Annex F
Oil Spill Modelling
Vessel Spill and 45-day Loss of Well Control
REPORT
Tamarind Resources - Tui Field Oil Spill Modelling
Prepared by: RPS AUSTRALIA WEST PTY LTD Prepared for: ERM NEW ZEALAND Suite E1, Level 4 117 Powderham St 140 Bundall Road New Plymouth 4310 Bundall, QLD 4217 New Zealand Australia T: +61 7 5574 1112 T: +64 (0)9 3034 664 E: [email protected] E: [email protected] W: www.erm.com Author: Nathan Benfer Reviewed: Sasha Zigic Approved: Nathan Benfer No.: MAQ0648J Version: 2 Date: 21/03/18
rpsgroup.com.au REPORT
Document Status
Version Purpose of Document Approved by Reviewed by Review Date
0 Draft issued to client Nathan Benfer Sasha Zigic 23/02/2018
1 Issued to client Nathan Benfer Sasha Zigic 7/01/2018
2 Issued to Client Nathan Benfer Sasha Zigic 21/03/2018
Approval for issue
Name Signature Date
Nathan Benfer 21/03/2018
This report was prepared by [RPS Australia West Pty Ltd (‘RPS’)] within the terms of its engagement and in direct response to a scope of services. This report is strictly limited to the purpose and the facts and matters stated in it and does not apply directly or indirectly and must not be used for any other application, purpose, use or matter. In preparing the report, RPS may have relied upon information provided to it at the time by other parties. RPS accepts no responsibility as to the accuracy or completeness of information provided by those parties at the time of preparing the report. The report does not take into account any changes in information that may have occurred since the publication of the report. If the information relied upon is subsequently determined to be false, inaccurate or incomplete then it is possible that the observations and conclusions expressed in the report may have changed. RPS does not warrant the contents of this report and shall not assume any responsibility or liability for loss whatsoever to any third party caused by, related to or arising out of any use or reliance on the report howsoever. No part of this report, its attachments or appendices may be reproduced by any process without the written consent of RPS. All enquiries should be directed to RPS.
MAQ0648J | Tamarind Resources - Tui Field | Oil Spill Modelling | 21/03/18 Page II
REPORT
Contents
EXECUTIVE SUMMARY ...... 1 Background ...... 1 Methodology ...... 1 Oil Properties ...... 1 Key Findings ...... 2
1 INTRODUCTION ...... 4
2 SCOPE OF WORK ...... 6
3 REGIONAL CURRENTS ...... 6 3.2 Tidal Currents ...... 8 3.2.1 Grid Setup ...... 8 3.2.2 Tidal Conditions ...... 10 3.2.3 Surface Elevation Validation ...... 10 3.3 Ocean Currents ...... 15 3.4 Currents at the Release Site ...... 16
4 WINDS ...... 18
5 WATER TEMPERATURE AND SALINITY ...... 21
6 OIL SPILL MODEL - SIMAP ...... 23 6.1 Stochastic Modelling ...... 23 6.2 Sea surface, Shoreline and In-Water Thresholds ...... 24 6.2.1 Sea surface Exposure Thresholds ...... 24 6.2.2 Shoreline Contact Threshold ...... 25 6.2.3 Water Column Exposure Thresholds ...... 26 6.3 Exposure Calculation...... 28
7 OIL PROPERTIES ...... 30
8 MODEL SETTINGS AND ASSUMPTIONS ...... 33
9 INTERPRETING MODELLING RESULTS ...... 35 9.1 February to May Stochastic Assessment ...... 35 9.2 Receptors Assessed ...... 37
10 RESULTS: AMOKURA-2H LOSS OF WELL CONTROL ...... 38 10.1 Stochastic Analysis ...... 38 10.1.1 Sea Surface Exposure and Shoreline Contact ...... 38 10.1.2 In-Water Exposure ...... 51
11 RESULTS: AMOKURA-2H 200 M3 MDO SURFACE RELEASE ...... 58 11.1 Stochastic Analysis ...... 58
MAQ0648J | Tamarind Resources - Tui Field | Oil Spill Modelling | 21/03/18 Page III
REPORT
12 REFERENCES ...... 66
MAQ0648J | Tamarind Resources - Tui Field | Oil Spill Modelling | 21/03/18 Page IV
REPORT
Tables Table 1 Location of the wells for the study...... 4 Table 2 Statistical comparison between the observed and HYDROMAP predicted surface elevations data from the 1st to 31st January 2014...... 11 Table 3 Predicted average and maximum surface current speeds at the study site. The data was derived by combining the HYCOM ocean data and HYDROMAP tidal data for 2008-2012 (inclusive)...... 16 Table 4 Predicted average and maximum winds for the wind node nearest the release site. Data derived from CFSR hindcast model from 2003-2012 (inclusive)...... 19 Table 5 Monthly average sea-surface temperature and salinity at the study area...... 21 Table 6 The Bonn Agreement Oil Appearance Code...... 24 Table 7 Thresholds used to classify the zones of sea surface exposure...... 25 Table 8 Thresholds use to assess shoreline contact ...... 26 Table 9 Dissolved aromatic threshold values applied as part of the modelling study ...... 27 Table 10 Entrained hydrocarbon threshold values applied as part of the modelling study. Thresholds based on OSPAR guidelines...... 27 Table 11 Physical characteristics...... 30 Table 12 Boiling point ranges...... 31 Table 13 Summary of the oil spill model settings used in this assessment...... 33 Table 14 Summary of potential zones of sea surface exposure at each surface oil threshold from a 45- day loss of well control...... 39 Table 15 Summary of shoreline contact at or above 10 g/m2 across all shorelines from a 45-day loss of well control...... 39 Table 16 Summary of the potential sea surface exposure to receptors from a 45-day loss of well control...... 40 Table 17 Summary of shoreline contact to individual shoreline receptors from a 45-day loss of well control...... 41 Table 18 Probability of low, moderate and high exposure to receptors at 0–10 m, 10 - 20m and 20 - 30m below the sea surface from dissolved aromatics from a 45-day loss of well control...... 52 Table 19 Probability of low, moderate and high exposure to marine based receptors from entrained hydrocarbons at 0-10 m below the sea surface from a 45-day loss of well control...... 56 Table 20 Summary of potential zones of sea surface exposure at each surface oil threshold from a 200 m3 MDO release...... 58
Figures Figure 1 Location of the wells for the study...... 5 Figure 2 Schematic showing the oceanic current circulation surrounding New Zealand (Image source: Brodie, 1960)...... 7 Figure 3 Map showing the regions of sub-gridding for the study area ...... 9 Figure 4 Bathymetry used in the hydrodynamic grid for the study region...... 9 Figure 5 Location of the nine tide stations around New Zealand used to validate the tidal model...... 11 Figure 6 Comparison between predicted (red line) and observed (blue line) surface elevation variation at Auckland (top), Bluff (middle) and Lyttelton (bottom), between the 1st and the 31st of January 2014...... 12
MAQ0648J | Tamarind Resources - Tui Field | Oil Spill Modelling | 21/03/18 Page V
REPORT
Figure 7 Comparison between predicted (red line) and observed (blue line) surface elevation variation at Napier (top), Nelson (middle) and Picton (bottom), between the 1st and the 31st of January 2014...... 13 Figure 8 Comparison between predicted (red line) and observed (blue line) surface elevation variation at Port Taranaki (top), Wellington (middle) and Westport (bottom), between the 1st and 31st of January 2014...... 14 Figure 9 Snapshot example of the predicted HYCOM ocean surface currents in the region. Colour of individual arrows indicate current speed (m/s)...... 15 Figure 10 Predicted monthly surface current rose plots at the study site. Data was derived by combining the HYDROMAP tidal currents and HYCOM ocean currents for 2008 – 2012. The colour key shows the current speed (m/s), the compass direction provides the current direction flowing TOWARDS and the length of the wedge gives the percentage of the record for a particular speed and direction combination...... 17 Figure 11 Sample of the CFSR modelled wind data used for the oil spill model...... 19 Figure 12 Modelled monthly wind rose distributions from 2003–2012 (inclusive), for the wind node closest to the release site. The colour key shows the wind magnitude, the compass direction provides the direction FROM and the length of the wedge gives the percentage of the record for a particular speed and direction combination...... 20 Figure 13 Monthly average sea temperature and salinity profiles at the study site...... 22 Figure 14 Photographs showing the difference between oil colour and thickness on the sea surface (source: adapted from OilSpillSolutions.org 2015) ...... 25 Figure 15 Example time series plot of concentration of entrained hydrocarbons (top) at a receptor, and exposure (bottom) with no depuration (blue) and with depuration (green) ...... 29 Figure 16 Weathering of Tui Crude under three static wind conditions. The results are based on a 19,803 bbl sub-surface release of Tui Crude over 24 hours, tracked for 75 days...... 31 Figure 17 Weathering of marine diesel oil under three static wind conditions. The results are based on a 200 m3 spill of marine diesel oil released over 6 hours, tracked for 20 days...... 32 Figure 18 Plot of the variable release rate over the 45-day period ...... 34 Figure 19 Receptors assessed for oil exposure...... 37 Figure 20 Zones of potential exposure on the sea surface from a 45-day loss of well control...... 42 Figure 21 Example snaphot of surface oil exposure at a given moment during a single simulation...... 43 Figure 22 Probability of oil exposure on the sea surface above low exposure (≥0.5 g/m2) from a 45-day loss of well control...... 44 Figure 23 Probability of oil exposure on the sea surface above moderate exposure (≥10 g/m2) from a 45- day loss of well control...... 45 Figure 24 Probability of oil exposure on the sea surface above high exposure (≥25 g/m2) from a 45-day loss of well control...... 46 Figure 25 Minimum time before oil exposure on the sea surface above low exposure (≥0.5 g/m2) from a 45-day loss of well control...... 47 Figure 26 Minimum time before oil exposure on the sea surface above moderate exposure (≥10 g/m2) from a 45-day loss of well control...... 48 Figure 27 Minimum time before oil exposure on the sea surface above high exposure (≥25 g/m2) from a 45-day loss of well control...... 49 Figure 28 Maximum potential shoreline loading from a 45-day loss of well control...... 50 Figure 29 Zones of potential dissolved aromatic exposure at 0-10 m below the sea surface from a 45-day loss of well control...... 53 Figure 30 Zones of potential dissolved aromatic exposure at 10-20 m below the sea surface from a 45-day loss of well control...... 54
MAQ0648J | Tamarind Resources - Tui Field | Oil Spill Modelling | 21/03/18 Page VI
REPORT
Figure 31 Zones of potential dissolved aromatic exposure at 20-30 m below the sea surface from a 45-day loss of well control...... 55 Figure 32 Zones of potential entrained hydrocarbon exposure at 0-10 m below the sea surface from a 45- day loss of well control...... 57 Figure 33 Zones of potential exposure on the sea surface from a 200 m3 MDO release...... 59 Figure 34 Probability of oil exposure on the sea surface above low exposure (≥0.5 g/m2) from a 200 m3 MDO release...... 60 Figure 35 Probability of oil exposure on the sea surface above moderate exposure (≥10 g/m2) from a 200 m3 MDO release...... 61 Figure 36 Probability of oil exposure on the sea surface above high exposure (≥25 g/m2) from a 200 m3 MDO release...... 62 Figure 37 Minimum time before oil exposure on the sea surface above low exposure (≥0.5 g/m2) from a 200 m3 MDO release...... 63 Figure 38 Minimum time before oil exposure on the sea surface above moderate exposure (≥10 g/m2) from a 200 m3 MDO release...... 64 Figure 39 Minimum time before oil exposure on the sea surface above high exposure (≥25 g/m2) from a 200 m3 MDO release...... 65
MAQ0648J | Tamarind Resources - Tui Field | Oil Spill Modelling | 21/03/18 Page VII
REPORT
Terms and Abbreviations ° – Degrees ‘– Minutes “– Seconds AMSA – Australian Maritime Safety Authority ANZECC – Australian and New Zealand Environment and Conservation Council. API – American Petroleum Institute gravity. A measure of how heavy or light a petroleum liquid is compared to water. Bonn Agreement Oil Appearance Code – An agreement for cooperation in dealing with pollution of the North Sea by oil and other harmful substances, 1983, includes: Governments of the Kingdom of Belgium, the Kingdom of Denmark, the French Republic, the Federal Republic of Germany, the Republic of Ireland, the Kingdom of the Netherlands, the Kingdom of Norway, the Kingdom of Sweden, the United Kingdom of Great Britain and Northern Ireland and the European Union. oC – degree Celsius (unit of temperature) CFSR – Climate Forecast System Reanalysis cm – Centimeter (unit of length) Decay – The process where oil components are changed either chemically or biologically (biodegradation) to another compound. It includes breakdown to simpler organic carbon compounds by bacteria and other organisms, photo- oxidation by solar energy, and other chemical reactions. Dissolved aromatic hydrocarbons – dissolved hydrocarbons within the water column with alternating double and single bonds between carbon atoms forming rings, containing at least one six-membered benzene ring. g/m2 – Grams per square meter (unit of surface area density related to thickness) EMBA – Environmental that may be affected Entrained oil – Droplets or globules of oil that are physically mixed (but not dissolved) into the water column. Physical entrainment can occur either during pressurised release from a sub-surface location, or through the action of breaking waves (>12 knots). Evaporation – The process whereby components of the oil mixture are transferred from the sea-surface to the atmosphere. GODAE – Global Ocean Data Assimilation Experiment HYCOM – Hybrid Coordinate Ocean Model is a data-assimilative, three-dimensional ocean model. HYDROMAP – Advanced ocean/coastal tidal model used to predict tidal water levels, current speed and current direction. IOA – Index of Agreement gives a non-dimensional measure of model accuracy or performance. Isopycnal layers – Water column layers with corresponding water densities ITOPF – The International Tanker Owners Pollution Federation JPDA – Joint Petroleum Development Area km – Kilometer (unit of length) km2 – Square Kilometers (unit of area) Knots – units of wind measurement (1 knot = 0.514 m/s)
LC50 – Median lethal dose. The dose required for mortality of 50% of a tested population after a specified test duration m – Meters (unit of length) m/s – Meters per Second (unit of speed) MAE – Mean Absolute Error is the average of the absolute values of the difference between the model predicted and observed surface elevations NASA – National Aeronautics and Space Administration NCEP – National Centres for Environmental Prediction NOAA – National Oceanic and Atmospheric Administration NOEC – No observed adverse effect level Ocean current – Large scale and continuous movement of seawater generated by forces such as breaking waves, wind, the Coriolis effect, and temperature and salinity gradients. It is the main flow of ocean waters.
MAQ0648J | Tamarind Resources - Tui Field | Oil Spill Modelling | 21/03/18 Page VIII
REPORT
OECD – The Organisation for Economic Co-operation and Development PNEC – Predicted no effect concentration. ppb – parts per billion (concentration) ppb.hrs – ppb multiplied for hours (concentration x time) PSU – Practical salinity units RPS APASA – RPS Asia-Pacific Applied Science Associates Sea surface exposure – Floating oil on the sea surface equal to or above reporting threshold (e.g. 0.5 g/m2) SIMAP – Spill Impact Mapping Analysis Program Shoreline contact – Stranded oil on the shoreline equal to or above reporting threshold (e.g. 10 g/m2) SRTM30_PLUS – Global topography /bathymetry dataset THC – Total hydrocarbons USCG – United States Coastguard USEPA – United States Environmental Protection Agency Visible oil – Floating oil on the sea surface equal to or above reporting threshold (e.g. 0.5 g/m2)
MAQ0648J | Tamarind Resources - Tui Field | Oil Spill Modelling | 21/03/18 Page IX
REPORT
Executive Summary
Background ERM New Zealand, on behalf of Tamarind Resources, requested RPS to carry out quantitative oil spill modelling for their Tui, Amokura and Pateke wells in the Taranaki Basin, New Zealand. Based on the proximity of the wells to land, and the potential maximum oil release rates, Amokura-2H was chosen as the representative (and worst case) scenario for this modelling study. The study considered the fate of 356,780 STB of Tui Crude released during a hypothetical loss of well control incident over 45 days and a 200m3 surface release of marine diesel from a vessel. The assessment was completed for the months of February to May, which covers the planned operational months of March and April. The SIMAP system, the methods and analysis presented herein use modelling algorithms which have been anonymously peer reviewed and published in international journals. Further, RPS APASA warrants that this work meets and exceeds the ASTM Standard F2067-13 “Standard Practice for Development and Use of Oil Spill Models”. Note that the modelling does not take into consideration any of the spill prevention, mitigation and response capabilities that might be in place during the operations. The modelling makes no allowance for intervention following a spill to reduce volumes and/or prevent hydrocarbons from reaching sensitive areas.
Methodology The modelling study was carried out in several stages. Firstly, a ten-year current dataset (2003–2012) that includes the combined influence of ocean currents from the HYCOM model and tidal currents from the HYDROMAP model was developed. Secondly, high-resolution local winds from the CFSR model and detailed hydrocarbon characteristics were used as inputs in the three-dimensional oil spill model (SIMAP) to simulate the drift, spread, weathering and fate of the spilled oils. As spills can occur during any set of wind and current conditions, modelling was conducted using a stochastic (random or non-deterministic) approach, which involved running 100 spill simulations initiated at random start times, using the same release information (spill volume, duration and composition of the oil). This ensured that each simulation was subject to different wind and current conditions and, in turn, movement and weathering of the oil.
Oil Properties Tui crude has a density of 808 kg/m3 (API gravity of 43.5) and a dynamic viscosity of 2.5 cP at 40ºC, and although it is classified as a Group II oil according to the International Tankers Owners Pollution Federation (ITOPF, 2014) classifications, it does have a high wax content of 17.3%, which will result in this oil turning in to wax after extended periods (weeks) in the marine environment. There is greater persistence of the Tui Crude on the sea surface with decreasing wind speeds, which correlates to increasing volumes of oil occurring in the water column with increasing wind speeds. Additionally, evaporative losses were greatest during the first day, while the oil was still being released, and then the loss rate reduced until at day 4 further evaporative losses were negligible.
MAQ0648J | Tamarind Resources - Tui Field | Oil Spill Modelling | 21/03/18 Page 1
REPORT
Marine diesel oil (MDO) was used for the surface release scenario from a loss of vessel containment. MDO has an API of 37.6, density of 829 kg/m3 (at 15 ºC) and a low viscosity of 4.0 cP at 25ºC, classifying it as a Group II oil according to the International Tankers Owners Pollution Federation (ITOPF, 2014) classifications. MDO is characterised by a large mixture (95%) of high and semi- to low-volatiles and contains 5% persistent hydrocarbons. It is important to note that some heavy components contained in marine diesel oil have a strong tendency to physically entrain into the upper water column in the presence of moderate winds (i.e. >12 knots) and breaking waves but can re-float to the surface if these energies abate.
Key Findings Amokura-2H Loss of Well Control Zones of potential low exposure (0.5-10 g/m2) were predicted to extend northeast, southeast and west, while zones of moderate (10-25 g/m2) and high (>25 g/m2) exposure predominantly extended south- southeast of the release site and southeast of the release sites. The maximum distance from a release site for the potential zone of low, moderate and high exposure were 284 km (north-northeast), 62 km (south-southeast) and 51 km (south-southeast). The probability of oil coming ashore was 98%, while the quickest time to shore for visible sheens was 56 hours and the maximum volume was 706 m3. The maximum lengths of shoreline exposure at low, moderate and high levels were 295 km, 158 km and 16 km, respectively. Both the South Taranaki shoreline and Marine Mammal Sanctuary had 99% probability of being exposed to visible floating oil. Only South Taranaki and New Plymouth shorelines had any probability of being exposed to surface oil of moderate thickness (10-25 g/m2). South Taranaki regional shoreline had the highest probability of contact, for all thresholds. Shoreline accumulation of greater than 100 g/m2 occurs within 65 hours for this section of coastline, eventually covering up to 94 km at this level. Some sections of the South Taranaki regional shoreline accumulated up to 2.885 kg/m2, while New Plymouth had a peak value of 3.144 kg/m2. Dissolved aromatics leeching away from the entrained oil while in the water column resulted in zones of potential low and moderate exposure in the top 30 m of the water column. There was no exposure above thresholds of concern below 30 m in the water column. The highest exposure in the water column occurred at adjacent to the South Taranaki regional shoreline. Zones of potential entrained hydrocarbon exposure were much smaller than the dissolved aromatics. The low exposure zone was also limited to the upper 10 m of the water column. Some shoreline areas were shown to have low exposure to entrained hydrocarbons, these were likely to occur from floating oil trapped in the coastal zone being forced in to the water column during a strong wind event.
MAQ0648J | Tamarind Resources - Tui Field | Oil Spill Modelling | 21/03/18 Page 2
REPORT
Amokura-2H Vessel Loss of Containment Zones of low (0.5-10 g/m2), moderate (10-25 g/m2) and high (>25 g/m2) potential exposure were predicted to extend predominantly south- southeast of the release sites. The maximum distance from a release site for the potential zone of low, moderate and high exposure were 46 km (southeast), 19 km (southeast) and 4 km (southeast), respectively. No shoreline contact was predicted for this hypothetical scenario. No entrained exposure (equal to or greater than the minimum reporting threshold 67,200 ppb.hrs) was predicted for this scenario. No dissolved aromatic exposure (equal to or greater than the minimum reporting threshold 576 ppb.hrs) was predicted for this scenario.
MAQ0648J | Tamarind Resources - Tui Field | Oil Spill Modelling | 21/03/18 Page 3
REPORT
1 Introduction
ERM New Zealand, on behalf of Tamarind Resources, requested RPS to carry out quantitative oil spill modelling for their Tui, Amokura and Pateke wells in the Taranaki Basin, New Zealand. Based on the proximity of the wells to land, and the potential maximum oil release rates, Amokura-2H was chosen as the representative (and worst case) scenario for this modelling study (Table 1 and Figure 1). The study considered the fate of 356,780 STB of Tui Crude released during a hypothetical loss of well control incident over 45 days and a 200m3 surface release of marine diesel from a vessel. The assessment was completed for the months of February to May, which covers the planned operational months of March and April. The modelling study was carried out in several stages. Firstly, a five-year current dataset (2003–2012) that includes the combined influence of ocean currents from the HYCOM model and tidal currents from the HYDROMAP model was developed. Secondly, high-resolution local winds from the CFSR model and detailed hydrocarbon characteristics were used as inputs in the three-dimensional oil spill model (SIMAP) to simulate the drift, spread, weathering and fate of the spilled oils. As spills can occur during any set of wind and current conditions, modelling was conducted using a stochastic (random or non-deterministic) approach, which involved running 100 spill simulations initiated at random start times, using the same release information (spill volume, duration and composition of the oil). This ensured that each simulation was subject to different wind and current conditions and, in turn, movement and weathering of the oil. The SIMAP system, the methods and analysis presented herein use modelling algorithms which have been anonymously peer reviewed and published in international journals. Further, RPS APASA warrants that this work meets and exceeds the ASTM Standard F2067-13 “Standard Practice for Development and Use of Oil Spill Models”. Note that the modelling does not take into consideration any of the spill prevention, mitigation and response capabilities that might be in place during the operations. The modelling makes no allowance for intervention following a spill to reduce volumes and/or prevent hydrocarbons from reaching sensitive areas.
Table 1 Location of the wells for the study.
Release Site Latitude Longitude Depth (m)
Amokura-2H 39° 25' 23" S 173° 12' 44" E 122.5
Pateke-3H 39° 22' 51" S 173° 12' 25" E 123.3
Patke-4H 39° 22' 31" S 173° 11' 46" E 124.0
Tui-2H 39° 26' 35" S 173° 14' 11" E 121.6
Tui-3H 39° 26' 34" S 173° 14' 9" E 121.8
MAQ0648J | Tamarind Resources - Tui Field | Oil Spill Modelling | 21/03/18 Page 4
REPORT
Figure 1 Location of the wells for the study.
