Report of the Terra Nova Development Project Environmental Assessment Panel

4.1 Overview

The physical environment, and most particularly the oceanic, atmospheric and ice regimes that dominate the Grand Banks represent formidable challenges to the Proponents. Design engineers will underestimate at their peril the forces of wind, ocean currents, waves, pack ice and icebergs that will be brought to bear upon both the floating and sub-sea elements of an oil production system. It is imperative that designers be concerned that their environmental data are as comprehensive as is reasonably possible; that their risk analyses are thorough; and that, if error is made, it must be error on the side of caution. The consequence of failure would be to place lives at risk and to court environmental disaster.

4.2 Winde and weather

The EIS derives its basic information concerning the nature of the Grand Banks weather from the comparable Hibernia EIS of 1985 which, in turn, drew upon studies that had been conducted in the early 1980s. Some more recent information has been included and, particularly, that deriving from the Canadian Atlantic Storms Program (CASP), Phases I and II, as well as the Genesis of Atlantic-Lows Experiment (GALE), and the Experiment on Rapidly Intensifying Cyclones over the Atlantic (ERICA). In general, climatological data are drawn from archival resources of the Atmospheric Environment Service of Environment Canada (AES), located at the Atlantic Weather Centre, Bedford, Nova Scotia, and the Canadian Climate Centre at Downsview, Ontario. Those data sets, some going back to the early 1800s, and some continuing through 1995, include hourly observations of weather conditions from certain local stations, some detailing wave, wind, ice cover, air and water temperatures, air pressure, and storm tracks obtained through marine or coastal station meteorological reports as summarized by the Marine Statistics System (MAST). One data set, located in the AES archive at Downsview and also at the U.S. Pacific Marine Environmental Laboratory (PMEL) at Seattle, although identified as the Comprehensive Ocean-Atmosphere Data Set (COADS), does not include certain Canadian marine observations since 1988. Furthermore, the range of outputs for data collected since the late 1980s are limited to monthly means. Of considerable importance is the Proponents' Terra Nova environmental data base, which consists of data on temperature, wind speed and direction, atmospheric pressure and variability, collected by oil industry and government vessels operating in or passing through the Terra Nova and Hibernia regions of the Grand Banks.

In short, it may be said that the best available data have been used by the Proponents in preparing their EIS. This is not to say, however, that current knowledge is definitive or that, in respect of forecasting abilities, the science is any way exact.

The data bases contain materials of varying provenance whose real value is also variable, ranging from relatively crude observations by untrained observers using unsophisticated measuring devices to carefully controlled measurements by qualified scientists or technologists using state-of-the-art equipment. Prior to the commencement of exploration activities on the Grand Banks, most climatological information from the area was provided by ships of opportunity and was in consequence somewhat random in terms of both time and quality. The usefulness of these data is further compromised because transocean shipping, though generally following standard shipping lanes, will avoid bad weather and ice where possible. The international fishing fleets that habitually frequented the Grand Banks did not maintain weather records in any systematic way. Over the next few years it will be possible to collect systematically additional data from the Hibernia platform and from drilling rigs operating in the area.

Nevertheless, it is clear, as the Proponents have recognized, that the ocean area to the south and east of Newfoundland is "the most active area of cyclonic activity in North America and the adjoining oceans". Data produced in the EIS indicate clearly that wind speeds over the Terra Nova area are consistently higher than the values reported from the nearest land stations. On average, one significant tropical storm per year passes within 300 km of the Newfoundland coast, tracking across the Terra Nova area, while less common but even more violent storms follow the same track. Indeed, it is recognized by the Proponents that the Project will be located in one of the harshest weather and sea-state environments on earth.

The calculations for expected extreme wind speeds at the Terra Nova location appear reasonable in the context of available data. The Det Novake Ventas calculations for gust factors as applied in the Seaconsult analysis appear to be consistent with recent studies of wind speed and gusting conducted off Canada's east coast, albeit not at the Terra Nova site. The CASP-I and II, GALE and ERICA research programs have added considerably to the knowledge of the physical processes that produce the major storms that skirt the coast of Newfoundland. These and other recent studies also have increased the knowledge and understanding of area scale variability in winds over the northern Grand Banks as influenced, for example, by the presence of pack ice.

