APPENDIX F
Comparative Study on Potential Oil Spill in Collision and/or Grounding— Different Tanker Designs

DET NORSKE VERITAS

The appendixes to this study are not included. Committee comments on assumptions and conclusions can be found at the end of the Det norske Veritas report on page 299 and following.



The National Academies | 500 Fifth St. N.W. | Washington, D.C. 20001
Copyright © National Academy of Sciences. All rights reserved.
Terms of Use and Privacy Statement



Below are the first 10 and last 10 pages of uncorrected machine-read text (when available) of this chapter, followed by the top 30 algorithmically extracted key phrases from the chapter as a whole.
Intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text on the opening pages of each chapter. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.

Do not use for reproduction, copying, pasting, or reading; exclusively for search engines.

OCR for page 238
Tanker Spills: Prevention by Design APPENDIX F Comparative Study on Potential Oil Spill in Collision and/or Grounding— Different Tanker Designs DET NORSKE VERITAS The appendixes to this study are not included. Committee comments on assumptions and conclusions can be found at the end of the Det norske Veritas report on page 299 and following.

OCR for page 238
Tanker Spills: Prevention by Design

OCR for page 238
Tanker Spills: Prevention by Design PREAMBLE The recent casualties of larger oil tankers resulting in severe oil spills, together with oil spills from a number of smaller tankers grounded, have caused growing concern in many countries. A review of the casualties indicates that human and operational aspects have been decisive factors in most casualties. In view of the serious consequences of every single oil spill, however, improvements in existing measures related to ship design and equipment will also have to be considered with the objective of reducing the potential for oil spills. In this report an assessment of the potential oil outflow from VLCCs and 40,000 dwt tankers in case of collision and grounding is presented. The objective of the study has been to investigate how double sides, double bottom, tank size and tank location influence the amount of oil escaping from a damaged VLCC. The study is based on available statistical information on damage location and extent. A novel probabilistic approach has been adopted to account for uncertainties in damage location and extent. Additional oil spill caused by a grounded tanker possibly breaking in parts has not been considered in this study as this would necessitate a rather detailed analysis on the ultimate strength margins in damaged condition. Possible additional oil spill when salvaging a grounded tanker has not been considered in the study. Neither has structural damage caused by explosion and fire resulting in oil spill been considered. The results presented in this report are applicable to tankers in general. The VLCC study is an extension of a joint Scandinavian study carried out earlier in 1990/8/. At the request of the National Research Council, two additional VLCCs - the intermediate oil tight deck and the hydrostatically balance loaded VLCC respectively, have been analysed and the results included in the graphs together with the other VLCCs. The 40,000 dwt tanker study was carried out exclusively for National Research Council. Det norske Veritas Classification A/S and the participants of the first VLCC study accept no liability for any loss, damage or expense allegedly caused directly or indirectly, when using or referring to the results presented in this report.

OCR for page 238
Tanker Spills: Prevention by Design List Of Contents     1 SCOPE OF WORK         2. ESTIMATION OF OIL OUTFLOW IN COLLISION AND GROUNDING         2.1 Oil outflow in collision         2.1.1 Basic Assumptions         2.1.2 Damage Analysis Procedure         2.2 Oil Outflow in Grounding         2.2.1 Basic Assumptions         2.2.2 Damage Analysis Procedure         3. PROBABILISTIC RANKING OF VLCC DESIGNS         3.1VLCC Designs Analysed         3.1.1 General Features         3.1.2 Particulars of VLCC Designs         3.2 Ranking of VLCC Designs         3.2.1 Ranking in Collision         3.2.2 Ranking in Grounding         3.2.3 Combined Ranking         3.2.4 Impact of a Vacuum System on Oil Outflow in Grounding         3.3 Conclusions and Recommendations         4. PROBABILISTIC RANKING OF 40,000 DWT TANKERS         4.1 The 40,000 dwt Designs Analysed         4.1.1 General Features         4.1.2 Particulars of 40,000 dwt Tankers         4.2 Probabilistic Ranking of 40,000 dwt Tankers         4.2.1 Ranking in Collision         4.2.2 Ranking in Grounding         4.2.3 Combined Ranking         4.3 Conclusions and Recommendations         5. ESTIMATED OIL OUTFLOW FROM A 80,000 DWT TANKER         5.1 Comparison of Performance for VLCCs vs 40,000 dwt Tankers         5.2 Conclusions and Recommendations         6. SUMMARY OF RESULTS         REFERENCES         APPENDIX 1: A Short Description of PROBAN         APPENDIX 2: PROBAN Limit State Functions for Design 1A         APPENDIX 3: Trim, Stability, Bending Moments and Shear Forces. Weight Estimation for VLCC    

