03)
0.36
(0.04)
0.22
(0.04)
0.08
(0.00)
0.20
(0.00)
0.15
(0.00)
0.28
(0.11)
0.17
(0.02)
0.23
(0.03)
0.32
(0.03)
--
Water-Soluble
(Otter Beach)
1m 10 m
0.20
(0.00)
0.15
(0.04)
0.34
(0.03)
0.20
(0.00)
0.37
(0.05)
0.25
(0.03)
0.10
(0.03)
0.26
(0.12)
0.12
(0.05)
0.35
(0.04)
0.16
(0.00)
0.28
(0.04)
0.22
(0.00)
--
Sample not collected.
226
-------
SNUG HARBOR
TABLE 6.24. PHOSPHATE (/i P/L) FROM NEARSHORE WATER OVER COBBLE BEACHES
AT SNUG HARBOR. MEAN OF FOUR REPLICATES (STANDARD DEVIATION).
METHOD DETECTION LIMIT - 0.20 /i P/L FOR SAMPLE DATE 6/10/89,
0.02 n P/L THEREAFTER.)
Sample Date
6/10/89
6/14/89
6/21/89
6/28/89
7/5/89
7/12/89
7/23/89
Untreated Control
(Fred Beach)
1 m 10 m
0.22
(0.03)
0.16
(0.02)
0.36
(0.02)
0.18
(0.02)
0.29
(0.05)
0.38
(0.00)
0.10
(0.03)
0.20
(0.00)
0.15
(0.00)
0.31
(0.03)
0.28
(0.03)
0.30
(0.04)
0.34
(0.03)
__•
Oleophilic
(Seal Beach)
1m 10m
0.20
(0.00)
0.15
(0.00)
0.35
(0.04)
0.16
(0.04)
0.32
(0.03)
0.25
(0.03)
0.09
(0.01)
0.20
(0.00)
0.12
(0.03)
0.25
(0.03)
0.15
(0.03)
0.25
(0.03)
0.23
(0.05)
--
Water-Soluble
(Seal Beach)
1m 10 m
0.22
(0.03)
0.14
(0.04)
0.26
(0.00)
0.20
(0.04)
0.34
(0.03)
0.25
(0.06)
0.09
(0.01)
0.20
(0.00)
0.14
(0.06)
0.27
(0.04)
0.24
(0.05)
0.30
(0.07)
0.22
(0.00)
--
a -- - Samples not collected.
227
-------
SNUG HARBOR
TABLE 6.25. TIDAL VARIATION IN MEASUREMENTS OF BACTERIAL ABUNDANCE
AND PLANKTON CHLOROPHYLL a FOR SAMPLING STATIONS AT SNUG HARBOR
ON 7/26-27/89. A SEQUENTIAL SERIES OF SAMPLES WAS COLLECTED
OVER A 24-HOUR PERIOD AT HIGH, MID, AND LOW TIDE.
REFER TO FIGURE 3.2 FOR SAMPLE STATION LOCATIONS
Bacterial Enumeration (cells x 100/L)
Station
Untreated ("ontrol
('•ravel
Gravel
Water-sol.
Cobble
Water-sol.
Untreated Control
Gravel
Gravel
Water-sol.
Cobble
Water-sol.
High
0.75
(0.06)
0.70
(0.04)
0.73
(0.02)
0.29
(0.49)
1.02
(0.26)
0.76
(0.04)
Mid
0.63
(0.03)
0.67
(0.06)
0.63
(0.02)
Mg
1.17
(0.30)
0.89
(0.340
0.42
(0.11)
Low
0.66
(0.05)
0.57
(0.04)
0.60
(0.03)
Chlorophyll a/L
0.82
(0.09)
0.80
(0.07)
0.92
(0.12)
Mid
0.58
(0.02)
0.62
(0.14)
0.63
(0.03)
0.62
(0.08)
0.77
(0.06)
0.89
(0.13)
High
0.66
(0.04)
0.64
(0.04)
0.66
(0.01)
0.87
(0.06)
0.90
(0.13)
0.90
(0.04)
Mid
0.61
(0.04)
0.73
(0.01)
0.74
(0.03)
0.70
(0.08)
0.88
(O.H)
0.95
(0.17)
Low
0.75
(0.04)
0.74
(0.02)
0.74
(0.02)
1.00
(0.09)
0.65
(0.18)
0.79
(0.07)
228
-------
SNUG HARBOR
TABLE 6.26. TIDAL VARIATION IN MEASUREMENTS OF BACTERIAL ABUNDANCF
AND PLANKTON CHLOROPHYLL a FOR SAMPLING STATIONS AT PASSAGE COVE
ON 8/7/89. VALUES ARE MEANS OF 4 REPLICATES WITH STANDARD
DEVIATION (). A SEQUENTIAL SERIES OF SAMPLES AS COLLECTED
OVER ONE TIDE AT HIGH, MID, AND LOW TIDE.