Despite this growing knowledge, observational techniques and operational forecasting skills still can be improved. Early identification of impending storms, and particularly of those intense winter storms that result from explosively deepening low pressure systems, can be improved through careful surface observations, remote sensing, and from numerical weather models. Happily, the Proponents are fully aware of the need for improved operational forecasting, especially of severe weather. They look for such improvements to the AES, which promises, at the very least, higher resolution wind field forecasts in the near future.

It was apparent from the hearings conducted by the Panel that the Proponents' EIS was deemed by the public to be generally satisfactory in respect of the impact of wind on the Project. Very few questions were raised in this domain and those that were raised, by Environment Canada, for example, were satisfactorily answered.

Nevertheless, the Proponents must not leave everything to the AES. Continuous monitoring of weather conditions at the production site in full collaboration with AES will add an important dimension to that agency's data gathering activities. Additionally, since accurate forecasting of weather must be a major concern in respect of both life safety and security of facilities and production, the Panel believes that a dedicated forecasting program with a research component should be an integral part of the Project.

Recommendation 33:

The Panel recommends that the Terra Nova Development should become, in

collaboration with the Atmospheric Environmental Service of Environment Canada and the Hibernia platform, an important centre for the collection of weather data both to enlarge and improve current data sets and to aid in the early identification of intense winter storms; and, that a collaborative weather program with a research component be designed and implemented to improve observational techniques and operational forecasting.

Other atmospheric weather phenomena that might impact upon the Project are thunderstorms, although those are comparatively rare in this area; fog, which will present a hazard in all seasons; and, freezing rain and spray icing that will, during winter and spring, be a regularly occurring danger to ships and aircraft.

Because the Terra Nova Development will involve ships, both stationary and in motion, as well as helicopters, and because the field is located near a major transatlantic shipping lane, the Great Circle Route, and in an area frequently traversed by fishing vessels, the prevalence of dense fogs must be the occasion for particular care. The price of safety in such an environment must be eternal vigilance in the context of a well-designed program, in collaboration with the Canadian Coast Guard and other appropriate authorities, for the monitoring and controlling of sea and airborne traffic.

Recommendation 34:

The Panel recommends that the development plan should include a program devised in consultation with the Canadian Coast Guard and other appropriate authorities for monitoring and controlling marine traffic and for the development of a set of protocols to obviate the danger of collision.

The data available in respect of ice accumulation on hulls and superstructural elements of ships and production facilities are relatively sparse, and the application of the best available models to the Newfoundland offshore has not been properly tested. It is known that spray icing, in conjunction with high winds, has historically represented an extreme hazard to shipping. Thus, in the context of the Project, the Panel believes that the Proponents' measures to ameliorate this hazard should be coupled with a monitoring and research program which would refine existing models, and would establish reliable design load estimates for the extreme conditions anticipated throughout the life of the Project.

Recommendation 35:

The Panel recommends that measures proposed by the Proponents to ameliorate spray icing or icing from freezing rain should be coupled with a research program designed to expand current knowledge and to refine existing models with the objective of establishing completely reliable design load estimates for the extreme conditions that may be encountered in the Terra Nova Development area.

4.3 Waves

The Proponents acknowledge that "a sound understanding of the mean and extreme wave climate at the Terra Nova site is required to support systems designs and operational planning." The Panel is satisfied that the EIS contains a reasonably comprehensive analysis of available data bases. In particular, the expanded wave hindcast data set, now including 80 specific storm events, indicate little change from earlier estimates. However, very large

waves are often episodic in their occurrence, exhibiting a large degree of variability extending over time scales of decades or longer. In this context, a value for extreme wave height derived from a study period of less than a single decade may well be invalid.

However, sea and wave conditions did not appear to be a concern to members of the public appearing before the Panel. Only the Natural History Society raised the issue and their comments did not directly address the Proponents' EIS, but rather provided a retrospective of comments by Environment Canada and the Department of Fisheries and Oceans raised in response to the Hibernia EIS. Notwithstanding what appears to be general public satisfaction with the Proponents' preparedness to deal with high wave conditions, the Panel believes that design criteria for vessels that will be on site for two decades or more must clearly recognize the possibility of extreme wave values higher than those predicted by the current model. There is also a necessity for continuing accurate observation and the regular updating of the wave hindcast data bases.

