Caribbean Disaster Mitigation Project
Implemented by the Organization of American States
Unit of Sustainable Development and Environment
for the USAID Office of Foreign Disaster Assistance and the Caribbean Regional Program

USAID Logo OAS Logo


Caribbean Electric Utility Services Corporation

Case Study of the Effects of Hurricane Luis on the Buildings and other Structures of the Electricity Section of the Antigua Public Utilities Authority


CONTENTS

Presentation
1. Introduction
2. Hurricane Luis
3. General Assessment of Damage in Antigua
4. Assessment of Damage to APUA-Elec Facilities
5. General Issues Affecting Failures and Successes
6. The Cost of Mitigating Damage
7. Recommendations
A1. Terms of Reference
A2. Figures
A3. Tables
A4. Photographs
A5. Barry Pinnock's Report
Notes


1996-02-23

CARILEC
Caribbean Electric Utility Services Corporation
Tile World Building
Bois d'Orange
P O Box 2056
Gros Islet
ST LUCIA

Dear Sirs,

Hurricane Damage to APUA (Elec) Facilities

In accordance with your instructions dated 18 September 1995, we have carried out field surveys and analyses of the facilities of the Antigua Public Utility Authority's Electricity Section (APUA-Elec) which were subjected to Hurricane Luis. Your stated Terms of Reference asked us to concentrate on the buildings. However, at your request and that of the Barbados Light & Power Co Ltd (BL&PC), we also reviewed those aspects of the Transmission and Distribution Systems within our non-specialist knowledge.

In addition to reporting specifically on the APUA-Elec facilities we felt it desirable to provide a general background to the event and to provide general comments and recommendations for future action. A power company does not exist in a vacuum. It is part of a community. If the community is badly damaged by a natural hazard the customer base of the power company is directly and adversely affected and business interruption becomes a real issue, even if the power company's facilities are undamaged. Therefore, it is in the interest of power companies not only to reduce the vulnerabilities of their own facilities but also to promote the general adoption of appropriate standards and codes in their communities.

The results of Hurricane Luis in Antigua indicate clearly that success is possible. By success we mean the limiting of losses in Category-3 hurricanes to tolerable levels (low, single-digit percentages) and ensuring that Category-4 hurricanes do not lead to national disasters with losses approaching the GDPs of the countries. We will not be able to eliminate losses completely so there will always be a role for insurance. However, insurance premium levels would sensibly be affordable with the benign scenarios described above.

We wish to thank the personnel of APUA-Elec (Eugene Benjamin in particular), BL&PC (Oliver Jones in particular) and CARILEC (Christopher Farrell in particular) for assisting us with this exercise. We wish also to acknowledge the support of Mr. Jan Vermeiren of the Organisation of American States and the presence of Dr Peter Vickery as an observer during most of the field surveys.

Our detailed report follows.

Yours faithfully,
CONSULTING ENGINEERS PARTNERSHIP LTD

Tony Gibbs

TG/acc


1 INTRODUCTION

1.1 Background

Antigua-Barbuda is in an area subject to multiple natural hazards. Of these, the hazards of earthquakes and hurricanes impact most on the buildings and other structures. In the past fifty years Antigua-Barbuda has suffered from four serious hurricanes and one serious earthquake. In addition, there have been many less-significant events which nevertheless caused damage to the built environment with the attendant disruption of the normal functioning of the Electric Utility Services.

Money spent on repairs and replacements of damaged and destroyed facilities inevitably leads to a pattern of "two steps forward and one step backward". In the past, catastrophe insurance cover was sufficiently inexpensive and available to permit the adequate funding of repairs and loss of profits. At present, and in the immediate future, insurance cover is both scarce and expensive. Some degree of co-insurance is indicated and this therefore focuses the attention of electric utilities on the reduction of vulnerability of their facilities to natural hazards. Also, in the future, there are likely to be increasing demands from the general population for their critical facilities to perform satisfactorily during, and immediately after, hurricanes and sub-catastrophic earthquakes.

The knowledge and materials are available to achieve success. What is lacking is the will.

The efforts of the Caribbean Electric Utility Services Corporation (CARILEC) and the Caribbean Disaster Mitigation Project (USAID/OAS-CDMP)1 in the area of mitigation are important in the region's thrust towards sustainable development. The recent event of Hurricane Luis in Antigua-Barbuda has provided opportunities, not only for collaborative exercises in reconstruction assistance, but also for learning from the failures and successes and for determining and demonstrating the feasibility of achieving almost total success in future hurricanes.

1.2 Terms of Reference

This Study responds directly to the formal request from CARILEC dated 26 September 1995. It also responds to the request from the Barbados Light & Power Company Ltd (BL&PC) for information specifically on the performance of the transmission and distribution (T&D) systems of APUA-Elec during Luis. The detailed Terms of Reference from CARILEC are reproduced in Appendix 1.


2 HURRICANE LUIS

2.1 General Meteorological Information

Hurricane Luis struck Antigua & Barbuda on 04 and 05 September 1995. Luis was a classical, Category-4 storm; almost perfectly formed; large in extent; loaded with moisture; with a very distinct eye of 70 kilometres in diameter and a forward motion of 17 kilometres per hour. (See the satellite photograph reproduced as a frontispiece.) Because of its overall size and slow forward motion, the hurricane impacted on Antigua for an uncommonly long period. Severe storm conditions lasted for about 30 hours during which time about 250 millimetres of rain fell.

The real environment in a hurricane consists of strong,
turbulent winds (sustained for many hours), that
change slowly in direction as the storm passes, and
carry large amounts of debris while accompanied by
torrential rains.

Prof Joseph Minor (modified by Tony Gibbs)

The above-quoted description of a hurricane was well-exemplified by Hurricane Luis in Antigua-Barbuda.

