Detailed Island Risk Assessment in Maldives
Volume III: Detailed Island Reports
Ga. Viligilli – Part 1
DIRAM team Disaster Risk Management Programme UNDP Maldives
December 2007
1 Table of contents
1. Geographic background 1.1 Location 1.2 Physical Environment 2. Natural hazards 2.1 Historic events 2.2 Major hazards 2.3 Event Scenarios 2.4 Hazard zones 2.5 Recommendation for future study 3. Environment Vulnerabilities and Impacts 3.1 General environmental conditions 3.2 Environmental mitigation against historical hazard events 3.3 Environmental vulnerabilities to natural hazards 3.4 Environmental assets to hazard mitigation 3.5 Predicted environmental impacts from natural hazards 3.6 Findings and recommendations for safe island development 3.7 Recommendations for further study 4. Structural vulnerability and impacts 4.1 House vulnerability 4.2 Houses at risk 4.3 Critical facilities at risk 4.4 Functioning impacts 4.5 Recommendations for risk reduction
2 1. Geographic Background
1.1 Location
Viligilli Island is located on the eastern rim of Gaafu Alifu atoll, at approximately 73° 26' 5"E and 0° 45' 22" N, about 380km from the nations capital Male’ and 72 km from the nearest airport, Gaafu Dhaalu Kadedhdhoo (Figure 1.1). The island forms part of the natural atoll called Huvadhoo Atoll, which is considered the second largest atoll in the world. Viligilli is the atoll capital amongst 10 other inhabited islands. It’s nearest inhabited islands are Maamendhoo (4 km) and Dhaandhoo (12.5 km). Villigili is also just 2 km from Koodoo, the main fish processing centre in the southern region of Maldives. Huvadhoo atoll is the nearest atoll in Maldives to the equator and sits along the southern half of the laccadive-chargoes ridge, exposing the entire atoll to direct wave action in Indian Ocean. However, it location in the heart of the doldrums makes the island relatively safe from major climatic hazard events. 73.25°E 73.5°E
Kolamaafushi
0.75°N Viligilli
Maamendhoo
N
Nilandhoo Location Map North Huvadhu Atoll Dhaandhoo of Viligilli (Gaafu Alifu Atoll) 0 5 10 kilometers Dhevvadhoo
0.5°N Kon'dey Dhiyadhoo Gemanafushi
Kanduhulhudhoo
Figure 1.1 Location map of Viligilli.
3 1.2 Physical Environment
Viligilli is a narrow and elongated island with a width ranging from 600 m to 180 m and a length of 1846 m. The total surface area of the island is 54.8 Ha (0.55 km 2). The island is oriented in a north south direction and the width of the island narrows northwards.
The reef of Viligilli is large with a surface area of 3000 Ha (30 km 2) extending to about 20.5 km. The reef also hosts three other uninhabited islands. Viligilli is located on the southern tip of reef system, next to a major reef entrance (Viligilli Kandu). The island of Falhuverreha, located off the north-eastern tip of the island is separated by a mere 20m ‘mini channel’. The depth of the reef flat is quite shallow averaging less than -1m MSL (EDC, 2006). The distance between the island shoreline and oceanward reef rim varies from 100m in the south to 750 m in the north. The average distance to reef edge is approximately 300 m.
Viligilli appears to have grown northwards and continues to do so with the constant supply of sediments and due to the funnelling effect caused by the orientation and location of Falhuverreha Island. Much of the original island was covered with wetland areas which have now been reclaimed on an adhoc basis by the inhabitants. Remnants of these wetland areas can still be found in the northern and southern part of the island. Hence, much of the settlement is located in low lying reclaimed areas. Over the years these fertile low areas of the island have become productive areas for vegetation and backyard agriculture. Much of the larger trees in the island, especially 95% of the mango trees are located in these low areas.
Viligilli, in spite of its size is highly urbanised. Due to the narrow width of the island, settlement has already expanded to the edges of coastline and new plots are still being developed. It was reported that Viligilli was facing land shortages for housing purposes. The island only had thin layer of depleted coastal vegetation around its settlement and the remaining vegetation could be further depleted with planned expansion of settlement. The coastal areas of the island have been heavily modified in the past, especially the northern western coastline of the island where harbour construction and reclamation activities have been undertaken twice. Environmental issues associate with urbanisation are being experienced by its inhabitants including, ground water contamination, improper waste disposal, degradation of coastal areas, depletion of vegetation and coastal erosion.
4 The tsunami of 2004 had a major impact on the islands terrestrial environment and entire settlement. The most notable environmental damages were the negative impacts on 50% of the island’s larger trees, including loss of 90% of all mango trees. The existing environmental characteristics and features of Viligilli appear to play a major role in exposing the island to natural hazards, especially flooding.
5 2. Natural hazards
This section provides the assessment of natural hazard exposure in G.A Viligilli Island. A severe event history is reconstructed and the main natural hazards are discussed in detail. The final two sections provide the hazard scenarios and hazard zone maps which are used by the other components of this study as a major input.
2.1 Historic events
The island of Viligilli has been exposed to multiple hazards in the past. A natural hazard event history was reconstructed for the island based on known historical events. As highlighted in methodology section, this was achieved using field interviews and historical records review. Table 3.1 below lists the known events and a summary of their impacts on the island.
The historic hazardous events for Viligilli showed that the island faced the following hazards: 1) flooding caused by heavy rainfall and 2) swell surges, 3) windstorms and 4) tsunami. Impacts and frequency of these events vary significantly. Flooding caused by rainfall and swell surges are the most commonly occurring hazard events. Windstorms have also been reported as frequent especially during the southwest monsoon. Since the elders from the island cannot recall events beyond 1978, it is plausible that severe events came to the attention of inhabitants only with the rapid expansion of settlement especially towards the hazard prone western coastline of the island and due to land reclamation activities.
Table 2.1: Known historic hazard events of Viligilli. Metrological Dates of the Impacts hazard recorded events
Flooding caused • Very frequent According to the island office records by Heavy rainfall (almost yearly heavy rainfall is most common during events) events mid to late South West (SW) monsoon. commonly These events are reported to cause occurring during heavy flooding of the old marsh (now SW monsoon. reclaimed) area in the central region of the island, which is approximately 30% of the presently populated area of the island. Heavy rainfalls are reported to cause flooding within this zone up to a height of 0.3m – 0.6m. It has been reported that the flood waters sometimes have lasted between 3 – 5 days. The major impacts of these flooding are:
6 - Blocking of the sewerage networks within the flooded zones (up to 75% of the houses could not flush their toilets during these floods) - Severe damages to the backyard crops such as bananas, chillies etc. - Damages to house furniture, children’s text books and other household goods.
Flooding caused • 5 July 1966 1 The island is reported to experience by swell surges • 24 August 1975 frequent (once every few years) flooding • 8 August 1979 caused by wave surges and sometimes • 17 Sept 1979 large swell waves generated far offshore • 14 July 1983 from the coasts of the Maldives. These • 06 April 1984 events are also reported to occur during • 28 June 1987 mid SW monsoon. Surge waters often reach more than 50m inland along much • June 2006 of the length of eastern shoreline to a maximum of 200m. The major impact of these events is damages to the backyard crops within the impact zone.
Windstorms • 28 Sept 1984 A number of windstorm events were • 27 March 1988 reported by elders; however, no dates could be determined. Major damages include falling trees, damage to crops, and structural damage to roofs.
Droughts No major event have been reported
Earthquake No major event have been reported
Tsunami 26 th Dec 2004 There has be en only one known event. This event flooded a large area of the island with great force. The tsunami wave was reported to have a height of approximately 3m when it reached the eastern shoreline of the island. Villingili is one of the most effected islands in the southern atolls. The impacts of the tsunami include: - Salinisation of groundwater - Damage to the sewerage network - Damage to large trees such as mangos, breadfruit, etc.
1 All dates in italics are adopted from MANIKU, H. A. (1990) Changes in the Topography of Maldives, Male', Forum of Writers on Environment of Maldives. And news paper reports.
7 - Destroyed backyard crops - Severe to minor structural damage to many houses
2.2 Major hazards
Based on the historical records, meteorological records, field assessment and Risk Assessment Report of Maldives (UNDP, 2006) the following meteorological, oceanic and geological hazards have been identified for Viligilli.
• Heavy rainfall (flooding) • Swell waves and wind waves • Windstorms • Tsunami • Earthquakes • Climate Change
2.2.1 Swell Waves and Wind Waves
Being located on the eastern rim of Huvadhoo atoll, Viligilli is relatively protected from the year round swell waves approaching from a west to southerly direction. There are no specific wave studies undertaken for Viligilli, but as Table 2.2 below shows, the predominant swell wave direction for the neighbouring Fuvahmulah Atoll is SW to South. These patterns have been reported throughout Maldives ( Kench et. al (2006), Young (1999), DHI(1999) and Binnie Black & Veatch (2000)). A similar pattern can be expected for Huvadhoo Atoll. Hence, the east coast of the Viligilli is generally exposed to wind generated waves during the NE Monsoon between November and March (EDC, 2006) with heights less than 2.0m and wave periods of 2-4 seconds. The west coast is exposed to wind generated waves originating within the atoll due to the 30-50 km fetch, usually about or less than or about 0.5m. The west coast could also receive residual swell waves penetrating the atoll through atoll passes (based on Kench et. al (2006)).
Despite, being located away from the predominant swell wave direction, Viligilli is still exposed to abnormal swell waves originating from intense storms in the southern hemisphere between 73°E and 130°E longitude. Waves generated from such abnormal events could travel against the predominant swell propagation patterns in the Indian
8 Ocean (Goda, 1998), causing flooding on the eastern rim island of Maldives. The historical flood events on the eastern coastline are most likely to be the result of such waves since the probability of storm surge is almost non existent due to the proximity to the equator.
Table 2.2 Wave regimes in neighbouring Fuvahmulah Atoll (source: DHI(1999)) Season Total Long Period Short Period
Predominantly from E-S. Mainly E-NE. High NE - Monsoon From S-SW High Waves from W waves from W
Transition Period 1 Mainly from SE-E From S-SW Mainly from NE-SE
From SE-SW. Mainly Mainly from SE-S. High SW - Monsoon from S. High Waves also From S-SW waves from West from W
From SE-W. Higher Transition Period 2 As SW monsoon From S-SW waves from West
The occurrence of abnormal swell waves on Viligilli reef flat is dependent on a number of factors such as the wave height, location of the original storm event with in the South Indian Ocean, tide levels and reef geometry. It is often difficult to predict occurrence of such abnormal events as there is only a small probability, even within storm events of similar magnitude, to produce waves capable of flooding islands.
Table 2.3 shows major flooding events in Viligilli and concurrent major storm events in South Indian Ocean.
Table 2.3 Historical flood events and possible links with storm events. Flooding Cyclone Date Maximum Distance Direction Tide Level event Name of Category Storm Event 5 July 1966 none Data not available 24 August none 1975
9 Flooding Cyclone Date Maximum Distance Direction Tide Level event Name of Category Storm Event 8 August 1979 none
17 Sept 1979 none
14 July 1983 11-07- 11-15 1 1500km S-SE Data not 1983 July available 1983 6th April 1984 7/03/1984 4 Apr – 3 1450km WSW- Data not 14 SSE available April 1984 28 June 1987 25/06/1987 25-27 1 1700km SE Median tide June 1987 10 September unknown NA 1987 June 2006 unknown Peak tide
Based on the current data available it is impossible to link the swell incidents specifically to the known cyclonic events. Moreover, much of the flood events are not associated with reported cyclonic events. Detailed assessment using synoptic charts of the South Indian Ocean corresponding to major flooding events are required to delineate any specific trends and exposure thresholds for Viligilli from southern swells. Unfortunately this study does not have the resources and time to undertake such an assessment but is strongly recommended for any future detailed assessments.
Flooding is also known to be caused in Viligilli by a gravity wave phenomenon known as Udha . These events are common throughout Maldives and especially in the southern atolls of Maldives. No specific research has been published on the phenomenon and has locally been accepted as resulting from local wind waves generated during the onset of southwest monsoon season. The relationship has probably been derived due to the annual occurrence of the events during the months of May or June. These events usually impact the western and southern coastline of the island.