MAQ0648J | Tamarind Resources - Tui Field | Oil Spill Modelling | 21/03/18 Page 5
REPORT
2 Scope of Work
The scope of work included the following components: 1. Generate ten years of net currents from 2003 to 2012 (inclusive) that include the combined influence of ocean and tidal currents. 2. Use high-resolution wind data, current data and hydrocarbon characteristics as input into the 3- dimensional oil spill model, SIMAP to model the movement, spreading, entrainment, weathering and potential shoreline contact by the hydrocarbon over time; 3. Use SIMAP’s stochastic model (also known as a probability model) to calculate exposure to surround waters and shoreline. This involved running 100 randomly selected single trajectory simulations for each scenario and/or season, with each simulation having the same spill information (spill volume, duration and composition of hydrocarbons) but varying start times. This ensured that each spill trajectory was subjected to unique wind and current conditions; and
3 Regional Currents
An extensive review of the ocean circulation surrounding New Zealand, which describes the ocean currents in the region, is provided by Heath (1985). The study describes two main surface water masses which surround New Zealand, which include the Subtropical and Subantarctic Surface waters. The Subtropical waters predominately originate from an extension of the East Australian Current (EAC), which is a western boundary current that flows from the South Equatorial Current (SEC), and down the eastern coast of Australia. Typically, the EAC carries warm waters to the south, before splitting off into the Tasman Sea approximately in line with Sydney (Coleman, 1984) and carrying the warmer waters eastwards towards New Zealand (Heath, 1985). The Subantarctic Water tends to flow northwards along the eastern side of the South Island, originating from the Circumpolar Current south of New Zealand. Figure 2 presents a schematic of the regional currents of New Zealand. The oceanic currents near the Taranaki Basin are predominately influenced by the typically eastward flowing D’Urville Current and the northward flowing Westland Current. The D’Urville Current consists of warm saline water that flows eastwards into the Cook Strait from the west-northwest (Brodie, 1960; Heath, 1985). The Westland Current has been observed to flow predominately northwards along the west coast of the South Island, where it then mixes with the D’Urville near the Cook Strait. An extension of the Westland current has been observed to extend further northwards along the western coast of the North Island. Previous work by Bowman et al. (1983) indicate that strong non-tidal flows through the Cook Strait may be influenced strongly by the prevailing winds, as shown by a pair of drifters which were observed to travel from the Taranaki Bight through the Cook Strait in a south easterly direction, with strong prevailing winds from the northwest.
MAQ0648J | Tamarind Resources - Tui Field | Oil Spill Modelling | 21/03/18 Page 6
REPORT
Figure 2 Schematic showing the oceanic current circulation surrounding New Zealand (Image source: Brodie, 1960).
MAQ0648J | Tamarind Resources - Tui Field | Oil Spill Modelling | 21/03/18 Page 7
REPORT
3.2 Tidal Currents The effects of tides were generated using RPS’s advanced ocean/coastal model, HYDROMAP. The HYDROMAP model has been thoroughly tested and verified through field measurements throughout the world over more than 20 years (Isaji and Spaulding, 1984; Isaji et al., 2001; Zigic et al., 2003). In fact, HYDROMAP tidal current data has been used as input for the OILMAP hydrocarbon spill modelling system, which forms part of the Incident Management System (IMS) operated by Maritime New Zealand (MNZ). HYDROMAP employs a sophisticated sub-gridding strategy, which supports up to six levels of spatial resolution, halving the grid cell size as each level of resolution is employed. The sub-gridding allows for higher resolution of currents within areas of greater bathymetric and coastline complexity, and/or of particular interest to a study. The numerical solution methodology follows that of Davies (1977a and 1977b) with further developments for model efficiency by Owen (1980) and Gordon (1982). A more detailed presentation of the model can be found in Isaji and Spaulding (1984) and Isaji et al. (2001).
3.2.1 Grid Setup HYDROMAP was set-up to cover the domain of interest, which was subdivided horizontally into a grid with 4 levels of resolution. The resolution of the primary level was set at 8 km. The resolution of the first, second and third levels were 4 km, 2 km, 1 km respectively. The finer grids were allocated in a step-wise fashion to more accurately resolve flows along the coastline, around islands and over more complex bathymetry. Figure 3 shows a zoomed in image in the region containing the Pateke-4H release site. A combination of datasets was used to describe the shape of the seabed within the high-resolution grid. Depths for the region were extracted from nautical charts and the SRTM30_PLUS dataset (see Becker et al., 2009), which provides a 30-arc second, or approximately 1 km, resolution (see Figure 4).
MAQ0648J | Tamarind Resources - Tui Field | Oil Spill Modelling | 21/03/18 Page 8
REPORT
Figure 3 Map showing the regions of sub-gridding for the study area
Figure 4 Bathymetry used in the hydrodynamic grid for the study region.
MAQ0648J | Tamarind Resources - Tui Field | Oil Spill Modelling | 21/03/18 Page 9
REPORT
3.2.2 Tidal Conditions The ocean boundary data for the regional model was obtained from satellite measured altimetry data (TOPEX/Poseidon 7.2) which provided estimates of the eight dominant tidal constituents at a horizontal scale of approximately 0.25 degrees. The eight major tidal constituents used were K2, S2, M2, N2, K1, P1, O1 and Q1. Using the tidal data, surface heights were firstly calculated along the open boundaries, at each time step in the model. The Topex-Poseidon satellite data is produced, and quality controlled by National Aeronautics and Space Administration (NASA). The satellites, equipped with two highly accurate altimeters, capable of taking sea level measurements accurate to less than ± 5 cm, measured oceanic surface elevations (and the resultant tides) for over 13 years (1992–2005). In total these satellites carried out 62,000 orbits of the planet. The Topex-Poseidon tidal data has been widely used amongst the oceanographic community, being the subject of more than 2,100 research publications (e.g. Andersen, 1995; Ludicone et al., 1998; Matsumoto et al., 2000; Kostianoy et al., 2003; Yaremchuk and Tangdong, 2004; Qiu and Chen 2010). As such the Topex/Poseidon tidal data is considered suitably accurate for this study.
3.2.3 Surface Elevation Validation To ensure that tidal predictions were accurate, predicted surface elevations were compared to data observed at five locations situated across the study region (Figure 5). Figure 6 to Figure 8 illustrate a comparison of the predicted and observed surface elevations for each location for January 2014. As shown on the graph, the model accurately reproduced the phase and amplitudes throughout the spring and neap tidal cycles. To provide a statistical measure of the model’s performance, the Index of Agreement (IOA – Willmott, 1981) and the Mean Absolute Error (MAE – Willmott, 1982; Willmott and Matsuura, 2005) were used. The MAE is the average of the absolute values of the difference between the model-predicted (P) and observed (O) variables. It is a more natural measure of the average error and more readily understood (Willmott and Matsuura, 2005). 푁 −1 푀퐴퐸 = 푁 ∑|푃푖 − 푂푖| 푖=1
The Index of Agreement (IOA) is determined by:
2 ∑|푋푚표푑푒푙 − 푋표푏푠| 퐼푂퐴 = 1 − 2 ∑(|푋푚표푑푒푙 − 푋̅̅̅표푏푠̅̅̅| + |푋표푏푠 − 푋̅̅̅표푏푠̅̅̅|)
Where: X represents the variable being compared and the time mean of that variable. A perfect agreement exists between the model and field observations if the index gives an agreement value of 1 and complete disagreement will produce an index measure of 0 (Wilmott, 1981). Willmott et al., (1985) also suggests that values meaningfully larger than 0.5 represent good model performance. Clearly, a greater IOA and lower MAE represent a better model performance. Table 2 shows the IOA and MAE values for the selected locations.
MAQ0648J | Tamarind Resources - Tui Field | Oil Spill Modelling | 21/03/18 Page 10
REPORT
Figure 5 Location of the nine tide stations around New Zealand used to validate the tidal model.
Table 2 Statistical comparison between the observed and HYDROMAP predicted surface elevations data from the 1st to 31st January 2014.
Tide Station IOA MAE (m)
Auckland (North Island) 0.95 0.29
Bluff (South Island) 0.93 0.25
Lyttelton (South Island) 0.91 0.26
Napier (North Island) 0.98 0.11
Nelson (South Island) 0.93 0.39
Picton (South Island) 0.93 0.15
Port Taranaki (North Island) 0.94 0.33
Wellington (North Island) 0.95 0.13
Westport (South Island) 0.94 0.30
Average 0.94 0.25
MAQ0648J | Tamarind Resources - Tui Field | Oil Spill Modelling | 21/03/18 Page 11
REPORT
Figure 6 Comparison between predicted (red line) and observed (blue line) surface elevation variation at Auckland (top), Bluff (middle) and Lyttelton (bottom), between the 1st and the 31st of January 2014.
MAQ0648J | Tamarind Resources - Tui Field | Oil Spill Modelling | 21/03/18 Page 12
REPORT
Figure 7 Comparison between predicted (red line) and observed (blue line) surface elevation variation at Napier (top), Nelson (middle) and Picton (bottom), between the 1st and the 31st of January 2014.
MAQ0648J | Tamarind Resources - Tui Field | Oil Spill Modelling | 21/03/18 Page 13
REPORT
Figure 8 Comparison between predicted (red line) and observed (blue line) surface elevation variation at Port Taranaki (top), Wellington (middle) and Westport (bottom), between the 1st and 31st of January 2014.
MAQ0648J | Tamarind Resources - Tui Field | Oil Spill Modelling | 21/03/18 Page 14
REPORT
3.3 Ocean Currents Data describing the flow of ocean currents was obtained from HYCOM (Hybrid Coordinate Ocean Model, (Chassignet et al., 2007), which is operated by the HYCOM Consortium, sponsored by the Global Ocean Data Assimilation Experiment (GODAE). HYCOM is a data-assimilative, three-dimensional ocean model that is run as a hindcast (for a past period), assimilating time-varying observations of sea-surface height, sea-surface temperature and in-situ temperature and salinity measurements (Chassignet et al., 2009). The HYCOM predictions for drift currents are produced at a horizontal spatial resolution of approximately 8.25 km (1/12th of a degree) over the region, at a frequency of once per day. HYCOM uses isopycnal layers in the open, stratified ocean, but uses the layered continuity equation to make a dynamically smooth transition to a terrain following coordinate in shallow coastal regions, and to z–level coordinates in the mixed layer and/or unstratified seas. For this study, the HYCOM hindcast currents were obtained for the year 2012. Figure 9 shows an example of the modelled surface ocean currents (HYCOM) for the region.
Figure 9 Snapshot example of the predicted HYCOM ocean surface currents in the region. Colour of individual arrows indicate current speed (m/s).
MAQ0648J | Tamarind Resources - Tui Field | Oil Spill Modelling | 21/03/18 Page 15
REPORT
3.4 Currents at the Release Site Table 3 displays the average and maximum combined surface current speeds (ocean plus tides) nearby the release sites. Figure 10 shows the monthly surface current rose distributions nearby the release sites. Note the convention for defining current direction is the direction the current flows towards, which is used to reference current direction throughout this report. Each branch of the rose represents the currents flowing to that direction, with north to the top of the diagram. Sixteen directions are used. The branches are divided into segments of different colour, which represent the current speed ranges for each direction. Speed intervals of 0.1 m/s are typically used in these current roses. The length of each coloured segment is relative to the proportion of currents flowing within the corresponding speed and direction. The data showed that the surface current speeds and directions were variable between the months, though with a predominant flow towards the south-southeast and north. The maximum and average surface current speeds were 1.44 m/s 0.24 m/s, respectively.
Table 3 Predicted average and maximum surface current speeds at the study site. The data was derived by combining the HYCOM ocean data and HYDROMAP tidal data for 2008-2012 (inclusive).
Average current Maximum current Month General Direction speed (m/s) speed (m/s)
January 0.24 1.17 Northeast to Southeast
February 0.20 0.96 Northeast to Southeast
March 0.23 1.37 Northwest and Southeast
April 0.25 1.44 Southeast
May 0.26 0.91 Southeast
June 0.25 1.04 Variable
July 0.24 1.10 Northwest to Northeast
August 0.23 1.21 South
September 0.27 1.32 Southeast
October 0.25 1.00 Northeast to southeast
November 0.23 1.07 Northeast
December 0.25 0.94 Southeast
Minimum 0.20 0.91
Maximum 0.27 1.44
MAQ0648J | Tamarind Resources - Tui Field | Oil Spill Modelling | 21/03/18 Page 16
REPORT
Figure 10 Predicted monthly surface current rose plots at the study site. Data was derived by combining the HYDROMAP tidal currents and HYCOM ocean currents for 2008 – 2012. The colour key shows the current speed (m/s), the compass direction provides the current direction flowing TOWARDS and the length of the wedge gives the percentage of the record for a particular speed and direction combination.
MAQ0648J | Tamarind Resources - Tui Field | Oil Spill Modelling | 21/03/18 Page 17
REPORT
4 Winds
High resolution wind data was sourced from the National Centre for Environmental Prediction (NCEP) Climate Forecast System Reanalysis (CFSR; see Saha et al., 2010). The CFSR wind model is a fully coupled, data- assimilative hind cast model representing the interaction between the earth’s oceans, land and atmosphere. The gridded wind data output is available at ¼ of a degree resolution (~33 km) and 1-hourly time intervals.
The CFSR wind data for the years 2003–2012 (inclusive) was compiled across the model domain. Figure 11 shows an example of the wind field used as input into the oil spill model.
Table 4 shows the monthly average and maximum winds derived from the CFSR model node nearest the release site.
Figure 12 shows the monthly wind roses derived from the CFSR model node closest to the release site.
Note that the atmospheric convention for defining wind direction, that is, the direction the wind blows from, is used to reference wind direction throughout this report. Each branch of the rose represents wind coming from that direction, with north to the top of the diagram. Sixteen directions are used. The branches are divided into segments of different colour, which represent wind speed ranges from that direction. Speed ranges of 10 knots are typically used in these wind roses. The length of each segment within a branch is proportional to the frequency of winds blowing within the corresponding range of speeds from that direction.
The wind node closest to the release site demonstrated two predominant (general) directions; 1) south- westerly winds during August to February and 2) south-westerly and south-easterly during March to July.
Winds in this region are moderate to strong. Monthly average wind speeds range from 15–19 knots and the monthly maximum wind speeds range from 41–54 knots (Error! Reference source not found.). The maximum wind speed occurred during July. Note these maximums do not include any short-term wind gusts during severe storms.
MAQ0648J | Tamarind Resources - Tui Field | Oil Spill Modelling | 21/03/18 Page 18
REPORT
Figure 11 Sample of the CFSR modelled wind data used for the oil spill model.
Table 4 Predicted average and maximum winds for the wind node nearest the release site. Data derived from CFSR hindcast model from 2003-2012 (inclusive).
Month Average wind Maximum wind General Direction (knots) (knots) (From)
January 16 43 Southwest February 15 41 Southwest March 16 51 Southwest and Southeast April 16 50 Southwest and Southeast May 18 46 Southwest and Southeast June 19 53 Southwest and Southeast July 19 54 Southwest and Southeast August 17 44 Southwest September 18 42 Southwest October 18 49 Southwest November 17 45 Southwest December 16 43 Southwest Minimum 15 41 Maximum 19 54
MAQ0648J | Tamarind Resources - Tui Field | Oil Spill Modelling | 21/03/18 Page 19
REPORT
Figure 12 Modelled monthly wind rose distributions from 2003–2012 (inclusive), for the wind node closest to the release site. The colour key shows the wind magnitude, the compass direction provides the direction FROM and the length of the wedge gives the percentage of the record for a particular speed and direction combination.
MAQ0648J | Tamarind Resources - Tui Field | Oil Spill Modelling | 21/03/18 Page 20
REPORT
5 Water Temperature and Salinity
The monthly temperature and salinity profiles of the water column near the release site was obtained from the World Ocean Atlas 2013 database produced by the National Oceanographic Data Centre (National Oceanic and Atmospheric Administration) and its co-located World Data Center for Oceanography (see Levitus et al., 2013). Monthly average sea-surface temperatures near the release site were found to vary over the course of the year from a minimum of 13.0°C (August) to a maximum of 19.1°C (March) (Table 5). Monthly average salinity of the upper water column near the release site varied only slightly throughout the year from a minimum of 35.0 PSU (January and March) to a maximum of 35.3 PSU (July) (Table 5). To accurately represent the sea temperature and salinity throughout the whole water column the modelling used monthly average sea temperature and salinity profiles, as presented in Figure 13.
Table 5 Monthly average sea-surface temperature and salinity at the study area.
Month Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec
Temperature (°C) 17.8 19.1 19.1 17.6 16.2 14.9 14.4 13.0 13.6 13.9 14.9 16.1
Salinity (PSU) 35.0 35.1 35.0 35.2 35.2 35.2 35.3 35.2 35.1 35.1 35.1 35.2
MAQ0648J | Tamarind Resources - Tui Field | Oil Spill Modelling | 21/03/18 Page 21
REPORT
Figure 13 Monthly average sea temperature and salinity profiles at the study site.
MAQ0648J | Tamarind Resources - Tui Field | Oil Spill Modelling | 21/03/18 Page 22
REPORT
6 Oil Spill Model - SIMAP
The oil spill modelling was performed using SIMAP. SIMAP is designed to simulate the fate and effects of spilled hydrocarbons for both the surface and subsurface releases (Spaulding et al., 1994; French et al., 1999; French-McCay, 2003; French-McCay, 2004; French-McCay et al., 2004; Spaulding, et al., 1994).
The SIMAP model calculates two components: (i) the transport, spreading, entrainment, evaporation and decay of surface oil slicks and, (ii) the entrained and dissolved hydrocarbons released from the slicks into the water column. Input specifications for oil-types include the density, viscosity, pour point, distillation curve (volume lost versus temperature) and the aromatic/aliphatic component ratios within given boiling point ranges.
The SIMAP trajectory model separately calculates the movement of the material that: (i) is on the water surface (as surface slicks), (ii) in the water column (as either entrained whole oil droplets or dissolved hydrocarbon), (iii) has stranded on shorelines, or (iv) that has precipitated out of the water column onto the seabed. The model calculates the transport of surface slicks from the combined forces exerted by surface currents and wind acting on the oil. Transport of entrained oil (oil that is below the water surface) is calculated using the currents only.
6.1 Stochastic Modelling As spills can occur during any set of wind and current conditions, SIMAP’s stochastic model was used to quantify the probability of exposure to the sea surface, in-water and shoreline contacts for a hypothetical spill scenario over a 10-year period. For this assessment, a total of 100 single spill trajectories were run for each hypothetical scenario. Each simulation had the same spill information (i.e. spill volume, duration and oil type) for each scenario but with varying start times, and in turn, the prevailing wind and current conditions. This approach ensures that the predicted transport and weathering of an oil slick is subject to a wide range of current and wind conditions. During each spill trajectory, the model records the grid cells exposed to hydrocarbons, as well as the time elapsed. Once all the spill trajectories have been run, the model then combines the results from the individual simulations to determine the following: Maximum exposure (or load) observed on the sea surface; Minimum time before sea surface exposure; Probability of contact to any shorelines; Probability of contact to individual sections of shorelines; Maximum volume of oil that may contact shorelines from a single simulation; Maximum load that an individual shoreline may experience; Maximum exposure from entrained hydrocarbons observed in the water column; and Maximum exposure from dissolved aromatic hydrocarbons observed in the water column.
The stochastic model output does not represent the extent of any one spill trajectory (which would be significantly smaller) but rather provides a summary of all trajectories run for the scenario.
MAQ0648J | Tamarind Resources - Tui Field | Oil Spill Modelling | 21/03/18 Page 23
REPORT
6.2 Sea surface, Shoreline and In-Water Thresholds
6.2.1 Sea surface Exposure Thresholds The SIMAP model is able to track hydrocarbons to levels lower than biologically significant or visible to the naked eye. Therefore, reporting thresholds have been specified (based on the scientific literature) to account for “exposure” on the sea surface and “contact” to shorelines at meaningful levels.
To better assess the potential for sea surface exposure, each of the 100 spill trajectories was tracked to a minimum of 0.5 g/m2, which equates approximately to an average thickness of ~0.5 μm. Oil of this thickness is described as a silvery to rainbow sheen in appearance, according to the Bonn Agreement Oil Appearance Code (Bonn Agreement, 2009) (refer to Table 6) and is also considered the practical limit of observing oil in the marine environment (AMSA, 2012). This threshold is considered below levels which would cause environmental harm and it is more indicative of the areas perceived to be affected due to its visibility on the sea surface and potential to trigger temporary closures of areas (i.e. fishing grounds) as a precautionary measure. Hence, the 0.5 g/m2 threshold has been selected to define the zone of potential low exposure on the sea surface. Ecological impact has been estimated to occur at 10 g/m2 (~10 µm) according to French et al. (1996) and French-McCay (2009) (see references therein) as this level of oiling has been observed to mortally impact birds and other wildlife associated with the water surface. The 10 g/m2 threshold has been selected to define the zone of potential moderate exposure on the sea surface.
Scholten et al. (1996) and Koops et al. (2004) indicated that a concentration of surface oil equal to 25 g/m2 or greater would be harmful for all birds that contact the slick. Exposure to oil concentrations at or above this threshold is used to define the zone of potential high exposure.
Table 7 defines the thresholds used to classify the zones of sea surface exposure. Figure 14 shows photographs highlighting the difference in appearance between a silvery sheen, rainbow sheen and metallic sheen.
Table 6 The Bonn Agreement Oil Appearance Code.
Layer Thickness Interval Code Description Appearance Litres per km2 (g/m2 or μm)
1 Sheen (silvery/grey) 0.04 – 0.30 40 – 300
2 Rainbow 0.30 – 5.0 300 – 5,000
3 Metallic 5.0 – 50 5,000 – 50,000
4 Discontinuous True Oil Colour 50 – 200 50,000 – 200,000
5 Continuous True Oil Colour 200 –> 200,000 –>
MAQ0648J | Tamarind Resources - Tui Field | Oil Spill Modelling | 21/03/18 Page 24
REPORT
Figure 14 Photographs showing the difference between oil colour and thickness on the sea surface (source: adapted from OilSpillSolutions.org 2015)
Table 7 Thresholds used to classify the zones of sea surface exposure.
Oil concentration Zone description (g/m2)
0.5 - 10 Low
10 - 25 Moderate
> 25 High
6.2.2 Shoreline Contact Threshold
There are many different types of shorelines, ranging from cliffs, rocky beaches, sandy beaches, mud flats and mangroves, and each of these influences the volume of oil that can remain stranded ashore and its thickness before the shoreline saturation point occurs. For instance, a sandy beach may allow oil to percolate through the sand, thus increasing its ability to hold more oil ashore over tidal cycles and various wave actions than an equivalent area of water; hence oil can increase in thickness onshore over time. A sandy beach shoreline was assumed as the default shoreline type for the modelling herein, as it allows for the highest carrying capacity of oil (of the available open/exposed shoreline types). Hence the results contained herein would be indicative of a worst-case scenario, where the highest volume of oil may be stranded on the shoreline (when compared to other shoreline types, such as exposed rocky shores). French et al. (1996) and French-McCay (2009) have defined an oil exposure threshold for shorebirds and wildlife (furbearing aquatic mammals and marine reptiles) on or along the shore at 100 g/m2, which is based on studies for sub-lethal and lethal impacts. These thresholds have been used in previous environmental risk assessment studies (see French-McCay, 2003; French-McCay et al., 2004; French-McCay et al., 2011; NOAA, 2013). The 100 g/m2 threshold is also recommended in the Australian Maritime Safety Authority’s (AMSA) foreshore assessment guide1 as the acceptable minimum thickness that does not inhibit the potential for recovery and is best remediated by natural coastal processes alone (AMSA, 2007). The 100 g/m2 threshold has been selected to define the zone of potential moderate contact on the shorelines.
1 Recommended for shoreline types including sandy beach, boulder shorelines, pebble shorelines, rock platforms and industry facility structures.
MAQ0648J | Tamarind Resources - Tui Field | Oil Spill Modelling | 21/03/18 Page 25
REPORT
Observations by Lin and Mendelssohn (1996), demonstrated that loadings of more than 1,000 g/m2 of oil during the growing season would be required to impact marsh plants significantly. Similar thresholds have been found in studies assessing oil impacts on mangroves (Grant et al., 1993; Suprayogi and Murray, 1999). The 1,000 g/m2 threshold has been selected to define the zone of potential high contact on the shorelines. Oil contact between 10 and 100 g/m2 represents the socio-economic (or low contact) threshold. The following thresholds (see Table 8) have therefore been derived to classify the shoreline contact.