It is always difficult to get reliable information on wind speeds in hurricanes. The subject event was no exception. At the US Satellite Tracking Station near to the airport the highest recorded gust was 129 knots or 66 metres per second (ms-1). This was at a height of 21 metres above adjacent ground2. A reasonable conclusion is that the maximum winds in the eye wall were equivalent to 65 to 70 ms-1 averaged over 3 seconds at a height of 10 metres at the coastline. The eye passed 50 kilometres north of Antigua so that it was the south-west, south and south-east eye walls that impacted on the northern part of this island. This meant that Antigua, in particular, was spared the full brunt of Luis. Indeed, the wind forces in the north eye wall would have been about 33% greater than those in the south eye wall. It can be concluded that the wind speeds in Antigua during Luis were no greater than are recommended by the current, non-mandated standards in the sub-region. The Caribbean Uniform Building Code (CUBiC) recommends a "reference pressure" equivalent to 56 ms-1 and the OAS/NCST/BNSI/BAPE3 Code recommends a "basic wind speed" of 64 ms-1.

2.2 Post-disaster Reconnaissance

The author (Tony Gibbs) travelled to Antigua in advance of the arrival of Hurricane Luis, once he was reasonably sure that Luis would make landfall in that location. At the invitation of its Director, Mr. Patrick Jeremiah, he spent the entire period (30 hours in all) of the storm passage in the Meteorological Station at VC Bird International Airport. There he was able to have a first-hand, gust-by-gust picture of the event, not only visually, but through the various instruments in use at the Station and through the special channels of communication available to such a facility.

Immediately after the storm the author proceeded to inspect the damage done to buildings and other structures in all parts of Antigua. Time is of the essence in these exercises since the evidence is soon tampered with and often cleared away completely. Tony Gibbs has been carrying out post-hurricane assessments for close on two decades and he is of the view that timing is an important factor in these matters.


3 GENERAL ASSESSMENT OF DAMAGE IN ANTIGUA

3.1 General Overview of the Damage

A fuller description of the post-Luis conditions of the APUA-Elec facilities in Antigua is given in the following Section 4 of this report.

The level of damage in Antigua was equivalent to two-thirds of the gross domestic product (GDP) of the country. Such an event has the potential to set back the development of a small island independent state by several years. In particular, much damage was done to essential facilities in the country. These facilities include telecommunications; water supply and distribution; electricity generation, transmission and distribution; and the public health services.

Subsequent to his visit at the time of the event the author was afforded the opportunity, through CARILEC and BL&PC, to review the damage done to the electricity sector. The analyses of causes of failures indicate quite clearly how most of the failures could have been reduced to manageable amounts and, in many cases, eliminated completely with little incremental effort and cost.

Damage to buildings was mainly due to weak connections of light-weight roofing and siding materials, impact damage to glazed openings from flying objects, inadequate fixings of windows and external doors and water damage from the torrential rains. There were also examples of catastrophic collapse of entire buildings due to unsound structural concepts. The lack of maintenance of building components contributed significantly to the damage. In the cases of structures not associated with buildings (e.g. telecommunication towers and transmission systems) inadequate specification of performance criteria at the procurement and design stages was an important factor in the failures. The actual wind speeds were not greater than should have been expected in a 1-in-50-year event. The introduction of mandatory building standards and codes would have a significant, positive impact in reducing losses in future hurricanes.

3.2 Natural Hazards and Disaster Mitigation

Antigua is not only in the regular path of severe hurricanes (See Figure 1 in Appendix 2 showing the region's isoline map of Category 3 hurricanes, produced by the University of the West Indies). It is also located in the most hazardous area of seismic activity in the Caribbean Archipelago. (See Fig2-Appendix2 showing the sub-region iso-acceleration map, provided by Dr John Shepherd - 1000 gals = gravity) The interrelationship of (and differences between) wind-resistant and earthquake-resistant design must not be lost sight of in the reconstruction process.

Earthquakes and hurricanes are not natural disasters, they are natural hazards which sometimes lead to manmade disasters. In these days of widespread technological education, sophisticated research, reliable building materials, computer-based geographical information systems and satellite-assisted warning programmes, hurricanes in the Caribbean should not lead to disasters. The one exception to this would be vulnerable agricultural crops, such as bananas. Although much less is known about earthquakes and there is effectively no warning system for these events, sufficient is known about the effects of earthquakes on structures to prevent disasters and to keep damage from earthquakes to tolerable levels.

It is now evident that disasters due to natural hazards are largely preventable and soon the public will demand deliberate actions to protect communities against such hazardous events. Disaster mitigation must therefore be made an essential ingredient in development planning and capital works projects. In the same way that environmental impact assessments (EIAs) have now become routine, so too should natural hazard impact assessments (NHIAs) be a standard requirement in the planning of projects.


4 ASSESSMENT OF DAMAGE TO APUA-ELEC FACILITIES

4.1 Field Visits

The first series of visits made specifically to investigate damage to APUA-Elec's facilities (under contract to CARILEC) was made between 28 September and 02 October 1995. On 28 September the team for the field trip comprised:

Oliver Jones BL&PC
Eugene Benjamin APUA-Elec T&D Supervisor4
Peter Vickery Observer from Applied Research Associates
Tony Gibbs CEP

Other persons with whom contact was made at the Cassada Gardens office were:

Earl Gardner APUA-Elec T&D Engineer
Peter Benjamin APUA-Elec General Manager
Hugo Ford Trinidad & Tobago Electricity Commission

The names of others are given under the headings for the relevant facilities.

Field visits also took place on 29 & 30 September (E Benjamin, Vickery, Gibbs) and on 02 October (E Benjamin, Gibbs).

On 30 October 1995 Barry Pinnock of CEP made further inspections of the three power-station sites in company with APUA-Elec's A B Segu to gather supplementary information for preparing cost estimates. (Mr Pinnock's report is reproduced as Appendix 5.)

(62 photographs, with captions, are reproduced in Appendix 4. They are important supplements to the main text of this report. Please note that only selected photos have been included in this digital version of the report.)