OCR for page 238
Tanker Spills: Prevention by Design 1. Scope Of Work The objective of this study is to shed some light on the merits of building tankers with double sides and/or double bottom in order to reduce the probability of oil outflow in case of collision and/or grounding. The analysis is focused on the cargo area only; - no crude oil is assumed to be located forward of the collision bulkhead, nor aft of the forward engine room bulkhead. Possible fuel oil leakage has not been considered in the study; - the fuel oil tanks are located aft of the forward engine room bulkhead. As the fuel tank location is assumed similar in all designs studied, pollution potential from fuel tanks is equal and hence not included in the study. In this study the potential oil outflow from 11 different VLCC designs with double sides and/or double bottom have been compared with the potential oil outflow from a modern conventional VLCC of 280,000 dwt having segregated ballast tanks. In addition, the potential oil outflow from a modern conventional VLCC SBT with hydrostatically balanced loading of cargo tanks has been evaluated. All VLCC designs have the same main particulars and body plan as the modern conventional VLCC. The length of the cargo area is equal for all VLCCs; the total cargo and ballast capacity respectively is approximately equal to that of the VLCC SBT within the cargo area for all VLCCs analysed. Eight 40,000 dwt designs have been analysed. All designs have the same main particulars and body plan. The length of the cargo area is equal for all designs; - the ballast capacity within the cargo area varies from approximately 14,000 to 17,5000 tons depending on arrangement. The oil outflow in collision and grounding has been estimated using available statistical information on damage extent, combined with basic laws of mechanics and PROBAN1 , a probabilistic analysis program developed by Veritas Research/1/. The calculated potential oil outflow in collision and grounding has been combined using general casualty frequency statistics to provide an overall ranking of the different designs. The influence of vacuum systems on the oil outflow in grounding has been investigated for VLCCs by amending the theory and formulae developed in connection with the DNVC - Pollution Prevention Class Notation work into the PROBAN oil outflow model. The results from the VLCC and 40,000 dwt tanker studies have been discussed and extrapolated for 80,000 dwt tankers. 1)   This program has been installed on computers at following organisations/companies in the United States: NASA, Boeing, ALCOA, Chevron and Conoco. A short description of PROBAN is given in Appendix 1.

OCR for page 238
Tanker Spills: Prevention by Design 2. Estimation Of Oil Outflow In Collision And Grounding The potential oil outflow is estimated separately for collision and grounding casualties, and combined using a weighing based on collision and grounding casualty frequencies for tankers. 2.1 Oil Outflow in Collision 2.1.1 Basic Assumptions Several simplifying assumptions have been made w.r.t. the collision process. In order to avoid optimistic results the assumptions made are conservative. It is assumed that oil begins to escape from a cargo tank when the bow of the striking ship touches and penetrates one of the sides or corners of the tank. No large deformations resulting in yielding of tank sides have been assumed. Hence, at contact the tank is penetrated. In the analysis it is postulated that the bow is wedge shaped and remains wedge shaped after having penetrated the hull plating. As the vertical bow penetrates the tank, the whole tank is ruptured from bottom to top. This assumption has been made in order to conform with the MARPOL/2/ assumption that all oil will escape from a damaged tank over time. A consequence of these assumptions is that the results presented for collision strictly apply for the final condition only i.e. when all oil has escaped from a holed tank. For more accurate estimates on oil outflow during the collision a detailed modeling of bow crushing behaviour and ship side structural response during the collision process, including the changes in ship speeds and headings, and added mass effects, would be necessary/4,5/. This has been beyond the scope for this study. It should be stressed that only damage in way of the collision contact point has been considered. Damages far away from the contact point, such as possible tearing in weld seems due to excessive tension potentially resulting in oil leakage, have not been considered. As for possible explosions due to friction heat and sparks generated in a collision, it is assumed that the inert gas system is effective; - should the system fail then structural damage may become extensive. These aspects have not been considered in the study. SEE COMMITTEE INTERPRETATIONS OF DNV COLLISION ASSUMPTIONS ON PAGE 299. 2.1.2 Damage Analysis Procedure The procedure for calculating the potential oil outflow in collision is shown in the flow chart below (Fig.2.1).