REFER TO FIGURE 3.3 FOR SAMPLE STATION LOCATIONS
Bacterial Enumeration (cells x 109/L)
Station
Station 3, 0.5 m
Station 3, 5.0 m
Station 5, 0.5 m
High
0.18
(0.01)
0.57
(0.03)
0.27
(0.01)
Mid
0.24
(0.04)
0.25
(0.020
0.50
(0.07)
Low
0.26
(0.04)
0.52
(0.08)
0.24
(0.02)
Chlorophyll a/L
Station 3, 0.5 m
Station 3, 5.0 m
Station 5, 0.5 mn
Station 7, 0.5 m
0.46
(0.12)
0.38
(0.55)
0.60
(0.28)
0.64
(0.19)
0.40
(0.03)
0.50
(0.03)
0.46
(0.04)
0.46
(0.11)
0.37
(0.19)
0.16
(0.05)
0.15
(0.39)
0.65
(0.40)
229
-------
1 2 .
o
c
o
o
COBBLE CONTROL
6-10 6-14 6-21 6-2» 7-6 7-12 7-23 6-9
COBBLE WATER-SOLUBLE
6-10 6-14 6-21 6-21 7-6 7-12 7-23 8-9
COBBLE OLEOPHILIC
QRAVEL CONTROL
• 1.0 meters
D 10.0 meters
6-10 6-14 6-21 6-28 74 7-12 7-23 8-9
GRAVEL WATER-SOLUBLE
6-10 6-14 6-21 6-28 7-C 7-12 7-23 8-9
GRAVEL OLEOPHILIC T
6-10 6-14 6-21 6-2» 7-« 7-12 7-23 6-9
6-10 6-14 6-21 6-28 7-6 7-12 7-23 6-9
SAMPLE DATES
Figure 6.87. Phytoplankton Chlorophyll Data (mg Chlorophyll a/L) from
Water Samples Collected Along Cobble and Gravel
Shorelines at Snug Harbor Following June 7 and 8, 1989,
Fertilizer Additions to Gravel Shorelines. Values are Means
(+SD) of 4 Replicates; Dark Bars are for Sample Sites 1 m
Offshore, Open Bars are For Sites 10m Offshore. Refer to
Figure 3.2 for Station Locations.
230
-------
SNUG HARBOR
greater than control values using statistical comparisons (Figure 6.88). However, these differences
generally were less than a factor of 2, inconsistent through time, and within the range of expected
ecological variability. If small changes in daily primary productivity were occurring, the lack of a
change in chlorophyll content suggests that the increased biomass associated with increased
productivity was not accruing faster than the dilution and transport of water masses associated with
tidal exchange for the basin.
Bacterial Abundance
Abundance of bacteria in the water column samples from Snug Harbor is reported in Figure
6.89. Mean bacterial abundances varied from 0.21 to 2.49 X 10° cells per liter. One week after the
nutrient additions, numbers were higher than Day 2 values for the fertilized shorelines, but the values
for control shorelines did not change from Day 2 to Week 1. Because the values for treated shorelines
did not increase beyond control shoreline values, the changes, by themselves, are not considered
ecologically significant. Differences observed between treated and control areas on other dates were
within the range of natural variability shown in Tables 6.25 and 6.26. Bacterial abundance showed
no trends associated with shoreline treatments, nearshore versus offshore comparisons, or changes
through time over the monitoring period.
Bacterial Production
Since the presence of cells alone may not represent the viability of planktonic microbes, bacterial
production estimates allow an evaluation of functional activity of the bacterial community and the
effect of nutrient enrichments. As seen in the bacterial abundance data, there were no consistent
changes or trends in bacterial production measurements that can be associated with fertilizer
application to the shoreline (Figure 6.90). An inspection of the data shows greater productivity
during the first two sampling periods compared to subsequent sampling. This was seen at control
samples as well as treated sites, indicative of a seasonal trend rather than a treatment effect. None
of these differences appeared to be ecologically significant.
231
-------
Legend
COBCTL - Cobble Control
COB OLE - Cobble Oleophilic
COBWS - Cobble Water-Soluble
GRAV CTL - Gravel Control
GRAV OLE - Gravel Oleophilic
GRAVWS - Gravel Water-Soluble
a ? i i 1 2 i ? i
nuum
a !
n
52
50
41
46
44
42
40
II
3$
34
92
30
21
je
24
22
20
tl
ie
1 4
12
1 0
01
OS
04
02
00
^ " «k
MM
8-2M9
Figure 6.88. Primary Productivity Estimates (From "C Uptake; mg
C/m3/hour) For Phytoplankton Samples From Snug Harbor at
Various Sample Dates Following the June 7 and 8,1989,
Fertilizer Additions Along Cobble and Gravel Shorelines.
Values are Means (+SD) of 4 Replicates. Refer to Figure 3.2
for Station Locations.
232
-------
o
O)
I
O
O
oc
QL
Q.
