ENVIRONMENTAL CONDITIONS IN THE SOUTHERN BARENTS SEA
© Arne Kvitrud, Sondre Nordheimsgate 9, 4021 Stavanger.
Paper presented in Stavanger in 1991, but put on Internet 25.9.2002.
The figures are not presented here.
During the last several years, exploration drilling have been performed in the Southern Barents Sea. This paper describes the main conclusions from the experience concerning the environmental conditions, as it concerns load bearing structures. The paper is restricted to conditions south of 740N 30'N in the Norwegian Barents Sea.
Since 1976 systematic collection of environmental data have been performed in the Barents Sea. All the data are publicly available and can be used without any restrictions. The meteorological data are stored at the Environmental Data Center in Oslo, and current data are stored at Oceanor in Trondheim.
If we look at measurements carried out by several offshore locations and extrapolate them to an annual probability of exceedence of 10¿², the result will be 30 - 36 m/s at a height of 10 m, and averaged over 10 min.
Use of hindcast data from Bjørnøya, Sentral- banken and Nordkappbanken gives approximately the same extreme values as for Tromsøflaket (Torsethaugen, 1989).
In the guidelines (1987), a recommended value of 41 m/s is suggested for the whole continental shelf. The recommendation should be on the safe side as far as the Barents Sea is concerned. This means that those who want to use other values may do so.
A major research project has been carried out by several Norwegian institutions and firms to examine polar lows (Lystad, 1986). The highest measured wind velocity in polar lows was 35 m/s at a height of 10 metres, averaged over 10 minutes. One of the conclusions was that it was not very likely that polar low pressures would give higher wind speed than normal storms. Extreme value statistics gave 38 m/s for Tromsøflaket, with an annual probability of exceedance of 10¿² (Houmb et al, 1986). For a given location the extreme values would be lower. As regards to the design of platforms, polar low pressure is not a significant problem. It will not give increased extreme estimates of wind speed, wave height, air temperature or icing (Houmb and others, 1986).
On the other hand they create problems for operations and planning of operations for a long period of time, as is normal for other places on the continental shelf. One example is Norsk Hydro's drilling on Block 7321/9 in the autumn of 1988. Two polar storms resulted in drilling stoppage for 22 hours (Halleraker, 1988).
Conclusion on wind
Exploration drilling platforms designed in accordance with norwegian rules with a recommended value of 41 m/s are not expected to have difficulties in the Barents Sea.
A general trend on the continental shelf is for mean wave heights to increase from the southern part (56 degree N), to the largest value in the central part (630 - 650 N), and then decline further north (Eide, 1986). The same applies for wave heights with an annual probability of exceedance of 10-².
On the basis of measured data, and supplemented with data from the hindcast model WINCH where no measurements exist, the 1987-guidelines for choice of wave heights to be used in early development phases. This is shown in figure 1. Barstow and others (1988) provide more information on WINCH.
Wave measurements have been made at several places in the Barents Sea.
There is a clear difference between the two western stations (Bjørnøya and Tromsøflaket) and the eastern stations (Sentralbanken and Nordkappbanken). The wave climate becomes less severe as we go east in the Barents Sea. This is also found by Torsethaugen (1989), based on hindcast data.
At Tromsøflaket the wave periods corresponding to a 100-year wave hight will be higher (17-19 sec) than what is usual in the North Sea (15-17 sec), see figure 1.
During the drilling with Ross Rig on 10 November 1988 near Bjørnøya, the wave period reached 18 seconds. This was very close to the resonance period of 20.3 seconds for the platform. This high wave period resulted in drilling stoppage for 8 hours (Halleraker, 1988).
The wave climate in the Barents Sea do not deviate significantly from our experience in other parts of the shelf. Platforms for exploration drilling, which are designed for conditions on Haltenbanken or Trænabanken for example, should be suitable for drilling in the Barents Sea, too.