Table 8 Thresholds use to assess shoreline contact
Shoreline concentration Zone description (g/m2)
10–100 Low
100-1,000 Moderate
> 1,000 High
6.2.3 Water Column Exposure Thresholds Sub-surface exposure to submerged habitats is better represented by estimates for entrained or dissolved hydrocarbons in the water column. Studies indicate that the dissolved aromatic compounds (typically the mono-aromatic hydrocarbons and the two and three ring poly-aromatic hydrocarbons) are commonly the largest contributor to the toxicity of solutions generated by mixing oil into water (Di Toro et al., 2007). The exposure level (threshold concentration over a given duration) was used to assess the potential for exposure to sub-sea habitats and species by entrained and dissolved aromatic hydrocarbons. The threshold value for species toxicity in the water column is based on global data from French et al. (1999) and French-McCay (2002, 2003), which showed that species sensitivity (fish and invertebrates) to dissolved aromatics exposure > 4 days (96-hour LC50) under different environmental conditions varied from 6 to 400 μg/l (ppb) with an average of 50 ppb. This range covered 95% of aquatic organisms tested, which included species during sensitive life stages (eggs and larvae). Based on scientific literature, a minimum threshold of 6 parts per billion (ppb) over 96-hours or equivalent was used to assess in-water low exposure zones (Engelhardt, 1983; Clark, 1984; Geraci and St. Aubin, 1988; Jenssen, 1994; Tsvetnenko, 1998). French-McCay, 2002 indicates that an average 96-hour LC50 of 50 ppb and 400 ppb could serve as an acute lethal threshold to 5% and 50% to biota, respectively. Hence, the thresholds were used to represent the moderate and high exposure zones, respectively. Given that the dissolved aromatics component of hydrocarbons in the water column are accounted for by the thresholds defined above, the environmental effects of the remaining undissolved hydrocarbons, essentially the entrained hydrocarbons in the water column, require different exposure thresholds. Considering that entrained oil has undergone processes analogous to weathering and/or water-washing (i.e., many of the toxic soluble hydrocarbons have been removed through evaporation and/or dissolution), its toxicity is representative of true ‘dispersed oil’ phase impacts. OSPAR (2012) has published predicted no effect concentrations (PNEC) for ‘dispersed oil’ in produced formation water (PFW) discharges. Dispersed oil in PFW discharges are small, discrete droplets suspended in the discharged water which are very similar to insoluble dispersed oil droplets formed from subsea blowouts. In essence, the oil has been partitioned
MAQ0648J | Tamarind Resources - Tui Field | Oil Spill Modelling | 21/03/18 Page 26
REPORT
(naturally separated) from gas/oil/water mixture by solubility (water washing) and vapour pressure (evaporation) based on the individual hydrocarbon chemical properties. The OSPAR PNEC for PFW is 70 ppb for protection of 95% of species, based on biomarker testing (i.e. whole organism responses) to total hydrocarbons (THC) by Smit et al., 2009. This PNEC represents an acceptable long term chronic exposure level from continuous point source discharges in the North Sea, which is one of the most concentrated areas in the world for oil and gas production. Appropriate threshold values can be extrapolated from the NOECs examined in Smit et al., 2009 based on effects ranging from oxidative stress to impacts on growth, reproduction and survival and are represented by: 7 µg/l (7ppb) (for 1% affected fraction of species), 70.5µg/l (70ppb) (for 5% affected fraction of species) and 804 µg/l (804 ppb) (for 50% affected fraction of species). Utilising methodologies contained in ANZECC (2000), which is based upon USEPA Guidelines, PNECs can be back-calculated to determine LC50 values by applying a factor of 100 to the PNEC values. This approach is supported by assessment factor criteria contained within the European Chemicals Agency (2008) and the OECD Existing Chemicals Programme 2002 (OECD, 2002). Employing these criteria, the following conservative threshold values for entrained hydrocarbons are applied:
LC50 (99% species protection): 700 µg/l (ppb)
LC50 (95% species protection): 7,050 µg/l (ppb); and
LC50 (50% species protection): 80,400 µg/l (ppb). The final exposure thresholds adopt a continuous exposure criterion of 96 hours. Table 9 and Table 10 provide a summary of the dissolved aromatic and entrained hydrocarbon threshold values used to define different levels of potential exposure in the modelling study.
Table 9 Dissolved aromatic threshold values applied as part of the modelling study
Threshold value for dissolved Equivalent exposure of Range of sensitive species Potential aromatic concentrations for a dissolved aromatics potentially impacted from level of 96-hour LC50 (ppb) over 96 hrs (ppb.hrs) acute exposure exposure
6 576 Very sensitive species Low
50 4,800 Average sensitive species Moderate
400 38,400 Tolerant sensitive species High
Table 10 Entrained hydrocarbon threshold values applied as part of the modelling study. Thresholds based on OSPAR guidelines
Threshold value for entrained Equivalent exposure of Range of sensitive species Potential hydrocarbon concentrations entrained hydrocarbons potentially impacted from level of for a 96-hour LC50 (ppb) over 96 hrs (ppb.hrs) acute exposure exposure
700 67,200 Very sensitive species Low
7,050 676,800 Average sensitive species Moderate
80,400 7,718,400 Tolerant sensitive species High
MAQ0648J | Tamarind Resources - Tui Field | Oil Spill Modelling | 21/03/18 Page 27
REPORT
6.3 Exposure Calculation The thresholds used for the in-water concentrations of entrained hydrocarbons were described as an exposure over time, rather than an instantaneous peak value. The exposure is expressed as a concentration multiplied by the number of hours exposed at that concentration, in the units of parts per billion multiplied by hours (ppb.hrs). There are two important and opposing mechanisms that will affect the cumulative exposure. The rate of uptake, due to the exposure concentration and the duration of exposure, and the rate of removal due to the ability of the organism to expel or metabolise hydrocarbons; a process referred to as depuration. Calculation for these natural removal processes are important so as to avoid falsely forecasting impacts by only allowing for the uptake of hydrocarbons, particularly over long duration release simulations. The uptake of entrained hydrocarbons was calculated over time for each model grid cell by addition of the concentrations calculated at each subsequent time step, multiplied by the time interval (typically hourly). Depuration was calculated by applying an exponential decay function to the previously accumulated exposure. A review of the literature describing the observed rates of depuration of hydrocarbons indicates that the reduction of concentration follows an exponential decay. For sub-lethal concentrations, depuration rates will be faster with increased concentration and then decrease as concentrations approach zero. Hence, depuration of the concentrations in a cell over each time step was calculated by applying an exponential decay function to the tissue concentration calculated by uptake. Observed rates of depuration show significant variation for different soluble hydrocarbons and different organisms, varying from a few days to a few weeks (Solbakken et al., 1984). For this study, the decay coefficient was set so that the exposure would fall to 1% of an initial concentration over 1 week, given no further exposure. Cumulative exposures at each time step were then compared to threshold exposures and any location where the exposure thresholds were ever exceeded during any simulation was mapped. To illustrate the effect of allowing for depuration of hydrocarbons over time, an example time-series plot of concentration and exposure at a receptor location is presented in Figure 15. The time-series of concentration shows intermittent contacts to hydrocarbons with gaps between contacts, of the order of ~5 days. Such an outcome might be expected, for example, from variation in the position of hydrocarbon plumes resulting from variations in the current field during an ongoing discharge. The lower panel shows the calculated exposure if depuration is not considered (blue line) and the exposure where an exponential depuration rate is allowed for, assuming a time-scale of 7 days for tissue concentrations to reduce to 1% of a starting concentration. In the case where depuration was ignored (blue line), the calculated exposure simply increases with addition of each hydrocarbon concentration and remains constant during times where there is no contact to hydrocarbons. In the case where depuration is allowed for (green line), exposure decreases exponentially between exposures. The maximum exposure occurs around day 50, where there was a prolonged (~5 days) exposure to a high concentration. The threshold exposure is not exceeded by the intermittent exposures to low concentrations but is exceeded at around day 48 when higher concentrations occurred for a longer duration.
MAQ0648J | Tamarind Resources - Tui Field | Oil Spill Modelling | 21/03/18 Page 28
REPORT
Figure 15 Example time series plot of concentration of entrained hydrocarbons (top) at a receptor, and exposure (bottom) with no depuration (blue) and with depuration (green)
MAQ0648J | Tamarind Resources - Tui Field | Oil Spill Modelling | 21/03/18 Page 29
REPORT
7 Oil Properties
For this oil spill modelling assessment Tui Crude was used as the oil type for the loss of well control scenario. Tui crude has a density of 808 kg/m3 (API gravity of 43.5) and a dynamic viscosity of 2.5 cP at 40ºC, and although it is classified as a Group II oil according to the International Tankers Owners Pollution Federation (ITOPF, 2014) classifications, it does have a high wax content of 17.3%, which will result in this oil turning in to wax after extended periods (weeks) in the marine environment. Figure 16 illustrates the weathering graphs a 19,803 bbl sub-surface release of Tui crude over 24 hours, tracked for 75 under 3 static wind conditions. The graphs illustrate greater persistence of the oil on the sea surface with decreasing wind speeds, which correlates to increasing volumes of oil occurring in the water column with increasing wind speeds. Additionally, evaporative losses were greatest during the first day, while the oil was still being released, and then the loss rate reduced until at day 4 evaporative losses were negligible. Marine diesel oil (MDO) was used for the surface release scenario from a loss of vessel containment. MDO has an API of 37.6, density of 829 kg/m3 (at 15 ºC) and a low viscosity of 4.0 cP at 25ºC, classifying it as a Group II oil according to the International Tankers Owners Pollution Federation (ITOPF, 2014) classifications. MDO is characterised by a large mixture (95%) of high and semi- to low-volatiles and contains 5% persistent hydrocarbons. It is important to note that some heavy components contained in marine diesel oil have a strong tendency to physically entrain into the upper water column in the presence of moderate winds (i.e. >12 knots) and breaking waves but can re-float to the surface if these energies abate. Figure 17 illustrates the weathering graphs of 200 m3 surface release of MDO under 3 static wind conditions. The graphs illustrate greater persistence of MDO on the sea surface with decreasing wind speeds, which correlates to increasing volumes of MDO occurring in the water column with increasing wind speeds. Additionally, evaporative losses over the 20-day period were greatest during the 5 knot wind conditions when the occurrence of MDO on the sea surface was greatest. Table 11 and Table 12 show the physical characteristics and boiling point ranges for the oil types used in this study.
Table 11 Physical characteristics.
Characteristic Tui Crude Marine Diesel Oil
Density (kg/m3) 808 @ 16˚C 829.1 at 25oC API 43.5 37.60 Dynamic viscosity (cP) 2.5 @ 40°C 4.0 at 25oC Pour Point (ºC) 24 -14 Oil Property Category Group II Group II
MAQ0648J | Tamarind Resources - Tui Field | Oil Spill Modelling | 21/03/18 Page 30
REPORT
Table 12 Boiling point ranges.
Volatiles Semi-volatiles Low volatiles Residual Characteristic (%) (%) (%) (%)
Boiling point (°C) <180 180 – 265 265 – 380 >380 Tui Crude 28 17 26 29 Marine Diesel Oil 6.0 34.6 54.4 5.0 Non-persistent Persistent
Figure 16 Weathering of Tui Crude under three static wind conditions. The results are based on a 19,803 bbl sub-surface release of Tui Crude over 24 hours, tracked for 75 days.
MAQ0648J | Tamarind Resources - Tui Field | Oil Spill Modelling | 21/03/18 Page 31
REPORT
Figure 17 Weathering of marine diesel oil under three static wind conditions. The results are based on a 200 m3 spill of marine diesel oil released over 6 hours, tracked for 20 days.
MAQ0648J | Tamarind Resources - Tui Field | Oil Spill Modelling | 21/03/18 Page 32
REPORT
8 Model Settings and Assumptions
The oil spill modelling study quantified the potential exposure to the surrounding waters and shorelines from two hypothetical, yet plausible, scenarios:
Subsea loss of well control for 45 days at Amokura-2H Surface loss of containment from a vessel The assessment was completed for the months of February to May, which covers the planned operational months of March and April. Table 13 provides a summary of the oil spill model settings and assumptions
Table 13 Summary of the oil spill model settings used in this assessment.
Parameter Sub-surface loss of well Loss of vessel control at Amokura-2H containment
Number of randomly selected 100 100 spill start times
Model Period Operational Period (February to May)
Oil Type Tui Crude Marine Diesel Oil (MDO)
Spill Volume 356,780 STB 200 m3 (56,721 m3)
Release Rate Variable (see Figure 18) (Day 1 – 19,803 bbl/day and Day 45 NA 5,509 bbl/day)
Release Duration 45 days 6 hours
Release Depth (m) 122.5 Surface
Simulation length (days) 75 20
Surface oil concentration 0.5, 10 and 25 thresholds (g/m2)
Shoreline load threshold (g/m2) 10, 100 and 1,000
Dissolved aromatic dosages to 576 (6 ppb x 96 hrs, potential low exposure) assess the potential exposure 4,800 (50 ppb x 96 hrs, potential moderate exposure) (ppb.hrs) 38,400 (400 ppb x 96 hrs, potential high exposure)
Entrained oil dosages to assess 67,200 (700 ppb x 96 hrs, potential low exposure) the potential exposure (ppb.hrs) 676,800 (7,050 ppb x 96 hrs, potential moderate exposure) 7,718,400 (80,400 ppb x 96 hrs, potential high exposure)
MAQ0648J | Tamarind Resources - Tui Field | Oil Spill Modelling | 21/03/18 Page 33
REPORT
Figure 18 Plot of the variable release rate over the 45-day period
MAQ0648J | Tamarind Resources - Tui Field | Oil Spill Modelling | 21/03/18 Page 34
REPORT
9 Interpreting Modelling Results
The results from the modelling study are presented in a number of tables and figures, which aim to provide an understanding of both the predicted sea surface exposure, shoreline contact and in-water exposure for the scenario.
9.1 February to May Stochastic Assessment The figures are based on the following principles: The potential zones of exposure (surface oil, entrained hydrocarbons and dissolved aromatics) – is determined by identifying the maximum loading (surface) or dosage (subsea) within a grid cell, and is then classified according to identified surface or subsea thresholds. The minimum time before oil exposure on the sea surface – is determined by recording the elapsed time before sea surface exposure to a grid cell, at a specified threshold. The probability of exposure/contact (surface oil, shoreline oil, entrained hydrocarbon or dissolved aromatic) – is calculated by dividing the number of spill trajectories passing over that given cell (surface, shoreline or subsea) by the total number of spill trajectories, above the specified threshold value. Maximum potential shoreline loading – is determined by identifying the maximum loading within a shoreline cell, and is then classified according to the identified thresholds (i.e. 100 g/m2 and 1,000 g/m2). The statistics are based on the following principles: The greatest distance travelled by a spill trajectory – is determined by: a) recording the maximum distance travelled by a single trajectory, within a scenario, from the release site to the identified exposure thresholds; and then b) report the greatest distance travelled by the 99th percentile spill trajectory (or second highest distance travelled by a single spill trajectory), along with the corresponding direction of travel from the release site. The probability of shoreline contact – is determined by recording to the number of spill trajectories to contact the shoreline, at a specific threshold, divided by the total number of spill trajectories within that scenario. The minimum time before oil exposure – is determined by recording the minimum time for a grid cell to record exposure, at a specific threshold. The average volume of oil ashore for a single spill – is determined by calculating the average volume of the all the single spill trajectories which were predicted to make shoreline contact within a scenario. The maximum volume of oil ashore from a single spill trajectory – is determined by identifying the single spill trajectory within a scenario/season, that recorded the maximum volume of oil to come ashore and presenting that value. The average length of shoreline contacted by oil - is determined by calculating the average of the length of shoreline (measured as grid cells) contacted by oil above a specified threshold. The maximum length of shoreline contacted by oil - is determined by recording the maximum length of shoreline (measured as grid cells) contacted by oil above a specified threshold. The probability of oil exposure to a receptor – is determined by recording the number of spill trajectories
MAQ0648J | Tamarind Resources - Tui Field | Oil Spill Modelling | 21/03/18 Page 35
REPORT
to reach a specified sea surface or subsea threshold within a receptor polygon, divided by the total number of spill trajectories within that scenario. The minimum time before oil exposure to a receptor– is determined by ranking the elapsed time before sea surface exposure, at a specified threshold, to grid cells within a receptor polygon and recording the minimum value. The probability of oil contact to a receptor– is determined by recording the number of spill trajectories to reach a specified shoreline contact threshold within a receptor polygon, divided by the total number of spill trajectories within that scenario. The minimum time before shoreline contact to a receptor – is determined by ranking the elapsed time before shoreline contact, at a specified threshold, to grid cells within a receptor polygon and recording the minimum value. The average potential oil loading within a receptor – is determined taking the average of the maximum loading to any grid cell within a polygon, for all simulations within a scenario/season, that recorded shoreline The maximum potential oil loading within a receptor – is determined by identifying the maximum loading to any grid cell within a receptor polygon, for a scenario. The average volume of oil ashore within a receptor – is determined by calculating the average volume of oil to come ashore within a receptor polygon, from all the single spill trajectories which were predicted to make shoreline contact within a scenario. The maximum volume of oil ashore within a receptor – is determined by recording the maximum volume of oil to come ashore within a receptor polygon, from all the single spill trajectories which were predicted to make shoreline contact within a scenario. The average length of shoreline contacted within a receptor is determined by calculating the average of the length of shoreline (measured as grid cells) contacted by oil within a receptor polygon, at a specified threshold, from all the single spill trajectories which were predicted to make shoreline contact within a scenario. The maximum length of shoreline contacted by oil is determined by recording the maximum length of shoreline (measured as grid cells) contacted by oil within a receptor polygon, at a specified threshold, from all the single spill trajectories which were predicted to make shoreline contact within a scenario
MAQ0648J | Tamarind Resources - Tui Field | Oil Spill Modelling | 21/03/18 Page 36
REPORT
9.2 Receptors Assessed
Figure 19 shows the Marine Reserves, Marine Mammal Sanctuaries and Regional Shorelines used in the modelling to assess for oil exposure.
Figure 19 Receptors assessed for oil exposure.
MAQ0648J | Tamarind Resources - Tui Field | Oil Spill Modelling | 21/03/18 Page 37
REPORT
10 Results: Amokura-2H Loss of Well Control
This scenario examined a hypothetical 45-day subsea release of 356,780 bbl of Tui Crude at a variable rate (see Figure 18), following a loss of well control incident at the Amokura-2H well. A total of 100 spill trajectories were simulated over the defined period of February to May.
10.1 Stochastic Analysis Section 10.1.1 presents the potential exposure to the sea surface and shoreline contact, while Section 10.1.2 presents potential in-water exposure. For the modelling study each spill trajectory was tracked to the following minimum thresholds: Visible sea surface oil – 0.5 g/m2 Shoreline oil contact – 10 g/m2 Dissolved aromatics – 576 ppb.hrs Entrained hydrocarbons – 67,200 ppb.hrs
10.1.1 Sea Surface Exposure and Shoreline Contact Figure 20 presents the zones of potential oil exposure on the sea surface for the modelled period. Zones of potential low exposure (0.5-10 g/m2) were predicted to extend northeast, southeast and west, while zones of moderate (10-25 g/m2) and high (>25 g/m2) exposure predominantly extended south-southeast of the release site and southeast of the release sites. Table 14 details the maximum distance travelled by oil on the sea surface at each threshold. The maximum distance from a release site for the potential zone of low, moderate and high exposure were 284 km (north- northeast), 62 km (south-southeast) and 51 km (south-southeast). Figure 21 provides a snapshot of what the sea surface oil may look like at a given moment during a simulation, rather than a summary of the entire sea surface contacted during 100 simulations shown in Figure 20. Shoreline contact for any shoreline is summarized in Table 15. The probability of oil coming ashore was 98%, while the quickest time to shore for visible sheens was 56 hours and the maximum volume was 706 m3. The maximum lengths of shoreline exposure at low, moderate and high levels were 295 km, 158 km and 16 km, respectively. Figure 22 to Figure 24 presents the probability of oil exposure on the sea surface reported at the low, moderate and high threshold for the modelling assessment period. Additionally, Figure 25 to Figure 27 present the minimum time before observing oil on the sea surface at the low, moderate and high thresholds. Table 16 summarises potential sea surface exposure to receptors during the model assessment period. Both the South Taranaki shoreline and Marine Mammal Sanctuary had 99% probability of being exposed to visible floating oil. However, only South Taranaki and New Plymouth shorelines had any probability of being exposed to surface oil of moderate thickness (10-25 g/m2). Table 17 presents the potential shoreline contact for individual regional shorelines. Again, South Taranaki regional shoreline had the highest probability of contact, for all thresholds. Shoreline accumulation of greater than 100 g/m2 occurs within 65 hours for this section of coastline, eventually covering up to 94 km at this level.
MAQ0648J | Tamarind Resources - Tui Field | Oil Spill Modelling | 21/03/18 Page 38
REPORT
Some sections of the South Taranaki regional shoreline accumulated up to 2.885 kg/m2, while New Plymouth had a peak value of 3.144 kg/m2. Figure 28 present the maximum potential shoreline loading at the low, moderate and high thresholds.
Table 14 Summary of potential zones of sea surface exposure at each surface oil threshold from a 45-day loss of well control.
Zones of potential sea surface exposure Period Distance and direction Low Moderate High (0.5–10 g/m2) (10–25 g/m2) (>25 g/m2)
Max. distance from release site (km) 284 62 51
Max. distance from release site (km) February - 238 57 43 May (99th percentile)
Direction North-Northeast South-Southeast South-Southeast
Table 15 Summary of shoreline contact at or above 10 g/m2 across all shorelines from a 45-day loss of well control.
Shoreline statistics February – May Period
Probability of contact to any shoreline (%) 98
Absolute minimum time for visible oil to shore (hours) 56
Maximum volume of hydrocarbons ashore (m3) 706
Average volume of hydrocarbons ashore (m3) 193
Maximum length of the shoreline at 10 g/m2 (km) 295
Average shoreline length (km) at 10 g/m2 (km) 126
Maximum length of the shoreline at 100 g/m2 (km) 158
Average shoreline length (km) at 100 g/m2 63
Maximum length of the shoreline at 1,000 g/m2 16
Average shoreline length (km) at 1,000 g/m2 (km) 5
MAQ0648J | Tamarind Resources - Tui Field | Oil Spill Modelling | 21/03/18 Page 39
REPORT
Table 16 Summary of the potential sea surface exposure to receptors from a 45-day loss of well control.
Probability of oil exposure on Minimum time before oil the sea surface exposure on the sea surface (%) (hours) Receptors Low Moderate High Low Moderate High (0.5–10 (10–25 (>25 (0.5–10 (10–25 (>25 g/m2) g/m2) g/m2) g/m2) g/m2) g/m2)
Mana Island 3 0 0 538 - -
Kapiti Island 13 0 0 325 - -
Wanganui 62 0 0 174 - -
Marlborough 1 0 0 545 - -
South Taranaki 99 2 0 56 212 -
New Plymouth 81 1 0 92 254 -
Waitomo 33 0 0 202 - -
Otorohanga 1 0 0 288 - -
Waikato 4 0 0 294 - -
Wellington 2 0 0 956 - -
Kapiti Coast 23 0 0 246 - -
Horowhenua 43 0 0 253 - -
Manawatu 31 0 0 252 - -
Rangitikei 54 0 0 185 - -
Porirua 2 0 0 995 - -
Franklin 5 0 0 390 - -
Parininihi Marine Reserve 16 0 0 194 - -
Kapiti Marine Reserve 11 0 0 266 - -
Tapuae Marine Reserve 49 0 0 153 - -
Marine Mammal Sanctuary - Taranaki 99 0 0 35 - -
Marine Mammal Sanctuary - Waikato 69 0 0 142 - -
Marine Mammal Sanctuary - Auckland 7 0 0 371 - -
MAQ0648J | Tamarind Resources - Tui Field | Oil Spill Modelling | 21/03/18 Page 40
REPORT
Table 17 Summary of shoreline contact to individual shoreline receptors from a 45-day loss of well control.