52
so.
41
48
44
42
40
H
36.
34
32
30
2»
28
24
22
20
I I
i 8
1 4
I 2
I 0
01
08
04
02
00
7-5-89
| i 1 § J
8 8 8 j 8
2 3
Si 3 a I
i 38 3
8 8 3 5 5
Figure 6.88. (Continued)
233
-------
1
QC
LU
ffi
D
COBBLE CONTROL
GRAVEL CONTROL
• 1.0 motors
D 10.0 maters
Ml ft-21 7-S 7-12 7-23 «-02
COBBLE WATER-SOLUBLE
GRAVEL
WATER-SOLUBLE
6-10 t-14 8-2' »-2»
COBBLE OLEOPHILIC
GRAVEL OLEOPHILIC
SAMPLE DATES
Figure 6.89. Abundance of Bacteria (cells x 10"/L) From Water Samples
Taken Along Cobble and Gravel Shorelines on Various
Sample Dates Following the June 7 and 8,1989, Fertilizer
Additions to Snug Harbor Shorelines. Plotted Values are
Means (+ SD) of 4 Replicates; Dark Bars are For Sample
Sites 1 m Offshore, Open Bars are For Sites 10 m Offshore.
Refer to Figure 3.2 for Station Locations.
2)4
-------
CO
T3
O
3
O
o
cc
Q.
ui
o
<
00
COBBLE WATER-SOLUBLE
t-10
*-u
T-tt
• 1.0 meters
D 10.0 meters
GRAVEL WATER-SOLUBLE
;••
SAMPLE DATES
Figure 6.90. Bacterial Productivity, as Measured by Tritiated Thymidine
Uptake (mM Thymidine/LVday), for Bacterial Samples
Collected on Various Sample Dates Adjacent to Cobble and
Gravel Shorelines at Snug Harbor. Fertilizer Additions Were
Completed on June 7 and 8,1989. Plotted Values are Means
(+SD) of 4 Replicates; Dark Bars Are For Sample Sites 1 m
Offshore, Open Bars Are For Sites 10 m Offshore. Refer to
Figure 3.2 for Station Locations.
235
-------
SNUG HARBOR
Cased Mussels
None of the mussel tissue samples had detectable residues of PAHs (Table 6.27). Application
of water-soluble nutrients (water) or oleophilic nutrients (Oleo) did not increase the input of
petroleum products into the nearshore zone to the extent that residues became detectable in nearshore
mussels. Oil components were also not detected either from the control areas. The detection limit
for these samples was low enough to identify the presence of PAHs before they might become an
i-cotoxicological problem. Bioaccumulation of PAHs by nearshore mussels was not detected at Snug
Harbor cither due to a lack of PAH input, or mixing and dilution of PAHs that did make it into the
nearshore zone.
TABLE 6.27. TOTAL PAH'S (/iG/G) IN CAGED MUSSELS AT SNUG HARBOR
AT 6 STATIONS OVER TIME
Cobble
Seal
Date
6/21
6/28
7/5
7/10
7/23
8/10
8/27
Water
ND
ND
ND
ND
ND
ND
ND
Oleo
ND
ND
ND
ND
ND
ND
ND
Sand
Otter
Water
ND
ND
ND
ND
ND
ND
ND
Oleo
ND
ND
ND
ND
ND
ND
ND
Cobble
Fred Nails
Control
ND
ND
ND
ND
ND
ND
No sample
Sand Detection
Rodney
Control
ND
ND
ND
ND
ND
No sample
ND
Limit
G*/g)
0.20
0.20
0.20
0.20
0.20
0.20
0.05
No. of
Replicates
Analyzed
4
4
4
4
4
4
4
ND - None detected
Problems were encountered using a small sample size (3 mussels/sample collected) and should
have been larger (10 mussels/sample collected). In addition, the number of "time zero" mussels
collected should have been equal to the number of mussels set out in all the test sites at the beginning
of (he test. This would ensure enough tissue for analytical method development and validation at the
beginning of the test, and enough for quality assurance analysis for the duration of the test.
236
-------
SNUG HARBOR
MICROCOSM STUDIES
Microcosm tank studies were performed in the summer of 1989 to simulate the field
demonstration project as a protection against possible loss of data, such as through a major storm
event or some other unforseen complications, and to provide a potential basis from which scale-up
decisions could be made. These studies were therefore designed to test the effects of the fertilizer
treatments under controlled conditions.
Tank Microcosms
Mixed sand and gravel microcosms were sampled 22 days post application (July 7) and cobble
microcosms were sampled 26 days post application and 41 days post application (July 11 and 26).
Visual observations at the time of sampling indicated that the oleophilic fertilizer-treated cobble
microcosms appeared to have the least surface oil on the cobble surfaces, but the difference with
other treated and control microcosms was not dramatic. Where oil was present it appeared mottled,
suggesting that the oil on the surface had been partially removed or degraded. Oil was apparent under
the rocks, but it was very black and viscid. This consistency appeared to be due to the oleophilic
fertilizer dissolving into the oil.