Recommendation 36:

The Panel recommends that the Proponents, in collaboration with Environment Canada and other relevant institutions, collect data and regularly update wave hindcast data bases.

Recommendation 37:

The Panel recommends that the Board ensure that design criteria for vessels that will be on site for two decades or more must clearly recognize the possibility of extreme wave values higher than those predicted by the current model.

4.4 Currents

In the event of oil spills at or near the production site, the pattern of drift that will determine the rate and spatial extent of dispersal is clearly a matter of importance. The Natural History Society, while acknowledging that the EIS contains an excellent and thorough summary of the oceanography of the area around the Terra Nova field, concludes that this "knowledge is not transferred into the assessment of environmental impact". Their principal criticism was that in the drift modeling done for the EIS, the Labrador current was apparently absent, even though the addition of the ocean flow of that current has an important impact on model results. If the studies reported by the Natural History Society correctly predict that an oil spill at the Terra Nova site "will drift southwards along the eastern flanks of the Grand Banks," the environmental impact will clearly be different from that arising in the case of a different pattern of dispersion. Additionally, the Panel is concerned that because of the extreme variability of surface currents in the area, it will be difficult to forecast the manner in which an oil spill would be dispersed. For this reason, the Panel recommends that a surface current monitoring program be established with the objective of adding to the knowledge base; improving assessment of the full impact of a spill at the Terra Nova site; and, to the development of contingency plans for amelioration.

Recommendation 38:

The Panel recommends that the Proponents be required to maintain a continuous surface current monitoring program at the Terra Nova site to enhance the predictability of oil dispersal patterns. The Panel further recommends that serious consideration be given to the incorporation of the data from the monitoring exercise with drift modeling.

4.5 Ice

Anchored FPSO technology has been proven in the harsh offshore environment of the North Sea. However, drifting ice, and most particularly icebergs, add a unique dimension of danger to both floating and sub-sea systems at the Terra Nova field that is difficult to overstate.

The EIS employs data compiled from approximately 40 years of airborne reconnaissance by both Canadian and American government agencies. This has been augmented, and in some cases replaced, by images from polar-orbiting satellites in daily and weekly ice charts produced at the AES Ice Centre.

The Proponents have relied also upon observations of the offshore oil exploration industry during the 1970s and 1980s, and as well upon the results of the Department of Fisheries and Oceans and AES research supported by the Panel on Energy Research and Development (PERD).

Data on icebergs has also been available from the International Ice Patrol (IIP), which was created subsequent to the Titanic disaster. Since 1989, Provincial Airlines Ltd. (PAL) has conducted airborne surveys on an approximate five-day schedule, which has allowed for a simpler process for counting icebergs within a given one-degree grid area than the more complex modeling of the IIP.

Significant differences between the figures derived from IIP estimates and the PAL counts indicate continuing uncertainties associated with both the survey and estimation procedures employed. In using a judicious interpretation of both sets of data, the Proponents are satisfied with the counting criterion for the particular one-degree grid in which the Terra Nova Development lies. Nevertheless, they must recognize that their conclusions are not absolute. Rate of drift and rate of deterioration are confounding elements of modeling equations. A single large iceberg, in an instant, might split in two or into a number of growlers and bergy bits, and thus confuse the counting process.

Of course, mere counting says nothing of physical dimensions and velocities nor, indeed, of propensity to scour. Data in respect of these phenomena are reliant upon monitoring and study programs carried out by the oil industry during exploration and preparation for production. Additionally, research supported by PERD and the Environmental Studies Revolving Fund have contributed to available knowledge.

Iceberg trajectories are notoriously difficult to forecast. They are subject to the influences of major ocean currents, tidal currents and by winds and wind-generated currents. The dimensions and shape of icebergs, the pack ice regimes in which they move, and the air, water and bottom drag coefficients also affect the ability to forecast. In consequence, the predictive values based on forecast wind and current values alone are somewhat suspect. Best estimates are derived only from near real-time wind and current data from the immediate area in which the iceberg is moving. This is to say, that the capacity to assemble such real-time data for instantaneous application will add an important empirical dimension to forecasting abilities.

The Panel believes that the problem of iceberg scouring has been adequately addressed in the EIS, although some uncertainties remain in respect of the available data and analysis based on them, and that glory holes and trenching to the indicated depths make sufficient allowance to ensure the safety of the wellheads and flow lines.

