Detailed Island Risk Assessment in Maldives
Volume III: Detailed Island Reports
L. Gan – Part 1
DIRAM team Disaster Risk Management Programme UNDP Maldives
December 2007 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
Gan is located on the eastern rim of Laamu Atoll, at approximately 73° 31' 50"E and 1° 52' 56" N, about 250 km from the nations capital Male’ and 3.5 km from the nearest airport, Kadhdhoo (Figure 1.1). Gan is the largest island in terms of land area and population amongst 13 inhabited islands of Laamu atoll. It’s nearest inhabited islands are Kalhaidhoo (7 km), Mundoo (10 km) and Atoll Capital Fonadhoo (10 km). Gan forms part of a stretch of 4 islands connected through causeways and bridges and is the second largest group of islands connected in this manner with a combined land area of 9.4km 2. The island is exposed to NE monsoon generated winds and waves, and occasional storm activities originating from the cyclone belt of Indian Ocean. Gan is also believed to be located in an area where offshore ocean bathymetry could create a ‘funnelling’ effect due to wave refraction during tsunami events originating from Sumatran Ridge (Shifaz, 2004).
E E
' '
5
0
1
3
°
°
3
3 7 Isdhoo
Maaen'boodhoo
2° 00' N Mundoo
Indian Ocean Hadhdhunmathi Atoll (Laamu Atoll) Gan
Maavah Kadhdhoo N Fonadhoo Location Map of Gan Gaadhoo 0 5 10 Hithadhoo kilometers Kunahandhoo
Figure 1.1 Location map of Gan.
3 1.2 Physical environment
Gan is the largest island in the Maldives with a surface area of 600 Ha (6 km 2). It has a length of 7.2km and a width of 1.5km at its widest point. The island is wider in the north (1500m) and narrower in the south (400m). There are three settlements on the island, Thundi (northeast), Mathimaradhoo (east) and Mukurimagu (south). In additional there is a zone designated as Industrial Development Zone, which has a number of structures, located within it. All the settlements are located along the coastline but only the Thundi settlement is located away from the oceanward coastline. Entire settlement of both the Mathimaradhoo and Mukurimagu are within 300m of oceanward coastline while Thundi is located approximately 900m away from it.
Gan has been connected to the adjacent island, Maandhoo through land reclamation. Together the two islands form a land area of 670 Ha (6.7 km 2) and covers 21km of coastline. In addition, the islands of Kadhoo (airport) and Funadhoo (Atoll Capital) are connected to Maandhoo Island through causeways and bridges. The total length of the island group is approximately 16km.
The reef of Gan is a large reef system with a surface area of 4500 Ha (45km 2), covering 70% of the eastern rim of Hadhunmathi Atoll and stretching to approximately 29km. The reef also hosts 5 inhabited islands, an Airport island (Kadhoo), 2 industrial islands and 8 uninhabited islands, totalling a 1220ha (12.2 km 2) of land. It is the largest concentration of land in a single reef and Gan comprises half of its land area.
Gan is oriented slightly in a northeast-southwest direction and is located in the middle of the reef system. The island is located approximately 250m from the oceanward reefline and 350m from the lagoonward reefline. The reef system is exposed to wind generated waves during NE monsoons and long distance swell waves from the southeast Indian Ocean.
In spite, of its size, Gan is a low lying island with an average height of +0.9m MSL. The oceanward coastline is long and low, exposing the island abnormal rises in sea level. Vegetation cover on the island is very high but large tracts of land have been cleared for agriculture and forestry. There are substantial variations in the topography of the island including a large wetland area, which plays a major role in the drainage system, especially during rainfall and ocean induced flooding events.
4 The proportion of Gan developed for human settlement is small. However, the impact of human settlement can be found throughout natural environment of the island. Parts of the natural environment have been modified to meet the development requirements of the settlement and the atoll population. Terrestrial modifications have been undertaken around the entire island for agricultural development, while coastal modifications have mainly been undertaken in three main points along the western shoreline which nonetheless have contributed to change coastal processes around Gan. Low areas within the island have been settled without proper levelling, leading to flooding in some of these areas.