10 The origins of the udha waves as yet remain scientifically untested. It is highly probable that waves originate as swell waves from the Southern Indian Ocean and is further fuelled by the onset of southwest monsoon during May. The timing of these events coincides as May marks the beginning of southern winter and the onset of southwest monsoon. The concurrent existence of these two forms of gravity waves during the southwest monsoon is confirmed by Kench et. al (2006) and DHI(1999) . It is also questionable whether the southwest monsoon winds waves alone could cause flooding in islands since the peak tide levels on average are low during May, June and July. Furthermore the strongest mean wind speeds in Kaadedhoo and Gan Airport has been observed for November and is more consistent during October to November than during May and June period (Naseer, 2003). This issue needs to be further explored based on long term wave and climatological data of the Indian Ocean before any specific conclusions can be made. However if the relationship does exists, this phenomena could prove to be a major hazard in the face of climate change since the intensity of southern Indian Ocean winter storms is expected to increase.
Processes controlling water levels around Viligilli
Waves undergo extreme and rapid transformations as they interact with reef crest, which control the character of hydrodynamic processes on adjacent reef flat. One of the products of such transformations is the water level setup created at the reef edge and currents generated by the wave setup. Current records made for various studied over reef flats (Aslam, 2004) have shown low frequency oscillations in the current speed. These low frequency oscillations in the current speed have been attributed to surf beat, edge wave and shear waves.
The degree to which wave energy is transformed or "filtered" by the process of wave breaking on the reef depends on several factors, including overall reef geometry, water depth at the reef crest, uniformity of depth along and across the reef, width of the reef flat and depth of the reef flat (Gourlay, 1994, Gourlay, 1996 ).
Strong winds can cause higher incident waves to break on the reef and the sea-level can rise locally due to shear force of wind on the water surface. The rise in water level due the shear force of winds and the wave setup created as a result of breaking waves on the reef edge can produce high water level set up on the reef flat. Similarly surges or swell waves beyond significant wave heights of 9m can cause water levels to rise 3.0m on the reef flat (based on (Department of Meteorology, 2007)). When such rises in water
11 level are combined with high tides there could be strong surges of water across the reef flat. Due to the low elevation of Viligilli coastline, such waves have the potential to flood.
Kench and Brander (2006) reported a relationship between wave energy propagation across a reef flat and, reef width and depth. Using their proposed Reef Energy Window Index, the percentage of occurrence of gravity wave energy at Viligilli reef flat is approximately 40%.
Historical surge related flood impacts
The common flooding area as a result of surges at present on the island is identified to be on the oceanward (eastern and southern) coastline of the island (Figure 2.1). The inland extent of flooding is greatest along the reclaimed old wetland areas. The reason could be attributed to the topographically lower elevations and absence of natural ridge system.
Historical Flood Events & Estimated Wave Propagation patterns around Thinadhoo
HISTORIC EVENTS
SW monsoon Wind waves Probable wave propagation patterns propagation wave Probable
0 150 300 meters
Figure 2.1 Historical flood events and estimated wave propagation patterns in Viligilli and its reef flat.
12 The highest wave height reported on the island during flooding events was 1.1m (3.5ft). This height is consistent with flood heights reported from swell or surge related waves in Maldives.
Future event prediction
It is known that Viligilli is exposed to abnormal swell waves originating from the Southern Indian Ocean. Due to its location in the southern half of the country, this should be considered amongst the most serious hazards facing the island. The exposure swell waves are mainly from south-easterly to southerly direction. There is also a low probability of storms in Bay of Bengal to generate swell waves. Events beyond these arcs may not influence Viligilli or could have reduced impact due to the protection offered by the southern and western rim of the atoll (Figure 2.2).
Possible range of swell wave direction in Viligilli: SE to S & NNE to NE
Historic storm events 1945 - 2007
Figure 2.2 Historical storm tracks (1945-2007) and possible direction of swell waves for Viligilli Island.
13 At present, it is very difficult to forecast the exact probability of swell hazard event and their intensities due to the unpredictability of swell events and lack of research into their impacts on Maldives. However, since the hazard exposure scenario is critical for this study a tentative exposure scenario has been developed based on the historical events. In this regard there is a probability of major swell events occurring every 10 years in Viligilli with probable water heights of 1.0m and every 5 years with probable water heights of 0.5-0.75 m. Events with water heights less than 0.5m and greater than 0.2 m are likely to occur once every 2-3 years. The timing of swell events is expected to be predominantly between April to October, based on historic events and storm event patterns (see Table 2.4).
The intensity of flooding in the inland areas may have been increased by improper wetland reclamation. The reclaimed areas are considerably lower than the existing island causing flood water to run-off towards the low areas more frequently.
Table 2.4 Variation of Severe storm events in South Indian Ocean between 1999 & 2003 (source: (Buckley and Leslie (2004)). Severe wind event variation Longitude band Wint er Summer 30 °E to 39 °E 12.5 17 40 °E to 49 °E 7.5 10 50 °E to 59 °E 7.5 26 60 °E to 69 °E 6 14 70 °E to 79 °E 6 6 80 °E to 89 °E 12 6 90 °E to 99 °E 12 8 100 °E to 109 °E 8 3 110 °E to 119 °E 15 7 120 °E to 130 °E 13.5 2
2.2.2 Heavy Rainfall
The rainfall pattern in the Maldives is largely controlled by the Indian Ocean monsoons. Generally the NE monsoon is dryer than the SW monsoon. Rainfall data from the three main meteorological stations, HDh Hanimaadhoo, K. Hulhule and S Gan shows an increasing average rainfall from the northern regions to the southern regions of the country (Figure 2.3). The average rainfall at S Gan is approximately 481mm more than that at HDh Hanimadhoo.
14 3500
3000
2500
2000
1500
1000
Meanannual rainfall (mm) 500
0 1975 1977 1979 1981 1983 1985 1987 1989 1991 1993 1995 1997 1999 2001 2003 Year
Gan Hulhule Hanimadhoo
Figure 2.3 Mean annual rainfall across the Maldives archipelago.
The closest meteorological station to Viligilli is Kaadedhoo airport which became operational in 1993. Unfortunately this study does not have access to Kaadedhoo data. Moreover, Kaadedhoo data may be limited for long term trend observation due smaller number of observation years. Hence, to resolve the issue, data from Gan has been used. It is recommended that further assessment be made once Kaadedhoo data becomes available.
The mean annual rainfall of Gan is 2299.3mm with a Standard Deviation of 364.8mm and the mean monthly rainfall is 191.6mm. Rainfall varies throughout the year with mean highest rainfall during October, December and May and lowest between February and April (See Figure 2.4).
15
Fig 2.4 Mean Monthly Rainfall (1978-2004).
Historic records of rainfall related flooding on the island of Viligilli indicates that this island is often flooded and its intensity is high in certain areas of the island. Records for all incidents have not been kept but interviews with locals and research into newspaper reports show that localised levels of flooding within areas of Viligilli has only become prominent since the 1980’s. This coincides with the extensive settlement expansion into the reclaimed wetland areas on the island. Heavy rainfall related flooding has been reported to reach up to 0.5m above the ground level. Based on available records, Viligilli is amongst the most intensely flooded islands in Maldives. One key feature of Viligilli island is that flood water recedes quite slowly in the low areas due to the shallow depth between surface and ground water lens.
Hence, Viligilli Islands' exposure to rainfall related flooding is a combination of being located in a high rainfall zone and human activities. Since 1960’s land reclamation has been undertaken in the central wetland areas on an ad-hoc basis known as baadi hikkun (see section on Physical Environment Vulnerability). Under this process, land is allocated in the uninhabitable wetland area and the land owner is responsible for reclaiming it. Occasionally reclamation was done as community work on holidays. Since the late 1980’s, harbour development projects facilitated large scale reclamation. Unfortunately, all these activity led to a substantial topographic low in the middle of the island establishing a natural drainage into this area. As a result, at times of heavy rainfall, the lowly reclaimed areas regularly get flooded. Furthermore, to remedy flooding
16 on roads, they were levelled and relevelled with extra sand without considering the flooding implications for surrounding houses. At present some houses are about 0.3m lower than the adjacent roads. With no artificial drainage system for the roads, the surrounding houses in the low areas are at constant risk of flooding.
The impacts of flooding so far reported has not been disastrous, but has had continued impacts on the community such as damage to personal belongings, crops and disruptions to daily life.
It would be possible to identify threshold levels for heavy rainfall for a single day that could cause flooding in Viligilli, through observation of daily rainfall data. Unfortunately, we were unable to acquire daily historical data. However, available records shows that Kaadedhoo received a maximum precipitation of 219.8mm for a 24 hour period on 10 th July 2002, the highest recorded anywhere in Maldives since recording began. This event caused widespread damage or disruption to personal property, road infrastructure, sewerage infrastructure, backyard crops, harbour quay wall, school operation and businesses. Flood heights in the southern part of the island were reported at 0.4-0.5m. The worst effected area was the south central part of the island. During the flooding events of November 2003, the recoded rainfall in Kaadedhoo for the 24 hour period was 64.4mm (DoM, , 2005). A month later, rainfall up to 60.3mm was observed. These two events caused disruption to businesses, school and minor damage to household goods.
The probable maximum precipitations predicted for Gan by UNDP (2006) are summarized in Table 2.5:
Table 2.5 Probable Maximum Precipitation for various Return periods in Gan Return Period 50 year 100 year 200 year 500 year 218.1 238.1 258.1 284.4
Given the high variations in rainfall in Kaadedhoo these figures may vary slightly. Based on the field observations and correlations with severe weather reports from Department of Meteorology ((DoM, 2005) the following threshold levels were identified for flooding. These figures must be revised once historical daily rainfall data becomes available (Table 2.6).
Quite often heavy rainfall is associated with multiple hazards especially strong winds and possible swell waves. It is therefore likely that a major rainfall event could inflict far more damages those identified in the table.
17 Table 2.6 Threshold levels for rainfall related flooding in Viligilli. Threshold level Impact (daily rainfall) 50mm Puddles on road, flooding in low houses, minor damage to household goods in most vulnerable locations, disruption to businesses and primary school in low areas. 100mm Moderate flooding in low houses; all low lying roads flooded; moderate damage to household items especially in the backyard areas 150mm Widespread flooding on roads and low lying houses. Moderate to major damage to household goods, School closure. 200mm Widespread flooding on roads and houses. Major damages to household goods, sewerage network, backyard crops, school closure, gullies created along shoreline, possible damage to road infrastructure. 250+mm Widespread flooding around the island. Major damages to household goods and housing structure, schools closed, businesses closed, damage to crops, damage to road infrastructure, sewerage network and quay wall.
2.2.3 Wind storms and cyclones
Maldives being located within the equatorial region of the Indian Ocean is generally free from cyclonic activity. There have only been a few cyclonic strength depressions that have tracked through the Maldives, all which occurred in the northern regions. According to the Hazard Risk Assessment Report (UNDP, 2006) Viligilli falls within the least hazardous zone for cyclone related hazards. There are no records of cyclones for the southern region, although a number of gale force winds have been recorded due to low depressions in the region. Winds exceeding 34 knots (gale to strong gale winds) were reported in Kaadedhoo annually between 2002 and 2005, all caused by known low pressure systems near Maldives rather than the monsoon.
Historic records for Viligilli have indicated that near gale force winds (see Table 2.7) have caused minor damage to property and trees on the island. Hence during the high winds between 2002 and 2005, a number of minor to moderate damages were reported to property, vegetation and backyard crops. Viligilli does have fairly lush vegetation dominated by larger trees such as mango and breadfruit, which acts to minimise the direct exposure of properties.
18 In order to perform a probability analysis of strong wind and threshold levels for damage, daily wind data is crucial. However, such data was unavailable for this study. Estimates have therefore been made using the only available data from nearby Gan: 2002 and 2003.