The amount of surface oil in the control and fertilizer powder-treated microcosms appeared to
be approximately the same. Cobble systems showed some rocks with clean surfaces, but there were
generally fewer than in the oleophilic fertilizer-treated systems. Oil on the rock surfaces appeared
gray and dried. Oil under the rocks was drier and less fluid than oil observed in the oleophilic
fertilizer-treated microcosms.
After sampling the microcosms it was noted that the inside walls of the fertilizer powder-treated
microcosms and the reference microcosms were spotted with oil smudges. This was not the case in
the-oleophilic fertilizer-treated set, where the walls generally appeared free of oil. Small particles of
white waxy material were also observed throughout the sand and gravel in the oleophilic fertilizer-
treated set of microcosms, even with the daily influx of fresh seawater. This material may have been
residual oleophilic fertilizer, suggesting possible over application.
237
-------
SNUG ItARBOR
In a sand and gravel microcosm sampled 17 days post fertilizer application, the nCI7/pristane
and nC18/phytane ratios in the oleophilic fertilizer-treated microcosms were the same as those in the
untreated microcosms (Table 6.28). Ratios for fertilizer powder-treated microcosms were almost half
the ratios for the other microcosms, and there was also approximately 20% less oil residue by weight.
These data suggest that the more rapid degradation of oil was occurring in the fertilizer powder treat-
ments, assuming oil concentration and composition were approximately the same in all microcosms
at the start of the experiment. Unfortunately, the t-0 samples were lost. However, since a single
batch of homogenized beach material was used to construct the microcosms and the consistency
between replicates was generally good, the effect of enhanced oil biodegradation seems reasonable.
Because of the large amount of readily degradable carbon added with the oleophilic fertilizer,
enhanced degradation of the oil may not occur until after much of this carbon is degraded.
TABLE 6.28. CHEMICAL ANALYSIS OF MIXED SAND AND GRAVEL MICROCOSMS
SAMPLED 17 DAYS AFTER INITIATION OF FERTILIZER APPLICATION
Treatment
Average
Oleophilic I
Oleophilic 2
Oleophilic 3
Average
Pert, powder 1
Pert, powder 2
Pert, powder 3
Average
Residue Weight
(mg/kg)
1,091
1,490
1,360
795
1,215
913
916
845
891
nC17/Pristane
0.4
0.4
0.4
0.4
0.4
0.3
O.I
0.3
0.2
nC18/Phytane
Control 1
Control 2
Control 3
1,570
913
790
0.5
0.4
0.4
0.8
0.6
0.6
0.7
0.7
0.8
0.7
0.7
0.4
0.3
0.3
0.3
238
-------
SNUG HARBOR
In the cobble microcosms sampled 26 days post application of fertilizer, similar results were
observed. The lowest oil residue weights (Table 6.29) and the greatest change in composition (as
measured by the nCI7/pristane and nCI8/phytane ratios in Table 6.30), appeared to be in the
fertilizer powder-treated systems. Oleophilic fertilizer-treated and untreated systems were
approximately the same. The results were relatively consistent between replicates and with different
layers in the microcosm. Thus, it is again tempting to assume that the lower residue weights and
ratios were due to fertilizer-enhanced biodegradation. Oil residue weights in the oleophilic fertilizer-
treated systems were as high as 6 times those in the control microcosms. This indicates that a
component of the oleophilic fertilizer was possibly contributing to the residue weight and that over
application had occurred.
In contrast to day 26 data, nC17/pristane and nC18/phytane ratios for the day 41 samples
suggest that oil was being degraded faster in the control microcosms than it was in the fertilizer
powder-treated microcosms (Table 6.31). However, the hydrocarbon ratios may yield false indications
of biodegradation if pristane or phytane are degraded along with straight chain hydrocarbons. Gas
chromatography/mass spectrometry analysis of the data provided sufficient data to evaluate this poss-
ibility (Table 6.31). By extracting and analyzing all microcosm samples using the same method, two
compounds whose concentrations did not change in any of the treatments were identified: norhopane
and hopane. The ratio of norhopane to hopane remained constant at 0.76 (Table 6.31). Constructing
nCI7/norhopane and pristane/norhopane ratios indicated that nC17 degradation was 5 times more in
the fertilizer powder-treated microcosms than in the control microcosms (Table 6.31). Based on the
same ratio method, pristane was also degraded in both the control and fertilizer powder-treated
microcosms, supporting the suggestion that nC17/pristane ratios could be misleading.
In addition, the ratios of the three major dibenzothiophene peaks to norhopane were also
examined using mass spectral analysis (Table 6.32). Further differences between the treatments were
apparent. Fertilizer powder-treated microcosm samples showed the lowest ratios. Interestingly, the
ratios for the oleophilic treatment indicated little change in the dibenzothiophene isomers, compared
with the ratios observed in a Prudhoe Bay crude oil standard. These observations are consistent with
the nC17/pristane data from previous samplings, which also indicated that oil degradation in the
oleophilic treatment was less than the degradation in both the fertilizer powder and control
treatments.