5 2. Natural hazards
This section provides the assessment of natural hazard exposure in Feydhoo 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 Gan has been exposed to multiple hazards in the past although its exposure has been limited. 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 2.1 below lists the known events and a summary of their impacts on the island.
The historic hazardous events for Gan showed that the island faced the following multiple 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 is the most commonly occurring hazard events. Windstorms have also been reported as frequent especially during the southwest monsoon. Swell surges have been reported as infrequent and as having little impact.
Table 2.1. Known historic hazard events of Gan Metrological Dates of the Impacts hazard recorded events
Flooding caused Events commonly There are areas in the 3 settlements by Heavy rainfall occurring during SW (Thundi, Mathimaradhoo and Mukurimagu) monsoon. which are prone to rainfall flooding. All these settlements have wetland areas in close proximity to the settlement. As settlements expand to the low areas exposure to flooding becomes imminent. Impacts from these events are usually minor with damage to household goods and disruption to daily activities such as businesses and schools.
Flooding caused • 1950’s (exact There was one major flooding event by swell surges date unknown) reported for Gan, which is dated back to • 5 July 1966 1950’s. Exact date is not known, but residents say there were reports of fish near the northern wetland area, which is located 400m inland. No substantial
6 damage to the settlements was reported.
Windstorms • 11 July 1966 1 No major recent events have been • 5 May 1977 reported. • 12 May 1978 Written records show damage to • 28 Sept 1984 vegetation and crops. Little damage to property was reported. Droughts No major event have been reported
Earthquake No major event have been reported
Tsunami 26 th Dec 2004 At least 70% of the island was flooded during the tsunami of 2004. Flood heights were recorded at 2.0m (maximum). Flood heights and their distances in Mathimaradhoo are as following 2.0m – at a distance of 30m from shoreline 1.5m at a distance of 100m from shoreline 1.0m at a distance of 150m from shoreline less than 0.5m – at a distance between 300m and 600m from shoreline The primary reason for tsunami inundation may be found in the low ridge of the island and the presence of very low areas towards the centre of the island.
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
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 2.2.1 Swell Waves and Wind Waves
Being located on the eastern rim of Laamu atoll, Gan is relatively protected from the year round swell waves approaching from a west to southerly direction. There are no specific wave studies undertaken for Gan, but studies undertaken around the country reports a predominantly southwest to a southerly direction for swell waves (Kench et. al (2006), Young (1999), DHI(1999) and Binnie Black & Veatch (2000)). A similar pattern could be expected for swell waves reaching Laamu Atoll. Laamu atoll is also one of the most closed atolls in Maldives with only 6 major reef passes, 5 of which are narrower than 600m. The widest channel (4km wide), is located in the southern rim facing a south easterly direction. Hence the probability of swell waves, approaching from the southwest, propagating through the atoll is very limited.
The east and west coastlines of Gan are exposed to wind waves, however. During the NE monsoon between November and March, the eastern (oceanward) coastline may receive strong waves. Wave studies done in similar settings in GA. Viligilli (EDC, 2006), and K.Hulhule’ (Binnie Black & Veatch, 2000) reported wave heights less than 2.0m and with wave periods of 2-4 seconds. The west coast is exposed to wind generated waves during SW monsoon, originating within the atoll due to the 30 km fetch and usually with wave heights less than or about 0.5m.
Despite, being located away from the predominant swell wave direction, Gan 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 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 low due to the proximity to the equator.
The occurrence of abnormal swell waves on Gan reef flat is dependent on a number of factors such as the wave height, location of the original storm event
8 within 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.
Based on the current data available it is impossible to link the swell incidents to the known cyclonic events in the Indian Ocean. 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 Gan 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.
Udha
Flooding is also known to be caused in Gan 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 coastline of the island and are probably caused by a combination of high tides and strong wind waves. Impacts from udha events are usually restricted to within 20m of the western coastline. Due to the comparatively high coastal ridges on the western coastline, the effects of udha incidents are further controlled.
The udha phenomena needs to be further explored based on long term wave and climatological data of the Indian Ocean. Udha events could prove to be a major hazard in the face of climate change since these events are very frequents, have a direct link to climate patterns and sea level.