OCR for page 238
Tanker Spills: Prevention by Design The interdependence between collision damage length and damage penetration used in this study is shown in Fig.2.2. No information is given on ship speeds and sizes, nor on the number of tankers included in the basic population of 296 ships. The damage length and damage penetration values in Fig.2.2 include implicitly elastic and plastic deformation of the bow and side structure, the speeds of both ships, changes in ship motions and accelerations as well as any mass effects. A multilognormal distribution for damage penetration and length has been calculated based on Fig.2.2. The distribution has been used in the PROBAN analysis for defining the extent of structural damage. The distribution of collision points along the ship hull is shown in Fig.2.3 for 332 collisions. The distribution includes struck and striking ships. This explains the large percentage of collisions in the bow region. Considering only struck ships, the distribution becomes bell shaped with about 30% of collisions aft and in the bow. No information on ship types or sizes is given; - it is assumed that the distribution is valid also for tankers. Fig.2.2 includes damages to struck ships as well as striking ships. No information is given on the number of damages to struck and striking ships respectively. Referring to the collision point distribution in Fig.2.3, damages to the relatively more rigid bow region are emphasized in Fig.2.2. The multilognormal distribution calculated may thus be somewhat optimistic for side damages i.e. the damage lengths and penetrations might be short. The effect of a double side structure on damage penetration has been considered by introducing an equivalent penetration depth. This equivalent penetration depth is obtained as a function of the single side penetration depth, the stiffness of the inner side structure in relation to the outer side structure and the distance between the outer and inner sides. For simplicity, elastic deformation only is assumed. In the analysis it has been assumed that the inner side has a mean stiffness of 0.7 of the outer side with a standard deviation of 0.1. The ship length has been divided into 50 sections of equal length. Using the damage extent distribution, the cargo volume punctured i.e. number and size of tanks, and the resulting oil outflow, has been calculated for the 51 locations (ends of each section) along the ship hull. The collision point distribution has been used for weighing the oil outflow calculated at the 51 locations. The oil outflow is obtained as volume (m3), and may be converted to tons using the specific gravity of oil. In Appendix 2 the limit state functions used for calculation of oil outflow in collision and grounding are given for design 1A as illustration.

OCR for page 238
Tanker Spills: Prevention by Design Fig.2.1. Flow Chart for Collision Analysis

OCR for page 238
Tanker Spills: Prevention by Design Fig.2.2. IMO Data on Stability and Subdivision - 296 Collisions/3/ Fig.2.3. Location of Damage for 332 Collisions