During the past few years the operators and several public institutions have carried out measurements of sea current in the Barents Sea. Measurements (closer to shore), over shorter periods, show extreme currents exceeding 1 m/s. But, in general, the currents are of the same magnitude as in other parts of the Norwegian continental shelf.
Comparing with the current speeds in the North Sea, which are often over 1 m/s, these currents are rather small and should not present any major difficulties.
Most of the seabed of the Barents Sea is 200-400 m below the water surface. This is comparatively deep, but it can be compared with Troll, Draugen, Gullfaks C and Snorre, where development has been made or is planned further south.
Measurements of the tidal water level in the Barents Sea do not deviate much from values found further south. However, in Vardø the M2 value is 104 m (Bjerke and Torsethaugen, 1989), which is somewhat higher than found other places.
The recommendation in the guidelines is based on a study made by the Otter group (1983) from Norsk Hydro for use on Oseberg. The valuesin the reoprt has been multiplied with a factor of two. Load case 1 gives the major contribution.
A theoretical analysis of the models which have been used has been made by Horje and Vefsmo (1985). Since the models:
a) have not taken into account the heat properties of the rig
b) are based on experience with small moving fishing boats,
the results of the models are probably on the conservative side. Although Tromsøflaket is not the place in the Barents Sea where the risk of icing is the largest (Bjerke and Torsethaugen, 1989), the guidelines (1987) are probably on the conservative side.
Conclusion on icing
Ice loads due to icing should not give major design problems, if they are considered at an early stage.
The most extreme known expansion of sea ice occurred in June 1881 (Kvitrud and Hønsi, 1990). The sea ice in a short period reached 20 km north of Berlevåg on the east coast of Finnmark. For a major wintertime period the sea ice was at 710 30'N to 720N.
The ice boundaries in the Barents Sea have been and are recorded weekly by Geir Kjærnli at DNMI on the basis of satellite observations.
Assuming that the annual extreme values are statistical distributed, the southernmost distribution of ocean ice with an annual probability of 10¿² of exceedance can be found (Kvitrud and Qvale, 1989 or Vefsmo et al, 1990b). The results are presented in figure 4. The southernmost position based on observations from 1898 to 1983 (Atlas of the Artic Ocean, 1985) correspond well with the extrapolations (Kvitrud and Nilsson, 1989).
If we extrapolate to the level of 10¿4, we will find that in most parts of the Barents Sea sea ice must be taken into account in the design of offshore structures.
The damage potential of sea ice depends on various parameters. The most important ones are the thickness of the ice, the relative velocity between ice and platform, the physical ice properties, and the size of the ice-fields (Mobil, 1988).
The oil companies in Norway have through several years done research to investigate these factors. Several years of future research must also be done before we have a safe way of managing sea ice.
Conclusion sea ice
In the most parts of the Barents Sea, sea ice has to be taken into account in design. The damage potential on the platforms is mainly on local elements in the splash zone. In addition it will cause station keeping problems for floating structures.
Icebergs have been observed at the coast of Norway a few times (Kvitrud and Hønsi, 1990 and Hønsi, 1988):
a) The first report of icebergs in the Barents Sea south of 740N is in February 1881. Two icebergs reached the coast at Kvaløya in Troms at 70013'N 19030'E. The larger iceberg of the two was 7 metres high.
b) In June 1881 several icebergs were observed at Gamvik, Berlevåg and Syltefjord at East- Finnmark. The largest iceberg was enormous, with a length said to be 10 km, and a sail height of 30 m.
c) During the period of April-June 1929, a number of icebergs reached the coast of Kola Peninsula and eastern Finnmark (from 240 to 440 E). The local newspapers in Finnmark reported that they reached up to 30 metres above sea level.
d) In 1939 two icebergs were observed at Koi- fjorden close to Gamvik.
Damage potential and risk
To investigate or quantify the risk of icebergs includes calculating the probability of collision and studying the consequences of collisions.