239
-------
SNUG HARBOR
TABLE 6.29. RESIDUE WEIGHT OF OIL IN COBBLE MICROCOSMS ANALYZED 26 DAYS
AFTER FERTILIZER APPLICATION
Residue Weights fme/kgl
Treatment
Top Cobble
Bottom Cobble
Gravel
Control 1
Control 2
Control 3
Average
Oleophilic 1
Oleophilic 2
Oleophilic 3
Average
Pert, powder 1
Pert, powder 2
Pert, powder 3
Average
1,420
889
1,040
1,116
1,770
1,260
2,340
1,790
161
1,240
383
595
1,120
1,090
722
977
1,910
2,460
3,550
1,640
1,310
725
664
900
889
1,090
1,030
1,993
6,350
5,580
6,960
6,297
1,020
714
814
849
TABLE 6.30. RATIOS OF HYDROCARBONS IN OIL FROM COBBLE MICROCOSMS
ANALYZED 26 DAYS AFTER FERTILIZER APPLICATION
nC17/Pristane
Treatment
Control 1
Control 2
Control 3
Average
Oleophilic 1
Oleophilic 2
Oleophilic 3
Average
Pert, powder 1
Pert, powder 2
Pert, powder 3
Average
Top
Cobble
0.8
0.9
Q.7
0.8
0.9
1.0
LQ
1.0
0.2
0.6
QA
0.5
Bottom
Cobble
0.7
0.6
QJ
0.7
0.8
0.9
1L2
0.9
0.3
0.1
JL2
0.2
Gravel
0.3
0.4
JL4
0.4
1.0
0.9
Q,8
0.9
0.5
0.6
Q*i
0.5
nC18/Pristane
Top
Cobble
1.3
1.3
L2
1.3
1.3
1.2
L2
1.2
0.4
1.1
JL2
0.8
Bottom
Cobble
1.1
1.0
U
1.0
1.4
1.5
L4
1.4
0.4
0.3
(U
0.3
Gravel
0.5
0.5
JLi
0.5
1.5
1.4
L2
1.4
0.5
0.7
QJ>
0.5
240
-------
SNUG HARBOR
TABLE 6.31. COMPARISON OF NCI7/PRISTANE RATIOS AND NC17/NORHOPANE RATIOS
AS MEASURES OF OIL DEGRADATION IN SAMPLES TAKEN FROM COBBLE MICROCOSM
42 DAYS AFTER INITIATION OF FERTILIZER APPLICATION
Microcosm
Control
Fert. powder
Fresh Prudhoe
Bay Crude Oil
nC17/
Pristane
0.19
0.49
1.7
nC17/
Norhopane
1.03
0.22
17.50
Pristane/
Norhopane
5.44
0.44
10.68
Norhopane/
Hopane
0.78
0.75
0.78
TABLE 6.32. USE OF DIBENZOTHIOPHENE PEAKS/NORHOPANE RATIOS
AS RELATIVE MEASURES OF THE DEGRADATION OF AROMATIC COMPONENTS
IN OIL SAMPLED FROM COBBLE MICROCOSMS 42 DAYS AFTER INITIATION
OF FERTILIZER APPLICATION
Dibenzothiophene Peaks'/Norhopane Ratios
Microcosms6 Peak 1 Peak 2 Peak 3
Control 1
Control 2
Control 3
.40
.49
.46
.54
.66
.70
.60
.71
.71
Fert. powder 1 .08 .13 .13
Fert. powder 2 .10 .12 .19
Fert. powder 3 .11 .13 .17
Oleophilic I .82 1.21 1.06
Oleophilic 2 .81 1.15 1.01
Oleophilic 3 .85 1.17 .99
Fresh Prudhoe Bay 1.06 1.84 1.54
Crude Oil
* In the mass spectral analysis of oil, C-2 dibenzothiophenes and their homologs show a scries of
peaks at mass ion 212. Three prominent peaks (labeled here 1,2, and 3) were selected for comparison.
b Average of three replicates.
241
-------
SNUG HARBOR
From these initial microcosm results, it can be concluded that enhanced oil biodegradation will
occur if sufficient nutrients are supplied to the microorganisms. Because microcosms represent test
systems that best reflect field conditions, a similar response could be expected in the field if nutrient
concentrations were maintained at adequate levels. The lack of any effect of the oleophilic fertilizer
was probably due to over application. The microcosm studies also showed that pristane and phytane
are biodegraded and therefore must be used with caution when assessing changes in oil composition.