OCR for page 238
Tanker Spills: Prevention by Design 2.2 Oil Outflow in Grounding 2.2.1 Basic assumptions The simplifying assumptions made in the grounding analysis relate to the damage extent and oil outflow. Again, the assumptions made are considered to be conservative. It is assumed that the ship has forward speed in grounding, i.e. the damage starts in the bow of the ship and develops towards the stern. Damage caused by grounding while the ship is adrift, when the ship is turning or going astern, has not been considered. The ground is assumed to be a solid rock which does not crush during the grounding process. The rock is assumed to be wedge shaped (i.e. triangular in the transverse yz-plane) having a constant breadth. A sensitivity analysis has been performed regarding the influence of the rock shape on damage extent (see below). Grounding on sand banks or mud bottom has not been considered as the damage in this case primarily would be local indentations, possibly combined with tearing of welds. As the rock comes into contact with one of the sides of a tank, the bottom or a corner of a tank, the hull plating is assumed to be penetrated. Oil begins to escape until hydrostatic equilibrium is achieved in the damaged tank. Should the ship side be damaged in grounding, then it is assumed that all oil which is below the intact side plating, will escape from the tank and be substituted with water. In addition, due to the reduced hydrostatic pressure, the oil level will drop in the tank proportionally to the specific gravities of water and oil, and height of side damage. In the analysis statistical information on maximum vertical extent of damage has been used for the whole damage length. This conservative assumption has been made due to lack of detailed information on the vertical damage extent, or penetration, as a function of damage length. Consequently, the damage lengths calculated may be too short as too much energy is absorbed vertically (see discussion below). Possible tearing of welds well away from the damage location causing leakage of oil have not been considered (ref. similar assumption in the collision analysis above). The influence of ship motions during the grounding process have not been considered in detail. During a grounding the ship may develop a forward trim due to shallow water i.e. downward suction of the bow, or due to sudden loss of buoyancy as the forward bottom plating is peeled off. The ship may also run up on the rock lifting the bow. In the analysis a variation in ship draught with ± 0.5 m has been included partly to cover these aspects.

OCR for page 238
Tanker Spills: Prevention by Design Tidal water effects on oil outflow have not been considered in this study. At low tide more oil may escape from a ship sitting on the rock because of reduced draught. Tidal water may also result in excessive hogging moments when the ship sits on the rock. These moments might result the hull breaking in parts with uncontrolled pollution as an outcome; - this problem may be studied separately in detail using available programs for ultimate strength analysis of the hull girder. SEE COMMITTEE INTERPRETATIONS OF GROUNDING ASSUMPTIONS ON PAGE 299. 2.2.2 Damage Analysis Procedure The flow chart for the grounding damage analysis procedure is shown in Fig.2.4. In analysing the extent of grounding damage, the energy absorbed by the bottom structure being crushed is estimated by following expression/6/: where Vs is the volume of material displaced As is the total area of fracture or tearing The kinetic energy to be absorbed by the bottom structure in grounding may be expressed as where m = ms (1+as) ms = mass of the laden ship as = added mass coefficient for surge v = ship's speed at grounding Assuming all kinetic is transformed into deformation energy, the extent of grounding damage may be expressed for a single bottom ship as follows: or where Ld is length of damage in longitudinal direction tp e is equivalent thickness of bottom plating tp a is actual thickness of bottom plating Bd is the breadth of damage

OCR for page 238
Tanker Spills: Prevention by Design Fig.4.18. Combined Ranking, 5 kn Fig.4.19. Combined Ranking, 10 kn

OCR for page 238
Tanker Spills: Prevention by Design CONCLUDING FROM THE ABOVE ANALYSIS ON 40,000 DWT TANKERS: narrow and long side tanks reduce the potential oil outflow in collision and grounding as does reduced tank volumes. introducing a centreline bulkhead reduces the potential oil outflow in collisions when compared with short wide cargo tank designs. for ships ‹ 50,000 dwt, the width of double side could be related to B/25 with a minimum of 760 mm according to the Chemical Code Type 2 ships in order not to reduce the cargo capacity (ref. limited influence of increased width on oil outflow), and allowing for centreline bulkheads. double bottom provides an effective barrier against oil pollution for the 40,000 dwt tankers analysed. The height of double bottom should not be less than B/15. double bottoms reduce oil outflow in groundings causing bilge and side damage single bottom designs should be fitted with a vacuum system, alternatively loaded only to a hydrostatic balance level to reduce oil outflow in grounding. the intermediate oil tight deck design performs quite well but will always leak some oil in collision and grounding SEE COMMITTEE COMMENTS ON DNV CONCLUSIONS FOR 40,000 DWT TANKERS ON PAGE 301.