Analysis of all the wind speed data for the years 2002 and 2003 indicates that the probability of occurrence of wind speeds greater than 23 knots is 1.3days (0.36%) in a year (Table 2.8). The analysis also indicated that highest winds blow from SSW – W (Fig 2.5).
Table 2.7 Beaufort scale and the categorisation of wind speeds. Average wind Cyclone Average wind speed Beau- fort No Description Specifications for estimating speed over land category speed (Knots) (kilometres per hour)
0 Calm Less than 1 less than 1 Calm, smoke rises vertically. Direction of wind shown by smoke drift, but not by wind 1 Light Air 1 -3 1 - 5 vanes. Wind felt on face; leaves rustle; ordinary wind vane moved 2 Light breeze 4 - 6 6 - 11 by wind. Leaves and small twigs in constant motion; wind extends 3 Gentle breeze 7 - 10 12 - 19 light flag. Moderate 4 breeze 11 - 16 20 - 28 Raises dust and loose paper; small branches moved. Small trees in leaf begin to sway; crested wavelets form on 5 Fresh breeze 17 -21 29 - 38 inland waters. Large branches in motion; whistling heard in telegraph 6 Strong breeze 22 - 27 39 - 49 wires; umbrellas used with difficulty. Whole trees in motion; inconvenience felt when walking 7 Near gale 28 - 33 50 - 61 against the wind.
8 Gale Category 1 34 - 40 62 - 74 Breaks twigs off trees; generally impedes progress. Slight structural damage occurs (chimney pots and slates 9 Strong gale Category 1 41 - 47 75 - 88 removed). Seldom experienced inland; trees uprooted; considerable 10 Storm Category 2 48 - 55 89 - 102 structural damage occurs. Very rarely experienced; accompanied by widespread 11 Violent storm Category 2 56 - 63 103 - 117 damage. 12 Hurricane Category 3,4,5 64 and over 118 and over Severe and extensive damage.
Table 2.8 Probability of occurrence of wind at different speeds in Addu Atoll (based on hourly records for the years 2002 and 2003)
19 Probability of occurance Direction Speed range <=10 kts >10 - 20kts >20 - 30kts >30kts 0 - 22.5 0.0881 0.0002 22.5 - 45 0.0529 0.0007 45 - 67.5 0.0278 0.0002 67.5 - 90 0.0304 0.0003 90 - 112.5 0.0216 0.0011 112.5 - 135 0.0253 0.0024 135 - 157.5 0.0246 0.0011 157.5 - 180 0.0419 0.0015 180 - 202.5 0.0615 0.0027 202.5 - 225 0.0655 0.0149 0.0002 0.0001 225 - 247.5 0.0645 0.0343 0.0002 247.5 - 270 0.1407 0.0838 0.0031 270 - 292.5 0.0769 0.0088 292.5 - 315 0.0619 0.0034 315 - 337.5 0.0545 0.0027 337.5 - 360 Total 0.8381 0.1583 0.0035 0.0001
Figure 2.5 Wind rose chart for Gan, Addu Atoll, using the hourly data for years 2002 and 2003.
These figures may not give the exact frequency and intensity for Viligilli, but it is a guide for the broader region.
20 The threshold levels for damage are predicted based on interviews with locals and housing structural assessments provided by risk assessment report (UNDP, 2006), as shown in Table 2.9.
Table 2.9 Threshold levels for wind damage based on interviews with locals and available meteorological data. Wind speeds Impact 1-10 knots No Damage 11 – 16 knots No Damage 17 – 21 knots Light damage to trees and crops 22 – 28 knots Breaking branches and minor damage to open crops, some weak roofs damaged 28 – 33 knots Minor damage to open crops and houses 34 - 40 knots Minor to Moderate to major damage to houses, crops and trees 40+ Knots Moderate to Major damage to houses, trees falling, crops damaged
2.2.4 Tsunami
UNDP (2006) reported the region where Viligilli is geographically located to be a very high tsunami hazard zone. The tsunami of December 2004 flooded parts of Viligilli. According to the official estimates, 33% of the island was flooded during this event. There were also extensive damage to properties and a significant percentage of the population lost much of their livelihood. The tsunami run-up height at the eastern shoreline of the island was reported to be approximately 4m above MSL. The field assessment of the site indicated that the flood waters reached a distance of approximately 350m inland from the eastern shoreline. This results in a flooding extent of nearly 70% of the island. However, the severest damage to the houses and structures were limited within approximately 150m from the eastern shoreline. The decay of the flood water for this tsunami showed a logarithmic decay function (Fig 2.6). Tsunami induced tide level within the lagoon predicted using the tide data from the nearest tide station at Gan, Addu Atoll shows why the island was not flooded from its lagoonward side near the harbour (Figures 2.6 and 2.7). It is evident that the tide level within the atoll lagoon did not rise above the elevation of the island. The impact on western shoreline along lower ridges was reported to be within 10m from the shoreline.
21 5 Tsunami flooding decay curve 4 Tsunami induced tide level recorded at 3 the nearest tide station (December 2004)
2
1
0 0 50 100 150 200 250 300 350 400 450 500
Height rel MSL rel Height (m) -1
-2
-3
-4 Distance from oceanward shoreline (m)
Figure 2.6 Maximum water level caused by tsunami of December 2004 plotted across the island profile. Graph also shows the logarithmically decaying flood water level.
200 150 100 50 0 -50 -100
Water depth (cm)MSL rel depth Water -150 -200 0 100 200 300 400 500 600 700 800 900 1000 1100 Elapsed time (min) since 00:00hrs (UTC) of 26-12-2004
Figure 2.7 Water level recordings from the tide gauge at Gan, Addu Atoll indicating the wave height of tsunami 2004.
Comparatively lower wave height recorded at Gan may be partially due to the refraction of the wave caused by the Indian Ocean bathymetry as it travelled westwards Maldives
22 (Ali, 2005). The Indian Ocean bathymetry (Figure 2.8) shows shallower water depths extended far offshore at around the central region of the Maldives (at around the atolls of Laamu – Meemu). This shallower area caused the wave to bend away from the southern atolls and focused towards the central region of the country. It is likely that a similar pattern may persist in any future event if the waves originate from the northern Sundra trench. Based on Ali’s (2005) findings, a tsunami originating from the southern Sundra trench could have a higher impact on Viligilli as exposure is more profound.
Figure 2.8 Submarine topography around Maldives archipelago and modelled wave refraction for the December 2004 tsunami (source: Ali (2005)).
The predicted probable maximum tsunami wave height for the area where Viligilli is located is 3.2 – 4.5 m (UNDP, 2006). Examination of the flooding that will be caused by a wave run-up of 4.5 m for the island of Villingili indicates that such a magnitude wave will flood the entire island. The first 100 – 200 m from the shoreline will be a severely
23 destructive zone (Figure 2.9). The theoretical tsunami flood decay curve was plotted for a wave that is applied only for the direct wave from the oceanward side of the island. However, it is well understood that the tsunami wave will also travel into the atoll lagoon which will cause the water level in the atoll lagoon to rise. Rising of water level in the atoll lagoon would also cause flooding of the island from the lagoonward side of the island, if the atoll lagoon water level rises above the height of the island. The maximum tsunami wave induced water level height predicted for the atoll lagoon near Villingili is 1.4 m. This would flood the island of Villingili not just from the oceanward side of the island but also from the lagoonward side and the entire island will be flooded.
5.0 Theoretical flood decay 4.0 curve
3.0 Threshold level of flooding for severe structural damage
2.0
1.0
0.0 0 50 100 150 200 250 300 350 400 450 500
Height rel MSL rel (m) Height -1.0
-2.0 Extent of most destructive zone Tsunami induced tide level within the atoll -3.0 lagoon for maximum predicted wave
-4.0 Distance from oceanward shoreline (m)
Figure 2.9 Tsunami induced tide level within the atoll lagoon near Viligilli will flood the entire island. Graph also shows the flooding decay curve for the maximum predicted tsunami at Viligilli.
2.2.5 Earthquakes
24 There hasn’t been any major earthquake related incident recorded in the history of Viligilli or even Maldives. However, there have been a number of anecdotally reported tremors around the country.
The Disaster Risk Assessment Report (UNDP 2006) highlighted that Huvadhoo Atoll is geographically located in the second highest seismic hazard zone of the Maldives. According to the report the rate of decay of Peak Ground Acceleration (PGA) for the zone 3 in which Viligilli is located has a value less than 0.07 for 475 years return period (see Table 2.10). PGA values provided in the report have been converted to Modified Mercalli Intensity (MMI) scale (see column ‘MMI’ in Table 2.11). The MMI is a measure of the local damage potential of the earthquake. See table 3.10 for the range of damages for specific MMI values. Limited studies have been performed to determine the correlation between structural damage and ground motion in the region. The conversion used here is based on United States Geological Survey findings. No attempt has been made to individually model the exposure of Viligilli Island as time was limited for such a detailed assessment. Instead, the findings of UNDP (2006) were used.
Table 2.10 Probable maximum PGA values in each seismic hazard zone of Maldives (modified from UNDP, 2006). Seismic PGA values for MMI 2 hazard zone 475yrs return period 1 < 0.04 I 2 0.04 – 0.05 I 3 0.05 – 0.07 I 4 0.07 – 0.18 I-II 5 0.18 – 0.32 II-III
Table 2.11 Modified Mercalli Intensity description (Richter, 1958). MMI Shaking Description of Damage Value Severity I Low Not felt. Marginal and long period effects of large earthquakes. II Low Felt by persons at rest, on upper floors, or favourably placed. III Low Felt indoors. Hanging objects swing. Vibration like passing of light trucks. Duration estimated. May not be recognized as an earthquake. IV Low Hanging objects swing. Vibration like passing of heavy trucks; or sensation of a jolt like a heavy ball striking the walls. Standing motor cars rock. Windows, dishes, doors rattle. Glasses clink. Crockery clashes. In the
2 Based on KATZFEY, J. J. & MCINNES, K. L. (1996) GCM simulation of eastern Australian cutoff lows. Journal of Climate , 2337-2355.
25 upper range of IV, wooden walls and frame creak. V Low Felt outdoors; direction estimated. Sleepers wakened. Liquids disturbed, some spilled. Small unstable objects displaced or upset. Doors swing, close, open. Shutters, pictures move. Pendulum clocks stop, start, change rate. VI-XII Light - Light to total destruction Catastrophe
According to these findings the threshold for damage is very limited even in a 475 year return earthquake. It should however be noted that the actual damage may be different in Maldives since the masonry and structural stability factors have not been considered at local level for the MMI values presented here. Usually such adjustments can only be accurately made using historical events, which is almost nonexistent in Maldives.
2.2.6 Climate Change
The debate on climate change, especially Sea Level Rise (SLR) is far from complete. Questions have been raised about SLR itself (Morner et al., 2004, Morner, 2004) and the potential for coral island environments to naturally adapt (Kench et al., 2005, Woodroffe, 1993). However the majority view of the scientific community is that climate is changing and that these changes are more likely to have far reaching consequences for Maldives. For a country like Maldives, who are most at risk from any climate change impacts, it is important to consider a cautious approach in planning by considering worst case scenarios. The findings presented in this section are based on existing literature. No attempt has been made to undertake detailed modelling of climate change impacts specifically on the island due to time limitations. Hence, the projection could change with new findings and should be constantly reviewed.
The most critical driver for future hazard exposure in Maldives is the predicted sea level rise and Sea Surface Temperature (SST) rise. Khan et al. (2002, Woodroffe, 1993) analysis of tidal data for Gan, Addu Atoll shows the overall trend of Mean Tidal Level (MTL) is increasing in the southern atolls of Maldives. Their analysis shows an increasing annual MTL at Gan of 3.9 mm/year. These findings have also been backed by a slightly higher increase reported for Diego Garcia south of Addu Atoll (Sheppard, 2002). These calculations are higher than the average annual rate of 5.0 mm forecasted by IPCC (2001), but IPCC does predict a likely acceleration as time passes. Hence, this indicates that the MTL at Viligilli by 2100 will be nearly 0.4m above the present day MTL .