In the public mind, the icebergs, growlers, bergy bits and pack ice represent the greatest hazard associated with development of the oil field. One participant, in his presentation to the Panel, concentrated on the possibilities of collisions of ice with floating elements of the production system. He felt that the EIS was too optimistic in respect of the ability to detect, monitor, and manage ice, and provide for "an orderly and controlled move off location, in the event an unmanageable iceberg that presents a hazard approaches too close to the facilities." His argument was predicated on the assumptions that ice is uncontrollable; that predicting iceberg drift is extremely difficult, if not impossible; that sometimes as many as 200 icebergs may be found in the Terra Nova area, and, on such occasions, it would be extremely difficult, if not impossible, to determine which individual icebergs should be selected for management; that the calculation of a median southerly ice edge is of little value; that growlers and bergy bits are virtually undetectable in wave heights commonly experienced at Terra Nova, but are nonetheless dangerous to ships; that the EIS makes no provision for the "management" of such growlers and bergy bits; and, that indeed such pieces of ice are not only hard to detect at long range, but practically impossible to deflect from their course. In support of his argument, the participant listed 52 significant incidents occurring offshore Eastern Canada since 1980 in respect of oil exploration and development activities. Of the 52 incidents, 10 involved ice management or avoidance. In no case was ice responsible for losses other than of time.

The Natural History Society of Newfoundland and Labrador in its submission echoed those concerns and systematically challenged the statistical methodologies applied in the EIS, asserting that the data used were collected for logistical rather than for scientific uses so that certain results were not scientifically reviewable. They complained about means reported without standard deviations, and other data sets in which considerable spatial and temporal averaging were masked. Environment Canada, too, concluded that the ice management plan assumes skill in the forecasting of iceberg motion that has yet to be demonstrated to their satisfaction. Environment Canada also expressed caution about the possibility of climate change causing a greater number of icebergs in the Project area.

The Panel is impressed with the degree of confidence expressed by the Proponents in respect of their capacity to deal with icebergs at the production site. However, there is strong public perception that difficulties and potential hazards may have been understated.

In view of the acknowledged hazard and the difficulties presented by analysis of available data bases, the continuing study of ice conditions and iceberg trajectories in the Terra Nova area, and a better means of identifying and tracking growlers and large bergy bits, must be implemented. This implies that a continuous program of observation and research, that leads to the improvement of radar and other remote-sensing devices making possible the early detection of even low-lying masses of floating ice, must be a significant component of the operational plan. Further, a carefully-designed third-party audit of the effectiveness of the Proponents' ice management plan would serve to validate the Proponents' optimism and satisfy public concerns.

While the EIS discussion of sea ice data drew similar public criticisms none of the participants deemed it to be as large a potential problem as icebergs, growlers and large bergy bits. Nevertheless, large pieces of hard multi-year ice will pose a threat to the integrity of the hulls of the FPSO, tankers and supply vessels. Ship designs for the Project should clearly recognize this hazard and the protocols proposed in respect of double-hull design, ice-strengthening, Canadian flagging and crewing should be mandatory. Furthermore, all marine crews should be properly trained and certified in safety and marine emergency procedures and, to this end, appropriate arrangements should be made with relevant training establishments in the Province.

Dangerous pack ice will be encountered at the Terra Nova site and avoidance would seem to be the only appropriate form of "management". The Panel should note the argument advanced by the Natural History Society that pack ice in the Flemish Pass area, just east of the Grand Banks, could quickly infest the Terra Nova site in the event of a strong easterly wind. The Panel believes that the ice management plan should clearly delineate the circumstances in which it will be appropriate to disconnect and remove all surface vessels to a safer environment.

Recommendation 39:

The Panel recommends that the ice management plan should:

1. allow for the difficulties in forecasting iceberg trajectories and provide for the acquisition of adequate real-time data that can add a substantial pragmatic element to model-driven projections;
2. clearly indicate a process for selecting the icebergs to be managed by towing, for example, when multiple icebergs are in the immediate area;
3. recognize that collisions with small growlers and bergy bits are definite hazards to shipping;
4. include a process for timely identification and management of threatening growlers;
5. recognize the potential for a substantial increase in the number of icebergs crossing the 48th parallel as a concomitant of global warming;
6. include provision for a third-party audit of its effectiveness;
7. clearly establish a set of protocols that will determine the conditions which will dictate disconnection and removal of all surface vessels to a safe area; and,
8. include a continuous program of observation and research that leads to the improvement of radar and other remote sensing devices that will make possible the early detection of even low-lying masses of floating ice.