OCR for page 238
Tanker Spills: Prevention by Design 5. ESTIMATED OIL OUTFLOW FROM A 80,000 DWT TANKER Based upon above studies for VLCCs and 40,000 dwt tankers, the potential oil outflow from a 80,000 dwt is discussed in this Chapter. No supporting calculations have been carried out with PROBAN. 5.1 Comparison of Performance for VLCCs vs. 40,000 dwt Tankers Original VLCC and 40,000 dwt tanker designs – single skin: The smaller tanker with large ballast side tanks forward performs relatively much better than the VLCC design which is penalised for not having ballast in the same side tanks. The high ballast/cargo volume ratio for the 40,000 dwt necessarily improves the performance of the original design in relation to other 40,000 dwt designs. Comparing the potential oil spill volume to total cargo volume ratio, the difference is only ˜ 1.3% in ratios for the VLCC and the 40,000 dwt tanker. In grounding, the overall performance is rather poor for both the VLCC and the 40,000 dwt tanker. Again ballast in forward side tanks reduces the oil spill, but due to long damage lengths at 10 kn, the effect is not very pronounced. Ballast in forward side tanks reduce also the potential for oil spill in bilge and side damages caused by grounding; – at low grounding speeds the reduction is considerable. The volume of oil spill in grounding in relation to total cargo volume is virtually the same for the VLCC and the 40,000 dwt tanker at 5 and 10 kn respectively. At 10 kn, the total amount oil both ships leak is 2.5–3 times more than at 5 kn. A vacuum system will reduce the oil outflow in grounding by ˜90% at 5 kn provided a vacuum can be maintained in 40% of the cargo tanks. At 10 kn the relative reduction will be smaller because damage lengths increase and more tanks are damaged. Increasing the vacuum capacity will reduce the oil outflow correspondingly. Hydrostatically balanced loading of cargo tanks will reduce oil outflow in grounding to less than 10% of the original amount. For a 80,000 dwt SBT tanker with conventional tank arrangement, much the same behaviour is expected.

OCR for page 238
Tanker Spills: Prevention by Design Double side and double bottom designs: Oil outflow from VLCCs with double sides and double bottom is radically lower compared with the VLCC SBT – the overall oil outflow is ˜33% of the outflow from the VLCC SBT at 10 kn. For the 40,000 dwt tanker the corresponding percentage is ˜37 at 10 kn. The same reduction in overall oil outflow may be expected for a 80,000 dwt tanker provided the ship meets with the B/20 double side and B/15 double bottom recommendations. Double sides and single bottom designs: In the VLCC and the 40,000 dwt tanker with double sides and single bottom, the double side width is appr. B/9. The probability for no oil outflow in collision is for both ships ˜42.0%. Both ships perform rather poorly in grounding. The VLCC is likely to leak 80% more than the VLCC SBT at 10 kn. The 40,000 dwt tanker 25 % more than the original 40,000 dwt tanker. A vacuum system reduces the potential oil spill in grounding at 5 kn for the VLCC by ˜70 %. Increasing the speed results in relatively more oil spill due to increased damage lengths. Hydrostatically balanced loading of cargo tanks would virtually eliminate all oil outflow in grounding as the probability for a bilge damage to extend into the cargo tanks is very low(assuming double side width B/9 as above). For the 80,000 dwt tanker, the oil spill in grounding may be 35–40% in comparison. This value depends on the L/D ratio for the design. Single side and double bottom designs: In collision, the 40,000 dwt tanker with ballast in double bottom only, leak considerably more than the original design. The increase in total oil outflow is some 65 % in spite of the ship having a centreline bulkhead and long tanks. In case of the VLCC, the single side/double bottom design is likely to leak ˜30% less than the original VLCC. As stated above, the reason for poor behaviour of the original VLCC is due to no ballast in forward side tanks. In grounding neither the VLCC(B/dbh = 8.6) nor the 40,000 dwt tanker(B/dbh=7.0) is expected to leak any oil. For a similar 80,000 dwt tanker no difference in performance is expected.