26 Similarly, Khan et al. (2002) reported air temperature at Addu Atoll is expected to rise at a rate of 0.4C per year, while the rate of rise in SST is 0.3C.
Predicted changes in extreme wind gusts related to climate change assumes that maximum wind gusts will increase by 2.5, 5 and 10 per cent per degree of global warming (Hay, 2006). Application of the rate of rise of SST to the best case assumption indicates a 15% increase in the maximum wind gusts by the year 2010 in Addu Atoll 120km south of where Viligilli is located.
The global circulation models predict an enhanced hydrological cycle and an increase in the mean rainfall over most of the Asia. It is therefore evident that the probability of occurrence and intensity of rainfall related flood hazards for the island of Viligilli will be increased in the future. It has also been reported that a warmer future climate as predicted by the climate change scenarios will cause a greater variability in the Indian monsoon, thus increasing the chances of extreme dry and wet monsoon seasons (Giorgi and Francisco, 2000). Global circulation models have predicted average precipitation in tropical south Asia, where the Maldives archipelago lies, to increase at a rate of 0.14% per year (Figure 2.10).
12
10
8
6 Rate of increase = 0.135% per year 4
2
Increase of precipitation (%) 0 2010 2020 2030 2040 2050 2060 2070 2080 2090 Year
Figure 2.10 Graph showing the rate of increase of averaged annual mean precipitation in tropical south Asia (Adger et al., 2004)
27 There are no conclusive agreements over the increase in frequency and intensity of Southern Indian Ocean Storms. However, some researchers have reported a possible increase in intensity and even a northward migration of the southern hemisphere storm belt (Kitoh et al., 1997) due rise in Sea Surface Temperatures (SST) and Sea Level Rise. If this is to happen in the Southern Indian Ocean, the frequency of and intensity of storms reaching Viligilli Island coastline will increase and thereby exposing the island more frequent damages from swell waves. The increase in sea level rise will also cause the storms to be more intense with higher flood heights.
The above discussed predicted climate changes for Viligilli and surrounding region is summarised in Table 2.12. It should be cautioned that the values are estimates based on most recent available literature on Gan which themselves have a number of uncertainties and possible errors. Hence, the values should only be taken as guide as it existed in 2006 and should be constantly reviewed. The first three elements are based climate change drivers while the bottom three is climatological consequences.
Table 2.12 Summary of climate change related parameters for various hazards. Element Predicted Predicted change (overall rise) Possible impacts on rate of Hazards in Viligilli Best Case Worst Case change
SLR 3.9-5.0mm Yr 2050: Yr 2050: +0.4m Tidal flooding, increase /yr +0.2m in swell wave flooding, Yr 2100: +0.88m reef drowning Yr 2100: +0.4m Air Temp 0.4°C / Yr 2050: decade +1.72° Yr 2100: +3.72° SST 0.3°C / Yr 2050: Increase in storm decade +1.29° surges and swell wave related flooding, Coral Yr 2100: bleaching & reduction +2.79° in coral defences Rainfall +0.14% / Yr 2050: Increased flooding, yr (or +1384mm Could effect coral reef +32mm/yr) growth Yr 2100: +2993mm Wind gusts 5% and Yr 2050: +3.8 Yr 2050: Increased windstorms, 10% / Knots +7.7Knots Increase in swell wave
28 degree of Yr 2100: +8.3 Yr 2100: +16.7 related flooding. warming Knots Knots Swell Frequency Increase in swell wave Waves expected related flooding. to change. Wave height in reef expected to be high
2.3 Event Scenarios
Based on the discussion provided in section 2.2 above, the following event scenarios have been estimated for Viligilli island (Tables 2.13. 2.14 and 2.15).
Table 2.13 Rapid onset flooding hazards
Hazard Max Impact thresholds Probability of Occurrence Prediction
Low Moderat Sever Low Moderate Severe e e Impact Impact Impact Swell Waves NA < 2.0m > 2.0m 3 > 3.0m Modera Low Very te Low (wave heights on reef flat – Average Island ridge height +1.7m above reef flat) Tsunami 4.5m < 2.0m > 2.0m > 3.0m Modera Low Very te low (wave heights on reef flat) SW monsoon 1.5m < 2.0m > 2.0m > 3.0m High Very low Unlikely high seas
Heavy Rainfall 284mm <60m > 60mm >175m High Moderate Low m m (For a 24 hour period)
3 Impact on southern western corner of the island will only be moderate if waves reach 2.5m, due to the high natural ridge. The rest of the reclaimed coastline is on average 1.5m higher than the reef flat.
29
Table 2.14 Slow onset flooding hazards (medium term scenario – year 2050)
Hazard Impact thresholds Probability of Occurrence
Low Moderate Severe Low Moderate Severe
SLR: Tidal < 2.0m > 2.0m > 3.0m Moderate Very Low Very Flooding Low
SLR: Swell < 2.0m > 2.0m > 3.0m Very high Moderate Low Waves
SLR: <60mm >60mm >175mm Very Moderate Low Heavy High Rainfall
Table 2.15 Other rapid onset events
Hazard Max Impact thresholds Probability of Occurrence Prediction
Low Moderate Severe Low Moderate Severe
Wind storm NA <30 > 30 knts > Very High Moderate knts 44Knts High Earthquake I < IV > IV > VI Very Unlikely none Low (MMI value 4)
2.4 Hazard zones
Hazard zones have been developed using a hazard intensity index. The index is based on a number of variables, namely historical records, topography, reef geomorphology, vegetation characteristics, existing mitigation measures and hazard impact threshold levels. The index ranges from 0 to 5 where 0 is considered as no impact and 5 is considered as very severe. In order to standardise the hazard zone for use in other components of this study only events above the severe threshold were considered. Hence, the hazard zones should be interpreted with reference to the hazard scenarios identified above.
4 Refer to earthquake section above
30 2.4.1 Swell waves and SW monsoon high Waves
The intensity of swell waves and SW monsoon udha is predicted to be highest 20 m from the western coastline on the lagoonward side (see Figure 2.11). Impacts of these waves are considered to be minor to negligible in most cases.
Swell waves higher than 3.0m on reef flat are predicted to reach the eastern coastline of the island. These waves may penetrate 100 to 300 m inland. The wave height on coastline is on average estimated to be 1.0-1.3 m and with rapid decline as it moves inland. The runoff on to the island is facilitated by the low topography in the reclaimed wetland area and due to the absence of coastal ridges. The presence of low wetlands in the south will restrict the flow on to the island. The impact is however not a result of vegetation but rather of the low elevation within the wetland.
The island lack natural ridges except for the south east corner of the island. These ridges themselves are low at approximately 1.5-1.8 m above MSL.
The western side of the island is relatively protected due to the cumulative effects of higher elevation of the area and the lower drainage basin on the east. Effects on the western side may be felt from wind waves and high seas (Udha), but will be limited to 50m from the coastline.
31 Hazrad Zoning Map Swell Waves and Udha
Intensity Index
Low 1 2 3 4 5 High Contour lines represent intensity index based on a severe event scenario (+3.0m on reef flat & +1.3m to +0.3m on land)
Predominant SW monsoon windwaves Predominant long distance Swell wave direction
0 150 300 meters
Figure 2.11 Hazard zoning map for swell waves and southwest monsoon high seas.
2.4.2 Tsunamis
When a severe threshold of tsunami hazard (>3.0 m on reef flat) is considered, the entire island predicted to be effected (Figure 2.12). If the waves reach beyond 4.0 m on reef flat the entire island is highly likely to be flooded. The intensity of flood waters will be most intense on the 100-150 m from the shoreline. Intensity is also expected to be high up to 300m inland where the flood waters will meet relatively higher ground in most of the mid to southern half of the island. Impact beyond 300 m is still considered to be moderate considering the possible surge from atoll lagoon due to rise in tide level.
The effected zone is dependent on the distance from coastline and minor variations in topography as it advances inland. Wave height around the island will vary based on the
32 original tsunami wave height, but the areas marked as low intensity is predicted to have proportionally lower heights compared to the coastline. A worst case scenario of 4.5 m tsunami (from MSL) is very likely to make the entire island a high intensity zone.
Hazrad Zoning Map Tsunami
Intensity Index
Low 1 2 3 4 5 High Contour lines represent intensity index based on a severe event scenario (+3.0m on reef flat & +2.0m to +0.5m on land)
0 150 300 meters
Figure 2.12 Hazard zoning map for tsunami flooding.
2.4.3 Heavy Rainfall
Heavy rainfall above the severe threshold is expected to flood most of the southern parts of the island where much of the settlement is located (Figure 2.13). The areas predicted for severe intensity are the reclaimed wetland areas in the south and the low areas along the newly reclaimed land adjacent to the harbour. These areas act as drainage basins for the surrounding higher areas. The reclaimed areas in general are lower than the existing island, and are within less than 0.3 m to the water table at highest tide. During
33 heavy rainfall, water table rises to the ground level leading to longer periods of flood water retention. The higher areas of the island are expected to have a lower intensity although the impact of flooding may be felt due to the increased height of roads compared to surrounding houses and lack of an artificial drainage system on the roads.
Hazrad Zoning Map Heavy Rainfall
Intensity Index
Low 1 2 3 4 5 High Contour lines represent intensity index based on a severe event scenario (+175mm in 24 hours)
0 150 300 meters
Figure 2.13 Hazard zoning map for heavy rainfall related flooding.
The rainfall hazard zones are approximate and based on the extrapolation of topographic data collected during field visits. A comprehensive topographic survey is required before these hazard zones could be accurately established.
2.4.4 Strong Wind
The intensity of the strong wind across the island is expected to remain fairly constant. Smaller variations may exist between the west and east side where by the west side
34 receives higher intensity due to the predominant westerly direction of abnormally strong winds. The entire island has been assigned an intensity index of 4 for strong winds.
2.4.5 Earthquakes
The entire island is a hazard zone with equal intensity. An intensity index of 1 has been assigned.
2.4.6 Climate Change
Establishing hazard zones specifically for climate change is impractical at this stage due to the lack of topographic and bathymetric data. However, the predicted impact patterns and hazard zones described above are expected to be prevalent with climate change as well, although the intensity is likely to slightly increase.
2.4.7 Composite Hazard Zones
A composite hazard zone map was produced using a GIS based on the above hazard zoning and intensity index (Figure 2.14). The coastal zone approximately 100m on the oceanward coastline and the reclaimed wetland areas are predicted are estimated to be the most intense regions for multiple hazards. The eastern side is particularly identified as a hazard zone due to the exposure to swell waves, tsunamis and wind damage. The reclaimed wetlands areas are have been identified due to its high exposure all hazards.
35 Hazard Zoning Map Multiple hazards
Intensity Index
Low 1 2 3 4 5 High Contour lines represent intensity index based on a severe event scenarios
0 150 300 meters
Fig 2.14 Composite hazard zone map
36 2.5 Limitations and recommendation for future study
The main limitation for this study is the incompleteness of the historic data for different hazardous events. The island authorities do not collect and record the impacts and dates of these events in a systematic manner. There is no systematic and consistent format for keeping the records. In addition to the lack of complete historic records there is no monitoring of coastal and environmental changes caused by anthropogenic activities such as road maintenance, beach replenishment, causeway building and reclamation works. It was noted that the island offices do not have the technical capacity to carryout such monitoring and record keeping exercises. It is therefore evident that there is an urgent need to increase the capacity of the island offices to collect and maintain records of hazardous events in a systematic manner.
The second major limitation was the inaccessibility to long-term meteorological data from the region. Historical meteorological datasets at least as daily records are critical in predicting trends and calculating the return periods of events specific to the site. The inaccessibility was caused by lack of resources to access them after the Department of Meteorology levied a substantial charge for acquiring the data. The lack of data has been compensated by borrowing data from alternate internet based resources such as University of Hawaii Tidal data. A more comprehensive assessment is thus recommended especially for wind storms and heavy rainfall once high resolution meteorological data is available.
The future development plans for the island are not finalised. Furthermore the existing drafts do not have proper documentations explaining the rationale and design criteria’s and prevailing environmental factors based on which the plan should have been drawn up. It was hence, impractical to access the future hazard exposure of the island based on a draft concept plan. It is recommended that this study be extended to include the impacts of new developments, especially land reclamations, once the plans are finalised.