4.2 Cassada Gardens Complex

(See Photographs 1,2,3,4,33)

4.2.1 Contact Persons

In addition to the persons mentioned in 4.1 contact was made with:

Ruvan Barnarde, Systems Control Engineer
Winston Smith5

4.2.2 General Description

The principal facilities at Cassada Gardens are:

4.2.3 Damage Assessment

The observations made at the Control Room at the time of the visit were:

The observations made at the Transmission & Distribution Section at the time of the visit were:

The observations made at the Office Building at the time of the visit was:

The observations made at the 11 kV Indoor Switchgear Room at the time of the visit were:

The observations made at the Canteen and Workshop at the time of the visit were:

The Fence was in reasonable condition.

(The estimates of repair costs for the buildings is given in Table 1a, Appendix 3.)

4.3 Friars Hill Power Station

4.3.1 Contact Person

In addition to the persons mentioned in 4.1 contact was made with:

Conrad Samuel, Station Superintendent

4.3.2 General Description

The principal facilities at Friars Hill are:

4.3.3 Damage Assessment

The observations made at the Power Station Building at the time of the visit were:

The observations made at the Offices, Stores and Workshop Buildings at the time of the visit were:

(The estimates of repair costs for the buildings is given in Table 1b, Appendix 3.)

4.4 Crabbs Complex

4.4.1 Contact Persons

Gordon Derrick, Mechanical Maintenance Engineer
Lyndon Francis, Plant Engineer

4.4.2 General Description

The principal facilities at Crabbs are:

4.4.3 Damage Assessment

The observations made at the Crabbs Diesel Power Station (See Photos 19,20,34) at the time of the visit were:

The observations made at the Reverse Osmosis Plant at the time of the visit were:

The observations made at the Desalination Cogeneration Plant at the time of the visit were:

The long-term effects of salt-laden winds cannot be established at this time.

(The estimates of repair costs for the buildings is given in Table 1c, Appendix 3.)

4.5 Substations

4.5.1 Belmont Substation (See Photo 36)

There was no wind damage nor did flying objects pose any problems.

Water got into the container-type equipment room. (This location has both the masonry&concrete-type and the container-type equipment rooms.)

4.5.2 Swetes Substation (See Photo 37)

The windows appeared to be double glazed. They were unbroken. (See Photo 23)

69/11 kV outdoor transformer (10MW) is on wheels on rails and not secured. It would be vulnerable in an earthquake. (See Photo 59)

4.5.3 Five Islands Substation (See Photos 24,25)

Construction work was in progress at the station.

4.5.4 Cassada Gardens, Friars Hill and Crabbs Substations

These were inspected at the same times as the relevant power stations and other facilities.

There was little damage to be seen.

4.6 Transmission and Distribution Systems

4.6.1 General Description

Wallaba poles were used up to 1987. During the British colonial period the use of H-frames was popular.

Wallaba poles are gradually being replaced with southern-yellow-pine poles.

In 1987 lattice-steel towers were introduced. The contractor was MAW spa of Milan, Italy.

4.6.2 Foundations

The planting of poles is generally in accordance with standard Antigua practice:

In volcanic rock, 5-foot depth
In other (normal) conditions, 6-foot depth

4.6.3 Typical Examples of Failures

In the Jolly Harbour Area poles carrying a single 3-phase 3/0 AAAC-11kV line were uprooted from their foundations (See Photos 48,49).

West of Old Road, after the planting of poles, the adjacent road was rebuilt. The rebuilding entailed reducing the overall level of the road. Thus the nearby poles ended up being founded in the embankment alongside the road. The embankment did not have sufficient lateral containment strength to hold the poles. (See Photo 52)

In the Jonas Road area white rot was present in many of the broken wallaba poles (See Photos 44,45).

In the Mental Hospital Road area the presence of knots at critical locations in the poles was a contributory factor in breakages.

In the Jonas Road area the overloading of poles with large telephone cables was evident (See Photo 56).

At the Falmouth - Piccadilly Road Junction the overloading of poles led to many breakages (See Photos 54,55).

4.6.4 Failure Modes and Causes

These can be summarised as follows:

With the exception of the last-listed item all of the causes of failures are preventable by convenient means.

4.6.5 Amount of Damage

The percentage losses (broken and leaning or uprooted poles) are 14.06% for the 11 kV system and 3.86% for the 69 kV system. There were no losses in the steel towers of the 69 kV system. (The details are given in Table 2, Appendix 3.)

It must be noted that the absence of trees in Antigua (relative to most Caribbean islands) was an important factor in keeping losses to such low figures.

4.6.6 Reinstatement of the T&D System

Several teams from other Caribbean companies (including the BL&PC) and from further afield were involved in the reinstatement of the T&D System. (The programme details are given in Table 3, Appendix 3 and in Photos 61 and 62. See the BL&PC team in Photo 60.)


5 GENERAL ISSUES AFFECTING FAILURES AND SUCCESSES

5.1 Conceptual Design

This is the single most-important factor determining success or failure of buildings. Once again this was demonstrated during Hurricane Luis in Antigua. With respect to hurricanes, suitable design concepts are particularly important for light-weight structures - timber and corrugated-metal walls and roofs.

Unfavourable features evident in Antigua were:

Favourable features evident in Antigua were:

5.2 Strength of Materials and Sizes of Construction Components

Building materials are supplied in wide ranges of strengths. For example, commonly in Antigua, the ranges of strengths (in Nmm-2) of basic building materials are as indicated below:

material lowest highest ratio
timber 17 105 6.2
corrugated metal roofing 70 410 5.9
reinforcing steel 210 460 2.2
concrete 17 35 2.0
concrete blocks 5 8 1.6

Clearly, these significant differences must be accounted for in construction. The lack of conscious appreciation of these differences can, and did, lead to failures. (It has to be said that such lack of appreciation also led to some accidental successes.)