Recommendation 40:

The Panel recommends that ship designs for the Project clearly recognize the hazard to hull integrity posed by growlers and bergy bits and meet the highest standards for navigation in ice as presented by the appropriate authorities.

Recommendation 41:

The Panel recommends that all marine crews be properly trained and certified in safety and marine emergency procedures and that the Proponents make appropriate arrangements with relevant establishments in the Province for such training.

4.6 Coincident Extremes

A matter of public concern raised on several occasions was the possibility of the coincidence of extreme wind force, extreme wave heights and extreme iceberg menace or rig icing. In the context of risk analysis, it is clear that there is a low probability of such a scenario. Nevertheless, the possibility must be recognized as a ponderable one.

In extreme wind and sea conditions the management of icebergs would be, given the current state of the art, an impossibility. It, therefore, follows that the ice management plan must adopt a policy of avoidance that eliminates any possibility of such a crisis arising. This implies not only an effective monitoring program and a completely reliable forecast capacity for wind and sea conditions, but also a well-designed and clearly understood decision-making process that will ensure the timely removal of the FPSO and support vessels from the area.

Recommendation 42:

The Panel recommends that operational planning should allow for the simultaneous occurrence of two or more 100-year events, involving combinations of wind, sea, and ice. This should include a well-designed and clearly understood decision-making process for the timely removal of the production vessel and all other vessels from the area.

4.7 Command Structures

This in turn implies a carefully conceived structure of command that permits no ambiguity nor any possibility of misunderstanding at times of emergency.

It is accepted under marine law and the Canada Shipping Act that vessels at sea are under the absolute authority of their masters. It is also accepted, throughout the oil industry, that a production platform is under the command of the offshore installations manager (OIM). The FPSO that will be used in developing the Terra Nova field will be a production platform when it is moored at the site and connected to the sub-sea production facilities. It will be a ship when it has been disconnected from those facilities and from its moorings.

There is, within the paradox of a vessel that becomes a ship, then ceases to be a ship, and then becomes a ship again, a clear and present possibility for conflict. For there are two authorities, each of whom has, from time to time, supreme and uncontradictable control over the vessel. It will be argued, of course, that there would be no conflict when the vessel was under way, for then the captain would be in command; nor any conflict when the vessel was in production mode, for then the OIM would be in command. From this, it might appear to follow that there would be no time at all when there was not a single person in command. However, a potential difficulty does exist when a decision must be taken to remove the FPSO from its moorings and change it into a ship under way. One might assume that the marine captain would be the person best qualified to make such a determination. However, until that decision to disconnect is actually taken, the person in command is the OIM.

The Proponents assured the Panel that there will be in place formal and well-structured provisions for appropriate consultation between senior oil production and ship handling specialists whenever wind, sea state or ice conditions warrant special concern. In eventualities such as those, when conditions of extreme hazard exist, the prime consideration must be the safety of the vessel and, more particularly, the safety of its crew. Since, in the Panel's view, the marine captain must be the person with the best appreciation of the impending dangers, and best able to interpret weather and sea-state data and how the vessel should be best prepared to meet severe storm conditions, the Panel believes that if a difference of opinion should arise as to whether to disconnect or not, the captain's decision should be final. The Panel recognizes, of course, the possibility that the OIM could be a qualified marine captain with appropriate sea experience and that the offices of ship's captain and OIM could be combined in one person.

The Panel does not, however, believe this to be an appropriate course to follow. The stresses of either job would be considerable. The additional stress of having to resolve the potentially severe conflict of competing interests might well be intolerable.

Recommendation 43:

The Panel recommends that the marine captain should be ultimately responsible for the safety of the vessel and her crew in respect of all weather or sea-state hazards. A mechanism for the formal and continuous consultation between the captain and the offshore installations manager should be clearly in place. The marine captain should be the one to implement, when it is necessary, the protocols to disconnect the vessel and remove it to a safe area.