The meteorological records in Maldives are based on 5 major stations and not at atoll level or island level. Hence all hazard predictions for Viligilli are based on regional data rather than localised data. Often the datasets available are short for accurate long term prediction. Hence, it should be noted that there would be a high degree of estimation and the actual hazard events could vary from what is described in this report. However, the findings are the closest approximation possible based on available data and time,
37 and does represent a detailed although not a comprehensive picture of hazard exposure in Viligilli.
References
ALI, S. (2005) December 26 2004 Tsunami Impact Assessment and a Tsunami Risk Assessment of the Maldives. School of Civil Engineering and the Environment. Southampton, United Kingdom, University of Southampton, . BINNIE BLACK & VEATCH (2000) Environmental / Technical study for dredging / reclamation works under Hulhumale' Project - Final Report. Male', Ministry of Construction and Public Works. BUCKLEY, B. W. & LESLIE, L. M. (2004) Preliminary climatology and improved modelling of South Indian Ocean and southern ocean mid-latitude cyclones. International Journal of Climatology, 24 , 1211-1230. DEPARTMENT OF METEOROLOGY (2007) The unusually strong swell, tidal waves hit Maldives Islands [sic]. Male', Maldives, Department of Meteorology. DEPARTMENT OF METEOROLOGY (DOM) (2005) Severe weather events in 2002 2003 and 2004. Accessed 1 November 2005,
38 KHAN, T. M. A., QUADIR, D. A., MURTY, T. S., KABIR, A., AKTAR, F. & SARKAR, M. A. (2002) Relative Sea Level Changes in Maldives and Vulnerability of Land Due to abnormal Coastal Inundation. Marine Geodesy, 25 , 133–143. KITOH, A., YUKIMOTO, S., NODA, A. & MOTOI, T. (1997) Simulated changes in the Asian summer monsoon at times of increased atmospheric CO2. Journal of Meteorological Society of Japan, 75 , 1019-1031. MANIKU, H. A. (1990) Changes in the Topography of Maldives, Male', Forum of Writers on Environment of Maldives. MORNER, N.-A. (2004) The Maldives project: a future free from sea-level flooding. Contemporary South Asia, 13 , 149-155. MORNER, N.-A., TOOLEY, M. & POSSNERT, G. (2004) New perspectives for the future of the Maldives. Global and Planetary Change, 40 , 177-182. NASEER, A. (2003) The integrated growth response of coral reefs to environmental forcing: morphometric analysis of coral reefs of the Maldives. Halifax, Nova Scotia, Dalhousie University. RICHTER, C. F. (1958) Elementary Seismology, San Francisco, W.H. Freeman and Company. SHEPPARD, C. R. C. (2002) Island Elevations, Reef Condition and Sea Level Rise in Atolls of Chagos, British Indian Ocean Territory. IN LINDEN, O., D. SOUTER, D. WILHELMSSON, AND D. OBURA (Ed.) Coral degradation in the Indian Ocean: Status Report 2002. Kalmar, Sweden, CORDIO, Department of Biology and Environmental Science, University of Kalmar. WOODROFFE, C. D. (1993) Morphology and evolution of reef islands in the Maldives. Proceedings of the 7th International Coral Reef Symposium, 1992. Guam, University of Guam Marine Laboratory. YOUNG, I. R. (1999) Seasonal variability of the global ocean wind and wave climate. International Journal of Climatology, 19 , 931–950.
39 3. Environment Vulnerabilities and Impacts
3.1 Environment Settings
3.1.1 Terrestrial Environment
Topography
The topography of Viligilli was assessed using three island profiles (see Figure 3.1). Given below are the general findings from this assessment. 0.75°N 0.76°N
73.43°E P3 P2
P1 Topographic Survey Locations Topographic profiles
P1 Profile reference
Figure 3.1 Topography field survey locations
Viligilli is one of the lowest islands studied under this project with an average elevation of +0.7 m MSL along the surveyed topographic profiles. This finding was reconfirmed from the shallow depths of ground water table around the island (on average approximately 0.5 m at high tide). The lower areas of the islands had the water table just 0.24 m below the ground level at average high tide.
Much of the low topography in the southern areas of the island can be attributed to the presence of wetland areas during the islands formation and subsequent land reclamation. As Figure 3.2 shows approximately 30% of the island was believed to be
40 wetland areas (based on historical aerial imagery, anecdotal information and field evidence). There are other accounts which report close to 45% (such as Luthfy (1994)), but their assessments were based entirely on anecdotal information. Since the 1950’s, almost the entire southern wetland areas were reclaimed on an ad hoc basis. New plots were allocated to individuals in the wetland areas and it was the up to the individuals to reclaim and develop them. Hence, it was reported that there was no organised reclamation of wetland except for the development of the roads during 1950’s and 60’s and that the reclamation activities were carried out without the use of any heavy machinery. During the first harbour development project in the early 1990’s, reclamation of small wetland areas in the south for housing and road development was carried out using heavy machinery. Furthermore during the second harbour development project in 2003, parts of the northern wetland areas were reclaimed for road development.
73.43°E
Wetland Areas Wetland Areas 1969 Potential Wetland Areas prior to 1969 0.76°N 0.75°N 0 150 300 meters Wetland Areas 2006
Figure 3.2 Past and present Wetland areas in Viligilli
The ad hoc reclamation practices meant that no consideration was given to the impacts on overall island topography. In fact the new newly reclaimed areas formed large depressions within the island. Figures 3.3-5 shows the topographic profiles of the island.
41 G
Modified Elevation Elevation Elevation Oceanward ridge Breakwater + 0.5m + 1.2m G’ + 0.4m + 0.9m Profile P1 1
Reclaimed land (harbour project) Reclaimed Wetland Area
0m Approximate Mean Sea Level Lagoonward Side Oceanward Side
0 100 200 300 400 500 600 700
Figure 3.3 Topographic profile P1
G Profile P2
G’ Partly artificial Harbour Oceanward Ridge Elevation Shallow Oceanward Ridge Quay + 1.0m Road + 0.6m watertable + 1.3m G’ G
1
Reclaimed Wetland Area
0 Approximate Mean Sea Level Lagoonward Side Oceanward Side
0 50 100 150 200 250 300 350 400 450
Figure 3.4 Topographic profile P2
G Profile P3
G’
Partly artificial Elevation Partly artificial Oceanward Ridge + 0.4m Constant slope Oceanward Ridge + 1.1m (reclaimed road) westward + 1.0m
1
Reclaimed Wetland Area 0 Approximate Mean Sea Level Lagoonward Side Oceanward Side
0 50 100 150 200 250 300 350 400
42 Figure 3.5 Topographic profile P3
While the low elevation in the south are associated with the wetland areas, the general low elevation in the northern half of the island may be associated with a rapid accretion and island formation process. The geologic age of the area was not determined in this study but based on the soil conditions and the vegetation cover it is generally evident that the northern half of the island is younger. The process of island building in the north is still very much active.
The implications of topographic patterns in Viligilli are considerable in terms of the exposure to natural hazards. Viligilli has frequent rainfall related flooding and the hazard zone has been identified as the low lying areas. Similarly, during times of ocean induced flooding, floodwaters tend to follow the topographic lows as was evident during the tsunami of 2004.
Vegetation 0.76°N 0.75°N 73.43°E
Thick Vegetation Vegetation Wetland Areas
Low and dispersed vegetation Reclaimed Areas 0 150 300 Settlement and sparse vegetation meters Large trees Figure 3.6 Vegetation distribution in Viligilli
Viligilli vegetation could be described based on two zones: The settlement area on the south and the uninhabited area in the north. The settlement area is sparsely vegetated
43 except for a patch of large trees (see Figure 3.6). These trees included common fruit bearing trees found in inhabited islands such as Mango and Breadfruit. The northern area of the island is mainly dominated by young and low trees.
Coastal vegetation around the island is depleted to a large extent and even in areas with a narrow strip of vegetation cover, the undergrowth has been removed. The remaining coastal vegetation is dominated by coconut palms. Coastal vegetation has also been effected due to beach replenishment activities on the eastern and western shoreline. Hence, the role of coastal vegetation as defence against natural hazards is very limited in the present settlement area of Viligilli.
One of the most noticeable characteristics of the Viligilli vegetation is the concentration of the larger trees in the south. Most of these trees have been planted after the 1950’s. When the vegetation distribution is analysed with the reclaimed wetland areas, it is apparent that much of these larger trees fall into the former wetland. Perhaps this concentration may be due to the proximity to water table and the fertility of soil in the area. While, these feature may be an advantage for vegetation growth it may be a disadvantage during an ocean induced flooding event. During the tsunami of 2004 almost 90% of the islands mango trees were destroyed. Incidentally, 95% of the mango trees are located within the reclaimed land. It was reported by the locals that tsunami floods were retained in the low-lying areas for a long period. Hence, the topography and shallow water table might have played a major role in allowing the salt water to be retained and subsequently damage the salt intolerant species in the area.
Ground Water and Soil.
No attempt was made in this study to do a quantitative analysis of the soil and ground water conditions but a visual assessment was based on similarities with other islands in Maldives.
The reclaimed wetland areas in the south had a substantial layer of humus underneath a thin layer of white sand, presumably from land reclamation. The northern half on the other hand represented a soil profile of a young island with a comparatively smaller layer of humus on top followed by fine and whiter material above the water table. The Viligili safe island development EIA (EDC, 2006) described the soil as loosely packed, silty, coral sand with pieces of coral shells’. Specific hard layers of up to 1 or 2m were also reported in 30% of the survey sites.
44 Viligilli ground water was reported to be generally in moderate condition (EDC, 2006). The EIA report also stated that freshwater lenses were significantly impacted during the tsunami due to salinisation and contamination. During the random survey of individual households, 30% reported bad smell and traces of saltiness. The random survey was done in the houses along the topographic survey locations. It was interesting to note the high proportion of houses within the reclaimed zone reporting good water quality. However, this finding is unreliable and could have been due to a number of factors such as heavy rainfall during the days prior to the survey and due differences in daily water usage. Similarly it was interesting to note the shallow watertable at 0.24m around high tide. This would mean that, at times of heavy rainfall, the water table would be higher than the ground level. Interviews with households in the low lying areas revealed a rise in well levels to about 0.5m from ground level. While this rise may be related to a number of factors, it shows the potential for water table to rapidly rise. Hence, it could be concluded that rainfall related flooding is imminent in the low lying areas during heavy rainfall, due to the shallow water table and the surface runoff patterns existing on the island.
The inhabitants reported no shortages of drinking water in the past perhaps due to the high rainfall in the region.
3.1.2 Coastal Environment
Beach and Beach Erosion
Viligilli island has undergone considerable erosion during the last 35 years (see Figure 3.7). This assessment is based on a snapshot of coastline during 1969 and 2004. During this period Viligilli, along with the adjacent uninhabited island (Falhuverrehaa) has lost approximately 2.7ha of land to erosion and gained 1.5ha. Hence, a net loss of 1.3ha of land is highly probable. It should be noted that this analysis is limited due the land reclamation done on the western coastline and potential errors in the 1969 aerial photographs. Significant erosion has been reported in historical records and during interviews with elderly locals, however. Extensive beach rock areas were also observed in the eroded areas indicating the retreat of beach.
45 0.75°N
73.43°E
Erosion and Accretion Erosion (1969 - 2004) Present severe erosion Accretion (1969 - 2004) Present Accretion
0 150 300 Land reclamation meters Figure 3.7 Erosion and Accretion in Viligilli
Much of the erosion and accretion has occurred in the northern part of the island which was observed to be highly dynamic due to the constant rapid flow of sediments and water in the area during high tide. It is unclear exactly when during the last 35 years the erosion of the northern tip took place, but interviews with locals suggest majority of the erosion and accretion in the region occurred in the 1990’s. While these reports remain to be substantiated, it raises the possibility that the erosion may have been part of the readjustment process following the harbour development and land reclamation in early 1990’s.