As well as strength, brittleness is a factor determining success and failure. The best evidence of this in Antigua was the breakage of corrugated asbestos sheeting used as roofs and sidings.

The sizes of construction components are greatly controlled by strengths of materials. Everything else being equal, the stronger the material the smaller the component size needs to be. Of course, practical considerations and aesthetics also have their influences on sizes. Such influences, when benign, lead to larger-than-necessary sizes from a strength point of view. Inadequate sizing was the contributory factor in some of the observed failures, principally those associated with light-weight roofing.

5.3 Analysis

Some of the APUA-Elec buildings are of such a small scale as may not normally warrant detailed, formal, engineering analysis.

Component sizes for small-scale construction are usually determined by tradition and rules of thumb. With the rapid introduction of new materials there isn't the time to develop new traditions. With the rapid expansion of the construction industry there isn't the time to train artisans and craftsmen through apprenticeships, as was done in the past. Rules of thumb are often not known by the new practitioners.

In such an environment some analysis is indicated, even for buildings of modest size. For critical facilities, including all APUA-Elec facilities, analysis must be used to determine or confirm the adequacy of component sizes.

The absence of a conscious engineered approach to many of the buildings was the cause of failures of some components and their connectors.

5.4 Detailing and Connections

In the words of the famous German architect, Mies van der Rohe, "God is in the details". It is difficult to overemphasise the importance of detailing. This is the process of arranging the details". It is difficult to overemphasise the importance of detailing. This is the process of arranging the structural and building elements in such a way that they perform their intended functions by carrying the applied loads safely. Thus the quantity of material may be sufficient but, if the arrangement of the material is inappropriate, failure may result. Thus was evident in some instances in Antigua.

Real-estate people talk about the three most important factors being: "location, location and location". Likewise, for hurricane resistance of lightweight structures, the three important issues can almost be said to be: "connections, connections and connections". The roof sheeting must be adequately connected to the purlins. The purlins must be adequately connected to the rafters. The rafters must be adequately connected to the wall plates. The wall plates must be adequately connected to the wall studs. The wall studs must be adequately connected to the base sleepers. The base sleepers must be adequately connected to the base walls or piers. The piers must be adequately founded. This litany simply says that the wind forces must be carried from wherever they impact on the building all the way into the ground without any weak links along the load path (See Figure 3 in Appendix 2). Hurricane Luis sought out weak links and found several.

5.5 Quality Control During Construction

All of the good work in the planning stages can (and often does) become unstuck by faulty construction. It is generally felt that poor construction is responsible for most of the damage in hurricanes. This is not the view of the author. However, poor construction is a contributory factor in a significant minority of the failures. It can also be said that whereas poor construction can undo good design, analysis and detailing; good construction cannot make up for bad design, analysis and detailing.

APUA-Elec's buildings and structures had their usual share of failures due to poor quality control during construction.

5.6 Non-structural Elements and Issues

Windows and external doors are the orphans of the construction industry and their acts of revenge for lack of attention can be very embarrassing. Usually engineers are not involved in the specification of these items. Usually architects are not equipped to determine the strength requirements for these items. Usually suppliers and contractors cannot be relied on to provide more than the commercial norm, which is inadequate for Category-4 hurricanes (a reasonable requirement for APUA-Elec buildings and structures). There were several failures to be seen, which is not surprising. The failures were sometimes of the fixings to the walls. At other times glass was broken by flying objects. The only ways to deal with vulnerability to breakage are hurricane shutters and laminated glass. The latter approach would still lead to breakage but the weather would be excluded during the hurricane.

The inadequacy of storm-water drainage can lead to much damage. There was some evidence of this at the APUA-Elec facilities. Not enough attention is paid to this aspect of design. Even where drainage is consciously engineered, design criteria are usually inadequate, especially for critical facilities.

The APUA-Elec facility at Crabbs is located on the coastline. Storm surge and wave action are hazards accompanying hurricanes. Predicting the severity of these marine hazards requires specialist advice which nevertheless is subject to much debate. This is compounded by the lack of reliable records of previous events. What is indisputable, however, is the need to be aware of marine hazards in locating critical facilities in coastal areas.

5.7 Maintenance

On the television programme CNN&Company of 13 December 1995 the main topic was urban infrastructure. The problem of decaying roads, bridges, water supply systems and sewerage systems was the subject of much debate. The cost, and political unpopularity, of preventative maintenance was recognised. It was also recognised that preventative maintenance was less expensive, in the long run, than emergency repairs and reconstruction brought about by inadequate maintenance. So it can be seen that even the wealthiest of nations won't willingly spend the funds necessary for proper preventative maintenance.

The inadequacy of preventative maintenance or, in some cases, the apparent total absence of maintenance is probably the second most important cause of much of the damage to be seen in APUA-Elec's buildings and structures.


6 THE COST OF MITIGATING DAMAGE

6.1 Case I - Favourable Concepts

For the single-storey and two-storey buildings used as APUA-Elec facilities in Antigua the cost of making the favourably-shaped ones virtually invulnerable to future Category-4 hurricanes would have been a maximum of 3% in initial capital cost. Most of this incremental cost would be used in protecting windows and external doors and securing the lightweight roofing and siding.

6.2 Case II - Unfavourable Concepts

Where the APUA-Elec facilities are of unfavourable shape the cost of virtual invulnerability to future Category-4 hurricanes could rise to about 7.5% in initial capital cost. This incremental cost would be used, not only in protecting windows and external doors, but also in adding strength to the entire building envelope because of the higher wind forces generated by the unfavourable shapes.

6.3 Maintenance Costs

These would increase gradually from almost zero for the first year of the life of a new building up to about 4% of the contemporary construction cost in year 20. Thereafter the figure of 4% of the contemporary construction cost should be used as an annual budget.

The above-estimated figures would, of course, vary with the materials of construction. Life-cycle costing should be employed in financial and economic planning of new facilities. Such an approach would likely lead to better quality construction (which could have higher first costs) and lower-cost maintenance. This strategy is usually the more favourable.