OCR for page 238
Tanker Spills: Prevention by Design Intermediate Oil Tight Deck Designs: In collision, the performance of the intermediate oil tight deck design is a little better for the VLCC and a little inferior for the 40,000 dwt tanker. In grounding, the performance is good for both sizes. For the 80,000 dwt, the same behaviour is expected. 5.2 Conclusions and Recommendations: Following general conclusions may be drawn regarding oil outflow from a 80,000 dwt tanker: long and narrow tanks are likely to leak less than short and wide tanks. A centreline girder should be fitted to reduce oil outflow in collisions a double bottom of B/15 should be fitted, an alternative may be a vacuum system with high capacity or hydrostatic loading of cargo tanks a double side should be considered to restricted pollution in low-energy collisions; preferred width B/20 high L/D reduces oil outflow in grounding SEE COMMITTEE COMMENTS ON DNV CONCLUSIONS FOR THE 80,000 DWT TANKER ON PAGE 302.

OCR for page 238
Tanker Spills: Prevention by Design 6. SUMMARY OF RESULTS The results should strictly be used only for ranking the VLCC and 40,000 dwt tanker designs as the statistics for damage extent used contain all types and sizes of ships. In addition, some simplifications have been made in the study influencing structural response in collision and grounding. The adopted procedure for assessing the potential oil outflow in collision and grounding seems promising, however, Provided more detailed statistics is applied for tanker damages, the results would improve. Summing up conclusions in Chapters 3–5, following measures should be considered for reducing the probability for oil outflow in collision and grounding: a low L/D ratio will increase oil outflow in grounding wide tanks are likely to spill more oil than long narrow tanks location of ballast forward in side tanks provides good collision protection double side protects against oil outflow in low-energy collision width of double side should tentatively be B/25 for ships ‹ 50,000 dwt, and B/20 for ships above, in order to be effective in ship/ship collision double bottom reduces oil outflow in grounding double bottom protects against bilge and side damage in grounding the double bottom should tentatively be B/15 vacuum systems reduce oil outflow in grounding hydrostatic loading of cargo tanks reduces oil outflow in collision and grounding intermediate oil tight deck reduces oil outflow in grounding

OCR for page 238
Tanker Spills: Prevention by Design REFERENCES /1/ PROBAN Theory Manual, Veritas Research 1986 /2/ IMO International Conference on Marine Pollution 1973 /3/ IMO Regulations on Subdivision and Stability of Passenger Ships, Resolution A.265(VIII) /4/ Valsgård, S. – Jørgensen, L.: Evaluation of Ship/Ship Collision Damage Using a Simplified Nonlinear Finite Element Procedure, Int. Symp. on Practical Design in Shipbuilding PRADS'83, Tokyo/Seoul, 1983 /5/ Køhler, P.E. – Jørgensen, L.: Ship Ice Impact Analysis, 4th Int. Symp. on Offshore Mechanics and Arctic Engineering OMAE, Dallas, 1985 /6/ Vaughan, H.: Bending and Tearing of Plate with Application to Ship Bottom Damage, The Naval Architect, May 1978 /7/ Wierzbicki, T. et. al.: Damage Estimates in High Energy Grounding of Ships, MIT, June 27, 1990 /8/ Køhler, P.E. et. al.: Potential Oil Spill from Tankers in Case of Collision and/or Grounding – A Comparative Study of Different VLCC Designs, DNVC Report 90–0074, May 1990 /9/ IMO International Code for the Construction and Equipment of Ships Carrying Dangerous Chemicals in Bulk, London 1986