The chance of collision is closely connected with the frequency of icebergs appearing in the area of interest. Vefsmo et al (1990) have developed a model computing the collision probability. Based on this model and the historical data, at least two areas are sensitive to iceberg collision, giving an annual probability of collision higher than 10-4:
a) South of Bjørnøya based on iceberg observation in recent years
b) East-Finnmark and the sea north of the coast, based on historical data.
Present exploration rigs (NMD, 1973) are designed for collisions with vessels with displacement of 5000 tonnes and a speed of 2 m/s. These collisions are assumed at sea level. The energy formula = 1/2 m v² gives 10 MJ when hydro- dynamic additional mass is not taken into account. A dynamical analysis of the condeep platform Gullfaks A shows that this platform can take a collision with a tanker of 150,000 tonnes at a speed of 2 m/s or an energy calculated the same way of 300 MJ. The energy distribution between iceberg and platform in a collision will be different for a vessel and an iceberg. The iceberg will be less rigid than the vessel, and a smaller part of the energy will be absorbed by the platform.
Between Bjørnøya and Hopen the mean speed of icebergs for a long period of time has been measured to be approximately 0.1 m/s (Bercha, 1989). The currents may be somewhat larger in this area than further south, but 0.45 m/s should be a reasonable estimate. To get an energy of 10 MJ, one must have an iceberg of approximately 100.000 tonnes. To get 300 MJ, the iceberg must be 3 million tonnes.
Measurements between Bjørnøya and Kong Karls Land in 1988 gave a mean value for icebergs of 570,000 tonnes, but there were large variations, (Løvås and Næss, 1989). Icebergs south of 740N will most likely be smaller.
Drilling experience north of 730N
During Mobil's drilling in August 1988 a survey was made by flying the route shown in figure 4. Four icebergs were found on this trip (Spring, 1988).
A similar survey carried out by Norsk Hydro in November did not give observations of icebergs. Nevertheless, on 23 November 1988 a remainder of an iceberg was found with a freeboard of 1 m and a weight of approx 500 tonnes, 9 km north of the drilling location (Engseth, 1989).
Conclusions on icebergs
Icebergs may cause serious damage to a platform. More information is needed about iceberg properties and behavior.
In the Barents Sea south of 740N temperatures down to -300 C have been registered (Iden and Tønnesen, 1988), see figure 9. Extrapolation at an annual probability of exceedence of 10-2 gives -350 C for Bjørnøya and -180 C for Tromsøflaket (Bjerke and Torsethaugen, 1989). These temperatures have an effect on the working environment, operations, design, choice of materials and testing of materials. Provided the temperature is taken into account, it should be a structural problem with the material properties which are available today.
Changes in sea temperature from the North Sea to the Barents Sea are minor. The differences have minor consequences for most purposes.
FORECASTING OF THE ENVIRONMENT
When the Barents Sea was opened for exploration drilling at winter time the operators provided reports on their experience on how the weather forecasting had be carried out.
They had all experienced the weather forecasting as being variable in quality. The general opinion was that winter forecasting in the Barents Sea was of a somewhat poorer quality than that of the North Sea (Aanstad, 1988 and Askedal, 1989). A verification of daily forecasts to "Ross Rig" in the Barents Sea and "Petrojarl" at Oseberg did not, however, show any appreciable differences. Comparisons made between forecasts from DNMI and forecasts from private companies do not show any systematic difference in quality (Aanstad, 1988). Bera (1988) was of the opinion that the use of a meteorologist onboard the rig gave a better forecast than those that came from shore.
In connection with exploration drilling north of the 73rd parallel, the operators established systems for monitoring ice that could represent a danger during drilling. This included both icebergs and sea ice.
The elements that Mobil (Armstrong, 1988) and Norsk Hydro (Engseth, 1989) employed in monitoring were:
a) satellite observations of the ice borders, conducted by DNMI
b) plane and helicopter missions to look for sea ice and icebergs
c) supply ship that went to the ice border and patrolled the area (Norsk Hydro)
d) satellite buoy on the ice to see how it moved (Norsk Hydro)
e) ice forecasting from DNMI and others
f) use of reports from the coast guard (Norsk Hydro)
One of the conclusions was that ice forecasting had a variable quality (Engseth, 1989).