This maintenance budget would cover management, other personnel, machines, materials for preventative maintenance and replacements of the entire fabric and "domestic" electrical/mechanical systems for the facility. Under such a desirable maintenance régime the buildings should last for up to two generations without noticeable deterioration.

6.4 The Cost of Controlling the Building Industry

6.4.1 Standards and Codes

In the overall context of the construction industry in general and the APUA-Elec facilities in particular, the cost of preparing standards and mandating codes is negligible.

6.4.2 Checking and Monitoring

If these functions are to be carried out effectively for design as well as for construction, an additional, one-off cost of 1% to 2% of the original construction sum is an average estimate. Spread over the (say) 50-year life of a building this figure becomes infinitesimally small.

6.4.3 Regulating the Professionals

This cost is almost zero. Antigua already has in place a Registration of Engineers Act. However, the Regulations have not yet been ratified by the Cabinet, so that the Act is not yet effective. Once the Regulations are ratified, the registering of engineers and the monitoring of the profession would be covered by annual registration fees paid by the said professionals. This would amount to no more than a few thousands of dollars, nothing that can be quantified as a measurable percentage of construction costs.


7 RECOMMENDATIONS

7.1 Mandating of Standards through Legal Codes

It is understood that there was a ministerial decision taken after Hurricane Hugo to institute a mandatory building code in Antigua-Barbuda. Every cloud has a silver lining. There are probably various arrangements still to be put in place to implement that decision. No effort should be spared, now that "the iron is hot", to keep the momentum going so that the Code becomes an integral part of the construction industry at a very early date. That Code is based on the OECS6 Building Code developed with funding from UNDP/UNCHS7 and referencing the regional standards of CUBiC.

7.2 Registration of Engineers

The necessary Regulations could be ratified at an early date so that the Registration Act can be put into force in time to support the mandatory Code implementation.

7.3 The Control System

7.3.1 Self Attestation

For buildings coming within the remit of the Engineers Registration Act it should be sufficient for designs, calculations and drawings to bear the stamp of a Registered Engineer and for that Engineer to attest in writing that the works have been designed in accordance with the provisions of the Antigua-Barbuda Building Code. A similar approach could be adopted for the construction process.

7.3.2 Random Audits

Such self-attestation as described above should be buttressed by in-depth audits of projects randomly selected and also selected where there is prima facie evidence of under-design or poor construction.

7.3.3 Check Consultants

For critical facilities, such as APUA-Elec facilities, independent check consultants are recommended. The use of check consultants is routine in France, including Martinique and Guadeloupe. It is also the practice in Colombia and, probably, in many other countries.

7.4 Maintenance

This could well be the most difficult recommendation to get acceptance. Facility owners have always balked at spending adequate amounts on this item. Also, budgeted amounts are frequently not drawn down because they get reallocated to other "more pressing" uses.

Notwithstanding these difficulties, however, it must be recognised how critical preventative maintenance is in protecting the APUA-Elec facilities from damage in major hurricanes.




Frontispiece: As seen by satellites 35, 000km above the earth, hurricanes appear as white clouds. These same clouds are seen from earth as dark and threatening. This image of Luis has thus been reproduced to represent better our experience.

SAT_IF SP1 02:30(02)GMT 05-SEPT-95 COPYRIGHT WSI CORPORATION


APPENDIX 1: TERMS OF REFERENCE: TO STUDY THE EFFECTS OF HURRICANE LUIS ON THE BUILDINGS AND OTHER STRUCURES OF THE ELECTRICITY SECTION OF APUA

1.0 GENERAL

This study will be an addition to the body of work that now exists within the CARILEC membership on the vulnerability of the electric utility facilities, and the remedial actions that should be taken on existing facilities which will be applicable to future construction.

Such studies now exist on the facilities of BL&P and LUCELEC. Additional risk management studies in respect of windstorms will be undertaken by the CARILEC membership this year in the Risk Management Survey which is part of the Risk Management Project being done by IDB/CDB and CARILEC.

The findings of this study will be included as part of the above project, and will have the special importance that can be placed on data and analysis done immediately after such an event.

2.0 FACILITIES TO BE EXAMINED:

The following facilities will be examined:

EFFECTS OF HURRICANE:
Where applicable, the consultant shall analyse and comment on the effects of high wind, flying debris, storm surge and abnormal rainfall, stating the vulnerability and corrective actions that can be taken with existing and future facilities.

3.0 SCOPE OF WORK:

  1. The consultant shall carry out inspection of the facilities listed under Section 2.0 to determine the damage caused by Hurricane Luis.
  2. Prepare a description of the repairs required, taking into account a reasonable standard of protection against future disasters. This description shall be general in nature and will not be used as detailed instructions to a contractor undertaking repairs to the damage.
  3. Draw conclusions from investigations in respect of inadequate design, where damage can be prevented by timely maintenance, failure of materials, and where the use of appropriate standards could have limited the damage.
  4. Prepare budgetary cost estimates for the repair of damage to the buildings and structure.
  5. Prepare comparative analyses, using examples of buildings that resisted the hurricane and those that sustained damage.
  6. Where possible, identify the reasons - technical and otherwise - for the different levels of damage sustained.
  7. Take photographs to illustrate the inspection, comments and analyses.

4.0 DELIVERABLES:

A report shall be submitted which will include inspection data, conclusions and recommendations. The report shall contain information and be of a format consistent with the reports previously done on BL&P and LUCELEC, and will add to or reinforce conclusions and recommendations that are specific to the affected utility and be applicable generally to other Caribbean electric utilities.