Maximum probability Minimum time before Load on Mean length of Maximum length of Minimum Volume on of shoreline loading shoreline shoreline shoreline contacted shoreline contacted time before shoreline (m3) (%) accumulation (hours) (g/m2) (km) (km) visible sea surface >10 >100 >1,000 >10 >100 >1,000 >10 >100 >1,000 >10 >100 >1,000 exposure Shoreline Mean Peak Mean Peak Receptor g/m2 g/m2 g/m2 g/m2 g/m2 g/m2 g/m2 g/m2 g/m2 g/m2 g/m2 g/m2 (day) Mana Island 2 2 0 539 585 - 174 309 3 5 2 2 0 2 2 0 22 Kapiti Island 12 9 0 335 410 - 108 786 8 40 5 4 0 13 9 0 14 Wanganui 60 43 0 176 188 - 76 532 9 32 10 5 0 34 13 0 7 Marlborough 1 0 0 553 - - 34 50 1 1 3 0 0 3 0 0 23 South Taranaki 98 95 23 59 65 156 163 2,885 152 590 79 44 4 150 94 16 2 New Plymouth 78 70 4 99 147 253 118 3,144 37 195 23 12 2 94 54 3 4 Waitomo 27 19 3 199 253 1,110 106 2,074 16 115 13 5 1 52 20 2 8 Otorohanga 1 1 0 291 333 - 52 164 7 7 12 1 0 12 1 0 12 Waikato 4 1 0 296 596 - 58 166 3 6 5 3 0 6 3 0 12 Wellington 1 1 0 848 904 - 78 114 3 3 3 1 0 3 1 0 40 Kapiti Coast 21 17 1 253 308 379 124 1,141 27 131 15 10 1 41 24 1 10 Horowhenua 43 33 6 252 272 386 122 1,826 27 163 15 8 3 42 33 4 11 Manawatu 32 11 0 240 255 - 74 486 7 28 7 6 0 14 12 0 11 Rangitikei 53 43 0 184 266 - 96 679 12 87 10 5 0 25 18 0 8 Porirua 2 2 0 866 1009 - 103 321 14 22 10 5 0 13 8 0 41 Franklin 5 5 0 395 513 - 85 559 20 30 21 5 0 25 9 0 5
MAQ0648J | Tamarind Resources - Tui Field | Oil Spill Modelling | 21/03/18 Page 41
REPORT
Figure 20 Zones of potential exposure on the sea surface from a 45-day loss of well control.
MAQ0648J | Tamarind Resources - Tui Field | Oil Spill Modelling | 21/03/18 Page 42
REPORT
Figure 21 Example snaphot of surface oil exposure at a given moment during a single simulation.
MAQ0648J | Tamarind Resources - Tui Field | Oil Spill Modelling | 21/03/18 Page 43
REPORT
Figure 22 Probability of oil exposure on the sea surface above low exposure (≥0.5 g/m2) from a 45-day loss of well control.
MAQ0648J | Tamarind Resources - Tui Field | Oil Spill Modelling | 21/03/18 Page 44
REPORT
Figure 23 Probability of oil exposure on the sea surface above moderate exposure (≥10 g/m2) from a 45-day loss of well control.
MAQ0648J | Tamarind Resources - Tui Field | Oil Spill Modelling | 21/03/18 Page 45
REPORT
Figure 24 Probability of oil exposure on the sea surface above high exposure (≥25 g/m2) from a 45-day loss of well control.
MAQ0648J | Tamarind Resources - Tui Field | Oil Spill Modelling | 21/03/18 Page 46
REPORT
Figure 25 Minimum time before oil exposure on the sea surface above low exposure (≥0.5 g/m2) from a 45-day loss of well control.
MAQ0648J | Tamarind Resources - Tui Field | Oil Spill Modelling | 21/03/18 Page 47
REPORT
Figure 26 Minimum time before oil exposure on the sea surface above moderate exposure (≥10 g/m2) from a 45-day loss of well control.
MAQ0648J | Tamarind Resources - Tui Field | Oil Spill Modelling | 21/03/18 Page 48
REPORT
Figure 27 Minimum time before oil exposure on the sea surface above high exposure (≥25 g/m2) from a 45-day loss of well control.
MAQ0648J | Tamarind Resources - Tui Field | Oil Spill Modelling | 21/03/18 Page 49
REPORT
Figure 28 Maximum potential shoreline loading from a 45-day loss of well control.
MAQ0648J | Tamarind Resources - Tui Field | Oil Spill Modelling | 21/03/18 Page 50
REPORT
10.1.2 In-Water Exposure Dissolved aromatics leeching away from the entrained oil while in the water column resulted in zones of potential low and moderate exposure in the top 30 m of the water column. There was no exposure above thresholds of concern below 30 m in the water column. Table 18 provides a summary of the dissolved aromatics exposure to receptors at each depth interval while Figure 29 to Figure 31 show the extent of the zones of potential low and moderate exposure through the water column. The highest exposure in the water column occurred at adjacent to the South Taranaki regional shoreline. Table 19 provides a summary of the probability of low, moderate and high exposures to marine based receptors from entrained hydrocarbons at 0-10 m below the sea surface. Zones of potential entrained hydrocarbon exposure were much smaller than the dissolved aromatics, reaching a maximum distance of approximately 30 km from the release site (Figure 32). The low exposure zone was also limited to the upper 10 m of the water column. Some shoreline areas were shown to have low exposure to entrained hydrocarbons, these were likely to occur from floating oil trapped in the coastal zone being forced in to the water column during a strong wind event.
MAQ0648J | Tamarind Resources - Tui Field | Oil Spill Modelling | 21/03/18 Page 51
REPORT
Table 18 Probability of low, moderate and high exposure to receptors at 0–10 m, 10 - 20m and 20 - 30m below the sea surface from dissolved aromatics from a 45-day loss of well control.
Dissolved Aromatic Exposure
0 – 10 m 10 – 20 m 20 – 30 m
Maximum Probability of exposure to Maximum Probability of exposure to Maximum Probability of exposure to Receptor exposure dissolved aromatics (ppb.hrs) exposure dissolved aromatics (ppb.hrs) exposure dissolved aromatic (ppb.hrs) to to to dissolved Low Moderate High dissolved Low Moderate High dissolved Low Moderate High aromatics (576 (4,800 (38,4000 aromatics (576 (4,800 (38,4000 aromatics (576 (4,800 (38,4000 (ppb.hrs) pph.hrs) ppb.hrs) ppb.hrs) (ppb.hrs) pph.hrs) ppb.hrs) ppb.hrs) (ppb.hrs) pph.hrs) ppb.hrs) ppb.hrs) Mana Island 634 1 0 0 917 3 0 0 787 1 0 0 Kapiti Island 1,794 3 0 0 2,025 5 0 0 1,373 2 0 0 Wanganui 3,292 4 0 0 1,084 1 0 0 0 0 0 0 South Taranaki 6,638 42 2 0 2,452 6 0 0 1,593 1 0 0 New Plymouth 3,792 20 0 0 2,194 3 0 0 0 0 0 0 Hutt City 0 0 0 0 1,003 1 0 0 0 0 0 0 Kapiti Coast 2,558 6 0 0 0 0 0 0 0 0 0 0 Horowhenua 4,016 8 0 0 2,901 5 0 0 0 0 0 0 Manawatu 3,657 6 0 0 1,176 3 0 0 0 0 0 0 Rangitikei 3,602 7 0 0 601 1 0 0 0 0 0 0 Porirua 0 0 0 0 1,464 2 0 0 0 0 0 0 Kapiti Marine Reserve 1,013 2 0 0 1,251 3 0 0 1,373 2 0 0 Tapuae Marine 3,017 10 0 0 1,628 3 0 0 0 0 0 0 Reserve Marine Mammal 4,546 39 0 0 4,374 22 0 0 2,153 6 0 0 Sanctuary - Taranaki Marine Mammal 1,552 2 0 0 1,329 2 0 0 876 1 0 0 Sanctuary - Waikato
MAQ0648J | Tamarind Resources - Tui Field | Oil Spill Modelling | 21/03/18 Page 52
REPORT
Figure 29 Zones of potential dissolved aromatic exposure at 0-10 m below the sea surface from a 45-day loss of well control.
MAQ0648J | Tamarind Resources - Tui Field | Oil Spill Modelling | 21/03/18 Page 53
REPORT
Figure 30 Zones of potential dissolved aromatic exposure at 10-20 m below the sea surface from a 45-day loss of well control.
MAQ0648J | Tamarind Resources - Tui Field | Oil Spill Modelling | 21/03/18 Page 54
REPORT
Figure 31 Zones of potential dissolved aromatic exposure at 20-30 m below the sea surface from a 45-day loss of well control.
MAQ0648J | Tamarind Resources - Tui Field | Oil Spill Modelling | 21/03/18 Page 55
REPORT
Table 19 Probability of low, moderate and high exposure to marine based receptors from entrained hydrocarbons at 0-10 m below the sea surface from a 45-day loss of well control.
0 – 10 m Receptor Maximum exposure to Probability of exposure to entrained hydrocarbons (ppb.hrs) entrained hydrocarbons Low Moderate High (ppb.hrs)) (67,200 pph.hrs) (676,800 ppb.hrs) (7,718,400 pph.hrs) South Taranaki 68,758 1 0 0 New Plymouth 80,041 1 0 0 Kapiti Coast 70,571 1 0 0 Horowhenua 78,172 3 0 0
MAQ0648J | Tamarind Resources - Tui Field | Oil Spill Modelling | 21/03/18 Page 56
REPORT
Figure 32 Zones of potential entrained hydrocarbon exposure at 0-10 m below the sea surface from a 45-day loss of well control.
MAQ0648J | Tamarind Resources - Tui Field | Oil Spill Modelling | 21/03/18 Page 57
REPORT
11 Results: Amokura-2H 200 m3 MDO Surface Release
This scenario examined a hypothetical 200 m3 surface release of MDO at the Amokura-2H. A total of 100 spill trajectories were simulated over the defined period of February to May.
11.1 Stochastic Analysis Section 10.1.1 presents the potential exposure to the sea surface and shoreline contact For the modelling study each spill trajectory was tracked to the following minimum thresholds: Visible sea surface oil – 0.5 g/m2 Shoreline oil contact – 10 g/m2 Dissolved aromatics – 576 ppb.hrs Entrained hydrocarbons – 67,200 ppb.hrs Figure 33 presents the zones of potential oil exposure on the sea surface for the modelled period. Zones of low (0.5-10 g/m2), moderate (10-25 g/m2) and high (>25 g/m2) potential exposure were predicted to extend predominantly south- southeast of the release sites. Table 20 details the maximum distance travelled by oil on the sea surface at each threshold. The maximum distance from a release site for the potential zone of low, moderate and high exposure were 46 km (southeast), 19 km (southeast) and 4 km (southeast), respectively. Figure 34 to Figure 36 presents the probability of oil exposure on the sea surface reported at the low, moderate and high threshold for the modelling assessment period. Additionally, Figure 37 to Figure 39 present the minimum time before observing oil on the sea surface at the low, moderate and high thresholds. No shoreline contact was predicted for this hypothetical scenario. No entrained exposure (equal to or greater than the minimum reporting threshold 67,200 ppb.hrs) was predicted for this scenario. No dissolved aromatic exposure (equal to or greater than the minimum reporting threshold 576 ppb.hrs) was predicted for this scenario.
Table 20 Summary of potential zones of sea surface exposure at each surface oil threshold from a 200 m3 MDO release.
Zones of potential sea surface exposure Period Distance and direction Low Moderate High (0.5–10 g/m2) (10–25 g/m2) (>25 g/m2)
Max. distance from release site (km) 46 19 4
Max. distance from release site (km) February - 45 18 4 May (99th percentile)
Direction Southeast Southeast Southeast
MAQ0648J | Tamarind Resources - Tui Field | Oil Spill Modelling | 21/03/18 Page 58
REPORT
Figure 33 Zones of potential exposure on the sea surface from a 200 m3 MDO release.
MAQ0648J | Tamarind Resources - Tui Field | Oil Spill Modelling | 21/03/18 Page 59
REPORT
Figure 34 Probability of oil exposure on the sea surface above low exposure (≥0.5 g/m2) from a 200 m3 MDO release.
MAQ0648J | Tamarind Resources - Tui Field | Oil Spill Modelling | 21/03/18 Page 60
REPORT
Figure 35 Probability of oil exposure on the sea surface above moderate exposure (≥10 g/m2) from a 200 m3 MDO release.
MAQ0648J | Tamarind Resources - Tui Field | Oil Spill Modelling | 21/03/18 Page 61
REPORT
Figure 36 Probability of oil exposure on the sea surface above high exposure (≥25 g/m2) from a 200 m3 MDO release.
MAQ0648J | Tamarind Resources - Tui Field | Oil Spill Modelling | 21/03/18 Page 62
REPORT
Figure 37 Minimum time before oil exposure on the sea surface above low exposure (≥0.5 g/m2) from a 200 m3 MDO release.
MAQ0648J | Tamarind Resources - Tui Field | Oil Spill Modelling | 21/03/18 Page 63
REPORT
Figure 38 Minimum time before oil exposure on the sea surface above moderate exposure (≥10 g/m2) from a 200 m3 MDO release.
MAQ0648J | Tamarind Resources - Tui Field | Oil Spill Modelling | 21/03/18 Page 64
REPORT
Figure 39 Minimum time before oil exposure on the sea surface above high exposure (≥25 g/m2) from a 200 m3 MDO release.
MAQ0648J | Tamarind Resources - Tui Field | Oil Spill Modelling | 21/03/18 Page 65
REPORT
12 References
American Society for Testing and Materials (ASTM) 2013. F2067-13 Standard Practice for Development and Use of Oil-Spill Trajectory Models, ASTM International, West Conshohocken (PA).
Australian Maritime Safety Authority (AMSA) 2007, ‘Foreshore Assessment, Termination of Clean-up and Rehabilitation Monitoring’, viewed 12 February 2014, https://www.amsa.gov.au/environment/maritime- environmental-emergencies/national- plan/ESC/documents/Foreshore_Assessment_and_Termination.pdf
Australian Maritime Safety Authority (AMSA) 2012, ‘Australian Maritime Safety Authority Technical Guideline for the Preparation of Marine Pollution Contingency Plans for Marine and Coastal Facilities Australian Maritime Safety Authority’, viewed 15 January 2015, https://www.amsa.gov.au/forms-and- publications/Publications/AMSA413_Contingency_Planning_Guidelines.pdf
Andersen, OB 1995, ‘Global ocean tides from ERS 1 and TOPEX/POSEIDON altimetry’, Journal of Geophysical Research: Oceans, vol. 100, no. C12, pp. 25249–25259.
Australian and New Zealand Environment and Conservation Council (ANZECC), 2000. Australian and New Zealand guidelines for fresh and marine water quality. Volume 1, The guidelines (National water quality management strategy; no.4). Australian and New Zealand Environment and Conservation Council, Agriculture and Resource Management Council of Australia and New Zealand.
Becker, JJ, Sandwell, DT, Smith, WHF, Braud, J, Binder, B, Depner, J, Fabre, D, Factor, J, Ingalls, S, Kim, S- H, Ladner, R, Marks, K, Nelson, S, Pharaoh, A, Trimmer, R, Von Rosenberg, J, Wallace, G & Weatherall, P 2009, ‘Global bathymetry and evaluation data at 30 arc seconds resolution: SRTM30_PLUS’, Marine Geodesy, vol. 32, no. 4, pp. 355–371.
Bonn Agreement 2009, Bonn Agreement aerial operations handbook, 2009 - Publication of the Bonn Agreement, London, viewed 13 January 2015, http://www.bonnagreement.org/site/assets/files/3947/ba- aoh_revision_2_april_2012.pdf
Bowman, MJ, Kibblewhite, AC, Murtagh, RA, Chiswell, SM & Sandersin, BG 1983, ‘Circulation and mixing in greater Cook Straight, New Zealand’, Oceanologica Acta, vol. 6, no. 4, pp. 383–391.
Brodie, JW 1960, ‘Coastal surface currents around New Zealand’, New Zealand Journal of Geology and Geophysics, vol. 3, no. 2, pp. 235–252.
Chassignet, EP, Hurlburt, HE, Smedstad, OM, Halliwell, GR, Hogan, PJ, Wallcraft, AJ, Baraille, R & Bleck, R 2007, ‘The HYCOM (hybrid coordinate ocean model) data assimilative system’, Journal of Marine Systems, vol. 65, no. 1, pp. 60–83.
MAQ0648J | Tamarind Resources - Tui Field | Oil Spill Modelling | 21/03/18 Page 66
REPORT
Chassignet, E, Hurlburt, H, Metzger, E, Smedstad, O, Cummings, J & Halliwell, G 2009, ‘U.S. GODAE: Global Ocean Prediction with the HYbrid Coordinate Ocean Model (HYCOM)’, Oceanography, vol. 22, no. 2, pp. 64–75.
Clark, RB, 1984 ‘Impact of oil pollution on seabirds’, Environmental Pollution, vol. 33, no. 1, pp. 1–22.
Coleman, R 1984, ‘Investigation of the Tasman Sea by satellite altimetry’, Australian Journal of Marine and Freshwater Research, vol. 35, no. 6, pp. 619–633.
Davies, AM 1977a, ‘The numerical solutions of the three-dimensional hydrodynamic equations using a B- spline representation of the vertical current profile’, in JC Nihoul (ed), Bottom Turbulence: Proceedings of the 8th Liège Colloquium on Ocean Hydrodynamics, Elsevier Scientific, Amsterdam, pp. 1–25.
Davies, AM 1977b, ‘Three-dimensional model with depth-varying eddy viscosity’, in JC Nihoul (ed), Bottom Turbulence: Proceedings of the 8th Liège Colloquium on Ocean Hydrodynamics, Elsevier Scientific, Amsterdam, pp. 27–48.
Di Toro, DM, McGrath, JA & Stubblefield, WA 2007, ‘Predicting the toxicity of neat and weathered crude oil: Toxic potential and the toxicity of saturated mixtures’, Environmental Toxicology and Chemistry, vol. 26, no. 1, pp. 24–36.
Engelhardt, FR 1983, ‘Petroleum effects on marine mammals’, Aquatic Toxicology, vol. 4, no. 3, pp. 199– 217.
European Chemicals Agency. (2008). Chapter R.10 - Characterisation of dose [concentration] -response for environment. In Guidance on information requirements and chemcial safety assessment (pp. 26-29). ECHA.
French, D, Reed, M, Jayko, K, Feng, S, Rines, H, Pavignano, S, Isaji, T, Puckett, S, Keller, A, French III, FW, Gifford, D, McCue, J, Brown, G, MacDonald, E, Quirk, J, Natzke, S, Bishop, R, Welsh, M, Phillips, M, Ingram, BS 1996, The CERCLA Type A natural resource damage assessment model for coastal and marine environments (NRDAM/CME), Technical Documentation, Volume I - Model Description, Final Report, Office of Environmental Policy and Compliance, U.S. Department of the Interior, Washington DC.
French, D, Schuttenberg, H & Isaji, T 1999, ‘Probabilities of oil exceeding thresholds of concern: examples from an evaluation for Florida Power and Light’, Proceedings of the 22nd Arctic and Marine Oil Spill Program (AMOP) Technical Seminar, Environment Canada, Alberta, pp. 243–270.
French-McCay, DP 2002, ‘Development and application of an oil toxicity and exposure model, OilToxEx’, Environmental Toxicology and Chemistry, vol. 21, no. 10, pp. 2080-2094.
French-McCay, DP 2003, ‘Development and application of damage assessment modelling: example assessment for the North Cape oil spill’, Marine Pollution Bulletin, vol. 47, no. 9, pp. 9–12.
MAQ0648J | Tamarind Resources - Tui Field | Oil Spill Modelling | 21/03/18 Page 67
REPORT
French-McCay, DP 2004, ‘Spill impact modelling: development and validation’, Environmental Toxicology and Chemistry, vol. 23, no.10, pp. 2441–2456.
French-McCay, DP 2009, ‘State-of-the-art and research needs for oil spill impact assessment modelling’, Proceedings of the 32nd Arctic and Marine Oil Spill Program (AMOP) Technical Seminar, Environment Canada, Ottawa, pp. 601–653.
French-McCay, D, Rowe, JJ, Whittier, N, Sankaranarayanan, S, & Etkin, DS 2004, ‘Estimate of potential impacts and natural resource damages of oil’, Journal of Hazardous Materials, vol. 107, no. 1, pp. 11– 25.
French-McCay, D, Reich, D, Rowe, J, Schroeder, M & Graham, E 2011, ‘Oil spill modeling input to the offshore environmental cost model (OECM) for US-BOEMRE's spill risk and costs evaluations’, Proceedings of the 34th Arctic and Marine Oil Spill Program (AMOP) Technical Siminar, Environment Canada, Ottawa.
Geraci, JR., & St. Aubin, DJ 1988, Synthesis of effects of oil on marine mammals. 292. Ventura, CA, USA: US Department of the Interior, Minerals Management Service, Atlantic OCS Region, OCS Study, MMS 880049.
Gordon, R 1982, ‘Wind driven circulation in Narragansett Bay’ PhD thesis, Department of Ocean Engineering, University of Rhode Island.
Grant, DL, Clarke, PJ & Allaway, WG 1993, ‘The response of grey mangrove (Avicennia marina (Forsk.) Vierh) seedlings to spills of crude oil,’ The Journal of Experimental Marine Biological Ecology, vol. 171, no. 2, pp. 273–295.
Heath, RA 1985, ‘Large-scale influence of the New Zealand seafloor topography on western boundary currents of the South Pacific Ocean’, Australian Journal of Marine and Freshwater Research, vol. 36, no. 1, pp. 1– 14.
Isaji, T & Spaulding, M 1984, ‘A model of the tidally induced residual circulation in the Gulf of Maine and Georges Bank’, Journal of Physical Oceanography, vol. 14, no. 6, pp. 1119–1126.
Isaji, T, Howlett, E, Dalton C, & Anderson, E 2001, ‘Stepwise-continuous-variable-rectangular grid hydrodynamics model’, Proceedings of the 24th Arctic and Marine Oil Spill Program (AMOP) Technical Seminar (including 18th TSOCS and 3rd PHYTO), Environment Canada, Edmonton, pp. 597–610.
International Tankers Owners Pollution Federation (ITOPF) 2014. Technical Information Paper 2 -Fate of Marine Oil Spills, International Tankers Owners Pollution Federation td, UK.
Jenssen, BM 1994, ‘Review article: Effects of Oil Pollution, Chemically Treated Oil, and Cleaning on the Thermal Balance of Birds’, Environmental Pollution, vol. 86, no. 2, pp. 207–215.
MAQ0648J | Tamarind Resources - Tui Field | Oil Spill Modelling | 21/03/18 Page 68
REPORT
Koops, W, Jak, RG & van der Veen, DPC 2004, ‘Use of dispersants in oil spill response to minimise environmental damage to birds and aquatic organisms’, Proceedings of the Interspill 2004: Conference and Exhibition on Oil Spill Technology, Trondheim, presentation 429.
Kostianoy, AG, Ginzburg, AI, Lebedev, SA, Frankignoulle, M & Delille, B 2003, ‘Fronts and mesoscale variability in the southern Indian Ocean as inferred from the TOPEX/POSEIDON and ERS-2 Altimetry data’, Oceanology, vol. 43, no. 5, pp. 632–642.
Levitus, S, Antonov, JI, Baranova, OK, Boyer, TP, Coleman, CL, Garcia, HE, Grodsky, AI, Johnson, DR, Locarnini, RA, Mishonov, AV, Reagan, JR, Sazama, CL, Seidov, D, Smolyar, I, Yarosh, ES & Zweng, MM 2013, ‘The World Ocean Database’, Data Science Journal, vol.12, no. 0, pp. WDS229–WDS234.
Lin, Q & Mendelssohn, IA 1996, ‘A comparative investigation of the effects of south Louisiana crude oil on the vegetation of fresh, brackish and Salt Marshes’, Marine Pollution Bulletin, vol. 32, no. 2, pp. 202– 209.
Ludicone, D, Santoleri, R, Marullo, S & Gerosa, P 1998, ‘Sea level variability and surface eddy statistics in the Mediterranean Sea from TOPEX/POSEIDON data. Journal of Geophysical ResearchI, vol. 103, no. C2, pp. 2995–3011.