OCR for page 238
Tanker Spills: Prevention by Design INTERPRETATIONS AND COMMENTS OF THE COMMITTEE ON TANK VESSEL DESIGN CONCERNING SECTIONS OF THE DET NORSKE VERITAS REPORT Following are the committee's interpretations of the most significant assumptions in the collision analysis. Statements drawn from the DnV report are in italics (some excerpts have been edited for brevity and style), and committee comments are in standard type. DNV Collision Assumptions (Section 2.1.1) Several simplifying assumptions were made with respect to the collision assessment. To avoid overly optimistic results, the assumptions were conservative. Therefore, the committee feels that outflow may be overstated on a consistent basis. The study assumes that oil begins to escape from a cargo tank when the bow of the striking ship touches a side or corner of the tank. No large deformations resulting in yielding of the tank sides were required. Thus, a tank is considered penetrated at contact. The striking bow is assumed to be wedge-shaped, and it remains in shape after penetrating hull plating. As the vertical bow penetrates a tank, the full height of the tank is ruptured, from bottom to top. This assumption was made to conform with the MARPOL assumption that all oil will escape from a damaged tank over time. One consequence of these assumptions is that the results apply strictly to the final condition — that is, after all oil has escaped from a holed tank. To obtain more accurate estimates on oil outflow during the collision, it would be necessary to conduct detailed modeling of bow crushing behavior and ship side structural response during the collision process, taking into account the changes in ship speeds and headings, and added mass effects. This is beyond the scope of the present study. Furthermore, the committee feels that additional research would be required even to develop the capability to obtain these estimates. In the present study, only damage in way of the collision contact point is considered. Damage elsewhere, such as possible tearing in weld seams due to excessive tension (potentially resulting in oil leakage), have not been considered. As for possible explosions due to friction heat and sparks generated in a collision, the study assumes that the inert gas system is effective and no fires ensue. Grounding Assumptions (Section 2.2.1) Following are the committee's interpretations of the most significant assumptions in the DnV grounding analysis. Statements drawn from the DnV

OCR for page 238
Tanker Spills: Prevention by Design report are in italics (some excerpts have been edited for brevity and style), and committee comments are in standard type. The simplifying assumptions relate to extent of damage and oil outflow. Again, the assumptions are conservative, so outflow may be overstated consistently. The analysis assumes that the grounding ship has forward speed, so that the damage is initiated at the bow and develops towards the stern. Damage caused by grounding while the ship is adrift, turning, or going astern, has not been considered. The bottom surface is defined as a solid rock, which does not crush during the grounding process. The rock is assumed to be wedge-shaped (triangular in the transverse plane), with a constant breadth. The Vaughan method outlined in Chapter 3 was used in this analysis. For purposes of this study, the hull plating is considered penetrated when the rock comes into contact with a tank side, bottom, or corner. Oil escapes until hydrostatic equilibrium is achieved in the damaged tank. Should the ship side be damaged, then all oil below the intact side plating will escape, to be replaced by water. Due to the reduced hydrostatic pressure, the oil level will drop proportionally to the specific gravities of water and oil, and the height of side damage. Statistics on maximum vertical extent of grounding damage were applied to the full length of damage. This conservative assumption was made due to lack of detailed information on vertical damage extent, or penetration, as a function of damage length. Consequently, the damage lengths calculated may be too short, due to overstatement of the energy absorbed vertically. Possible tearing of welds well away from the direct grounding damage have not been considered. Nor was the influence of ship motions during the grounding process considered in detail. During a grounding, a ship may develop a forward trim due to either downward suction of the bow, or sudden loss of buoyancy as the forward bottom plating is peeled off. Or, the ship may run up on the rock, lifting the bow. In this analysis, a variation in ship draft (+ 0.5 m) was included, partly to account for these factors. Tidal effects on oil outflow were not considered. At low tide, more oil may escape from a ship sitting on a rock, due to reduced draft. Tides also may result in excessive hogging moments, which might result the hull breaking apart, with uncontrolled pollution. This matter requires a separate study, using available computer programs for ultimate strength analysis of the hull girder. DNV Conclusions for VLCCs (Section 3.3) Following are the committee's comments on each of DnV's conclusions for the VLCC designs. DnV conclusions are in italics, and the committee's comments are in standard type.