APPENDIX 2: FIGURES

Figure 1: Isolines Depicting Geographical Distribution of Category 3 Hurricanes (Windspeeds 96 to 113 Knots) During the Period 1886-1992

Click Here to see Figure 1

The Chart shown here is derived from a detailed study of the frequency of occurrence of Category 3 Hurricanes during the period 1886-1992. The isolines shown are lines which connect points representing a particular frequency of occurrence of Category 3 Hurricanes. These lines are similar to contour lines on a topographical map, the interval representing one frequency of occurrence. The heavy coloured lines represent 5 frequencies of occurrence in ascending order, as shown by the numbering on the Chart. The isolines are used to determine the average number of Hurricanes which have affected the location during the period. Thus, a two-degree square box appropriate to that location is chosen and the isoline(s) nearest to that location is(are) found. The numbers appropriate to those isolines provide the average number of Category 3 Hurricanes which have affected the location during the period. For example, the islands of St. Kitts/Nevis are located within a box bounded by 17 to 19 degrees latitude and 62 to 64 degrees longitude and have an isoline with the number 13 going through them. Thus, the average number of Category 3 Hurricanes affecting those islands during the period was 13, giving a return period of approximately 8 years.

Figure 2: Regional isoacceleration map

Source PAIGH: contour interval 50 gals

Figure 3: Continuous Load Path

From Simpson Strong-Tie connectors publication


APPENDIX 3: TABLES

Table 1a: Cost Estimates for Repairs to Buildings, Cassada Gardens Complex

Building Element Activity Estimate Subtotal
Generator Building Sheeting Roof and side sheeting $10,000 $10,000
         
Control Building Roofing Replace corroded sheets $10,800  
  Ceiling Replace tiles $9,200  
  Vinyl tiling Replace tile in affected rooms $1,800  
  Painting Paint interior and exterior $28.100 $50,000
         
Canteen Demolition Remove remaining walls $7,200  
  Roofing Rebuild roof and ceiling $94,300  
  Walls Rebuild $92,000  
  Carpentry Cupboards, windows and doors $27,700  
  Painting Paint interior and exterior $25,100  
  Tiling Retile in 12"x12" tiles $44,000  
  Electrical Re-wire building $28,700 $319,000
         
Transmission and Distribution Roofing Replace damage sheeting $8,100  
  Siding Replace corroded sheets $9,900  
  Ceiling Replace damage ceilings $5,400  
  Tiling Relay loose vinyl tiles $14,400  
  Painting Paint siding to match new $18,000  
  Electrical Lighting to new ceiling $1,800 $57,000
         
Total       $436,000

Notes:
No allowance is made for mechanical equipment
No allowance is made for electrical re-wiring of the buildings

Table 1b: Cost Estimates for Repairs to Buildings: Friars Hill Power Station

Building Element Activity Estimate Subtotal
Site Fencing Re-erect as necessary $40,200 $40,200
         
Generator Building Roofing Re-sheet roof $116,700  
  Siding Repair side sheeting $53,900  
  Sliding doors Rebuild $43,100  
  Masonry Patch masonry as required $3,600  
  Carpentry Re-build ceilings $39,500  
  Tiling Relay vinyl tiles $4,600  
  Painting General redecoration $53,900 $315 300
         
Office and Warehouse Roofing Repair corroded sheets $26,900  
  Siding Repair corroded sheets $9,900  
  Masonry Patch where necessary $3,000  
  Carpentry Rebuild ceilings $3,600  
  Tiling Relay vinyl tiles $4,700  
  Painting General redecoration $ 14,400 $62,500
         
Industrial Workshop Roofing Re-sheet roof $175,500  
  Walls Side sheeting $89,000  
  Carpentry Windows and doors $38,000  
  Painting General redecoration $15,000 $317,500
         
Total       $735,500

Notes:
No allowance is made for mechanical equipment
No allowance is made for electrical re-wiring of the buildings

Table 1c: Cost Estimates for Repairs to Buildings: Crabbs Complex

Building Element Activity Estimate Subtotal
Site Fencing Re-erect fencing $32,300 $32,300
         
Generator Building Roofing Re-sheet N-S generator hall $140,000  
  Siding Repair siding $59,300  
  Masonry Repair blockwall $2,400  
  Tiling Relay vinyl tiles $25,100  
  Carpentry Repair doors, louvres etc $82,400  
  Doors Roller shutters $71,800  
  Ventilators Replace $32,300  
  Painting General redecoration $11,800 $425,100
         
Machine Shop Roofing Replace $35,900  
  Masonry Repair Blockwall $1,800  
  Carpentry Windows and doors $21,500  
  Miscellaneous General repairs $43,100  
  Painting General redecoration $6,700 $109,000
         
Administration and Control building Roofing General repairs $18,000  
  Miscellaneous General repairs $53,900 $71,900
         
Total       $638,300

Notes:
No allowance is made for mechanical equipment
No allowance is made for electrical re-wiring of the buildings

Table 2: Damage to Transmission and Distribution Systems

 

Feeder Number

 
Substation

1

2

3

4

5

6

Total

 

Number of Poles

 
Swetes  

96

559

119

   

774

Union Road  

146

 

182

   

328

Friars Hill

423

74

189

278

   

964

Cassada Gardens

249

131

 

253

242

 

875

Crabbs      

386

   

386

Livingston

193

372

347

269

   

1181

Belmont  

247

558

     

805

Total

865

1066

1653

1487

242

0

5313


Total number of poles on 11 kV system = 5313
Add 2% for poles planted after count = 106
Total = 5419
Damage to 11 kV system:
Broken poles = 440 % = 8.12
Leaning poles = 322 % = 5.94
Broken conductors = 160
Detached conductors = 364
Total number of poles on 69 kV system = 200
Total number of towers on 69 kV system = ?
Damage to 69 kV system:
Broken poles = 6 % = 3.00
Leaning poles = 45 % = 0.83
Damaged towers = 0 % = ?