Matsumoto, K, Takanezawa, T & Ooe, M 2000, ‘Ocean tide models developed by assimilating TOPEX/POSEIDON altimeter data into hydrodynamical model: A global model and a regional model around Japan’, Journal of Oceanography, vol. 56, no.5, pp. 567–581.
National Oceanic and Atmospheric Administration (NOAA) 2013, Screening level risk assessment package Gulf state, Office of National Marine Sanctuaries & Office of Response and Restoration, Washington DC.
The Organisation for Economic Co-operation and Development (OECD) 2002, Chapter 4: Initial Assessment of Data. In OECD, Manual for Investigation of HPV Chemicals, pp. 1-11.
Oil Spill Solutions 2015, Evaluation - The Theory of Oil Slick Appearances, viewed 6 January 2015, http://www.oilspillsolutions.org/evaluation.htm
OSPAR Commission (OSPAR) 2012, OSPAR guidelines in support of recommendation 2012/5 for risk- based approach to the management of produced water discharges from offshore installations. OSPAR Commission, p. 21.
Owen, A 1980, ‘A three-dimensional model of the Bristol Channel’, Journal of Physical Oceanography, vol. 10, pp. 1290–1302.
Qiu, B & Chen, S 2010, ‘Eddy-mean flow interaction in the decadally modulating Kuroshio Extension system’, Deep-Sea Research II, vol. 57, no. 13, pp. 1098–1110.
MAQ0648J | Tamarind Resources - Tui Field | Oil Spill Modelling | 21/03/18 Page 69
REPORT
Saha, S, Moorthi, S, Pan, H-L, Wu, X, Wang, J & Nadiga, S 2010, ‘The NCEP Climate Forecast System Reanalysis’, Bulletin of the American Meteorological Society, vol. 91, no. 8, pp. 1015–1057.
Scholten, MCTh, Kaag, NHBM, Dokkum, HP van, Jak, R.G., Schobben, HPM & Slob, W 1996, Toxische effecten van olie in het aquatische milieu, TNO report TNO-MEP – R96/230, Den Helder.
Smit, MG, Bechmann, RK, Hendriks, AJ, Skadsheim, A, Larsen, BK, Baussant, T, Shaw, B & Sanni, S 2009, ‘Relating biomarkers to whole-organism effects using species sensitivity distributions: A pilot study for marine species exposed to oil’, Environmental Toxicology and Chemistry, vol. 28, no. 5, pp. 1104–1109.
Spaulding, ML., Kolluru, VS, Anderson, E & Howlett, E 1994, ‘Application of three-dimensional oil spill model (WOSM/OILMAP) to hindcast the Braer Spill’, Spill Science and Technology Bulletin, vol. 1, no. 1, pp. 23–35.
Solbakken, JE, Ingebrigtsen, K & Palmork, KH 1984, ‘Comparative study on the fates of the polychlorinated biphenyl 2, 4, 5, 20, 40, 50-hexachlorobiphenyl and the polycyclic aromatic hydrocarbon phenanthrene in flounder (Platichthys flesus) determined by scintillation counting and autoradiography’, Marine Biology, vol. 83, pp. 239–246.
Suprayogi, B & Murray, F 1999, ‘A field experiment of the physical and chemical effects of two oils on mangroves’, Environmental and Experimental Botany, vol. 42, no. 3, pp. 221–229.
Tsvetnenko, Y 1998, ‘Derivation of Australian tropical marine water quality criteria for protection of aquatic life from adverse effects of petroleum hydrocarbons’, Environmental Toxicology and Water Quality, vol. 13, no. 4, pp. 273–284.
Willmott, CJ 1981, ‘On the validation of models’, Physical Geography, vol. 2, no. 2, pp. 184–194.
Willmott, CJ 1982, ‘Some comments on the evaluation of model performance’, Bulletin of the American Meteorological Society, vol. 63, no. 11, pp. 1309–1313.
Willmott CJ, Ackleson SG, Davis RE, Feddema JJ, Klink, KM, Legates, DR, O’Donnell, J & Rowe, CM 1985, ‘Statistics for the evaluation of model performance’, Journal of Geophysical Research, vol. I 90, no. C5, pp. 8995–9005.
Willmott, CJ & Matsuura, K 2005, ‘Advantages of the mean absolute error (MAE) over the root mean square error (RMSE) in assessing average model performance’, Journal of Climate Research, vol. 30, no. 1, pp. 79–82.
Yaremchuk, M & Tangdong, Q 2004, ‘Seasonal variability of the large-scale currents near the coast of the Philippines’, Journal of Physical Oceanography, vol. 34, no., 4, pp. 844–855.
Zigic, S, Zapata, M, Isaji, T, King, B, & Lemckert, C 2003, Modelling of Moreton Bay using an ocean/coastal circulation model, Auckland, NZ: Proceedings of the Coasts and Ports Australasian Conference.
MAQ0648J | Tamarind Resources - Tui Field | Oil Spill Modelling | 21/03/18 Page 70
Oil Spill Modelling
110-day Loss of Well Control
REPORT
Tamarind Resources - Tui Field 110-Day Oil Spill Modelling
Prepared by: RPS AUSTRALIA WEST PTY LTD Prepared for: ERM NEW ZEALAND Suite E1, Level 4 117 Powderham St 140 Bundall Road New Plymouth 4310 Bundall, QLD 4217 New Zealand Australia T: +61 7 5574 1112 T: +64 (0)9 3034 664 E: [email protected] E: [email protected] W: www.erm.com Author: Nathan Benfer Reviewed: Sasha Zigic Approved: Nathan Benfer No.: MAQ0648J Version: 1 Date: 21/03/2018
rpsgroup.com.au REPORT
Document Status
Version Purpose of Document Approved by Reviewed by Review Date
0 Draft issued to client Nathan Benfer Sasha Zigic 28/02/2018
1 Issued to client Nathan Benfer Sasha Zigic 21/03/2018
Approval for issue
Name Signature Date
Nathan Benfer 21/03/2018
This report was prepared by [RPS Australia West Pty Ltd (‘RPS’)] within the terms of its engagement and in direct response to a scope of services. This report is strictly limited to the purpose and the facts and matters stated in it and does not apply directly or indirectly and must not be used for any other application, purpose, use or matter. In preparing the report, RPS may have relied upon information provided to it at the time by other parties. RPS accepts no responsibility as to the accuracy or completeness of information provided by those parties at the time of preparing the report. The report does not take into account any changes in information that may have occurred since the publication of the report. If the information relied upon is subsequently determined to be false, inaccurate or incomplete then it is possible that the observations and conclusions expressed in the report may have changed. RPS does not warrant the contents of this report and shall not assume any responsibility or liability for loss whatsoever to any third party caused by, related to or arising out of any use or reliance on the report howsoever. No part of this report, its attachments or appendices may be reproduced by any process without the written consent of RPS. All enquiries should be directed to RPS.
MAQ0648J | Tamarind Resources - Tui Field | 110-Day Oil Spill Modelling | 21/03/2018 Page II
REPORT
Contents
EXECUTIVE SUMMARY ...... 1 Background ...... 1 Methodology ...... 1 Oil Properties ...... 1 Key Findings ...... 2
1 INTRODUCTION ...... 3
2 SCOPE OF WORK ...... 5
3 REGIONAL CURRENTS ...... 5 3.2 Tidal Currents ...... 7 3.2.1 Grid Setup ...... 7 3.2.2 Tidal Conditions ...... 9 3.2.3 Surface Elevation Validation ...... 9 3.3 Ocean Currents ...... 14 3.4 Currents at the Release Site ...... 15
4 WINDS ...... 17
5 WATER TEMPERATURE AND SALINITY ...... 20
6 OIL SPILL MODEL - SIMAP ...... 22 6.1 Stochastic Modelling ...... 22 6.2 Sea surface, Shoreline and In-Water Thresholds ...... 23 6.2.1 Sea surface Exposure Thresholds ...... 23 6.2.2 Shoreline Contact Threshold ...... 24 6.2.3 Water Column Exposure Thresholds ...... 25 6.3 Exposure Calculation...... 27
7 OIL PROPERTIES ...... 29
8 MODEL SETTINGS AND ASSUMPTIONS ...... 31
9 INTERPRETING MODELLING RESULTS ...... 33 9.1 February to May Stochastic Assessment ...... 33 9.2 Receptors Assessed ...... 35
10 RESULTS: AMOKURA-2H LOSS OF WELL CONTROL ...... 36 10.1 Stochastic Analysis ...... 36 10.1.1 Sea Surface Exposure and Shoreline Contact ...... 36 10.1.2 In-Water Exposure ...... 48
11 REFERENCES ...... 52
MAQ0648J | Tamarind Resources - Tui Field | 110-Day Oil Spill Modelling | 21/03/2018 Page III
REPORT
Tables Table 1 Location of the wells for the study...... 3 Table 2 Statistical comparison between the observed and HYDROMAP predicted surface elevations data from the 1st to 31st January 2014...... 10 Table 3 Predicted average and maximum surface current speeds at the study site. The data was derived by combining the HYCOM ocean data and HYDROMAP tidal data for 2008-2012 (inclusive)...... 15 Table 4 Predicted average and maximum winds for the wind node nearest the release site. Data derived from CFSR hindcast model from 2003-2012 (inclusive)...... 18 Table 5 Monthly average sea-surface temperature and salinity at the study area...... 20 Table 6 The Bonn Agreement Oil Appearance Code ...... 23 Table 7 Thresholds used to classify the zones of sea surface exposure ...... 24 Table 8 Thresholds use to assess shoreline contact ...... 25 Table 9 Dissolved aromatic threshold values applied as part of the modelling study ...... 26 Table 10 Entrained hydrocarbon threshold values applied as part of the modelling study. Thresholds based on OSPAR guidelines...... 26 Table 11 Physical characteristics...... 29 Table 12 Boiling point ranges...... 29 Table 13 Summary of the oil spill model settings used in this assessment...... 31 Table 14 Summary of potential zones of sea surface exposure at each surface oil threshold from a 110- day loss of well control...... 37 Table 15 Summary of shoreline contact at or above 10 g/m2 across all shorelines from a 110-day loss of well control...... 37 Table 16 Summary of the potential sea surface exposure to receptors from a 110-day loss of well control...... 38 Table 17 Summary of shoreline contact to individual shoreline receptors from a 110-day loss of well control...... 39 Table 18 Probability of low, moderate and high exposure to receptors at 0–10 m, 10 - 20m and 20 - 30m below the sea surface from dissolved aromatics from a 110-day loss of well control...... 49 Table 19 Probability of low, moderate and high exposure to marine based receptors from entrained hydrocarbons at 0-10 m below the sea surface from a 110-day loss of well control...... 51
Figures Figure 1 Location of the wells for the study...... 4 Figure 2 Schematic showing the oceanic current circulation surrounding New Zealand (Image source: Brodie, 1960)...... 6 Figure 3 Map showing the regions of sub-gridding for the study area ...... 8 Figure 4 Bathymetry used in the hydrodynamic grid for the study region...... 8 Figure 5 Location of the nine tide stations around New Zealand used to validate the tidal model...... 10 Figure 6 Comparison between predicted (red line) and observed (blue line) surface elevation variation at Auckland (top), Bluff (middle) and Lyttelton (bottom), between the 1st and the 31st of January 2014...... 11
MAQ0648J | Tamarind Resources - Tui Field | 110-Day Oil Spill Modelling | 21/03/2018 Page IV
REPORT
Figure 7 Comparison between predicted (red line) and observed (blue line) surface elevation variation at Napier (top), Nelson (middle) and Picton (bottom), between the 1st and the 31st of January 2014...... 12 Figure 8 Comparison between predicted (red line) and observed (blue line) surface elevation variation at Port Taranaki (top), Wellington (middle) and Westport (bottom), between the 1st and 31st of January 2014...... 13 Figure 9 Snapshot example of the predicted HYCOM ocean surface currents in the region. Colour of individual arrows indicate current speed (m/s)...... 14 Figure 10 Predicted monthly surface current rose plots at the study site. Data was derived by combining the HYDROMAP tidal currents and HYCOM ocean currents for 2008 – 2012. The colour key shows the current speed (m/s), the compass direction provides the current direction flowing TOWARDS and the length of the wedge gives the percentage of the record for a particular speed and direction combination...... 16 Figure 11 Sample of the CFSR modelled wind data used for the oil spill model...... 18 Figure 12 Modelled monthly wind rose distributions from 2003–2012 (inclusive), for the wind node closest to the release site. The colour key shows the wind magnitude, the compass direction provides the direction FROM and the length of the wedge gives the percentage of the record for a particular speed and direction combination...... 19 Figure 13 Monthly average sea temperature and salinity profiles at the study site...... 21 Figure 14 Photographs showing the difference between oil colour and thickness on the sea surface (source: adapted from OilSpillSolutions.org 2015) ...... 24 Figure 15 Example time series plot of concentration of entrained hydrocarbons (top) at a receptor, and exposure (bottom) with no depuration (blue) and with depuration (green) ...... 28 Figure 16 Weathering of Tui Crude under three static wind conditions. The results are based on a 19,803 bbl sub-surface release of Tui Crude over 24 hours, tracked for 75 days...... 30 Figure 17 Plot of the variable release rate over the 110-day period ...... 32 Figure 18 Receptors assessed for oil exposure...... 35 Figure 19 Zones of potential exposure on the sea surface from a 110-day loss of well control...... 40 Figure 20 Probability of oil exposure on the sea surface above low exposure (≥0.5 g/m2) from a 110-day loss of well control...... 41 Figure 21 Probability of oil exposure on the sea surface above moderate exposure (≥10 g/m2) from a 110- day loss of well control...... 42 Figure 22 Probability of oil exposure on the sea surface above high exposure (≥25 g/m2) from a 110-day loss of well control...... 43 Figure 23 Minimum time before oil exposure on the sea surface above low exposure (≥0.5 g/m2) from a 110-day loss of well control...... 44 Figure 24 Minimum time before oil exposure on the sea surface above moderate exposure (≥10 g/m2) from a 110-day loss of well control...... 45 Figure 25 Minimum time before oil exposure on the sea surface above high exposure (≥25 g/m2) from a 110-day loss of well control...... 46 Figure 26 Maximum potential shoreline loading from a 110-day loss of well control...... 47 Figure 27 Zones of potential dissolved aromatic exposure at 0-10 m below the sea surface from a 110-day loss of well control...... 50
MAQ0648J | Tamarind Resources - Tui Field | 110-Day Oil Spill Modelling | 21/03/2018 Page V
REPORT
Terms and Abbreviations ° – Degrees ‘– Minutes “– Seconds AMSA – Australian Maritime Safety Authority ANZECC – Australian and New Zealand Environment and Conservation Council. API – American Petroleum Institute gravity. A measure of how heavy or light a petroleum liquid is compared to water. Bonn Agreement Oil Appearance Code – An agreement for cooperation in dealing with pollution of the North Sea by oil and other harmful substances, 1983, includes: Governments of the Kingdom of Belgium, the Kingdom of Denmark, the French Republic, the Federal Republic of Germany, the Republic of Ireland, the Kingdom of the Netherlands, the Kingdom of Norway, the Kingdom of Sweden, the United Kingdom of Great Britain and Northern Ireland and the European Union. oC – degree Celsius (unit of temperature) CFSR – Climate Forecast System Reanalysis cm – Centimeter (unit of length) Decay – The process where oil components are changed either chemically or biologically (biodegradation) to another compound. It includes breakdown to simpler organic carbon compounds by bacteria and other organisms, photo- oxidation by solar energy, and other chemical reactions. Dissolved aromatic hydrocarbons – dissolved hydrocarbons within the water column with alternating double and single bonds between carbon atoms forming rings, containing at least one six-membered benzene ring. g/m2 – Grams per square meter (unit of surface area density related to thickness) EMBA – Environmental that may be affected Entrained oil – Droplets or globules of oil that are physically mixed (but not dissolved) into the water column. Physical entrainment can occur either during pressurised release from a sub-surface location, or through the action of breaking waves (>12 knots). Evaporation – The process whereby components of the oil mixture are transferred from the sea-surface to the atmosphere. GODAE – Global Ocean Data Assimilation Experiment HYCOM – Hybrid Coordinate Ocean Model is a data-assimilative, three-dimensional ocean model. HYDROMAP – Advanced ocean/coastal tidal model used to predict tidal water levels, current speed and current direction. IOA – Index of Agreement gives a non-dimensional measure of model accuracy or performance. Isopycnal layers – Water column layers with corresponding water densities ITOPF – The International Tanker Owners Pollution Federation JPDA – Joint Petroleum Development Area km – Kilometer (unit of length) km2 – Square Kilometers (unit of area) Knots – units of wind measurement (1 knot = 0.514 m/s)
LC50 – Median lethal dose. The dose required for mortality of 50% of a tested population after a specified test duration m – Meters (unit of length) m/s – Meters per Second (unit of speed) MAE – Mean Absolute Error is the average of the absolute values of the difference between the model predicted and observed surface elevations NASA – National Aeronautics and Space Administration NCEP – National Centres for Environmental Prediction NOAA – National Oceanic and Atmospheric Administration NOEC – No observed adverse effect level Ocean current – Large scale and continuous movement of seawater generated by forces such as breaking waves, wind, the Coriolis effect, and temperature and salinity gradients. It is the main flow of ocean waters.
MAQ0648J | Tamarind Resources - Tui Field | 110-Day Oil Spill Modelling | 21/03/2018 Page VI
REPORT
OECD – The Organisation for Economic Co-operation and Development PNEC – Predicted no effect concentration. ppb – parts per billion (concentration) ppb.hrs – ppb multiplied for hours (concentration x time) PSU – Practical salinity units RPS APASA – RPS Asia-Pacific Applied Science Associates Sea surface exposure – Floating oil on the sea surface equal to or above reporting threshold (e.g. 0.5 g/m2) SIMAP – Spill Impact Mapping Analysis Program Shoreline contact – Stranded oil on the shoreline equal to or above reporting threshold (e.g. 10 g/m2) SRTM30_PLUS – Global topography /bathymetry dataset THC – Total hydrocarbons USCG – United States Coastguard USEPA – United States Environmental Protection Agency Visible oil – Floating oil on the sea surface equal to or above reporting threshold (e.g. 0.5 g/m2)
MAQ0648J | Tamarind Resources - Tui Field | 110-Day Oil Spill Modelling | 21/03/2018 Page VII
REPORT
Executive Summary
Background ERM New Zealand, on behalf of Tamarind Resources, requested RPS to carry out quantitative oil spill modelling for their Tui, Amokura and Pateke wells in the Taranaki Basin, New Zealand. Based on the proximity of the wells to land, and the potential maximum oil release rates, Amokura-2H was chosen as the representative (and worst case) scenario for this modelling study. The study considered the fate of 654,516 STB of Tui Crude released during a hypothetical loss of well control incident over 110 days. The assessment was completed for the months of February to May, which covers the planned operational months of March and April. The SIMAP system, the methods and analysis presented herein use modelling algorithms which have been anonymously peer reviewed and published in international journals. Further, RPS APASA warrants that this work meets and exceeds the ASTM Standard F2067-13 “Standard Practice for Development and Use of Oil Spill Models”. Note that the modelling does not take into consideration any of the spill prevention, mitigation and response capabilities that might be in place during the operations. The modelling makes no allowance for intervention following a spill to reduce volumes and/or prevent hydrocarbons from reaching sensitive areas.
Methodology The modelling study was carried out in several stages. Firstly, a ten-year current dataset (2003–2012) that includes the combined influence of ocean currents from the HYCOM model and tidal currents from the HYDROMAP model was developed. Secondly, high-resolution local winds from the CFSR model and detailed hydrocarbon characteristics were used as inputs in the three-dimensional oil spill model (SIMAP) to simulate the drift, spread, weathering and fate of the spilled oils. As spills can occur during any set of wind and current conditions, modelling was conducted using a stochastic (random or non-deterministic) approach, which involved running 100 spill simulations initiated at random start times, using the same release information (spill volume, duration and composition of the oil). This ensured that each simulation was subject to different wind and current conditions and, in turn, movement and weathering of the oil.
Oil Properties Tui crude has a density of 808 kg/m3 (API gravity of 43.5) and a dynamic viscosity of 2.5 cP at 40ºC, and although it is classified as a Group II oil according to the International Tankers Owners Pollution Federation (ITOPF, 2014) classifications, it does have a high wax content of 17.3%, which will result in this oil turning in to wax after extended periods (weeks) in the marine environment. There is greater persistence of the Tui Crude on the sea surface with decreasing wind speeds, which correlates to increasing volumes of oil occurring in the water column with increasing wind speeds. Additionally, evaporative losses were greatest during the first day, while the oil was still being released, and then the loss rate reduced until at day 4 further evaporative losses were negligible.
MAQ0648J | Tamarind Resources - Tui Field | 110-Day Oil Spill Modelling | 21/03/2018 Page 1
REPORT
Key Findings Amokura-2H Loss of Well Control Zones of potential low exposure (0.5-10 g/m2) were predicted to extend north-west, while zones of moderate (10-25 g/m2) and high (>25 g/m2) exposure predominantly extended east-northeast of the and south-southeast of the release sites. The maximum distance from a release site for the potential zone of low, moderate and high exposure were 395 (north-west), 63 km (east-northeast) and 46 km (south-southeast). The probability of oil coming ashore was 100%, while the quickest time to shore for visible sheens was 56 hours and the maximum volume was 778 m3. The maximum lengths of shoreline exposure at low, moderate and high levels were 375 km, 237 km and 5 km, respectively. Both the South Taranaki shoreline and Marine Mammal Sanctuary had 100% and New Plymouth had 99% probability of being exposed to visible floating oil. Only South Taranaki and New Plymouth shorelines had any probability of being exposed to surface oil of moderate thickness (10-25 g/m2). South Taranaki regional shoreline had the highest probability of contact, for all shoreline accumulation thresholds. Shoreline accumulation of greater than 100 g/m2 occurs within 65 hours for the South Taranaki section of coastline, eventually covering up to 101 km at this level. Some sections of the South Taranaki regional shoreline accumulated up to 3.091 kg/m2 load on shoreline, while New Plymouth had a peak value of 2.815 kg/m2. Dissolved aromatics leeching away from the entrained oil while in the water column resulted in zones of potential low and moderate exposure in the top 30 m of the water column. There was no exposure above thresholds of concern below 30 m in the water column. The highest exposure in the water column occurred at adjacent to the South Taranaki regional shoreline. Zones of potential entrained hydrocarbon exposure were much smaller than the dissolved aromatics. The low exposure zone was also limited to the upper 10 m of the water column. Some shoreline areas were shown to have low exposure to entrained hydrocarbons, these were likely to occur from floating oil trapped in the coastal zone being forced in to the water column during a strong wind event.
MAQ0648J | Tamarind Resources - Tui Field | 110-Day Oil Spill Modelling | 21/03/2018 Page 2
REPORT
1 Introduction
ERM New Zealand, on behalf of Tamarind Resources, requested RPS to carry out quantitative oil spill modelling for their Tui, Amokura and Pateke wells in the Taranaki Basin, New Zealand. Based on the proximity of the wells to land, and the potential maximum oil release rates, Amokura-2H was chosen as the representative (and worst case) scenario for this modelling study (Table 1 and Figure 1). The study considered the fate of 654,516 STB of Tui Crude released during a hypothetical loss of well control incident over 110 days. The assessment was completed for the months of February to May, which covers the planned operational months of March and April. The modelling study was carried out in several stages. Firstly, a five-year current dataset (2003–2012) that includes the combined influence of ocean currents from the HYCOM model and tidal currents from the HYDROMAP model was developed. Secondly, high-resolution local winds from the CFSR model and detailed hydrocarbon characteristics were used as inputs in the three-dimensional oil spill model (SIMAP) to simulate the drift, spread, weathering and fate of the spilled oils. As spills can occur during any set of wind and current conditions, modelling was conducted using a stochastic (random or non-deterministic) approach, which involved running 100 spill simulations initiated at random start times, using the same release information (spill volume, duration and composition of the oil). This ensured that each simulation was subject to different wind and current conditions and, in turn, movement and weathering of the oil. The SIMAP system, the methods and analysis presented herein use modelling algorithms which have been anonymously peer reviewed and published in international journals. Further, RPS APASA warrants that this work meets and exceeds the ASTM Standard F2067-13 “Standard Practice for Development and Use of Oil Spill Models”. Note that the modelling does not take into consideration any of the spill prevention, mitigation and response capabilities that might be in place during the operations. The modelling makes no allowance for intervention following a spill to reduce volumes and/or prevent hydrocarbons from reaching sensitive areas.