Jar Microcosms
The results from these studies indicated that the addition of oleophilic fertilizer led to a
substantial increase in the number of organisms capable of growth on oleic acid-agar plates
(Figure 6.91). High background concentrations of oleic acid-degrading bacteria were observed in the
water even before oleophilic fertilizer treatment.
Since the aqueous phase at each water change was sterilized, the number of oleic acid-degraders
may reflect those that sloughed off the oiled rocks during a 24-hour period. However, no obvious
differences were observed for the different aqueous phases. Similar results were observed in systems
that did not have daily water changes.
Results from the enumeration of oil-degrading organisms indicated that in all cases the
populations increased to a high value by Day 3 and then decreased to an intermediate but variable
level for the following 6 days (Figure 6.92). Similar results were seen in those jars that did not have
a daily water change. Although all samples showed a peak after 3 days of incubation, jars containing
only seawater appeared to have the fewest microorganisms in the 6 days following the 3-day peak.
Chemical analysis of the water samples is being performed. Information on how effectively the
enriched oleic acid-degraders can degrade the oil also is forthcoming.
Oleophilic fertilizer increased the number of oleic acid-degrading bacteria in flask studies
designed to approximate field conditions. This situation would theoretically result in competition for
available nutrients between oleic acid-degrading and oil-degrading bacteria. This competition could
explain the decrease in oil-degrading bacteria following their initial rise after initiation of the
experiment. Supplying dissolved nutrients in addition to nutrients in the oleophilic fertilizer did not
seem to affect the oleic acid- and oil-degrading bacterial populations.
242
-------
Control
INIPOL-Treated
c
3 I
U
Artificial 5 :
Seawater | ~
I i
o ;
u —
I A
>, ••
8 "f
Seawater ^ 4 ^
I 1
2 '•=
o ;
Q
8 —
u ;
o
x 6 ;
Saline | 5~
Solution „ 4bi
s 4 :
! r
i _
i
0 -
—
—
—
I
1
1
—
"•""
1
1
2
1
—
3
1
1
3
1
—
i
—
—
1
I
4
\
—
1
Da
—
D,
—
1
lys
I
5
ays
I
—
6
—
—
I
EOT,
i
P3S
—
7
1
1
1
1
4
I
a
1
— i
i.
—
—
1
I
t
I
Days
Figure 6.91.
Effect of INIPOL on the Relative Numbers of Oleic
Acid-Degrading Microorganisms in Jars Containing Oiled
Rocks and Artificial Seawater, Seawater, or Saline
Solution. Incubated with Dally Change of Water.
241
-------
10 —
to « —
1 "i
cc
<» 7 -
I -
Artificial f «-
Seawater ? , _
a
r i -
10 —
* B —
S '-
e
15 •
a * -
21
Seawater § 5-
! '-
1 3-
I-
1 i_
0 •
10 —
3> * ~
? 7-
e
IS =
0. S -
? i
Saline | s^
Solution | 4^
a H
t** -
J-
-
1 ,^
0 —
1 — 1
—
1 —
I
I
1
I
1
1
?
1
'
1
1
1
—
1
1
1
4
1
—
5
!
1
1
1
e
(
1
1
i
1
• —
rar_,
I
1
H
3
I
Cc
INI
ntrol
POL-
9
1
1
Tre
4
i
ate
1
1
1
id
Days
Figure 6.92. Effect of INIPOL on the Relative Numbers of Oil-Degrading
Microorganisms in Jars Containing Oiled Rocks and
Artificial Seawater, Seawater, or Saline Solution.
Incubated with Daily Change of Water.
-------
SNUG HARBOR
MUTAGENICITY TESTS
Experiments were initiated at Snug Harbor to determine the potential mutagenic ucliviiy
associated with biodegradation of oil. The typos of health hazards for which monitoiinit is most
difficult are those with chronic, delayed effects such as carcinogenicity, neurotoxicity, and
mutagenicity. Fortunately, for mutagenicity there are short-term in vitro tests that demonstrate
whether or not a pollutant interacts in a detrimental manner with DNA. Due to the mechanistic
research with oncogenes, available evidence shows that oncogene activity can be initiated by mutation.
Mutation assays, although not definitive, can be used as screening tests for the presence of potential
carcinogens. When performed in a quantitative, dose-responsive fashion, these bioassays can be used
to detect alterations in the quantity of mutagens present within complex mixture samples.
Potential health effects from the oil spill were assessed by examining the mutagenicity associated
with the oil spill, the weathered oil, and the products associated with bioremediation. The most
commonly used mutation assay is the Salmonella typhimurium/mzmmalian microsome assay developed
by Ames. An early pilot study had demonstrated that extracts of spilled oil were mutagenic in the
Salmonella typhimurium bioassay for mutagenicity. This meant that the removal of genotoxic
components from the oil by biodegradation could be monitored with this assay.
Both the Prudhoe Bay crude oil and the weathered oils tested were weakly mutagenic using
TA100. The commercial fertilizer formulations were non-mutagenic. Figure 6.93 shows the
mutagenicity of oil samples collected from an untreated control site, an oleophilic fertilizer-treated
site, and a water-soluble fertilizer-treated site within sandy/gravel and cobblestone beach areas.