OCR for page 238
Tanker Spills: Prevention by Design Narrow and long cargo tanks and increased double bottom height clearly reduces the oil outflow as does reduced tank volumes. The committee agrees. Ballast forward in side tanks will reduce the potential oil spill in collision, and in low speed groundings. The committee agrees. VLCCs should have a double bottom height approaching B/15 in order to be effective against pollution in grounding. The committee agrees, as one approach. In addition, double sides having a width of at least B/20 should be considered. The committee prefers wider double sides. Single bottom designs should preferably be fitted with a vacuum system in order to reduce the amount of oil escaping in grounding. The committee feels that the vacuum system needs further evaluation. Hydrostatic loading may be an alternative to vacuum systems, the lost cargo is not prohibitive at A-freeboard. The increased draft may create problems in coastal waters and loading/discharge terminals. While the committee sees some benefit to hydrostatic loading, as noted in this statement, it perceives additional problems and feels hydrostatic loading should be considered for existing but not new vessels (except in conjunction with structural designs employing double hulls or double sides). Intermediate oil-tight decks may be effective against oil pollution if adequate collision protection is provided. The committee feels that the intermediate oil-tight deck, when combined with double sides, shows promise and deserves further consideration. Low L/D ratios (=10) contribute to excessive pollution in grounding due to high freeboard. A higher L/D ratio (=12) would significantly reduce the amount of oil escaping in grounding. The committee feels that high freeboard, not L/D ratio, is the relevant issue. DNV Conclusions for 40,000 DWT Tankers (Section 4.3) Following are the committee's comments on each of the DnV conclusions for the 40,000 DWT tankers. DnV conclusions are in italics, and committee comments are in standard type. Narrow and long side tanks reduce the potential oil outflow in collision and grounding as does reduced tank volume. The committee agrees. Introducing a centreline bulkhead reduces the potential oil outflow in collisions when compared with short wide cargo tank designs. The committee agrees but points out that centerline bulkheads may introduce possible asymmetric damage scenarios. For ships over 50,000 DWT, the width of double side could be related to B/25 with a minimum of 760 mm according to the Chemical Code Type 2 ships in order not to reduce the cargo capacity (ref. limited influence of increased width on oil outflow), and allowing for centreline bulkheads. The committee prefers wider double sides.

OCR for page 238
Tanker Spills: Prevention by Design Double bottom provides an effective barrier against oil pollution for the 40,000 DWT tankers analysed. The height of double bottom should not be less than B/15. The committee agrees that double bottoms are beneficial in terms of protection from groundings. Double bottoms reduce oil outflow in groundings causing bilge and side damage. The committee agrees that double bottoms, as well as double sides, would reduce outflow. Single bottom designs should be fitted with a vacuum system, alternatively loaded only to a hydrostatic balance level to reduce oil outflow in grounding. The committee feels that the vacuum system needs further evaluation and that hydrostatic loading should be considered for existing but not new vessels except in conjunction with structural designs employing double hulls or double sides. It does concur with the DnV conclusion that some oil leakage will occur if the hull is penetrated. The intermediate oil-tight deck with double sides will provide a reduction of outflow in high energy groundings. The intermediate oil-tight deck design performs quite well but will always leak some oil in collision and grounding. The committee feels that the intermediate oil-tight deck, when combined with double sides, shows promise and deserves further consideration. DNV Conclusions for the 80,000 DWT Tanker (Section 5.2) Long and narrow tanks are likely to leak less than short and wide tanks. A centreline girder should be fitted to reduce oil outflow in collisions. The committee agrees with the benefits of long and narrow tanks and centerline bulkheads, but points out that centerline bulkheads may introduce possible asymmetric damage scenarios. A double bottom of B/15 should be fitted, an alternative may be a vacuum system with high capacity or hydrostatic loading of cargo tanks. The committee agrees with the benefits of double bottoms in terms of grounding protection, but feels that the vacuum system needs further evaluation, and recommends that hydrostatic loading be considered for existing but not new tank vessels except in conjunction with structural designs employing double hulls or double sides. A double side should be considered to restrict pollution in low-energy collisions; preferred width B/20. The committee prefers wider double sides. High L/D reduces oil outflow in groundings. The committee feels that freeboard, not L/D ratio, is the relevant issue.