Table 1 Location of the wells for the study.
Release Site Latitude Longitude Depth (m)
Amokura-2H 39° 25' 23" S 173° 12' 44" E 122.5
Pateke-3H 39° 22' 51" S 173° 12' 25" E 123.3
Patke-4H 39° 22' 31" S 173° 11' 46" E 124.0
Tui-2H 39° 26' 35" S 173° 14' 11" E 121.6
Tui-3H 39° 26' 34" S 173° 14' 9" E 121.8
MAQ0648J | Tamarind Resources - Tui Field | 110-Day Oil Spill Modelling | 21/03/2018 Page 3
REPORT
Figure 1 Location of the wells for the study.
MAQ0648J | Tamarind Resources - Tui Field | 110-Day Oil Spill Modelling | 21/03/2018 Page 4
REPORT
2 Scope of Work
The scope of work included the following components: 1. Generate ten years of net currents from 2003 to 2012 (inclusive) that include the combined influence of ocean and tidal currents. 2. Use high-resolution wind data, current data and hydrocarbon characteristics as input into the 3- dimensional oil spill model, SIMAP to model the movement, spreading, entrainment, weathering and potential shoreline contact by the hydrocarbon over time; 3. Use SIMAP’s stochastic model (also known as a probability model) to calculate exposure to surround waters and shoreline. This involved running 100 randomly selected single trajectory simulations for each scenario and/or season, with each simulation having the same spill information (spill volume, duration and composition of hydrocarbons) but varying start times. This ensured that each spill trajectory was subjected to unique wind and current conditions; and
3 Regional Currents
An extensive review of the ocean circulation surrounding New Zealand, which describes the ocean currents in the region, is provided by Heath (1985). The study describes two main surface water masses which surround New Zealand, which include the Subtropical and Subantarctic Surface waters. The Subtropical waters predominately originate from an extension of the East Australian Current (EAC), which is a western boundary current that flows from the South Equatorial Current (SEC), and down the eastern coast of Australia. Typically, the EAC carries warm waters to the south, before splitting off into the Tasman Sea approximately in line with Sydney (Coleman, 1984) and carrying the warmer waters eastwards towards New Zealand (Heath, 1985). The Subantarctic Water tends to flow northwards along the eastern side of the South Island, originating from the Circumpolar Current south of New Zealand. Figure 2 presents a schematic of the regional currents of New Zealand. The oceanic currents near the Taranaki Basin are predominately influenced by the typically eastward flowing D’Urville Current and the northward flowing Westland Current. The D’Urville Current consists of warm saline water that flows eastwards into the Cook Strait from the west-northwest (Brodie, 1960; Heath, 1985). The Westland Current has been observed to flow predominately northwards along the west coast of the South Island, where it then mixes with the D’Urville near the Cook Strait. An extension of the Westland current has been observed to extend further northwards along the western coast of the North Island. Previous work by Bowman et al. (1983) indicate that strong non-tidal flows through the Cook Strait may be influenced strongly by the prevailing winds, as shown by a pair of drifters which were observed to travel from the Taranaki Bight through the Cook Strait in a south easterly direction, with strong prevailing winds from the northwest.
MAQ0648J | Tamarind Resources - Tui Field | 110-Day Oil Spill Modelling | 21/03/2018 Page 5
REPORT
Figure 2 Schematic showing the oceanic current circulation surrounding New Zealand (Image source: Brodie, 1960).
MAQ0648J | Tamarind Resources - Tui Field | 110-Day Oil Spill Modelling | 21/03/2018 Page 6
REPORT
3.2 Tidal Currents The effects of tides were generated using RPS’s advanced ocean/coastal model, HYDROMAP. The HYDROMAP model has been thoroughly tested and verified through field measurements throughout the world over more than 20 years (Isaji and Spaulding, 1984; Isaji et al., 2001; Zigic et al., 2003). In fact, HYDROMAP tidal current data has been used as input for the OILMAP hydrocarbon spill modelling system, which forms part of the Incident Management System (IMS) operated by Maritime New Zealand (MNZ). HYDROMAP employs a sophisticated sub-gridding strategy, which supports up to six levels of spatial resolution, halving the grid cell size as each level of resolution is employed. The sub-gridding allows for higher resolution of currents within areas of greater bathymetric and coastline complexity, and/or of particular interest to a study. The numerical solution methodology follows that of Davies (1977a and 1977b) with further developments for model efficiency by Owen (1980) and Gordon (1982). A more detailed presentation of the model can be found in Isaji and Spaulding (1984) and Isaji et al. (2001).
3.2.1 Grid Setup HYDROMAP was set-up to cover the domain of interest, which was subdivided horizontally into a grid with 4 levels of resolution. The resolution of the primary level was set at 8 km. The resolution of the first, second and third levels were 4 km, 2 km, 1 km respectively. The finer grids were allocated in a step-wise fashion to more accurately resolve flows along the coastline, around islands and over more complex bathymetry. Figure 3 shows a zoomed in image in the region containing the Pateke-4H release site. A combination of datasets was used to describe the shape of the seabed within the high-resolution grid. Depths for the region were extracted from nautical charts and the SRTM30_PLUS dataset (see Becker et al., 2009), which provides a 30-arc second, or approximately 1 km, resolution (see Figure 4).
MAQ0648J | Tamarind Resources - Tui Field | 110-Day Oil Spill Modelling | 21/03/2018 Page 7
REPORT
Figure 3 Map showing the regions of sub-gridding for the study area
Figure 4 Bathymetry used in the hydrodynamic grid for the study region.
MAQ0648J | Tamarind Resources - Tui Field | 110-Day Oil Spill Modelling | 21/03/2018 Page 8
REPORT
3.2.2 Tidal Conditions The ocean boundary data for the regional model was obtained from satellite measured altimetry data (TOPEX/Poseidon 7.2) which provided estimates of the eight dominant tidal constituents at a horizontal scale of approximately 0.25 degrees. The eight major tidal constituents used were K2, S2, M2, N2, K1, P1, O1 and Q1. Using the tidal data, surface heights were firstly calculated along the open boundaries, at each time step in the model. The Topex-Poseidon satellite data is produced, and quality controlled by National Aeronautics and Space Administration (NASA). The satellites, equipped with two highly accurate altimeters, capable of taking sea level measurements accurate to less than ± 5 cm, measured oceanic surface elevations (and the resultant tides) for over 13 years (1992–2005). In total these satellites carried out 62,000 orbits of the planet. The Topex-Poseidon tidal data has been widely used amongst the oceanographic community, being the subject of more than 2,100 research publications (e.g. Andersen, 1995; Ludicone et al., 1998; Matsumoto et al., 2000; Kostianoy et al., 2003; Yaremchuk and Tangdong, 2004; Qiu and Chen 2010). As such the Topex/Poseidon tidal data is considered suitably accurate for this study.
3.2.3 Surface Elevation Validation To ensure that tidal predictions were accurate, predicted surface elevations were compared to data observed at five locations situated across the study region (Figure 5). Figure 6 to Figure 8 illustrate a comparison of the predicted and observed surface elevations for each location for January 2014. As shown on the graph, the model accurately reproduced the phase and amplitudes throughout the spring and neap tidal cycles. To provide a statistical measure of the model’s performance, the Index of Agreement (IOA – Willmott, 1981) and the Mean Absolute Error (MAE – Willmott, 1982; Willmott and Matsuura, 2005) were used. The MAE is the average of the absolute values of the difference between the model-predicted (P) and observed (O) variables. It is a more natural measure of the average error and more readily understood (Willmott and Matsuura, 2005). 푁 −1 푀퐴퐸 = 푁 ∑|푃푖 − 푂푖| 푖=1
The Index of Agreement (IOA) is determined by:
2 ∑|푋푚표푑푒푙 − 푋표푏푠| 퐼푂퐴 = 1 − 2 ∑(|푋푚표푑푒푙 − 푋̅̅̅표푏푠̅̅̅| + |푋표푏푠 − 푋̅̅̅표푏푠̅̅̅|)
Where: X represents the variable being compared and the time mean of that variable. A perfect agreement exists between the model and field observations if the index gives an agreement value of 1 and complete disagreement will produce an index measure of 0 (Wilmott, 1981). Willmott et al., (1985) also suggests that values meaningfully larger than 0.5 represent good model performance. Clearly, a greater IOA and lower MAE represent a better model performance. Table 2 shows the IOA and MAE values for the selected locations.
MAQ0648J | Tamarind Resources - Tui Field | 110-Day Oil Spill Modelling | 21/03/2018 Page 9
REPORT
Figure 5 Location of the nine tide stations around New Zealand used to validate the tidal model.
Table 2 Statistical comparison between the observed and HYDROMAP predicted surface elevations data from the 1st to 31st January 2014.
Tide Station IOA MAE (m)
Auckland (North Island) 0.95 0.29
Bluff (South Island) 0.93 0.25
Lyttelton (South Island) 0.91 0.26
Napier (North Island) 0.98 0.11
Nelson (South Island) 0.93 0.39
Picton (South Island) 0.93 0.15
Port Taranaki (North Island) 0.94 0.33
Wellington (North Island) 0.95 0.13
Westport (South Island) 0.94 0.30
Average 0.94 0.25
MAQ0648J | Tamarind Resources - Tui Field | 110-Day Oil Spill Modelling | 21/03/2018 Page 10
REPORT
Figure 6 Comparison between predicted (red line) and observed (blue line) surface elevation variation at Auckland (top), Bluff (middle) and Lyttelton (bottom), between the 1st and the 31st of January 2014.
MAQ0648J | Tamarind Resources - Tui Field | 110-Day Oil Spill Modelling | 21/03/2018 Page 11
REPORT
Figure 7 Comparison between predicted (red line) and observed (blue line) surface elevation variation at Napier (top), Nelson (middle) and Picton (bottom), between the 1st and the 31st of January 2014.
MAQ0648J | Tamarind Resources - Tui Field | 110-Day Oil Spill Modelling | 21/03/2018 Page 12
REPORT
Figure 8 Comparison between predicted (red line) and observed (blue line) surface elevation variation at Port Taranaki (top), Wellington (middle) and Westport (bottom), between the 1st and 31st of January 2014.
MAQ0648J | Tamarind Resources - Tui Field | 110-Day Oil Spill Modelling | 21/03/2018 Page 13
REPORT
3.3 Ocean Currents Data describing the flow of ocean currents was obtained from HYCOM (Hybrid Coordinate Ocean Model, (Chassignet et al., 2007), which is operated by the HYCOM Consortium, sponsored by the Global Ocean Data Assimilation Experiment (GODAE). HYCOM is a data-assimilative, three-dimensional ocean model that is run as a hindcast (for a past period), assimilating time-varying observations of sea-surface height, sea-surface temperature and in-situ temperature and salinity measurements (Chassignet et al., 2009). The HYCOM predictions for drift currents are produced at a horizontal spatial resolution of approximately 8.25 km (1/12th of a degree) over the region, at a frequency of once per day. HYCOM uses isopycnal layers in the open, stratified ocean, but uses the layered continuity equation to make a dynamically smooth transition to a terrain following coordinate in shallow coastal regions, and to z–level coordinates in the mixed layer and/or unstratified seas. For this study, the HYCOM hindcast currents were obtained for the year 2012. Figure 9 shows an example of the modelled surface ocean currents (HYCOM) for the region.
Figure 9 Snapshot example of the predicted HYCOM ocean surface currents in the region. Colour of individual arrows indicate current speed (m/s).
MAQ0648J | Tamarind Resources - Tui Field | 110-Day Oil Spill Modelling | 21/03/2018 Page 14
REPORT
3.4 Currents at the Release Site Table 3 displays the average and maximum combined surface current speeds (ocean plus tides) nearby the release sites. Figure 10 shows the monthly surface current rose distributions nearby the release sites. Note the convention for defining current direction is the direction the current flows towards, which is used to reference current direction throughout this report. Each branch of the rose represents the currents flowing to that direction, with north to the top of the diagram. Sixteen directions are used. The branches are divided into segments of different colour, which represent the current speed ranges for each direction. Speed intervals of 0.1 m/s are typically used in these current roses. The length of each coloured segment is relative to the proportion of currents flowing within the corresponding speed and direction. The data showed that the surface current speeds and directions were variable between the months, though with a predominant flow towards the south-southeast and north. The maximum and average surface current speeds were 1.44 m/s 0.24 m/s, respectively.
Table 3 Predicted average and maximum surface current speeds at the study site. The data was derived by combining the HYCOM ocean data and HYDROMAP tidal data for 2008-2012 (inclusive).
Average current Maximum current Month General Direction speed (m/s) speed (m/s)
January 0.24 1.17 Northeast to Southeast
February 0.20 0.96 Northeast to Southeast
March 0.23 1.37 Northwest and Southeast
April 0.25 1.44 Southeast
May 0.26 0.91 Southeast
June 0.25 1.04 Variable
July 0.24 1.10 Northwest to Northeast
August 0.23 1.21 South
September 0.27 1.32 Southeast
October 0.25 1.00 Northeast to southeast
November 0.23 1.07 Northeast
December 0.25 0.94 Southeast
Minimum 0.20 0.91
Maximum 0.27 1.44
MAQ0648J | Tamarind Resources - Tui Field | 110-Day Oil Spill Modelling | 21/03/2018 Page 15
REPORT
Figure 10 Predicted monthly surface current rose plots at the study site. Data was derived by combining the HYDROMAP tidal currents and HYCOM ocean currents for 2008 – 2012. The colour key shows the current speed (m/s), the compass direction provides the current direction flowing TOWARDS and the length of the wedge gives the percentage of the record for a particular speed and direction combination.
MAQ0648J | Tamarind Resources - Tui Field | 110-Day Oil Spill Modelling | 21/03/2018 Page 16
REPORT
4 Winds
High resolution wind data was sourced from the National Centre for Environmental Prediction (NCEP) Climate Forecast System Reanalysis (CFSR; see Saha et al., 2010). The CFSR wind model is a fully coupled, data- assimilative hind cast model representing the interaction between the earth’s oceans, land and atmosphere. The gridded wind data output is available at ¼ of a degree resolution (~33 km) and 1-hourly time intervals.
The CFSR wind data for the years 2003–2012 (inclusive) was compiled across the model domain. Figure 11 shows an example of the wind field used as input into the oil spill model.
Table 4 shows the monthly average and maximum winds derived from the CFSR model node nearest the release site.
Figure 12 shows the monthly wind roses derived from the CFSR model node closest to the release site.
Note that the atmospheric convention for defining wind direction, that is, the direction the wind blows from, is used to reference wind direction throughout this report. Each branch of the rose represents wind coming from that direction, with north to the top of the diagram. Sixteen directions are used. The branches are divided into segments of different colour, which represent wind speed ranges from that direction. Speed ranges of 10 knots are typically used in these wind roses. The length of each segment within a branch is proportional to the frequency of winds blowing within the corresponding range of speeds from that direction.
The wind node closest to the release site demonstrated two predominant (general) directions; 1) south- westerly winds during August to February and 2) south-westerly and south-easterly during March to July.
Winds in this region are moderate to strong. Monthly average wind speeds range from 15–19 knots and the monthly maximum wind speeds range from 41–54 knots (Table 4). The maximum wind speed occurred during July. Note these maximums do not include any short-term wind gusts during severe storms.
MAQ0648J | Tamarind Resources - Tui Field | 110-Day Oil Spill Modelling | 21/03/2018 Page 17
REPORT
Figure 11 Sample of the CFSR modelled wind data used for the oil spill model.
Table 4 Predicted average and maximum winds for the wind node nearest the release site. Data derived from CFSR hindcast model from 2003-2012 (inclusive).
Month Average wind Maximum wind General Direction (knots) (knots) (From)
January 16 43 Southwest February 15 41 Southwest March 16 51 Southwest and Southeast April 16 50 Southwest and Southeast May 18 46 Southwest and Southeast June 19 53 Southwest and Southeast July 19 54 Southwest and Southeast August 17 44 Southwest September 18 42 Southwest October 18 49 Southwest November 17 45 Southwest December 16 43 Southwest Minimum 15 41 Maximum 19 54
MAQ0648J | Tamarind Resources - Tui Field | 110-Day Oil Spill Modelling | 21/03/2018 Page 18
REPORT
Figure 12 Modelled monthly wind rose distributions from 2003–2012 (inclusive), for the wind node closest to the release site. The colour key shows the wind magnitude, the compass direction provides the direction FROM and the length of the wedge gives the percentage of the record for a particular speed and direction combination.
MAQ0648J | Tamarind Resources - Tui Field | 110-Day Oil Spill Modelling | 21/03/2018 Page 19
REPORT
5 Water Temperature and Salinity
The monthly temperature and salinity profiles of the water column near the release site was obtained from the World Ocean Atlas 2013 database produced by the National Oceanographic Data Centre (National Oceanic and Atmospheric Administration) and its co-located World Data Center for Oceanography (see Levitus et al., 2013). Monthly average sea-surface temperatures near the release site were found to vary over the course of the year from a minimum of 13.0°C (August) to a maximum of 19.1°C (March) (Table 5). Monthly average salinity of the upper water column near the release site varied only slightly throughout the year from a minimum of 35.0 PSU (January and March) to a maximum of 35.3 PSU (July) (Table 5). To accurately represent the sea temperature and salinity throughout the whole water column the modelling used monthly average sea temperature and salinity profiles, as presented in Figure 13.
Table 5 Monthly average sea-surface temperature and salinity at the study area.
Month Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec
Temperature (°C) 17.8 19.1 19.1 17.6 16.2 14.9 14.4 13.0 13.6 13.9 14.9 16.1
Salinity (PSU) 35.0 35.1 35.0 35.2 35.2 35.2 35.3 35.2 35.1 35.1 35.1 35.2
MAQ0648J | Tamarind Resources - Tui Field | 110-Day Oil Spill Modelling | 21/03/2018 Page 20
REPORT
Figure 13 Monthly average sea temperature and salinity profiles at the study site.
MAQ0648J | Tamarind Resources - Tui Field | 110-Day Oil Spill Modelling | 21/03/2018 Page 21
REPORT
6 Oil Spill Model - SIMAP
The oil spill modelling was performed using SIMAP. SIMAP is designed to simulate the fate and effects of spilled hydrocarbons for both the surface and subsurface releases (Spaulding et al., 1994; French et al., 1999; French-McCay, 2003; French-McCay, 2004; French-McCay et al., 2004).
The SIMAP model calculates two components: (i) the transport, spreading, entrainment, evaporation and decay of surface oil slicks and, (ii) the entrained and dissolved hydrocarbons released from the slicks into the water column. Input specifications for oil-types include the density, viscosity, pour point, distillation curve (volume lost versus temperature) and the aromatic/aliphatic component ratios within given boiling point ranges.
The SIMAP trajectory model separately calculates the movement of the material that: (i) is on the water surface (as surface slicks), (ii) in the water column (as either entrained whole oil droplets or dissolved hydrocarbon), (iii) has stranded on shorelines, or (iv) that has precipitated out of the water column onto the seabed. The model calculates the transport of surface slicks from the combined forces exerted by surface currents and wind acting on the oil. Transport of entrained oil (oil that is below the water surface) is calculated using the currents only.
6.1 Stochastic Modelling As spills can occur during any set of wind and current conditions, SIMAP’s stochastic model was used to quantify the probability of exposure to the sea surface, in-water and shoreline contacts for a hypothetical spill scenario over a 10-year period. For this assessment, a total of 100 single spill trajectories were run for each hypothetical scenario. Each simulation had the same spill information (i.e. spill volume, duration and oil type) for each scenario but with varying start times, and in turn, the prevailing wind and current conditions. This approach ensures that the predicted transport and weathering of an oil slick is subject to a wide range of current and wind conditions. During each spill trajectory, the model records the grid cells exposed to hydrocarbons, as well as the time elapsed. Once all the spill trajectories have been run, the model then combines the results from the individual simulations to determine the following: Maximum exposure (or load) observed on the sea surface; Minimum time before sea surface exposure; Probability of contact to any shorelines; Probability of contact to individual sections of shorelines; Maximum volume of oil that may contact shorelines from a single simulation; Maximum load that an individual shoreline may experience; Maximum exposure from entrained hydrocarbons observed in the water column; and Maximum exposure from dissolved aromatic hydrocarbons observed in the water column.
The stochastic model output does not represent the extent of any one spill trajectory (which would be significantly smaller) but rather provides a summary of all trajectories run for the scenario.
MAQ0648J | Tamarind Resources - Tui Field | 110-Day Oil Spill Modelling | 21/03/2018 Page 22
REPORT
6.2 Sea surface, Shoreline and In-Water Thresholds
6.2.1 Sea surface Exposure Thresholds The SIMAP model is able to track hydrocarbons to levels lower than biologically significant or visible to the naked eye. Therefore, reporting thresholds have been specified (based on the scientific literature) to account for “exposure” on the sea surface and “contact” to shorelines at meaningful levels.
To better assess the potential for sea surface exposure, each of the 100 spill trajectories was tracked to a minimum of 0.5 g/m2, which equates approximately to an average thickness of ~0.5 μm. Oil of this thickness is described as a silvery to rainbow sheen in appearance, according to the Bonn Agreement Oil Appearance Code (Bonn Agreement 2009) (refer to Table 6) and is also considered the practical limit of observing oil in the marine environment (AMSA, 2012). This threshold is considered below levels which would cause environmental harm and it is more indicative of the areas perceived to be affected due to its visibility on the sea surface and potential to trigger temporary closures of areas (i.e. fishing grounds) as a precautionary measure. Hence, the 0.5 g/m2 threshold has been selected to define the zone of potential low exposure on the sea surface. Ecological impact has been estimated to occur at 10 g/m2 (~10 µm) according to French et al. (1996) and French-McCay (2009) (see references therein) as this level of oiling has been observed to mortally impact birds and other wildlife associated with the water surface. The 10 g/m2 threshold has been selected to define the zone of potential moderate exposure on the sea surface.
Scholten et al. (1996) and Koops et al. (2004) indicated that a concentration of surface oil equal to 25 g/m2 or greater would be harmful for all birds that contact the slick. Exposure to oil concentrations at or above this threshold is used to define the zone of potential high exposure.
Table 7 defines the thresholds used to classify the zones of sea surface exposure. Figure 14 shows photographs highlighting the difference in appearance between a silvery sheen, rainbow sheen and metallic sheen.
Table 6 The Bonn Agreement Oil Appearance Code
Layer Thickness Interval Code Description Appearance Litres per km2 (g/m2 or μm)
1 Sheen (silvery/grey) 0.04 – 0.30 40 – 300
2 Rainbow 0.30 – 5.0 300 – 5,000
3 Metallic 5.0 – 50 5,000 – 50,000
4 Discontinuous True Oil Colour 50 – 200 50,000 – 200,000
5 Continuous True Oil Colour 200 –> 200,000 –>
MAQ0648J | Tamarind Resources - Tui Field | 110-Day Oil Spill Modelling | 21/03/2018 Page 23
REPORT
Figure 14 Photographs showing the difference between oil colour and thickness on the sea surface (source: adapted from OilSpillSolutions.org 2015)
Table 7 Thresholds used to classify the zones of sea surface exposure
Oil concentration Zone description (g/m2)
0.5 - 10 Low
10 - 25 Moderate
> 25 High
6.2.2 Shoreline Contact Threshold
There are many different types of shorelines, ranging from cliffs, rocky beaches, sandy beaches, mud flats and mangroves, and each of these influences the volume of oil that can remain stranded ashore and its thickness before the shoreline saturation point occurs. For instance, a sandy beach may allow oil to percolate through the sand, thus increasing its ability to hold more oil ashore over tidal cycles and various wave actions than an equivalent area of water; hence oil can increase in thickness onshore over time. A sandy beach shoreline was assumed as the default shoreline type for the modelling herein, as it allows for the highest carrying capacity of oil (of the available open/exposed shoreline types). Hence the results contained herein would be indicative of a worst-case scenario, where the highest volume of oil may be stranded on the shoreline (when compared to other shoreline types, such as exposed rocky shores). French et al. (1996) and French-McCay (2009) have defined an oil exposure threshold for shorebirds and wildlife (furbearing aquatic mammals and marine reptiles) on or along the shore at 100 g/m2, which is based on studies for sub-lethal and lethal impacts. These thresholds have been used in previous environmental risk assessment studies (see French-McCay, 2003; French-McCay et al., 2004; French-McCay et al., 2011; NOAA, 2013). The 100 g/m2 threshold is also recommended in the Australian Maritime Safety Authority’s (AMSA) foreshore assessment guide1 as the acceptable minimum thickness that does not inhibit the potential for recovery and is best remediated by natural coastal processes alone (AMSA, 2007). The 100 g/m2 threshold has been selected to define the zone of potential moderate contact on the shorelines.