The inhabitants reported erosion in the southern and eastern coastline as their main concern. This may be due to the proximity of settlement to coastline in these areas. The most notable feature in Viligilli in relation to hazard exposure is the retreat of south western coastline, which is the closest region to the oceanward reef line. Between 2004 and 2006, a retreat of 4m was observed based on aerial photography and high resolution GPS observations during field visits. Similarly a retreat of about 10m was observed between 1969 and 2004, suggesting rapid erosion during last few years. Given the strong wave action in the region and the potential wave exposure from the east
46 (based on Naseer (2003)), a strong coastal ridge is expected. However, the ridge appears to have been depleted to a large extent and the island profiles confirmed the low elevation of the ridge (see Figure 3.3). Further studies are required to understand the geomorphologic changes in the area. At present it appears that the ridge has been gradually eroded and perhaps has undergone geomorphologic change during the tsunami of 2004.
3.1.3 Marine environment
General Reef Conditions
The reef environment of Viligilli has been assessed in detail during the Viligili safe island development EIA (EDC, 2006). The report states that the oceanward reef is in poor condition with average live coral cover at less than 10% and with a low number of fish species. On the other hand the atoll lagoonward side was reported to be in very good condition with live coral cover averaging around 40% and high number of fish species. The decline in quality of the oceanward reef was attributed to the coral bleaching event of 1998, and slow regeneration following the event.
Interviews with a number of fishermen and young snorkelers revealed similar findings. The general agreement amongst the interviewees were that the quality of reef areas on the oceanward side declined rapidly over last 10 years ad along with it a lowering of coral cover and reduction in fish numbers. Reef conditions on the atoll lagoonward reef were reported to be in relatively good condition and as a general reef fishing spot.
Almost 80%of oceanward reef flat is covered with sea grass. Sea grass patches were also observed elsewhere around the island, but was less abundant with coverage averaging 10-20%. Sea grass is known to stabilise beach areas due to sediment trapping capabilities. However, sea grass may also deprive the island coastline of a consistent supply of sediments, possibly resulting in gradual erosion. This assumption needs to be further substantiated using detailed assessment but the erosion patterns and timings do coincide with the growth and establishment of sea grass patches suggesting a role in sea grass growth on depriving the coastline of sediment supply.
3.1.4 Environmental Issues
The urbanisation of Viligilli Island has led to a number of environmental issues common in inhabited islands of Maldives.
47 • Waste management: Two separate waste sites have been established on both ends of the islands, but the lack of a collection method and distance to the waste sites has lead to adhoc dumping of waste on the shoreline and around the island.
• Viligilli island has sewerage system based on a combination of septic tanks and a piped systems. Most of the outlets are directly along the oceanward shoreline with discharges of untreated sewages. As a result the much of the coastal environment around the settlement is in hygienically poor condition. The overgrowth of seagrass may also be related to the frequent nutrient supply from these pipes.
• Coastal erosion: Severe erosion in selected points around the settlement.
• Ground water: Declining ground water quality due to salinisation and contamination.
3.1.5 Modifications to Natural Environment
Coastal Modifications
• As in most inhabited islands of Maldives, access infrastructure has been developed in Viligilli Island (see Figure 3.8). These include harbour entrance channels, breakwaters, dredged areas for boat landing and land reclamation as method of dredge material disposal. Almost all the development activities have been located along the eastern coastline of the island. Two specific projects were undertaken for harbour development: the first in early 1990’s and a renovation project in 2003.
• A number of coastal areas around the island have been replenished using dredge material from the harbour development projects. These include the development of an artificial ridge (approximately 180m long) along the western shoreline to mitigate erosion and potential North east monsoon related flooding.
• As a result of the modifications on the eastern side of the island, coastal processes around the area appear to have changed considerably, leading to rapid erosion in the reclaimed areas.
48 0.75°N
Harbour
Breakwater
73.43°E
Beach replenishment
Coastal and terrestrial Modifications Land reclamtion during Possible wetland reclamation Harbour development
0 150 300 Wetland Reclamation Harbour dredging meters
Figure 3.8 Coastal and terrestrial modifications
Terrestrial Modifications
• As described earlier, large areas of wetland areas in the island have been reclaimed for housing development (see figure 9). Reclamation was done in an ad hoc basis with individuals and community undertaking reclamation on need basis. 80% of the island’s wetland areas are known to have been reclaimed. Based on the interviews with locals, it appeared that these wetlands had little biodiversity significance as was evident in the present wetland areas. It may be difficult to ascertain the biodiversity importance of the old wetlands, but the areas do play a dominant role in the islands drainage system. Improper reclamation in the past has led to the prevalence of this drainage system and hence leading to frequent rainfall related flooding in the reclaimed area
• As presented earlier, substantial changes to the vegetation was necessitated due to the expanding human settlement. Vegetation cover has been reduced considerably and is expected to continue to decline in the future due to planned expansion of the settlement.. The re-vegetation in the newly reclaimed settlement
49 areas were much organised in Viligilli than any other island assessed under this project. Larger trees such as mango and breadfruit trees were grown in and around the settlement for the past 30 years.
• The increase in rainfall related flooding prompted the authorities to undertake road maintenance activities, which primarily involved levelling and raising roads. This has led to some houses in the island to be lower than the road causing flooding in these houses during heavy rainfall.
3.2 Environmental mitigation against historical hazard events
3.2.1 Natural Adaptation
There is little evidence that Viligilli was in the past exposed to severe storm events or intense wave activity. The beach content on the south eastern end of the island bears evidence of strong wave activity, but could be associated with the area being close to the reef edge. The heights of the ridges along the rest of the island are quite low and material quite fine, indicating lack of intense storm activity or strong wave action.
3.2.2 Human Adaptation
Viligilli has relatively few modifications undertaken to directly prevent natural hazards. The main activities include use of construction debris and beach replenishment to mitigate coastal erosion in along the eastern coastline and construction to breakwaters along the local harbour to minimise the impact of wave activity. Actions have also been undertaken to prevent rainfall relation flooding in newly reclaimed low areas and the main roads by raising the road surface and some housing plots.
3.3 Environmental vulnerabilities to natural hazards
3.3.1 Natural Vulnerabilities
• Villigili Island has one of the lowest elevations amongst the nine islands studied under this project. This makes Villigili amongst the most vulnerable islands to flooding events.
• The narrow width of the island in its northern half exposes the area to severe impacts during major flooding events like a tsunami.
50 • The narrow width combined with a north-south orientation and location on eastern rim has proven to be a major natural vulnerability during flooding events, in terms of the exposure of human settlements. It has been found in this study that waves run-up on land at least to a specific distance (see section 3). It has generally been established that the narrower the island the higher the likelihood of the island getting entirely inundated.
• There are substantial topographic variations within Villigili. The low lying areas of the island are exposed to heavy rainfall associated flooding and create conditions for flood run-up during ocean induced flooding events. Currently the low areas experience rainfall related flooding almost regularly effecting island functions such as schools and economic activities. Analysis of flood extents during tsunami shows that the effects of topographic lows were prominent in the wave run up.
• The ridges around the island are quite low, about +1.5m MSL maximum and averaging just +1.2m MSL. Such heights can easily be over topped by tsunamis and predicted storm surges.
• The presence of an adjacent island (Falhuverrehaa) may contribute to the increase in intensity of flooding in the area. There is a probability that the initial funnelling effect of water towards the northern edge of Viligilli and western edge of Falhuverreha and subsequent blockage of water flow may cause the water to rise and inundate both the islands. Such a pattern was observed in Laamu Gan during the tsunami of 2004. Interviews with locals in Viligilli partly confirmed this pattern during the tsunami of 2004 but the large extent of inundation in the northern half could have been influenced by a number of other factors including island width, low ridge and lack of vegetation.
• Villigili Island is exposed to swell waves from South East Indian Ocean (Naseer 2003).
• The island is located high rainfall zone. Combined with substantial variations in topography, parts of the island are frequently exposed to rainfall related flooding.
51 • Reef width appears to play an important role increasing or decreasing the impacts of ocean induced wave activity. The impact of gravity waves such as tsunami’s for example has its impacts reduced based on the length of reef. As has been discussed in the natural hazards chapter, these findings are preliminary and needs further inquiry using detailed empirical research.
3.3.2 Human induced vulnerabilities
• Improper land reclamation in wetland areas has created substantial topographic variations within Villigili, exposing the island to rainfall related flooding and ocean induced flooding. The original wetland areas formed the natural drainage system of the island and continued to function in a similar manner after reclamation. At times of ocean induced flooding the extent of inundation has been influenced by the topographic lows within the island. The structures and vegetation within the low areas have been specifically exposed to flooding impacts, as was evident from the tsunami of 2004. It is noteworthy, that only a few houses located on the islands original land was affected during the tsunami.
• The lack of coastal vegetation in certain parts of the coastline is a major concern in terms of exposure to natural hazards. Coastal vegetation including the undergrowth acts as natural barrier against tsunami’s, other ocean induced flooding events and wind storms. A wider coastal vegetation belt would absorb wave energy from a tsunami or a flooding event reducing the impact on infrastructure and human settlement. This has been proven from findings across the nine islands studied under this project. In Villigili the coastal vegetation has been substantially removed. On the eastern coastline on average there is barely 10m of vegetation. Amongst this narrow stretch, the under growth has been cleared leaving only the coconut palms and larger trees. Coupled with the low elevation of the ridge, the eastern side is therefore very vulnerable to ocean induced flooding events.
• Similar to the lack of coastal vegetation, the removal of vegetation from the settlement area exposes the structures to the direct effects of strong wind. The effects of climate change and global warming could be felt more strongly due to the apparent increase in temperature within the settlement area.
52 • The western coastline of the island has been modified to develop a harbour. Since its development and associated land reclamation, the newly reclaimed land has undergone considerable erosion probably in search of equilibrium in coastal processes. It was also observed that the southern half of the island remains more exposed during the South West monsoon. This may relate to the fact that not enough sediment is being transported to the southern side during SW monsoon due to the modifications to the western side. It may be a long time before the coastline adapts to the new modifications. Substantial changes to the northern tip of the island and western lagoon has also been observed from remote sensing analysis, which is highly likely to have been influenced by coastal modification on the western side.
3.4 Environmental assets to hazard mitigation
• The relatively large size of Villigili could be considered an asset against natural hazards specifically in the southern half of the island. The north-south orientation of the island and the narrow width in the northern part exposes the area to damage during tsunamis.
• Viligilli has two major routes of sediment supply due to its location on the reef system. This has allowed the island to grow northward consistently. More detailed studies are required to confirm this finding.
• The presence of Falhuverrehaa (uninhabited island in the north east corner) may act as a buffer during strong winds of North East monsoon. It has also been reported by the inhabitants that the effect of tsunami was also reduced by the presence of this island, an effect which appears plausible but limited given the geography and approach of the tsunami waves.
• The area on the southeast corner of the island close to reef edge appears to perform a major role in island protection and building processes. Based on the topographic evidence, this zone could have performed functions similar to a barrier island in the past during the formation and stabilisation of Viligilli. The shape and geomorphologic features in the northern areas also suggests that the island is being formed north ward and relies on the protection provided by the zone.
53 • Reef width appears to play an important role increasing or decreasing the impacts of ocean induced wave activity. The impact of gravity waves such as tsunami’s for example has its impacts reduced based on the length of reef. As has been discussed in the natural hazards chapter, these findings are preliminary and needs further inquiry using detailed empirical research.
3.5 Predicted environmental impacts from natural hazards
The natural environment of Viligilli and islands in Maldives archipelago in general appear to be resilient to most natural hazards. The impacts on island environments from major hazard events are usually short-term and insignificant in terms of the natural or geological timeframe. Natural timeframes are measured in 100’s of years which provides ample time for an island to recover from major events such as tsunamis. The recovery of island environments, especially vegetation, ground water and geomorphologic features in tsunami effected islands like Laamu Gan provides evidence of such rapid recovery. Different aspects of the natural environment may differ in their recovery. Impacts on marine environment and coastal processes may take longer to recover as their natural development processes are slow. In comparison, impacts on terrestrial environment, such as vegetation and groundwater may be more rapid. However, the speed of recovery of all these aspects will be dependent on the prevailing climatic conditions.