Table 3: Hurricane Luis Feeder Restoration Schedule

Substation

Feeder Number

Planned Start

Planned Finnish

Planned Days

% Done

Cassada Gardens

1

11 -Sep-95

18-Sep-95

7

100

Cassada Gardens

2

06-Sep-95

12-Sep-95

6

100

Cassada Gardens

3

18-Sep-95

27-Sep-95

9

0

Cassada Gardens

4

08-Sep-95

15-Sep-95

7

75

Cassada Gardens

5

18-Sep-95

27-Sep-95

9

85

Belmont

2

18-Sep-95

23-Sep-95

5

0

Belmont

3

14-Sep-95

27-Sep-95

13

0

Swetes

2

23-Sep-95

07-Oct-95

14

0

Swetes

3

23-Sep-95

07-Oct-95

14

0

Swetes

4

28-Sep-95

19-Oct-95

21

0

Lavington

1

11 -Sep-95

12-Sep-95

1

100

Lavington

2

2

0

Five Islands

4

17-Sep-95

24-Sep-95

7

50

Friars Hill: St John's Central  

11 -Sep-95

15-Sep-95

4

100

Friars Hill: St John's South  

11 -Sep-95

23-Sep-95

12

100

Friars Hill: St John's West  

16-Sep-95

22-Sep-95

6

100

Friars Hill: St John's North  

22-Sep-95

03-Oct-95

11

5

Friars Hill / WlOC Interconnector  

11 -Sep-95

11 -Sep-95

0

100

Friars Hill / Garden Interconnector  

11 -Sep-95

13-Sep-95

2

100

Baker / Tango Interconnector  

08-Sep-95

10-Sep-95

2

100



APPENDIX 4: PHOTOGRAPHS

Please note that only selected photos have been included in this digital version of the report.

  1. Cassada Gardens Complex - overall aerial view
  2. Cassada Gardens Complex - Note the badly rusted but undamaged Old Power Station and the destroyed Canteen/Workshop.
  3. Cassada Gardens Stores Building
  4. Cassada Gardens Stores Building - impact damage to the siding (Shows a detail from photo 3)
  5. Cassada Gardens Canteen - roof structure completely removed; ceiling structure badly damaged, but in place.
  6. Cassada Gardens Canteen - removed roof trusses on the ground
  7. Cassada Gardens Canteen - cracked walls (Shows a detail from photo 5)
  8. Friars Hill Power Station - the south gable-end cladding has been replaced since the hurricane.
  9. Friars Hill Power Station - the south gable-end cladding showing repairs after Hugo (1989) and Luis (1995) - note that the latest repairs are not even as well secured as those which failed in Luis.
  10. Friars Hill Power Station - Repairs to sidings show the absence of horizontal rails at the overlap of sheeting. This would lead to inevitable failure in the next hurricane.
  11. Friars Hill Workshop, Stores and Office Building
  12. Friars Hill Workshop
  13. Friars Hill Workshop
  14. Friars Hill Workshop - note the torn corrugated steel sheets
  15. Friars Hill Workshop Surprisingly, the sheeting failed away from the fixing locations.
  16. Friars Hill Workshop - view from inside showing loss of gable-end and ridge sheeting
  17. Friars Hill Stores - loss of roof sheeting
  18. Friar's Hill Office Building - loss of roof sheeting
  19. Crabbs Diesel Power Station - damage to gable-end sheeting
  20. Crabbs Diesel Power Station - damage to gable-end wall and roof sheeting
  21. Crabbs Reverse Osmosis Plant Building - loss of gable flashing
  22. Crabbs Reverse Osmosis Plant Building - impact damage probably from a flying pipe
  23. Typical container-type Substation Control Room showing door and window Openings
  24. Typical container-type Substation Control Room showing door removed by the wind
  25. Close-up of damaged hinge of the door in No 24
  26. Crabbs Complex - destruction of roller shutter door, a common occurrence in hurricanes
  27. Cassada Gardens Complex - hurricane shutters were in place during Luis
  28. Cassada Gardens Complex - bolts for fixing hurricane shutters
  29. Crabbs Complex - damage to insulation of Fairbanks Morse stack
  30. Crabbs Complex - damage to cladding and insulation around intake duct of Cogeneration Plant
  31. Crabbs Complex - close-up of damage to cladding and insulation in No 30
  32. Crabbs Complex - damage to cladding and insulation around intake duct of Cogeneration Plant
  33. Cassada Gardens Complex - building with no damage whatsoever, proving that success is possible
  34. Crabbs Diesel Power Station - control panels were protected by tarpaulin during Luis
  35. Cassada Gardens Complex - control panels did not need additional protection during Luis
  36. Belmont Substation - showing old and new types of Control Rooms
  37. Swetes Substation - tie-downs were installed in advance of Luis to prevent toppling of the containers
  38. Crabbs Complex - destruction of marine facility at the root of the Jetty due to wave action
  39. Crabbs Complex - destruction of Jetty due to wave action
  40. Crabbs Complex - the helmetted man is standing at the location where it was claimed that the storm surge reached
  41. Crabbs Complex - gabion cofferdam protecting equipment pit outside of Diesel Power Station (note gap in the wall)
  42. Failure of southern-yellow-pine pole due to rotting at the soil line
  43. Failure of southern-yellow-pine pole (Mental Hospital Road) due to rotting at the soil line
  44. White rot causes deterioration of wallaba poles starting from the inside
  45. Close-up of failed wallaba pole (Jonas Road) due to white rot
  46. Failure of southern-yellow-pine pole due to overstress
  47. Failure of new southern-yellow-pine pole due to overstress
  48. Failure of braced-pole assembly by uprooting (Jolly Harbour location)
  49. Close-up of the failure in No 48
  50. Poles are marked with manufacturer's recommendations for embedment
  51. End of the line at Five Islands Substation - note the leaning poles indicating incipient failure of the anchorages
  52. The leaning poles were planted before the road was excavated (west of Old Road). This led to less eventual foundation depth than was intended.
  53. The pole broke at the location of a drilled-through tie anchorage
  54. Total chaos at the Falmouth-Piccadilly Junction
  55. Close-up of broken pole in No 54
  56. Many of the poles were overloaded with telephone and television cables (Jonas Road)
  57. Damage due to trees was not common-place in Antigua (John Hughes area)
  58. There were 160 examples of broken conductors
  59. Heavy equipment may not be vulnerable in hurricanes, but need to be anchored for the earthquake hazard.
  60. The Barbados Light & Power crew taking a lunch break in Antigua.
  61. Progress chart for Feeder Restoration
  62. Allocation of foreign resources to the restoration programme