Examination of the mutagenicity per amount of soil (Figure 6.93) shows that the overall mutagenic
activity of the gravel beach is higher than the cobblestone beach. This activity tends to persist for
the four month period. On both types of beaches, however, the mutagenicity declines with time. It
is interesting to note that the decline was observed for both the untreated control and treated beaches.
By the time the first samples were collected in June, natural processes may have already initiated
natural bioremediation activities (e.g., the snow melt on the mountain may have washed organic
matter into the beach area). In addition, the fresh water outflow and wave action may have removed
oil spill organics from the beach.
The chemical analyses, however, did not determine whether or not mutagens were removed from
the soil at a rate commensurate with other residue chemicals. Figure 6.93B shows that the
245
-------
Gravel Beach
Cobblettone Beach
i no
110
40
-5-5-
4-r
_ji
«
0
«
i.
w
M
w
t1
U.I
0 01
0 001
"B
« 6 8 I
mi
6 n
as0? as1? &S0? a
3 S * * S fc « S » *
Sept.
S »
« I
June
July
Gravel Batch
June
July
Sept.
Cobblestone Beach
Legend
Rei - Untreated
Control
Oleo - Oleophilic
Fertilizer
W-S = Water-Soluble
Fertilizer
Figure 6.93. Mutageniclty of Soil Extracts Using the Spiral Salmonella
typhlmurlum Assay (Houk et al., In Press) with Strain TA98
with an Aroclor 1254-lnduced CD-1 Rat Liver Homogenate
(Atlas and Pramer, 1990) Exogenous Activation System. The
Spiral Assay Analysis Provided a Net Increase In Revertant
Colonies per Amount of Substance Tested. (A) Mutagenicity
Per Gram of Soil for Samples Taken From Sandy/Gravel and
Cobblestone Beaches. Each Beach Type was Represented
by an Untreated Control Site (No Fertilizer); Oleo (Oleophilic
Fertilizer); and W-S Site (Water-Soluble Fertilizer). Numerals
1 and 2 for Each Treatment Indicate the Two Randomly
Selected Grids Used for Sampling. Each Sample was
Bloassayed Twice and the Means are Presented. (B)
Mutageniclty of the Extracted Organic Material Expressed as
Revertants per mg Organic Material. The Values Represent
the Means of Samples Taken from Two Representative Grids
with Each Sample Tested Twice. (C) The Mean Percent of
Extractable Mass for the Samples Represented In Graph B.
Samples Were Extracted with Dlchloromethane.
-------
SNUG HARBOR
mutagenicity per mg of extractable matter (oil residue) decreased with time. Since this is not
expressed on a total volume or weight of soil basis, it demonstrates that the mutagenic components
were being depleted at a faster rate than the overall organic content. Figures 6.93B and 6.93C also
show that both the mutagenic activity per amount of extractable organic matter and the percent of
extractable mass decreased over time.
These mutagenicity studies show that mutagenic toxins associated with spills of Prudhor Bay
crude oil were lost over time. In conjunction with chemical analysis, these studies demonstrated that
decreases in toxicity were due to both fertilizer-enhanced bioremediation and other natural processes.
In addition to ongoing laboratory studies that are examining the mutagenicity of other oils and their
degradation products, these studies will assist in selecting appropriate bioremediation procedures for
environmental oil spills.
SUMMARY AND CONCLUSIONS
Based on analysis of the data from the bioremediation field demonstration in Snug Harbor in
the summer of 1989 the following general discussion and conclusions can be drawn:
a) Visual inspection of beaches treated with oleophilic fertilizer showed that oil was removed
from the beach surface approximately 2 to 3 weeks after fertilizer application. The effect
was most apparent on cobble beaches, where initially much of the surface oil was removed.
No visible decreases in the oil occurred, however, on the beaches treated with the slow-
release fertilizer briquettes or on the untreated control beaches. Disappearance of oil on
oleophilic-treated plots continued over time, eventually leading to the disappearance of oil
from most of the beach material surfaces.
b) No oil slicks or oily materials were observed in the seawater following application of the
fertilizers, and no oil or petroleum hydrocarbons were detected in mussels contained in cages
just offshore from the fertilizer-treated beaches. This suggested that removal of oil from the
beaches did not appear to be a result of dispersing phenomena.