The resilience of coral islands to impacts from long-term events, especially predicted sea level rise is more difficult to predict. On the one hand it is generally argued that the outlook for low lying coral island is ‘catastrophic’ under the predicted worst case scenarios of sea level rise (IPCC 1990; IPCC 2001), with the entire Maldives predicted to disappear in 150-200 years. On the other hand new research in Maldives suggests that ‘contrary to most established commentaries on the precarious nature of atoll islands Maldivian islands have existed for 5000 yr, are morphologically resilient rather than fragile systems, and are expected to persist under current scenarios of future climate change and sea-level rise’ (Kench, McLean et al. 2005). A number of prominent scientists have similar views to the latter (for example, Woodroffe (1993), Morner (1994)).
In this respect, it is plausible that Viligilli may continue to naturally adapt to rising sea level. There are two scenarios for geological impacts on Viligilli. First, if the sea level continues to rise as projected and the coral reef system keep up with the rising sea level
54 and survive the rise in Sea Surface Temperatures, then the negative geological impacts are expected to be negligible, based on the natural history of Maldives (based on findings by Kench et. al (2005), Woodroffe (1993)). Second, if the sea level continues to rise as projected and the coral reefs fail to keep-up, then their could be substantial changes to the land and beaches of Viligilli (based on (Yamano 2000)). The question whether the coral islands could adjust to the latter scenario may not be answered convincingly based on current research. However, it is clear that the highly, modified environments of Viligilli stands to undergo substantial change or damage (even during the potential long term geological adjustments), due to potential loss of land through erosion, increased inundations, and salt water intrusion into water lens (based on Pernetta and Sestini (1989), Woodroffe (1989), Kench and Cowell (2002)).
Viligilli has particular vulnerability to sea level rise due to the presence of wetland areas. Since wetland areas in coral islands are linked to the tide and sea level, an increase in sea level may result in increase in size of such areas and a subsequent reduction in land (Woodroffe 1989). Although land has been reclaimed from the wetland areas, the low elevation of reclamation is highly likely to undergo impacts of a rising water table.
As noted earlier, environmental impacts from natural hazards will be apparent in the short-term and will appear as a major problem in inhabited islands due to a mismatch in assessment timeframes for natural and socio-economic impacts. The following table presents the short-term impacts from hazard event scenarios predicted for Viligilli.
Hazard Scenario Probability Poten tial Major Environmental Impacts at Location Tsunami (maximum scenario) 2.5m Low • Moderate to major damage to coastal vegetation (Short-term) • Long term or permanent damage to selected inland vegetation especially common backyard species such as mango and breadfruit trees. • Salt water intrusion into present wetland areas and island water lens causing loss of some flora and fauna. • Contamination of ground water if the sewerage system is damaged or if liquid contaminants such as diesel and chemicals are leaked. • Damage to waste management site and subsequent dispersion of debris in northern half of the island and pollution (land and
55 Hazard Scenario Probability Poten tial Major Environmental Impacts at Location ground water) • Salinisation of ground water lens causing ground water shortage. Salinisation expected to last until next rainy season possibly causing major temporary loss of flora. If the rainwater collection facilities are destroyed, potable water shortage would be critical. Impacts on reclaimed wetland areas will be higher due to low water table. • Widespread damage to backyard crops. Certain species may be completely and permanently destroyed. • Widespread damage to coastal protection and island access infrastructure such as breakwaters and ports. • Short-medium term loss of soil productivity • Moderate damage to coral reefs (based on UNEP (2005)) Storm Surge (based on UNDP, (2005)) 0.60m (1.53m Low • Minor damage to coastal vegetation storm tide) • Minor loss of crops • Minor to moderate damage to coastal protection infrastructure • Minor geomorphologic changes in the north western shoreline and lagoon 1.32m (2.30m Very Low • Moderate to major damage to coastal storm tide) vegetation (Short-term) • Long term or permanent damage to selected inland vegetation especially common backyard species such as mango and breadfruit trees. • Salt water intrusion into present wetland areas and island water lens causing loss of some flora and fauna. • Contamination of ground water if the sewerage system is damaged or if liquid contaminants such as diesel and chemicals are leaked. • Damage to waste management site and subsequent dispersion of debris in northern half of the island and pollution (land and ground water) • Salinisation of ground water lens causing ground water shortage. Salinisation expected to last until next rainy season possibly causing major temporary loss of flora. If the rainwater collection facilities are destroyed, potable water shortage would be
56 Hazard Scenario Probability Poten tial Major Environmental Impacts at Location critical. Impacts on reclaimed wetland areas will be higher due to low water table. • Widespread damage to backyard crops. Certain species may be completely and permanently destroyed. • Widespread damage to coastal protection and island access infrastructure such as breakwaters and ports. • Short-medium term loss of soil productivity • Minor-moderate damage to coral reefs Strong Wind 28-33 Knots Very High • Minor damage to very old and young fruit trees • Debris dispersion near waste sites. 34-65 Knots Low • Moderate damage to vegetation with falling branches and occasionally whole trees • Debris dispersion near waste sites. • Moderate-high damage to open field crops • Minor changes to coastal ridges 65+ Knots Very Low • Widespread damage to inland vegetation • Debris dispersion near waste sites. • Minor changes to coastal ridges Heavy rainfall 187mm Moderate • Minor to moderate flooding in low areas, including roads and houses. 284mm Very Low • Widespread flooding but restricted to low areas of the island, especially reclaimed wetlands. Drought • Minor damage to backyard fruit trees Earthquake • Minor-moderate geomorphologic changes Sea Level Rise by year 2100 (effects of single flood event) Medium Moderate • Widespread flooding during high tides and (0.41m) storm surges. • Loss of land due to erosion. • Loss of coastal vegetation • Major changes to coastal geomorphology. • Saltwater intrusion into wetland areas and salinisation of ground water leading to water shortage and loss of flora and fauna. • Minor to moderate expansion of wetland areas • Rise in water table of reclaimed wetland areas forcing additional reclamation to raise the areas.
3.6 Findings and Recommendations for safe island development plan
57 • The proposed safe island development project in Villigili proposes considerable change to the existing natural environment, converting it to a predominantly artificial environment. The implications of these changes are numerous for the existing environment both in the short and long term. The proposed modifications may require considerable time for the island to achieve equilibrium in different forces controlling coastal processes. During this period considerable alterations to the existing coastal environment may be imminent.
• The proposed land reclamation project is expected to have the biggest impact on the island environment and island exposure to natural hazards. The following points were noted on the existing reclamation project.
o Reclamation is being conducted on the oceanward side and reaches to within 100m of the reefline. The implications for moving the coastline close to the reef line needs to be clearly understood. There is a possibility that the reduced distance may increase the chances of wave overtopping and flooding during severe weather events. The extent of land reclamation needs to be reviewed.
o Reclaiming the oceanward side and protecting the reclaimed area with breakwaters may cause considerable changes in the unprotected coastlines around the island. This could include rapid onset of erosion at specific points around the island, especially at the end points of coastal protection and a possible prolonged continuation of the erosion and accretion until equilibrium in coastal processes are achieved.
o The reclamation is highly likely to cause considerable damage to the outer reef due to its proximity and current land reclamation practices. This would reduce the defensive capacity of the reef system and expose Viligilli to long term climate hazards.
o The soil composition of a reclaimed area may need to be properly established. Soil in coral islands of Maldives has specific profiles which dictate the suitability vegetation and perhaps drainage.
o A coral island requires a properly functioning drainage system to mitigate rainfall hazards. Newly reclaimed land should consider establishing either a natural or an artificial drainage system to mitigate these hazards. Past
58 reclamation projects in islands like GDh. Thinadhoo has caused increased flooding hazards due to improper drainage systems.
o The elevation of the newly reclaimed area should be inline with the existing island topography or should consider establishing a functioning drainage system to mitigate flooding hazards resulting from modified topography. Specific attention should be given where the new reclamation joins the existing island. A drainage zone has been planned in the area where the existing island (Villigili) joins the new reclamation. However, the width of the drainage zone, the sloping of newly reclaimed land to facilitate drainage in this zone and modifying existing ridge of Viligilli to facilitate runoff need to be considered and properly planned. Furthermore, the low areas where the new reclamation joins the uninhabited island (Falhuverrehaa) will need to be flood proofed with a proper artificial drainage system.
o The proposed shape of the reclamation zone is artificial and does not represent the coastline shapes of large natural islands. There may be implications for wave action and foreshore currents, especially wave refraction. Hence, there is a probability of implication for ocean induced flooding hazards. These aspects need to be properly studied before the proposed shape is approved.
o The flat elevation of a +1.4m above MSL for the reclaimed land may not be the most efficient topography for a functioning drainage system. The costs involved in establishing and maintaining an artificial drainage system without the assistance of natural slopes may be considerably higher.
• Viligilli Island has a number of vulnerabilities resulting from past improper reclamation activities in the wetland areas. These vulnerabilities need to be reduced in any planned safe island development programme.
• The function of the low drainage areas in the Environment Protection Zone (EPZ) needs to be reviewed. Given the limited topographic variations within the newly proposed reclaimed land, the proposed 0.1m variation and the 25m width in the drainage area may not have the desired effects on flood control. The function of
59 a low area near the high ridges has best been performed in other islands if the width of the area is large and if an appropriate variation in height between the low area and the high areas exists. Hence it is recommended that a review of the function and characteristics of the floodway, reconsideration of the flat elevation of +1.4m for the island and reconsideration of the 0.1m variation for the floodway be undertaken.
• Based on the 9 islands studied in this project, it has been observed that strong coastal vegetation is amongst most reliable natural defences of an island at times of ocean induced flooding, strong winds and against coastal erosion. The design of EPZ zone needs to be reviewed to consider the important characteristics of coastal vegetation system that is required to be replicated in the safe island design. The width of the vegetation belt, the composition and layering of plant species and vegetation density needs to be specifically looked into, if the desired outcome from the EPZ is to replicate the coastal vegetation function of a natural system. Based on our observations, the proposed width of coastal vegetation may not be appropriate for reducing certain ocean induced hazard exposures. The timing of vegetation establishment also needs to be clearly identified in the safe island development plan.
• The constant height of the ridge proposed in the present safe island development concept needs to be reviewed to identify a suitable height to the wave conditions prevailing around Viligilli Island and predicted hazard scenarios for the region.
• A re-vegetation plan needs to be incorporated into the safe island development plan to ensure minimal exposure to strong winds and future climate change related temperature increases.
• The EPZ zones needs to be extended around the island.
3.7 Limitations and recommendations for further study
• The main limitation of this study is the lack of time to undertake more empirical and detailed assessments of the island. The consequence of the short time limit is the semi-empirical mode of assessment and the generalised nature of findings.
• The lack of existing survey data on critical characteristics of the island and reef, such as topography and bathymetry data, and the lack of long term survey data
60 such as that of wave on current data, limits the amount of empirical assessments that could be done within the short timeframe.
• The topographic data used in this study shows the variations along three main roads of the island. Such a limited survey will not capture all the low and high areas of the island. Hence, the hazard zones identified may be incomplete due to this limitation.
• This study however is a major contribution to the risk assessment of safe islands. It has highlighted several leads in risk assessment and areas to concentrate on future more detailed assessment of safe islands. This study has also highlighted some of the limitations in existing safe island concept and possible ways to go about finding solutions to enhance the concept. In this sense, this study is the foundation for further detailed risk assessment of safe islands.
• There is a time scale mismatch between environmental changes and socio- economic developments. While we project environmental changes for the next 100 years, the longest period that a detailed socio-economic scenario is credible is about 10 years.
• Uncertainties in climatic predictions, especially those related Sea Level Rise and Sea Surface Temperature increases. It is predicted that intensity and frequency of storms will increase in the India Ocean with the predicted climate change, but the extent is unclear. The predictions that can be used in this study are based on specific assumptions which may or may not be realized.