Forecasting of icebergs is not possible unless each iceberg is equipped with instruments so that one may know where they are at any given time. What one must rely on are good radar systems that can give warning of icebergs when they are at a certain distance from the platforms. Knowledge of experience from other countries seems essential.
Conclusions on forecasting
The forecasting service in the Barents Sea is not as good as that found further south. This gives a lower safety level than what one is used to for operations sensitive to weather.
There are only few registrations of earthquakes in the Barents Sea (Bungum, 1988). Since 1987 there have, however, been local measurements that show that the Ringvassøy - Loppa fracture is earthquake-active. However, the earthquakes here are so small that they are not registered by the regular measuring stations.
The Barents Sea has low values for design earthquakes compared with other parts of the shelf, (Bungum and Selnes, 1988).
Topography and geotechnical conditions
Major parts of the seabed in the Barents Sea show signs of ploughing caused by icebergs. Local pockmarks have been recorded along some of the plough marks.
Surveys of the geotechnical conditions in and outside the plough marks on Tromsøflaket show that the ploughing process produces plastic deformations and stirring of the earth, and thus local changes in the geographical characteristics. Installation of platforms in plough marks may be difficult. It may also be difficultto establish the correct geotechnical characteristics of the soil.
As in other parts of the shelf there is stratification with sediment from clay to stones (Gunleiksrud, 1986 and Elverhøy and others, 1988). Drilling done by Statoil at Askeladden shows low shear strength in the upper 2 m, and then harder clay further down. From 10-25 m the soils are very stiff. The geotechnical characteristics of the clay that were examined at Askeladden correspond to what is found in the North Sea (Tjelta and others, 1983).
At Tromsøflaket there are sedimentary strata at about 100-200 m (Gunleiksrud, 1986). Further north and east it seems that the sediment depth diminishes (Solheim and Elverhøy, 1988).
Conclusion on soil conditions
The Barents Sea differs from the North Sea in that it has large plough marks on the seabed. This is a problem also found at the Haltenbanken. The sediment strata in the Barents Sea are considerably smaller than what is usual. This could provide simpler foundation solutions.
Exploration drilling and possible future production of oil and gas in the Barents Sea has presented on the shelf. With regard to knowledge about the environmental conditions, it is probably in connection with icebergs that we have the largest gaps. Icebergs also may cause the greatest damage. Wave loads, however, will dominate design in the southern Barents Sea as they do further south.
Aanstad, K., 1988, "Naturdatainnsamling i Barentshavet", letter from Operatørkomitè Nord
Armstrong, R.P., 1987, "Collection of environmental data in the Barents Sea".
Askedal, S.A., 1989, "Erfaringer fra leteboring med "Ross Rig" i Barentshavet, brønn 7321/9-1".
Atlas of the Artic, 1985, USSR State Committy for Hydrometeorlogy and Natural Enviroment, Moscow.
Barstow, S., Vardal, L., Guddal, J. og Reistad, M., 1988, "Evaluation of the WINCH hindcast for the Barent Sea. Oceanor report OCN 88015", Trondheim.
Barstow, S.F., Matisen, J.P., and Schjølberg, P., 1989, "Miljøforhold i Barentshavet, Sentralbanken, Bjørnøya, Nordkapbanken, Stasjon B og C, Periode 1985-88." Oceanor, Trondheim
Bera, J., 1988, "Natural data collection - Barents Sea". Letter from Total Marine Norsk A/S
Bercha International, 1989, "SAR imagery analysis, 1988 IDAP project", Calgary.
Bjerke, P.L., and Torsethaugen, K., 1989, "Environmental Conditions on the Norwegian Continental Shelf, Barents Sea", NHL report STF 60 A89052, Trondheim.
Bungum, H., 1988, "Earthquake occurrence and seismotectonics in Norway and surrounding areas", Norsar, Oslo.