c) Analysis of oil extracted from all beach plots showed that the effect of fertilizer application
on the loss of oil residues from the test plots was only apparent in oil samples taken from the
cobble surface of the IN1POL fertilizer-treated plots. Results were statistically significant
247
-------
SNUG HARBOR
at the 95% confidence interval. The enhancement effect, however, lasted for only four
weeks, suggesting that the supply of nitrogen and phosphorus from the fertilizer was depleted
during this period. This oil disappearance corresponded to both visual observations and the
detection of elevated nutrient concentrations on this plot. Although it is tempting to attribute
this toss to biodcgradation, further analysis of changes in oil composition must be conducted
to substantiate the potential role of biodcgradation. None of the other types of beach samples
(oil from mixed sand and gravel under cobble and oil from mixed sand and gravel alone) or
fertilizer treatments showed enhanced oil removal relative to the untreated control. The
absence of any effect on the oil in the mixed sand and gravel under cobble may have been
due to very low initial oil concentrations (little room to see changes) and highly erratic
distribution in the beach material. If we entertain the possibility of a prominent role of
biodegradation in determining oil fate, it is quite likely in the mixed sand and gravel plots
that not enough nutrients were delivered to oil-degrading microbial communities in the beach
material to promote a sufficient enhancement of oil degradation and. thus, cause a significant
decrease in oil concentration. On the other hand, oil-degrading microbial communities
within the beach material may not have had time to become as enriched as those on the
surface of the cobblestone. Thus, we would suggest that INIPOL works best on surface oil
and that additional fertilizer should be added to supply nutrients to developing microbial
communities in the subsurface.
d) Due to the very heterogeneous distribution of oil on the beaches, imprecise methods for
sampling unconfined gravel and cobble, and high amounts of natural oil biodegradation, it
was not possible to statistically link visual changes with enhanced removal of oil residues.
However, trends in the data strongly suggest that the most substantial loss of oil residue
occurred in the cobble surface samples taken from the oleophilic fertilizer-treated beaches,
particularly during the first 20 to 30 days of the test. During the latter part of the test
period, oil residue losses on all beaches seemed to dramatically increase beginning sometime
in the middle of July, leading to a virtual cleaning of all test beaches. The factors
responsible for this increase are not known.
e) Samples of oil from fertilizer-treated beaches, particularly from cobble surfaces, that were
taken around the time the oil was visually disappearing showed substantial changes in
hydrocarbon composition. This indicated extensive biodegradation, and suggested that
biodcgrudution might also be affecting oil removal, both through direct decomposition and
248
-------
SNUG HARBOR
possibly through the production of biochemical products (bioemulsificrs) known to be
produced by bacteria as they consume oil and hydrocarbons as food sources. Changes in oil
composition, including the standard measure or oil biodegradation involving the ratio or
specific branched alkanes to rapidly degraded straight-chain alkanes, were greatest on the
beach treated with fertilizer briquettes. The effect was most pronounced during the first
20 to 30 days of the test in cobble plot samples, suggesting that the fertilizer-enhanced
changes were short-lived. Depending on the hydrocarbons measured, the application of
oleophilic fertilizer also significantly enhanced changes in oil composition, but to a lesser
extent. It is hypothesized that the presumed ability of the oleophilic fertilizer to hold
nutrients within the oil-degrading microbial communities led to a greater mass of oil
degraded. This degradation included mineralization to CO2 and conversion into microbial
biomass. This in turn changed the physical consistency of the oil, thereby allowing the
degraded oil to be sloughed off the beach material by tidal action. Thus, oil biodegradation
in Prince William Sound was nutrient limited and rates were enhanced by the addition of
fertilizer. These results also lead to the conclusion that fertilizer briquettes, or a similar
formulation that releases inorganic nitrogen and phosphorus, would likely affect changes in
oil composition on both the cobble surface and within the mixed sand and gravel matrix.
f) All changes in oil composition were accompanied by large decreases in the nCI8/phytane
ratio. This represents a differential change in chemically similar hydrocarbons, and can only
be attributed to biodegradation processes. Thus, fertilizer application appeared to enhance
oil biodegradation.
g) Numbers of oil-degrading microorganisms did not appear to increase as a result of fertilizer
application. However, large heterogeneity in the microbial population precluded observing
statistical differences. This was further complicated by the awareness-that the numbers of
oil-degrading bacteria in the oiled beach material before exposure to fertilizers were very
high, averaging 1 to 10% of the total bacterial population. The high numbers represented
an enrichment of oil-degrading microorganisms of approximately 10s to I08 as compared
to beach microorganisms not exposed to oil. These results demonstrate that the beaches were
well primed for bioremediation.
h) Extensive monitoring studies indicated that the addition of fertilizer to oiled shorelines did
not cause ecologically significant increases in planktonic algae or bacteria, or any measurable
249
-------
SNUG HARBOR
nutrient enrichment in adjacent embayments. In addition, mutagenicity studies showed that
mutagenic materials associated with Prudhoe Bay crude oil were lost over time from both
treated and untreated control plots. In conjunction with chemical analysis, these studies
demonstrated that decreases in mutagenicity were due to both fertilizer-enhanced
biodegradation and other natural processes.
U S. Environmental Protection Agency
Region 5, Library (PL-12J)
77 West Jackson Boulevard, 12th Hoor
Chicago, IL 60604-3590
250
------- |