1 Recommended for shoreline types including sandy beach, boulder shorelines, pebble shorelines, rock platforms and industry facility structures.
MAQ0648J | Tamarind Resources - Tui Field | 110-Day Oil Spill Modelling | 21/03/2018 Page 24
REPORT
Observations by Lin and Mendelssohn (1996), demonstrated that loadings of more than 1,000 g/m2 of oil during the growing season would be required to impact marsh plants significantly. Similar thresholds have been found in studies assessing oil impacts on mangroves (Grant et al., 1993; Suprayogi and Murray, 1999). The 1,000 g/m2 threshold has been selected to define the zone of potential high contact on the shorelines. Oil contact between 10 and 100 g/m2 represents the socio-economic (or low contact) threshold. The following thresholds (see Table 8) have therefore been derived to classify the shoreline contact.
Table 8 Thresholds use to assess shoreline contact
Shoreline concentration Zone description (g/m2)
10–100 Low
100-1,000 Moderate
> 1,000 High
6.2.3 Water Column Exposure Thresholds Sub-surface exposure to submerged habitats is better represented by estimates for entrained or dissolved hydrocarbons in the water column. Studies indicate that the dissolved aromatic compounds (typically the mono-aromatic hydrocarbons and the two and three ring poly-aromatic hydrocarbons) are commonly the largest contributor to the toxicity of solutions generated by mixing oil into water (Di Toro et al., 2007). The exposure level (threshold concentration over a given duration) was used to assess the potential for exposure to sub-sea habitats and species by entrained and dissolved aromatic hydrocarbons. The threshold value for species toxicity in the water column is based on global data from French et al. (1999) and French-McCay (2002, 2003), which showed that species sensitivity (fish and invertebrates) to dissolved aromatics exposure > 4 days (96-hour LC50) under different environmental conditions varied from 6 to 400 μg/l (ppb) with an average of 50 ppb. This range covered 95% of aquatic organisms tested, which included species during sensitive life stages (eggs and larvae). Based on scientific literature, a minimum threshold of 6 parts per billion (ppb) over 96-hours or equivalent was used to assess in-water low exposure zones (Engelhardt, 1983; Clark, 1984; Geraci and St. Aubin, 1988; Jenssen, 1994; Tsvetnenko, 1998). French-McCay, 2002 indicates that an average 96-hour LC50 of 50 ppb and 400 ppb could serve as an acute lethal threshold to 5% and 50% to biota, respectively. Hence, the thresholds were used to represent the moderate and high exposure zones, respectively. Given that the dissolved aromatics component of hydrocarbons in the water column are accounted for by the thresholds defined above, the environmental effects of the remaining undissolved hydrocarbons, essentially the entrained hydrocarbons in the water column, require different exposure thresholds. Considering that entrained oil has undergone processes analogous to weathering and/or water-washing (i.e., many of the toxic soluble hydrocarbons have been removed through evaporation and/or dissolution), its toxicity is representative of true ‘dispersed oil’ phase impacts. OSPAR (2012) has published predicted no effect concentrations (PNEC) for ‘dispersed oil’ in produced formation water (PFW) discharges. Dispersed oil in PFW discharges are small, discrete droplets suspended in the discharged water which are very similar to insoluble dispersed oil droplets formed from subsea blowouts. In essence, the oil has been partitioned
MAQ0648J | Tamarind Resources - Tui Field | 110-Day Oil Spill Modelling | 21/03/2018 Page 25
REPORT
(naturally separated) from gas/oil/water mixture by solubility (water washing) and vapour pressure (evaporation) based on the individual hydrocarbon chemical properties. The OSPAR PNEC for PFW is 70 ppb for protection of 95% of species, based on biomarker testing (i.e. whole organism responses) to total hydrocarbons (THC) by Smit et al., 2009. This PNEC represents an acceptable long term chronic exposure level from continuous point source discharges in the North Sea, which is one of the most concentrated areas in the world for oil and gas production. Appropriate threshold values can be extrapolated from the NOECs examined in Smit et al., 2009 based on effects ranging from oxidative stress to impacts on growth, reproduction and survival and are represented by: 7 µg/l (7ppb) (for 1% affected fraction of species), 70.5µg/l (70ppb) (for 5% affected fraction of species) and 804 µg/l (804 ppb) (for 50% affected fraction of species). Utilising methodologies contained in ANZECC (2000), which is based upon USEPA Guidelines, PNECs can be back-calculated to determine LC50 values by applying a factor of 100 to the PNEC values. This approach is supported by assessment factor criteria contained within the European Chemicals Agency (2008) and the OECD Existing Chemicals Programme 2002 (OECD, 2002). Employing these criteria, the following conservative threshold values for entrained hydrocarbons are applied:
LC50 (99% species protection): 700 µg/l (ppb)
LC50 (95% species protection): 7,050 µg/l (ppb); and
LC50 (50% species protection): 80,400 µg/l (ppb). The final exposure thresholds adopt a continuous exposure criterion of 96 hours. Table 9 and Table 10 provide a summary of the dissolved aromatic and entrained hydrocarbon threshold values used to define different levels of potential exposure in the modelling study.
Table 9 Dissolved aromatic threshold values applied as part of the modelling study
Threshold value for dissolved Equivalent exposure of Range of sensitive species Potential aromatic concentrations for a dissolved aromatics potentially impacted from level of 96-hour LC50 (ppb) over 96 hrs (ppb.hrs) acute exposure exposure
6 576 Very sensitive species Low
50 4,800 Average sensitive species Moderate
400 38,400 Tolerant sensitive species High
Table 10 Entrained hydrocarbon threshold values applied as part of the modelling study. Thresholds based on OSPAR guidelines
Threshold value for entrained Equivalent exposure of Range of sensitive species Potential hydrocarbon concentrations entrained hydrocarbons potentially impacted from level of for a 96-hour LC50 (ppb) over 96 hrs (ppb.hrs) acute exposure exposure
700 67,200 Very sensitive species Low
7,050 676,800 Average sensitive species Moderate
80,400 7,718,400 Tolerant sensitive species High
MAQ0648J | Tamarind Resources - Tui Field | 110-Day Oil Spill Modelling | 21/03/2018 Page 26
REPORT
6.3 Exposure Calculation The thresholds used for the in-water concentrations of entrained hydrocarbons were described as an exposure over time, rather than an instantaneous peak value. The exposure is expressed as a concentration multiplied by the number of hours exposed at that concentration, in the units of parts per billion multiplied by hours (ppb.hrs). There are two important and opposing mechanisms that will affect the cumulative exposure. The rate of uptake, due to the exposure concentration and the duration of exposure, and the rate of removal due to the ability of the organism to expel or metabolise hydrocarbons; a process referred to as depuration. Calculation for these natural removal processes are important so as to avoid falsely forecasting impacts by only allowing for the uptake of hydrocarbons, particularly over long duration release simulations. The uptake of entrained hydrocarbons was calculated over time for each model grid cell by addition of the concentrations calculated at each subsequent time step, multiplied by the time interval (typically hourly). Depuration was calculated by applying an exponential decay function to the previously accumulated exposure. A review of the literature describing the observed rates of depuration of hydrocarbons indicates that the reduction of concentration follows an exponential decay. For sub-lethal concentrations, depuration rates will be faster with increased concentration and then decrease as concentrations approach zero. Hence, depuration of the concentrations in a cell over each time step was calculated by applying an exponential decay function to the tissue concentration calculated by uptake. Observed rates of depuration show significant variation for different soluble hydrocarbons and different organisms, varying from a few days to a few weeks (Solbakken et al., 1984). For this study, the decay coefficient was set so that the exposure would fall to 1% of an initial concentration over 1 week, given no further exposure. Cumulative exposures at each time step were then compared to threshold exposures and any location where the exposure thresholds were ever exceeded during any simulation was mapped. To illustrate the effect of allowing for depuration of hydrocarbons over time, an example time-series plot of concentration and exposure at a receptor location is presented in Figure 15. The time-series of concentration shows intermittent contacts to hydrocarbons with gaps between contacts, of the order of ~5 days. Such an outcome might be expected, for example, from variation in the position of hydrocarbon plumes resulting from variations in the current field during an ongoing discharge. The lower panel shows the calculated exposure if depuration is not considered (blue line) and the exposure where an exponential depuration rate is allowed for, assuming a time-scale of 7 days for tissue concentrations to reduce to 1% of a starting concentration. In the case where depuration was ignored (blue line), the calculated exposure simply increases with addition of each hydrocarbon concentration and remains constant during times where there is no contact to hydrocarbons. In the case where depuration is allowed for (green line), exposure decreases exponentially between exposures. The maximum exposure occurs around day 50, where there was a prolonged (~5 days) exposure to a high concentration. The threshold exposure is not exceeded by the intermittent exposures to low concentrations but is exceeded at around day 48 when higher concentrations occurred for a longer duration.
MAQ0648J | Tamarind Resources - Tui Field | 110-Day Oil Spill Modelling | 21/03/2018 Page 27
REPORT
Figure 15 Example time series plot of concentration of entrained hydrocarbons (top) at a receptor, and exposure (bottom) with no depuration (blue) and with depuration (green)
MAQ0648J | Tamarind Resources - Tui Field | 110-Day Oil Spill Modelling | 21/03/2018 Page 28
REPORT
7 Oil Properties
For this oil spill modelling assessment Tui Crude was used as the oil type for the loss of well control scenario. Tui crude has a density of 808 kg/m3 (API gravity of 43.5) and a dynamic viscosity of 2.5 cP at 40ºC, and although it is classified as a Group II oil according to the International Tankers Owners Pollution Federation (ITOPF, 2014) classifications, it does have a high wax content of 17.3%, which will result in this oil turning in to wax after extended periods (weeks) in the marine environment. Figure 16 illustrates the weathering graphs a 19,803 bbl sub-surface release of Tui crude over 24 hours, tracked for 75 under 3 static wind conditions. The graphs illustrate greater persistence of the oil on the sea surface with decreasing wind speeds, which correlates to increasing volumes of oil occurring in the water column with increasing wind speeds. Additionally, evaporative losses were greatest during the first day, while the oil was still being released, and then the loss rate reduced until at day 4 evaporative losses were negligible. Table 11 and Table 12 show the physical characteristics and boiling point ranges for crude oil used in this study.
Table 11 Physical characteristics.
Characteristic Tui Crude
Density (kg/m3) 808 @ 16˚C API 43.5 Dynamic viscosity (cP) 2.5 @ 40°C Pour Point (ºC) 24 Oil Property Category Group II
Table 12 Boiling point ranges.
Volatiles Semi-volatiles Low volatiles Residual Characteristic (%) (%) (%) (%)
Boiling point (°C) <180 180 – 265 265 – 380 >380 Tui Crude 28 17 26 29 Non-persistent Persistent
MAQ0648J | Tamarind Resources - Tui Field | 110-Day Oil Spill Modelling | 21/03/2018 Page 29
REPORT
Figure 16 Weathering of Tui Crude under three static wind conditions. The results are based on a 19,803 bbl sub-surface release of Tui Crude over 24 hours, tracked for 75 days.
MAQ0648J | Tamarind Resources - Tui Field | 110-Day Oil Spill Modelling | 21/03/2018 Page 30
REPORT
8 Model Settings and Assumptions
The oil spill modelling study quantified the potential exposure to the surrounding waters and shorelines from a hypothetical, yet plausible, scenarios, a subsea loss of well control for 110 days at Amokura-2H. The assessment was completed for the months of February to May, which covers the planned operational months of March and April. Table 13 provides a summary of the oil spill model settings and assumptions
Table 13 Summary of the oil spill model settings used in this assessment.
Parameter Sub-surface loss of well control at Amokura-2H
Number of randomly selected 100 spill start times
Model Period Operational Period (February to May)
Oil Type Tui Crude
Spill Volume 654,516 STB (104,068 m3)
Release Rate Variable (see Figure 17) (Day 1 – 19,803 bbl/day and Day 110 – 4,092 bbl/day)
Release Depth 122.5 m
Release Duration 110 days
Simulation length 140 days
Surface oil concentration 0.5, 10 and 25 thresholds (g/m2)
Shoreline load threshold (g/m2) 10, 100 and 1,000
Dissolved aromatic dosages to 576 (6 ppb x 96 hrs, potential low exposure) assess the potential exposure 4,800 (50 ppb x 96 hrs, potential moderate exposure) (ppb.hrs) 38,400 (400 ppb x 96 hrs, potential high exposure)
Entrained oil dosages to assess 67,200 (700 ppb x 96 hrs, potential low exposure) the potential exposure (ppb.hrs) 676,800 (7,050 ppb x 96 hrs, potential moderate exposure) 7,718,400 (80,400 ppb x 96 hrs, potential high exposure)
MAQ0648J | Tamarind Resources - Tui Field | 110-Day Oil Spill Modelling | 21/03/2018 Page 31
REPORT
Figure 17 Plot of the variable release rate over the 110-day period
MAQ0648J | Tamarind Resources - Tui Field | 110-Day Oil Spill Modelling | 21/03/2018 Page 32
REPORT
9 Interpreting Modelling Results
The results from the modelling study are presented in a number of tables and figures, which aim to provide an understanding of both the predicted sea surface exposure, shoreline contact and in-water exposure for the scenario.
9.1 February to May Stochastic Assessment The figures are based on the following principles: The potential zones of exposure (surface oil, entrained hydrocarbons and dissolved aromatics) – is determined by identifying the maximum loading (surface) or dosage (subsea) within a grid cell, and is then classified according to identified surface or subsea thresholds. The minimum time before oil exposure on the sea surface – is determined by recording the elapsed time before sea surface exposure to a grid cell, at a specified threshold. The probability of exposure/contact (surface oil, shoreline oil, entrained hydrocarbon or dissolved aromatic) – is calculated by dividing the number of spill trajectories passing over that given cell (surface, shoreline or subsea) by the total number of spill trajectories, above the specified threshold value. Maximum potential shoreline loading – is determined by identifying the maximum loading within a shoreline cell, and is then classified according to the identified thresholds (i.e. 100 g/m2 and 1,000 g/m2). The statistics are based on the following principles: The greatest distance travelled by a spill trajectory – is determined by: a) recording the maximum distance travelled by a single trajectory, within a scenario, from the release site to the identified exposure thresholds; and then b) report the greatest distance travelled by the 99th percentile spill trajectory (or second highest distance travelled by a single spill trajectory), along with the corresponding direction of travel from the release site. The probability of shoreline contact – is determined by recording to the number of spill trajectories to contact the shoreline, at a specific threshold, divided by the total number of spill trajectories within that scenario. The minimum time before oil exposure – is determined by recording the minimum time for a grid cell to record exposure, at a specific threshold. The average volume of oil ashore for a single spill – is determined by calculating the average volume of the all the single spill trajectories which were predicted to make shoreline contact within a scenario. The maximum volume of oil ashore from a single spill trajectory – is determined by identifying the single spill trajectory within a scenario/season, that recorded the maximum volume of oil to come ashore and presenting that value. The average length of shoreline contacted by oil - is determined by calculating the average of the length of shoreline (measured as grid cells) contacted by oil above a specified threshold. The maximum length of shoreline contacted by oil - is determined by recording the maximum length of shoreline (measured as grid cells) contacted by oil above a specified threshold. The probability of oil exposure to a receptor – is determined by recording the number of spill trajectories
MAQ0648J | Tamarind Resources - Tui Field | 110-Day Oil Spill Modelling | 21/03/2018 Page 33
REPORT
to reach a specified sea surface or subsea threshold within a receptor polygon, divided by the total number of spill trajectories within that scenario. The minimum time before oil exposure to a receptor– is determined by ranking the elapsed time before sea surface exposure, at a specified threshold, to grid cells within a receptor polygon and recording the minimum value. The probability of oil contact to a receptor– is determined by recording the number of spill trajectories to reach a specified shoreline contact threshold within a receptor polygon, divided by the total number of spill trajectories within that scenario. The minimum time before shoreline contact to a receptor – is determined by ranking the elapsed time before shoreline contact, at a specified threshold, to grid cells within a receptor polygon and recording the minimum value. The average potential oil loading within a receptor – is determined taking the average of the maximum loading to any grid cell within a polygon, for all simulations within a scenario/season, that recorded shoreline The maximum potential oil loading within a receptor – is determined by identifying the maximum loading to any grid cell within a receptor polygon, for a scenario. The average volume of oil ashore within a receptor – is determined by calculating the average volume of oil to come ashore within a receptor polygon, from all the single spill trajectories which were predicted to make shoreline contact within a scenario. The maximum volume of oil ashore within a receptor – is determined by recording the maximum volume of oil to come ashore within a receptor polygon, from all the single spill trajectories which were predicted to make shoreline contact within a scenario. The average length of shoreline contacted within a receptor is determined by calculating the average of the length of shoreline (measured as grid cells) contacted by oil within a receptor polygon, at a specified threshold, from all the single spill trajectories which were predicted to make shoreline contact within a scenario. The maximum length of shoreline contacted by oil is determined by recording the maximum length of shoreline (measured as grid cells) contacted by oil within a receptor polygon, at a specified threshold, from all the single spill trajectories which were predicted to make shoreline contact within a scenario
MAQ0648J | Tamarind Resources - Tui Field | 110-Day Oil Spill Modelling | 21/03/2018 Page 34
REPORT
9.2 Receptors Assessed
Figure 18 shows the Marine Reserves, Marine Mammal Sanctuaries and Regional Shorelines used in the modelling to assess for oil exposure.
Figure 18 Receptors assessed for oil exposure.
MAQ0648J | Tamarind Resources - Tui Field | 110-Day Oil Spill Modelling | 21/03/2018 Page 35
REPORT
10 Results: Amokura-2H Loss of Well Control
This scenario examined a hypothetical 110-day subsea release of 654,516 STB of Tui Crude at a variable rate (see Figure 17), following a loss of well control incident at the Amokura-2H well. A total of 100 spill trajectories were simulated over the defined period of February to May.
10.1 Stochastic Analysis Section 10.1.1 presents the potential exposure to the sea surface and shoreline contact, while Section 10.1.2 presents potential in-water exposure. For the modelling study each spill trajectory was tracked to the following minimum thresholds: Visible sea surface oil – 0.5 g/m2 Shoreline oil contact – 10 g/m2 Dissolved aromatics – 576 ppb.hrs Entrained hydrocarbons – 67,200 ppb.hrs
10.1.1 Sea Surface Exposure and Shoreline Contact Figure 19 presents the zones of potential oil exposure on the sea surface for the modelled period. Zones of potential low exposure (0.5-10 g/m2) were predicted to extend northeast, southeast and west, while zones of moderate (10-25 g/m2) and high (>25 g/m2) exposure predominantly extended south-southeast of the release site and southeast of the release sites. Table 14 details the maximum distance travelled by oil on the sea surface at each threshold. The maximum distance from a release site for the potential zone of low, moderate and high exposure were 395 km (north- west), 63 km (east - northeast) and 46 km (south-southeast). Shoreline contact for any shoreline is summarized in Table 15. The probability of oil coming ashore was 100%, while the quickest time to shore for visible sheens was 56 hours and the maximum volume was 778 m3. The maximum lengths of shoreline exposure at low, moderate and high levels were 237 km, 104 km and 20 km, respectively. Figure 20 to Figure 22 presents the probability of oil exposure on the sea surface reported at the low, moderate and high threshold for the modelling assessment period. Additionally, Figure 23 to Figure 25 present the minimum time before observing oil on the sea surface at the low, moderate and high thresholds. Table 16 summarises potential sea surface exposure to receptors during the model assessment period. Both the South Taranaki shoreline and Marine Mammal Sanctuary had 100% and New Plymouth had 99% probability of being exposed to visible floating oil. However, only South Taranaki and New Plymouth shorelines had any probability of being exposed to surface oil of moderate thickness (10-25 g/m2). Table 17 presents the potential shoreline contact for individual regional shoreline receptors. Again, South Taranaki regional shoreline had the highest probability of contact, for all thresholds. Shoreline accumulation of greater than 100 g/m2 occurs within 65 hours for this section of coastline, eventually covering up to 65 km at this level. Some sections of the South Taranaki regional shoreline accumulated up to 3.091 kg/m2, while New Plymouth had a peak value of 2.815 kg/m2.
MAQ0648J | Tamarind Resources - Tui Field | 110-Day Oil Spill Modelling | 21/03/2018 Page 36
REPORT
Table 14 Summary of potential zones of sea surface exposure at each surface oil threshold from a 110-day loss of well control.
Zones of potential sea surface exposure Period Distance and direction Low Moderate High (0.5–10 g/m2) (10–25 g/m2) (>25 g/m2)
Max. distance from release site (km) 395 63 46
Max. distance from release site (km) February - 296 56 42 May (99th percentile)
Direction North-west East-Northeast South-Southeast
Table 15 Summary of shoreline contact at or above 10 g/m2 across all shorelines from a 110-day loss of well control.
Shoreline statistics February – May Period
Probability of contact to any shoreline (%) 100
Absolute minimum time for visible oil to shore (hours) 56
Maximum volume of hydrocarbons ashore (m3) 778
Average volume of hydrocarbons ashore (m3) 308
Maximum length of the shoreline at 10 g/m2 (km) 375
Average shoreline length (km) at 10 g/m2 (km) 183
Maximum length of the shoreline at 100 g/m2 (km) 237
Average shoreline length (km) at 100 g/m2 104
Maximum length of the shoreline at 1,000 g/m2 20
Average shoreline length (km) at 1,000 g/m2 (km) 5
MAQ0648J | Tamarind Resources - Tui Field | 110-Day Oil Spill Modelling | 21/03/2018 Page 37
REPORT
Table 16 Summary of the potential sea surface exposure to receptors from a 110-day loss of well control.
Probability of oil exposure on Minimum time before oil the sea surface exposure on the sea surface (%) (hours) Receptors Low Moderate High Low Moderate High (0.5–10 (10–25 (>25 (0.5–10 (10–25 (>25 g/m2) g/m2) g/m2) g/m2) g/m2) g/m2)
Mana Island 11 0 0 551 - -
Kapiti Island 23 0 0 332 - -
Wanganui 80 0 0 168 - -
Marlborough 2 0 0 591 - -
South Taranaki 100 3 0 56 212 -
New Plymouth 99 1 0 91 252 -
Waitomo 52 0 0 202 - -
Otorohanga 4 0 0 323 - -
Waikato 8 0 0 280 - -
Wellington 5 0 0 545 - -
Kapiti Coast 47 0 0 247 - -
Horowhenua 68 0 0 235 - -
Manawatu 43 0 0 214 - -
Rangitikei 71 0 0 174 - -
Porirua 10 0 0 908 - -
Franklin 11 0 0 390 - -
Parininihi Marine Reserve 17 0 0 194 - -
Kapiti Marine Reserve 19 0 0 277 - -
Tapuae Marine Reserve 72 0 0 153 - -
Marine Mammal Sanctuary - Taranaki 100 0 0 35 - -
Marine Mammal Sanctuary - Waikato 88 0 0 142 - -
Marine Mammal Sanctuary - Auckland 15 0 0 386 - -
Marine Mammal Sanctuary - Northland 1 0 0 722 - -
MAQ0648J | Tamarind Resources - Tui Field | 110-Day Oil Spill Modelling | 21/03/2018 Page 38
REPORT
Table 17 Summary of shoreline contact to individual shoreline receptors from a 110-day loss of well control.