• The following data and assessments need to be included in future detailed environmental risk assessment of safe islands.
o A topographic and bathymetric survey for all assessment islands prior to the risk assessment. The survey should be at least at 0.5m resolution for land and 1.0m in water.
o Coral reef conditions data of the ‘house reef’ including live coral cover, fish abundance and coral growth rates.
o At least a year’s data on island coastal processes in selected locations of Maldives including sediment movement patterns, shoreline changes, current data and wave data.
61 o Detailed GIS basemaps for the assessment islands.
o Coastal change, flood risk and climate change risk modelling using GIS.
o Quantitative hydrological impact assessment.
o Coral reef surveys
o Wave run-up modelling on reef flats and on land for gravity waves and surges.
References
Enviornment and Dredging Consultancy (EDC) (2006). Envrionmental Impact Assessment of Construction of Safe Island Viligilli, Gaafu Alifu Atoll, Maldives. Male', Maldives, Ministry of Planning and National Development.
IPCC (1990). Strategies for Adaptation to Sea-Level Rise: Report of the Coastal Management Subgroup. Strategies for Adaptation to Sea-Level Rise: Report of the Coastal Management Subgroup . IPCC Response Strategies Working Group. Cambridge, University of Cambridge.
IPCC (2001). Climate Change 2001: Impacts, Adaptation, and Vulnerability . Cambridge, United Kingdom and New York, NY, USA, Cambridge University Press.
Kench, P. S. and P. J. Cowell (2002). "Erosion of low- lying reef islands." Tiempo 46 : 6- 12.
Kench, P. S., R. F. McLean, et al. (2005). "New model of reef-island evolution: Maldives, Indian Ocean." Geology 33 (2): 145-148.
Luthfy, M. I. (1994). Dhivehirajjeyge Geographeege Vanavaru (The Geogrpahy of Maldives) . Male', Novelty Printers and Publishers.
Naseer, A. (2003). The integrated growth response of coral reefs to environmental forcing: morphometric analysis of coral reefs of the Maldives. Halifax, Nova Scotia, Dalhousie University : 275.
Pernetta, J. and G. Sestini (1989). The Maldives and the impact of expected climatic changes. UNEP Regional Seas Reports and Studies No. 104 . Nairobi, UNEP.
UNEP (2005). Maldives: Post-Tsunami Environmental Assessment, United Nations Environment Programme.
United Nations Development Programme (UNDP) (2005). Disaster Risk Profile for Maldives. Male', UNDP and Government of Maldives.
62 Woodroffe, C. D. (1989). Maldives and Sea Level Rise: An Environmental Perspective. Male', Ministry of Planning and Environment : 63.
Woodroffe, C. D. (1993). Morphology and evolution of reef islands in the Maldives. Proceedings of the 7th International Coral Reef Symposium, 1992 . Guam, University of Guam Marine Laboratory. 2: 1217-1226.
Yamano, H. (2000). Sensitivity of reef flats and reef islands to sea level change . Bali, Indonesia.
63 4. Structural vulnerability and impacts
Ga. Viligilli Island experiences frequent natural events, some of which have resulted in substantial losses. It is predominantly exposed to ocean-originated floods, rainfall flood and strong winds (occasionally). In summary this island is exposed to frequent flooding due to its topography and past human activities. These issues will continue to be a problem in the future, although with the proposed reclamation, the frequency and impact of ocean induced flooding could be minimised.
4.1 House vulnerability
Around 90 houses were identified as vulnerable, which accounts for about 23% of the total houses on the island. Among the vulnerable houses identified, houses with extremely poorly physical conditions account for less than 10% of the total houses. In contrast, 15% of houses are very close to shoreline and without proper protection measures. Houses that are vulnerable due to low elevation account for 5% only.
4.1.1 Vulnerability type
The house vulnerability of the island belongs to a PP-WB-LE type and most houses are vulnerable due to their weak physical conditions and poor protection with respect to ocean-originated floods ( Fig. 4.1 ). Of 91 vulnerable houses identified, more than 60% are found to be located close to shoreline and without proper protection and 40% are structurally-weak. However, low elevation does not contribute too much to the house vulnerability.
4.1.2 Vulnerable houses
64 The vulnerable houses of Viligili Island can be divided into 3 groups in terms of vulnerability indicator groups. As shown in Fig. 4.2, around 60% of the vulnerable houses are not well protected facing to the threats from the ocean. About 25% of the vulnerable houses are physically weak and 11% are vulnerable in all 3 vulnerable indicator group.
70.0 60.0 50.0 40.0 30.0 Houses 20.0 10.0 %of Total Vulnerable 0.0 WB PP LE Indicator group
Fig. 4.1 The type of house vulnerability.
Viligili WB
WBPP 4% 2% WBLE 25% WBPPLE
PP 7% 49% 2% LE 11% PPLE
Fig. 4.2 Distribution of vulnerable houses.
4.2 Houses at risk
4.2.1 Tsunami floods
Tsunami flood is the most destructive hazard on the island. A 3-4 m tsunami wave can inundate 2/3 of the island. At present, around 320 houses are exposed
65 to tsunami hazard, accounting for 80% the total houses on the island, and 85 are vulnerable due to their weak structure and poor protection (Fig. 4.4). Given a 3-4 m tsunami wave, 4 houses may be subjected to serious damage, 55 to moderate damage, and 25 to slight damage (Table 4.1).
Currently, land reclamation oceanward is in planning. After reclamation, the impact severity of tsunami flood will be significantly reduced. As shown in Table 4.1, the number of the houses exposed to tsunami floods will reduced by 50%, from 320 to 124, and the number of vulnerable houses decreases to 27. Given the same tsunami event, damage will be significant reduced. However, it must be pointed out that this calculation is based on the assumption that no new exposure is added after reclamation.
4.2.2 Rainfall flood
Reclaimed by filling a wetland since the 1950’s, the rainfall flood-prone areas of Viligili account for 27% of the island, with an area of 14.6 hectares (Fig. 4.4, left). At present, more than 30% of the houses and population are exposed to frequent rainfall floods (Table 4.1); 32 houses are vulnerable, of which 5 may be subjected to slight damage due to their poor physical conditions and 27 may be affected in their contents due to low elevation, during moderate to heavy rainfall.
Rainfall floods in these areas will be further enhanced by improperly raising the road surface up to 30 cm. Elevating the roads originally is supposed to mitigate flooding over the road, however, it ends up with increased flooding in adjacent households. The water depth can reach up to 30-50 cm and last for several days during the raining season. Extended inundation has caused disruption of sewerage system and a widespread damage to backyard crops (bananas, chillies etc.) .
66 4.2.3 Swell wave flood
Swell wave/surge floods prevail in the eastern coast and southwestern side of the island. It can intrude up to 100 m inland along most part of the eastern shoreline, whereas on the southwestern side, floods of 0.5 m may reach up to 400 m inland. At present, around 54 houses are exposed to swell wave / surge flooding (Fig. 4.4, right and Table 4.1), accounting for 14% of the total houses on the island. 23 houses are considered as vulnerable and most of them vulnerable are located on the eastern coast. Given a flooding with a water depth of 0.5 m, 5 of the 23 vulnerable houses will subjected to a slight damage and 18 affected with their contents.
4.2.4 Earthquake
Located in Seismic Hazard Zone 5, Viligili is exposed to a PGA between 0.05 and 0.07. Therefore, most houses on the island are resistant to earthquake of such degree. However, around 40 houses may be still subjected to a slight to moderate damage. In worse case, some houses may be completely destroyed during an earthquake.
Table 4.1 Houses at risk on Ga.Viligilli. Exposed Vulnerable Potential Damage Hazard houses houses Serious Moderate Slight Content type # % # % # % # % # % # %
TS(p) 317 80.1 85 21.7 4 1.3 55 17.4 25 7.9 233 73.5 TS(f) 124 31.6 27 21.8 0 0 5 4.0 14 11.3 105 84.7 W/S 54 13.8 23 42.6 0 0 0 0 5 9.3 49 90.1
RF 141 36.0 32 22.7 0 0 0 0 5 3.5 136 96.5 Flood Earthquake 392 100 41 10.5 Wind 392 100 41 10.5 ------Erosion
67
Fig. 4.4 Houses at risk associated with rainfall-induced floods (left) and swell wave-induced floods (right).
68
Fig. 4.5 Houses at risk associated with tsunami floods: Left-Prior to land reclamation; Right-After land reclamation.
69 4.3 Critical facilities at risk
The exposure of critical facilities is shown in Fig. 4.6 and 4.7, and summarized in Table 4.1.
4.3.1 Rainfall flood Few critical facilities are located in the rainfall flood-prone area of Viligili Island (Fig. 4.7, left).
4.3.2 Swell wave / surge flood
Hospital is located in the swell wave flood-prone area, but it is not subjected any damage to its structure.
4.3.3 Tsunami flood
Most critical facilities of the targeted island, such as schools, mosques, and island office, are located in the rainfall flood-prone area, whereas only a few in the ocean-originated flood-prone area (Table 4.2, Fig. 4.4 and 4.5 ). Physically, most buildings of critical facilities are not vulnerable to flood hazards identified on the island and subjected to little damage during flooding, given the water depth recorded. All facility buildings have strong foundations and are well structured, with an age of less than 10 years. However, contents of some critical facility buildings may be affected and subjected to some degree of damage or loss, due to the low elevation relative to their adjacent roads. For example, the plinth level of schools, i.e. KPS pre-school and Viligilli school, is just 10-30 cm above their adjacent road surface and entrances just at road level. A moderate heavy rainfall can cause flooding in school yards and disturb school activities. Under some circumstances, schools may be closed for days. Located in the northeastern low- lying area of the island, on the other hand, buildings of Cable TV and power distribution stations may be subjected to frequent floods with the plinths at road
70 level. However, most mosques on the island may not be affected by most flooding events because of their high plinth level up to 40-60 cm, except for some that are relatively close to southwestern shoreline and subjected to higher floods.
Therefore, critical facilities on Viligilli Island are at low risk, although located in hazard-prone areas.
Table 4.2 Critical facilities at risk on Ga. Viligilli Island. Critical facilities Potential damage/loss Hazard type Monetary Exposed Vulnerable Physical damage value Hospital, power Hospital, Tsunami house, Atholhuge power (prior to Dhiraagu and house, reclamation) Wataniya sites, waste site waste site Tsunami Hospital, Dhiraagu Hospital (after and Wataniya sites, Flood reclamation) Atholhuge Wave/surge Oil storage, hospital None N/A
1 mosque, 1 None N/A Rainfall Atholhuge, hospital, wataniya site Earthquake All facilities No No No Wind - - - - Erosion - - - - Note: “-“ means “not applicable”.
71
Fig. 4.5 Critical facilities at risk associated with rainfall floods (left) and swell / surge floods (right).
72
Fig. 4.6 Critical facilities at risk associated with swell wave/surge floods.
73
4.4 Functioning impacts
Although causing no physical damage to most critical facility buildings, major flooding events may impact the functioning of some critical facilities. Some potential functioning impacts are summarized in Table 4.3 . As one of the serious functioning impacts, the sewerage system on the island may fail to operate days during flooding, whereas school activities may be interrupted.
Table 4.3 Potential functioning disruption matrix Flood Earth- Function Wind Tsunami Swell wave Ra infall quake Administration 1)
Health care a few weeks days
Education
Religion
Sanitation 3) 3-5 days
Water supply
Power supply A week
Transportation
Communication 2)
Note: 1) Administration including routine community management, police, court, fire fighting; 2) Communication refers to telecommunication and TV; 3) Sanitation issues caused by failure of sewerage system and waste disposal.
4.5 Recommendations for risk reduction
According to the physical vulnerability and impacts in the previous sections, the following options are recommended for risk reduction of Ga. Viligili:
• Set up an EPZ with a buffer zone of proper width on the eastern coast to reduce the exposure of houses to ocean-originated flooding. • Avoid protecting roads from rainfall flooding by raising the road surface to reduce household-wide flooding. • Enhance building codes in the newly-reclaimed area on the eastern coast. • Retrofit critical facilities such as hospital, power house, and communication sites.
75