APPENDIX 5: BARRY PINNOCK'S REPORT: CEP SURVEY OF ANTIGUA PUBLIC UTILITIES AUTHORITY (APUA) POWER GENERATING FACILITIES

Survey was conducted by Barry Pinnock of Consulting Engineers Partnership Ltd (CEP). Those present at all times were Mr A B Segu, Manager of the Electrical Division of APUA. The survey was carried out on the 30th October, 1995. Transport was provided by Barry Pinnock of CEP. The first installation visited was Friars Hill, the report for which follows:

1. FRIARS HILL

1.1 Main Building Housing Generating Equipment

Only one generator was operating at Mirrlees Blackstone - 4.3w. Other generating equipment was under regular maintenance, which has been further inhibited by the salt spray brought inland by the hurricane windforce contaminating machinery housings and other sensitive parts.

1.2 External Fittings

It was difficult to assess whether the exhaust laggings were damaged as a result of the hurricane, since they appeared well weathered in their damaged state. No other external fittings appeared disturbed.

1.3 The Roof and Cladding Material is a galvanised sheeting

Roof sheets to the south end were lost and are in the process of being replaced. Sheet cladding to the west side, lost, has now been replaced. Sheet cladding to the north elevation, lost, have been replaced.

It is to be noted that fixings holding the cladding have been applied in an irregular manner spacings between fixings appear to be from 9" apart to 14" apart. Sheeting lost appears to have been as a result of high windshear on the smaller panels, located at the bottom of the side cladding, being ripped away, by penetrating wind force allowing further and more severe damage to occur.

It would therefore appear that continuous sheeting from the eaves to the footings with additional angle brackets on the main steel frame placed at closer spacings at footings and eaves may prove more successful should similar circumstance occur again.

Much of the guttering lost with it's fixings has now been replaced but is still insufficient, as it is. Friars Hill Installation includes the APUA main workshop that comprised of both open and closed in facilities, commanding 35/45 metres in length and 10 metres wide approximately. Approximately 70% of the roofing has been lost. Roofing to the open area has been completely lost, leaving the enclosed Stores and parts of areas north and south, partially with without roofing, but which now has a temporary covering, to protect parts and materials.

It was noticed that the roof purlins of timber were light and where intact carried spacings of 24" at the eaves and widened to 48" in the center. Ridge spacings were not evident, having been lost. A more ridged form of construction is recommended. The concrete flooring has not suffered damage from the hurricane and appears to be intact.

With the workshop not functioning as it should, transportation and plant facilities are badly affected, downgrading maintenance services.

Damage and repair costs based on a direct purchasing procedure, free of all duties, taxes and levies appears to be in the amount of EC$950,000.00

2. CASSADA GARDENS

2.1 Main Building Housing Generation Equipment

All generating equipment was inoperative due to maintenance. Standby power generated by Cummings alternator plant is available but was not in use.

2.2 The Roof and Cladding material is a galvanised sheeting

Roof sheeting does not appear to have been damaged other than minor superficial loosing of sheet work, now re-secured. Damage has occurred to the side cladding on the south-west elevation, which has been replaced.

The Instrument Workshop suffered damage to its timber framed roof. Construction was limited to the use of light timbers, which collapsed internally causing much water damage to instruments and other sensitive equipment.

2.3 The Control & Systems Room

This building is almost totally undamaged, its construction indicates quality workmanship. The construction was achieved by Italian input. Even with such quality construction, some water damage occurred inside the building, mainly causing rust to the metal duct floor covers and sensitive equipment. This sensitive area operates the 11Kv buss bar conductor system. Some minor outside damage is evident on the west p>These areas suffered internal damage as a result of water entering via roof and eaves but outside damage was limited to sheeting being lifted by wind force and allowing water to enter.

Whilst the desalinisation plant did not form part of the survey, it should be noted that the fuel line is a marine fixture to the concrete jetty, which collapsed as a result of the armour rock it was built on shifting from high seas during the hurricane. (Considerable damage was very apparent around the Antigua coast line, as a result of the high seas.) Piling was only evident at the jetty's seaward threshold.

Damage and repair costs based on a direct purchasing procedure, free of all duties, taxes and levies appears to be in the amount of EC$550,000.00.

Jetty rebuild costs will depend on design specifications. An estimate of not less than EC$700,000.00 is envisaged.


NOTES

1 The USAID/OAS-CDMP is funded by the United States Agency for International Development and managed by the Organisation of American States.

2 The hot-wire anemometer was mounted on a 21-metre tower adjacent to a 9-metre-high building on a small knoll. The equipment is a Case Indicator, CY-2732/GMQ-20, Stock No 6660-00-805-5729, Part No 5114-100, Airflo Instrument Co, Glastonbury, Connecticut, USA. The instrument stopped working at 02:15 on Tuesday, 05 September.

3 This code was originally prepared by the Barbados Association of Professional Engineers (BAPE) in 1970. In 1981 the (Barbados) National Council for Science and Technology (NCST) commissioned a revision funded by the Organisation of American States (OAS). More recently, the Barbados National Standards Institution (BNSI) adopted the document as a national standard.

4 (work: 462 1391/1470/2229; fax: 462 2573; home: 463 2118)

5 (461 3746)

6 Organisation of East Caribbean States

7 United Nations Development Programme through United Nations Centre for Human Settlements


CDMP home page: http://www.oas.org/en/cdmp/ Project Contacts

Page Last Updated: 20 April 2001