Bungum, H., and Selnes, P.B., 1988, "Earthquake loading on the Norwegian Continental Shelf", summary report, Norsar/NGI, Oslo.
Det norske meteorologiske institutt (DNMI), "Monthly mean sea ice charts", Oslo, January 1970 - December 1988.
Eide, L.I., 1986, "Plans for future wave measurements and analysis", proceedings ANODAODAP seminar, Sintef, Trondheim.
Elverhøi, A., and Solheim, A., 1988, "The Barents Sea ice sheet - a sedimentological discussion", Norsk Polarinstitutt, Skrift nr 47, Oslo.
Engseth, B., 1989, "Boreerfaring med sjøis og isfjell ved Bjørnøya", proceedings Naturdatakonferansen i Harstad, 26-28.9.1989, Harstad.
Gunleiksrud, T., 1986, "Geologi og grunnforhold på norsk sokkel - oversikt", proceedings kursdagene ved NTH, Norske Sivilingeniørers Forening, Oslo.
Halleraker, H., 1988, "Operasjon med "Ross Rig", brønn 7321/9-1, License 141", Norsk Hydro, Sandnes.
Horjen, I., and Vefsnmo, S., 1985, "Computer modelling of sea spray iceing on marine structures", proceedings of the IFIP/IFAC Symposium, Automation for safety in shipping and offshore petroleum operations - ASSOPO 85, Trondheim.
Houmb, O.G., Lønseth, L., Schjølberg, P., and Vollan, B., 1986: "Environmental conditions under polar lows". Oceanor report OCN86022, Trondheim.
Hønsi, I., 1988, "Isfjell i Barenthavet", Stavanger.
Iden, K. A., and Tønnesen, H., 1988, "Kart over lufttemperatur i området mellom Norge og Svalbard", DNMI, Oslo.
Kvitrud, A., and Hønsi, I., 1991, "Icebergs in the Norwegian Continetal Shelf in 1880-1881", Proceedings OMAE-91, Stavanger.
Kvitrud, A., and Nilsson, K.N., 1989 "Naturtilhøva i Barentshavet", Proceedings Naurdatakonferanse , Harstad
Kvitrud, A., and Qvale, A., 1989, "Isgrenser i Barentshavet", Stavanger.
Lystad, M., 1986, "Polar lows in the Norwegian, Greenland and Barents Sea", DNMI, Oslo.
Løvås, S.M., and Næss, T., 1989 : "1988 IDAP projects in the Barents Sea". Aerial stereo photography, Fjellanger Widerøe og NHL, Oslo og Trondheim.
Mobil Oil Canadaog Mobil Exploration Norway, 1988, "A synopsis of Canadian cold water environmental research", Stavanger.
Norwegian Maritime Directorate, 1973, "Regulations for mobile drilling platforms with installations and equipment", Oslo.
Otter group, 1983, "Oseberg Conseptual Study, Estimation of Loads due to platform icing", Trondheim.
Spring, W., 1988, "Trip report, iceberg reconnaissance", 20. august 1988, Mobil, Stavanger.
Tjelta, T.I., Watt, P.H. and Senner, D.W., 1983, "Deep water soil investigation : planning to results", Statoil.
Torsethaugen, K, 1989, "Wind and waves in the Barents Sea", Proceedings Naturdatakonferansen i Harstad, 26.-28. September 1989, Harstad.
Vefsnmo, S., Johansen, Ø., Løvås, S.M., and others, 1990a: "Platform and foundation behavior during combined hydrodynamic and ice loading - location A and B" - ESARC report no 21, Sintef report STF71F90015, Trondheim.
Vefsnmo, S.,Mathiesen, M.,and Løvås, S.M., 1990b, "IDAP 90 - Statistical analysis of the sea ice data. Volume 1 : Extreme southern Limit analysis", Sintef report STF60 F90079, Trondheim
Vinje, T., 1985: "The physical environment, Western Barents Sea, Drift, composition, morphology and distribution of the sea ice fields in the Barents Sea", Norsk Polarinstitutt skrifter no 179c, Oslo.