Maximum probability Minimum time before Load on Mean length of Maximum length of Minimum Volume on of shoreline loading shoreline shoreline shoreline contacted shoreline contacted time before shoreline (m3) (%) accumulation (hours) (g/m2) (km) (km) visible sea surface >10 >100 >1,000 >10 >100 >1,000 >10 >100 >1,000 >10 >100 >1,000 exposure Shoreline Mean Peak Mean Peak Receptor g/m2 g/m2 g/m2 g/m2 g/m2 g/m2 g/m2 g/m2 g/m2 g/m2 g/m2 g/m2 (day) Mana Island 8 8 0 661 1,223 - 146 286 6 24 2 2 - 3 2 - 23 Kapiti Island 23 18 0 335 413 - 160 791 42 86 6 5 - 13 9 - 14 Wanganui 79 69 0 170 187 - 90 558 39 60 10 4 - 33 19 - 7 Marlborough 2 0 0 584 - - 60 79 1 7 2 - - 3 0 - 25 South Taranaki 100 100 43 59 65 146 220 3,091 566 1,418 104 65 5 156 101 16 2 New Plymouth 98 97 18 127 147 254 156 2,815 220 329 32 16 2 97 55 4 4 Waitomo 50 37 10 199 242 1,110 126 1,914 115 117 11 5 1 54 21 3 8 Otorohanga 3 2 0 325 336 - 60 141 3 17 4 1 - 5 1 - 13 Waikato 8 3 0 281 583 - 68 266 5 15 4 1 - 11 2 - 12 Wellington 5 4 0 543 592 - 96 198 4 16 3 1 - 6 2 - 23 Kapiti Coast 45 41 0 247 307 - 133 852 125 136 13 8 - 40 22 - 10 Horowhenua 66 62 18 236 267 379 159 1,999 158 220 18 10 2 42 36 4 10 Manawatu 46 26 0 199 267 - 109 488 57 77 8 7 - 14 14 - 9 Rangitikei 71 59 1 170 269 2,505 139 1,356 130 137 12 8 2 31 26 - 7 Porirua 11 8 0 660 677 - 92 410 16 48 6 4 - 13 6 - 38 Franklin 11 9 0 380 434 - 84 439 34 98 16 6 - 33 11 - 5
MAQ0648J | Tamarind Resources - Tui Field | 110-Day Oil Spill Modelling | 21/03/2018 Page 39
REPORT
Figure 19 Zones of potential exposure on the sea surface from a 110-day loss of well control.
MAQ0648J | Tamarind Resources - Tui Field | 110-Day Oil Spill Modelling | 21/03/2018 Page 40
REPORT
Figure 20 Probability of oil exposure on the sea surface above low exposure (≥0.5 g/m2) from a 110-day loss of well control.
MAQ0648J | Tamarind Resources - Tui Field | 110-Day Oil Spill Modelling | 21/03/2018 Page 41
REPORT
Figure 21 Probability of oil exposure on the sea surface above moderate exposure (≥10 g/m2) from a 110-day loss of well control.
MAQ0648J | Tamarind Resources - Tui Field | 110-Day Oil Spill Modelling | 21/03/2018 Page 42
REPORT
Figure 22 Probability of oil exposure on the sea surface above high exposure (≥25 g/m2) from a 110-day loss of well control.
MAQ0648J | Tamarind Resources - Tui Field | 110-Day Oil Spill Modelling | 21/03/2018 Page 43
REPORT
Figure 23 Minimum time before oil exposure on the sea surface above low exposure (≥0.5 g/m2) from a 110-day loss of well control.
MAQ0648J | Tamarind Resources - Tui Field | 110-Day Oil Spill Modelling | 21/03/2018 Page 44
REPORT
Figure 24 Minimum time before oil exposure on the sea surface above moderate exposure (≥10 g/m2) from a 110-day loss of well control.
MAQ0648J | Tamarind Resources - Tui Field | 110-Day Oil Spill Modelling | 21/03/2018 Page 45
REPORT
Figure 25 Minimum time before oil exposure on the sea surface above high exposure (≥25 g/m2) from a 110-day loss of well control.
MAQ0648J | Tamarind Resources - Tui Field | 110-Day Oil Spill Modelling | 21/03/2018 Page 46
REPORT
Figure 26 Maximum potential shoreline loading from a 110-day loss of well control.
MAQ0648J | Tamarind Resources - Tui Field | 110-Day Oil Spill Modelling | 21/03/2018 Page 47
REPORT
10.1.2 In-Water Exposure Dissolved aromatics leeching away from the entrained oil while in the water column resulted in zones of potential low and moderate exposure in the top 30 m of the water column. There was no exposure above thresholds of concern below 30 m in the water column. Table 18 provides a summary of the dissolved aromatics probability of low, moderate and high exposure to receptors at each depth interval while Figure 27 shows the extent of the zones of potential low and moderate exposure from 0-10 m below the sea surface. The highest exposure in the water column occurred at adjacent to the New Plymouth regional shoreline. Table 19 provides a summary of the probability of low, moderate and high exposures to marine based receptors from entrained hydrocarbons at 0-10 m below the sea surface. Zones of potential entrained hydrocarbon exposure were much smaller than the dissolved aromatics, reaching a maximum distance of approximately 30 km from the release site. The low exposure zone was also limited to the upper 10 m of the water column. Some shoreline areas were shown to have low exposure to entrained hydrocarbons, these were likely to occur from floating oil trapped in the coastal zone being forced in to the water column during a strong wind event.
MAQ0648J | Tamarind Resources - Tui Field | 110-Day Oil Spill Modelling | 21/03/2018 Page 48
REPORT
Table 18 Probability of low, moderate and high exposure to receptors at 0–10 m, 10 - 20m and 20 - 30m below the sea surface from dissolved aromatics from a 110-day loss of well control.
Dissolved Aromatic Exposure
0 – 10 m 10 – 20 m 20 – 30 m
Maximum Probability of exposure to Maximum Probability of exposure to Maximum Probability of exposure to Receptor exposure dissolved aromatics (ppb.hrs) exposure dissolved aromatics (ppb.hrs) exposure dissolved aromatic (ppb.hrs) to to to dissolved Low Moderate High dissolved Low Moderate High dissolved Low Moderate High aromatics (576 (4,800 (38,4000 aromatics (576 (4,800 (38,4000 aromatics (576 (4,800 (38,4000 (ppb.hrs) pph.hrs) ppb.hrs) ppb.hrs) (ppb.hrs) pph.hrs) ppb.hrs) ppb.hrs) (ppb.hrs) pph.hrs) ppb.hrs) ppb.hrs) Mana Island - - - - 577 1 ------Kapiti Island 1,749 6 - - 1,292 3 - - 582 1 - - Wanganui 1,287 2 ------0 0 - - South Taranaki 6,139 55 2 - 2,318 9 - - 937 1 - - New Plymouth 3,081 34 - - 1,936 9 - - 693 1 - - Hutt City 800 2 - - 1,002 2 - - 596 1 - - Kapiti Coast - - - - 643 1 - - 1,028 1 - - Horowhenua 1,758 6 - - 1,034 3 ------Manawatu 2,745 14 - - 2,818 5 ------Rangitikei 3,667 9 - - 1,435 2 ------Porirua 4,616 8 - - 932 ------Kapiti Marine Reserve 615 1 ------Tapuae Marine 1,975 15 - - 1,334 9 ------Reserve Marine Mammal 7,614 50 1 - 4,011 23 - - 2,601 8 - - Sanctuary - Taranaki Marine Mammal 1,302 2 - - 1,501 2 - - 889 1 - - Sanctuary - Waikato
MAQ0648J | Tamarind Resources - Tui Field | 110-Day Oil Spill Modelling | 21/03/2018 Page 49
REPORT
Figure 27 Zones of potential dissolved aromatic exposure at 0-10 m below the sea surface from a 110-day loss of well control.
MAQ0648J | Tamarind Resources - Tui Field | 110-Day Oil Spill Modelling | 21/03/2018 Page 50
REPORT
Table 19 Probability of low, moderate and high exposure to marine based receptors from entrained hydrocarbons at 0-10 m below the sea surface from a 110-day loss of well control.
0 – 10 m Receptor Maximum exposure to Probability of exposure to entrained hydrocarbons (ppb.hrs) entrained hydrocarbons (ppb.hrs)) Low (67,200 pph.hrs) Moderate (676,800 ppb.hrs) High (77,184 pph.hrs) South Taranaki 70,048 1 0 0 New Plymouth 83,035 1 0 0 Kapiti Coast 0 0 0 0 Horowhenua 78,485 3 0 0
MAQ0648J | Tamarind Resources - Tui Field | 110-Day Oil Spill Modelling | 21/03/2018 Page 51
REPORT
11 References
American Society for Testing and Materials (ASTM) 2013. F2067-13 Standard Practice for Development and Use of Oil-Spill Trajectory Models, ASTM International, West Conshohocken (PA).
Australian Maritime Safety Authority (AMSA) 2007, ‘Foreshore Assessment, Termination of Clean-up and Rehabilitation Monitoring’, viewed 12 February 2014, https://www.amsa.gov.au/environment/maritime- environmental-emergencies/national- plan/ESC/documents/Foreshore_Assessment_and_Termination.pdf
Australian Maritime Safety Authority (AMSA) 2012, ‘Australian Maritime Safety Authority Technical Guideline for the Preparation of Marine Pollution Contingency Plans for Marine and Coastal Facilities Australian Maritime Safety Authority’, viewed 15 January 2015, https://www.amsa.gov.au/forms-and- publications/Publications/AMSA413_Contingency_Planning_Guidelines.pdf
Andersen, OB 1995, ‘Global ocean tides from ERS 1 and TOPEX/POSEIDON altimetry’, Journal of Geophysical Research: Oceans, vol. 100, no. C12, pp. 25249–25259.
Australian and New Zealand Environment and Conservation Council (ANZECC), 2000. Australian and New Zealand guidelines for fresh and marine water quality. Volume 1, The guidelines (National water quality management strategy; no.4). Australian and New Zealand Environment and Conservation Council, Agriculture and Resource Management Council of Australia and New Zealand.
Becker, JJ, Sandwell, DT, Smith, WHF, Braud, J, Binder, B, Depner, J, Fabre, D, Factor, J, Ingalls, S, Kim, S- H, Ladner, R, Marks, K, Nelson, S, Pharaoh, A, Trimmer, R, Von Rosenberg, J, Wallace, G & Weatherall, P 2009, ‘Global bathymetry and evaluation data at 30 arc seconds resolution: SRTM30_PLUS’, Marine Geodesy, vol. 32, no. 4, pp. 355–371.
Bonn Agreement 2009, Bonn Agreement aerial operations handbook, 2009 - Publication of the Bonn Agreement, London, viewed 13 January 2015, http://www.bonnagreement.org/site/assets/files/3947/ba- aoh_revision_2_april_2012.pdf
Bowman, MJ, Kibblewhite, AC, Murtagh, RA, Chiswell, SM & Sandersin, BG 1983, ‘Circulation and mixing in greater Cook Straight, New Zealand’, Oceanologica Acta, vol. 6, no. 4, pp. 383–391.
Brodie, JW 1960, ‘Coastal surface currents around New Zealand’, New Zealand Journal of Geology and Geophysics, vol. 3, no. 2, pp. 235–252.
Chassignet, EP, Hurlburt, HE, Smedstad, OM, Halliwell, GR, Hogan, PJ, Wallcraft, AJ, Baraille, R & Bleck, R 2007, ‘The HYCOM (hybrid coordinate ocean model) data assimilative system’, Journal of Marine Systems, vol. 65, no. 1, pp. 60–83.
MAQ0648J | Tamarind Resources - Tui Field | 110-Day Oil Spill Modelling | 21/03/2018 Page 52
REPORT
Chassignet, E, Hurlburt, H, Metzger, E, Smedstad, O, Cummings, J & Halliwell, G 2009, ‘U.S. GODAE: Global Ocean Prediction with the HYbrid Coordinate Ocean Model (HYCOM)’, Oceanography, vol. 22, no. 2, pp. 64–75.
Clark, RB, 1984 ‘Impact of oil pollution on seabirds’, Environmental Pollution, vol. 33, no. 1, pp. 1–22.
Coleman, R 1984, ‘Investigation of the Tasman Sea by satellite altimetry’, Australian Journal of Marine and Freshwater Research, vol. 35, no. 6, pp. 619–633.
Davies, AM 1977a, ‘The numerical solutions of the three-dimensional hydrodynamic equations using a B- spline representation of the vertical current profile’, in JC Nihoul (ed), Bottom Turbulence: Proceedings of the 8th Liège Colloquium on Ocean Hydrodynamics, Elsevier Scientific, Amsterdam, pp. 1–25.
Davies, AM 1977b, ‘Three-dimensional model with depth-varying eddy viscosity’, in JC Nihoul (ed), Bottom Turbulence: Proceedings of the 8th Liège Colloquium on Ocean Hydrodynamics, Elsevier Scientific, Amsterdam, pp. 27–48.
Di Toro, DM, McGrath, JA & Stubblefield, WA 2007, ‘Predicting the toxicity of neat and weathered crude oil: Toxic potential and the toxicity of saturated mixtures’, Environmental Toxicology and Chemistry, vol. 26, no. 1, pp. 24–36.
Engelhardt, FR 1983, ‘Petroleum effects on marine mammals’, Aquatic Toxicology, vol. 4, no. 3, pp. 199– 217.
European Chemicals Agency 2008, Chapter R.10 - Characterisation of dose [concentration] -response for environment. In Guidance on information requirements and chemcial safety assessment (pp. 26-29). ECHA.
French, D, Reed, M, Jayko, K, Feng, S, Rines, H, Pavignano, S, Isaji, T, Puckett, S, Keller, A, French III, FW, Gifford, D, McCue, J, Brown, G, MacDonald, E, Quirk, J, Natzke, S, Bishop, R, Welsh, M, Phillips, M, Ingram, BS 1996, The CERCLA Type A natural resource damage assessment model for coastal and marine environments (NRDAM/CME), Technical Documentation, Volume I - Model Description, Final Report, Office of Environmental Policy and Compliance, U.S. Department of the Interior, Washington DC.
French, D, Schuttenberg, H & Isaji, T 1999, ‘Probabilities of oil exceeding thresholds of concern: examples from an evaluation for Florida Power and Light’, Proceedings of the 22nd Arctic and Marine Oil Spill Program (AMOP) Technical Seminar, Environment Canada, Alberta, pp. 243–270.
French-McCay, DP 2002, ‘Development and application of an oil toxicity and exposure model, OilToxEx’, Environmental Toxicology and Chemistry, vol. 21, no. 10, pp. 2080-2094.
French-McCay, DP 2003, ‘Development and application of damage assessment modelling: example assessment for the North Cape oil spill’, Marine Pollution Bulletin, vol. 47, no. 9, pp. 9–12.
MAQ0648J | Tamarind Resources - Tui Field | 110-Day Oil Spill Modelling | 21/03/2018 Page 53
REPORT
French-McCay, DP 2004, ‘Spill impact modelling: development and validation’, Environmental Toxicology and Chemistry, vol. 23, no.10, pp. 2441–2456.
French-McCay, DP 2009, ‘State-of-the-art and research needs for oil spill impact assessment modelling’, Proceedings of the 32nd Arctic and Marine Oil Spill Program (AMOP) Technical Seminar, Environment Canada, Ottawa, pp. 601–653.
French-McCay, D, Rowe, JJ, Whittier, N, Sankaranarayanan, S, & Etkin, DS 2004, ‘Estimate of potential impacts and natural resource damages of oil’, Journal of Hazardous Materials, vol. 107, no. 1, pp. 11– 25.
French-McCay, D, Reich, D, Rowe, J, Schroeder, M & Graham, E 2011, ‘Oil spill modeling input to the offshore environmental cost model (OECM) for US-BOEMRE's spill risk and costs evaluations’, Proceedings of the 34th Arctic and Marine Oil Spill Program (AMOP) Technical Siminar, Environment Canada, Ottawa.
Geraci, JR., & St. Aubin, DJ 1988, Synthesis of effects of oil on marine mammals. 292. Ventura, CA, USA: US Department of the Interior, Minerals Management Service, Atlantic OCS Region, OCS Study, MMS 880049.
Gordon, R 1982, ‘Wind driven circulation in Narragansett Bay’ PhD thesis, Department of Ocean Engineering, University of Rhode Island.
Grant, DL, Clarke, PJ & Allaway, WG 1993, ‘The response of grey mangrove (Avicennia marina (Forsk.) Vierh) seedlings to spills of crude oil,’ The Journal of Experimental Marine Biological Ecology, vol. 171, no. 2, pp. 273–295.
Heath, RA 1985, ‘Large-scale influence of the New Zealand seafloor topography on western boundary currents of the South Pacific Ocean’, Australian Journal of Marine and Freshwater Research, vol. 36, no. 1, pp. 1– 14.
Isaji, T & Spaulding, M 1984, ‘A model of the tidally induced residual circulation in the Gulf of Maine and Georges Bank’, Journal of Physical Oceanography, vol. 14, no. 6, pp. 1119–1126.
Isaji, T, Howlett, E, Dalton C, & Anderson, E 2001, ‘Stepwise-continuous-variable-rectangular grid hydrodynamics model’, Proceedings of the 24th Arctic and Marine Oil Spill Program (AMOP) Technical Seminar (including 18th TSOCS and 3rd PHYTO), Environment Canada, Edmonton, pp. 597–610.
International Tankers Owners Pollution Federation (ITOPF) 2014. Technical Information Paper 2 -Fate of Marine Oil Spills, International Tankers Owners Pollution Federation td, UK.
Jenssen, BM 1994, ‘Review article: Effects of Oil Pollution, Chemically Treated Oil, and Cleaning on the Thermal Balance of Birds’, Environmental Pollution, vol. 86, no. 2, pp. 207–215.
MAQ0648J | Tamarind Resources - Tui Field | 110-Day Oil Spill Modelling | 21/03/2018 Page 54
REPORT
Koops, W, Jak, RG & van der Veen, DPC 2004, ‘Use of dispersants in oil spill response to minimise environmental damage to birds and aquatic organisms’, Proceedings of the Interspill 2004: Conference and Exhibition on Oil Spill Technology, Trondheim, presentation 429.
Kostianoy, AG, Ginzburg, AI, Lebedev, SA, Frankignoulle, M & Delille, B 2003, ‘Fronts and mesoscale variability in the southern Indian Ocean as inferred from the TOPEX/POSEIDON and ERS-2 Altimetry data’, Oceanology, vol. 43, no. 5, pp. 632–642.
Levitus, S, Antonov, JI, Baranova, OK, Boyer, TP, Coleman, CL, Garcia, HE, Grodsky, AI, Johnson, DR, Locarnini, RA, Mishonov, AV, Reagan, JR, Sazama, CL, Seidov, D, Smolyar, I, Yarosh, ES & Zweng, MM 2013, ‘The World Ocean Database’, Data Science Journal, vol.12, no. 0, pp. WDS229–WDS234.
Lin, Q & Mendelssohn, IA 1996, ‘A comparative investigation of the effects of south Louisiana crude oil on the vegetation of fresh, brackish and Salt Marshes’, Marine Pollution Bulletin, vol. 32, no. 2, pp. 202– 209.
Ludicone, D, Santoleri, R, Marullo, S & Gerosa, P 1998, ‘Sea level variability and surface eddy statistics in the Mediterranean Sea from TOPEX/POSEIDON data. Journal of Geophysical Research, vol. 103, no. C2, pp. 2995–3011.
Matsumoto, K, Takanezawa, T & Ooe, M 2000, ‘Ocean tide models developed by assimilating TOPEX/POSEIDON altimeter data into hydrodynamical model: A global model and a regional model around Japan’, Journal of Oceanography, vol. 56, no. 5, pp. 567–581.
National Oceanic and Atmospheric Administration (NOAA) 2013, Screening level risk assessment package Gulf state, Office of National Marine Sanctuaries & Office of Response and Restoration, Washington DC.
The Organisation for Economic Co-operation and Development (OECD) 2002, Chapter 4: Initial Assessment of Data. In OECD, Manual for Investigation of HPV Chemicals, pp. 1-11.
Oil Spill Solutions 2015, Evaluation - The Theory of Oil Slick Appearances, viewed 6 January 2015, http://www.oilspillsolutions.org/evaluation.htm
OSPAR Commission (OSPAR) 2012, OSPAR guidelines in support of recommendation 2012/5 for risk- based approach to the management of produced water discharges from offshore installations. OSPAR Commission, p. 21.
Owen, A 1980, ‘A three-dimensional model of the Bristol Channel’, Journal of Physical Oceanography, vol. 10, pp. 1290–1302.
Qiu, B & Chen, S 2010, ‘Eddy-mean flow interaction in the decadally modulating Kuroshio Extension system’, Deep-Sea Research II, vol. 57, no. 13, pp. 1098–1110.
MAQ0648J | Tamarind Resources - Tui Field | 110-Day Oil Spill Modelling | 21/03/2018 Page 55
REPORT
Saha, S, Moorthi, S, Pan, H-L, Wu, X, Wang, J & Nadiga, S 2010, ‘The NCEP Climate Forecast System Reanalysis’, Bulletin of the American Meteorological Society, vol. 91, no. 8, pp. 1015–1057.
Scholten, MCTh, Kaag, NHBM, Dokkum, HP van, Jak, R.G., Schobben, HPM & Slob, W 1996, Toxische effecten van olie in het aquatische milieu, TNO report TNO-MEP – R96/230, Den Helder.
Smit, MG, Bechmann, RK, Hendriks, AJ, Skadsheim, A, Larsen, BK, Baussant, T, Shaw, B & Sanni, S 2009, ‘Relating biomarkers to whole-organism effects using species sensitivity distributions: A pilot study for marine species exposed to oil’, Environmental Toxicology and Chemistry, vol. 28, no. 5, pp. 1104-1109.
Spaulding, ML., Kolluru, VS, Anderson, E & Howlett, E 1994, ‘Application of three-dimensional oil spill model (WOSM/OILMAP) to hindcast the Braer Spill’, Spill Science and Technology Bulletin, vol. 1, no. 1, pp. 23–35.
Solbakken, JE, Ingebrigtsen, K & Palmork, KH 1984, ‘Comparative study on the fates of the polychlorinated biphenyl 2, 4, 5, 20, 40, 50-hexachlorobiphenyl and the polycyclic aromatic hydrocarbon phenanthrene in flounder (Platichthys flesus) determined by scintillation counting and autoradiography’, Marine Biology, vol. 83, pp. 239-246.
Suprayogi, B & Murray, F 1999, ‘A field experiment of the physical and chemical effects of two oils on mangroves’, Environmental and Experimental Botany, vol. 42, no. 3, pp. 221–229.
Tsvetnenko, Y 1998, ‘Derivation of Australian tropical marine water quality criteria for protection of aquatic life from adverse effects of petroleum hydrocarbons’, Environmental Toxicology and Water Quality, vol. 13, no. 4, pp. 273–284.
Willmott, CJ 1981, ‘On the validation of models’, Physical Geography, vol. 2, no. 2, pp.184–194.
Willmott, CJ 1982, ‘Some comments on the evaluation of model performance’, Bulletin of the American Meteorological Society, vol. 63, no. 11, pp. 1309–1313.
Willmott CJ, Ackleson SG, Davis RE, Feddema JJ, Klink, KM, Legates, DR, O’Donnell, J & Rowe, CM 1985, ‘Statistics for the evaluation of model performance’, Journal of Geophysical Research, vol. I 90, no. C5, pp. 8995–9005.
Willmott, CJ & Matsuura, K 2005, ‘Advantages of the mean absolute error (MAE) over the root mean square error (RMSE) in assessing average model performance’, Journal of Climate Research, vol. 30, no. 1, pp. 79–82.
Yaremchuk, M & Tangdong, Q 2004, ‘Seasonal variability of the large-scale currents near the coast of the Philippines’, Journal of Physical Oceanography, vol. 34, no., 4, pp. 844–855.
Zigic, S, Zapata, M, Isaji, T, King, B, & Lemckert, C 2003, Modelling of Moreton Bay using an ocean/coastal circulation model, Auckland, NZ: Proceedings of the Coasts and Ports Australasian Conference.
MAQ0648J | Tamarind Resources - Tui Field | 110-Day Oil Spill Modelling | 21/03/2018 Page 56