Processes controlling water levels around Gan
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
9 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 level are combined with high tides there could be strong surges of water across the reef flat. Due to the low elevation of Gan coastline, such waves have the potential to create flooding.
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 Gan 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) coastline of the island. The inland extent of flooding is greatest towards the northern wetland. The reason could be attributed to the topographically lower elevations and absence of natural ridge system.
10 Predominant NE monsoon windwaves
SW wind waves
Predominant Long distance swell wave direction
Swell wave Propagation Wetland Flood Extent
Historical Flood Extents & Probable wave propagation patterns around Gan
0 500 1,000 metres
Figure 2.1 Historical flood events and probable wave propagation patterns in Gan and its reef flat.
The highest wave height reported on the island during flooding events was 1.0m (3.0ft). This height is consistent with flood heights reported from swell or surge related waves in Maldives.
Future event prediction
It is known that Gan 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
11 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 Gan or could have reduced impact due to the protection offered by the southern and western rim of the atoll.
Possible range of swell wave direction in L.Gan: 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 Gan Island
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 15 years in Gan with probable water heights (on land) of 1.0m and every 8 years with probable water heights of 0.5-0.75m. Events with water heights less than 0.5m and greater than 0.2m
12 are likely to occur once every 5years. The timing of swell events is expected to be predominantly between April to October, based on historic events and storm event patterns (see Table 2.2).
Table 2.2 Variation of Severe storm events in South Indian Ocean between 1999 & 2003 (source: (Buckley and Leslie (2004)) Severe wind event variation Longitude band Winter 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
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 island more frequently.
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.
13 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 L.Gan is Kadhoo airport which became operational in 1986. Unfortunately this study does not have access to Kadhoo data. Moreover, Kadhoo data may be limited for long term trend observation due smaller number of detailed observation years. Hence, to resolve the issue, data from Hulhule’ has been used. It is recommended that further assessment be made once Kadhoo data becomes available.
The mean annual rainfall of Hulhule’ is 1991.5mm with a Standard Deviation of 316.4mm 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 below).
14
Figure 2.4 Mean Monthly Rainfall in Hulhule’(1975-2004).
Historic records of rainfall related flooding on the island of Gan 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 sections of Thundi, Mathimaradhoo and Mukurimagu. These areas usually correspond to wetland edges in Thundi and Mathimaradhoo settlement, and reclaimed wetland areas in Mukurimagu settlement. Moreover, substantial topographic variations exist within the Gan Island, as is common on larger islands of Maldives. Settlement expansion into and along the edges of these low lying areas have exposed them to flood impact. Furthermore, to remedy flooding 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 in all three settlements. With no artificial drainage system for the roads, the surrounding houses in the low areas are at constant risk of flooding. Heavy rainfall related flooding has been reported to reach up to 0.35m above the ground level in Thundi and Mukurimagu. In addition, construction of the ‘main road’ along the length of the island has caused blockages for water runoff towards the northern wetland areas. As a result the areas on either side of the main road are usually flooded during heavy rainfall.
15 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 Gan, through observation of daily rainfall data in Kadhoo. Unfortunately, we were unable to acquire daily historical data. However, available limited severe weather reports shows that Kadhoo received a maximum precipitation of 110.8mm for a 24 hour period on 21 th November 2004 (DoM, 2005). Based on interviews with locals, this event caused minor to moderate levels of flooding in all three settlements. Damages in Thundi and Mukurimagu settlements were reported for personal property, backyard crops and some open field crops. Flood heights in the northern half was reported at 0.2-0.35m. The worst affected area was the southern and western part of the Thundi, western part of Mathimaradhoo and northern part of Mukurimagu island, at low lying areas close to wetlands. Schools in Thundi Island were closed due to flood waters. Similarly an event during January 2003 caused 78.mm of rainfall during a 24 hour period and led minor damages to personal property in Mukurimagu island.
The probable maximum precipitations predicted for Hulhule’ and S.Gan by UNDP (2006) are as follows:
Table 2.3 Probable Maximum Precipitation for various Return periods in Hulhule’ and Gan Station Return Period 50 year 100 year 200 year 500 year Hulhule’ 187.4 203.6 219.8 241.1 Gan 218.1 238.1 258.1 284.4
Given the high variations in rainfall in Kadhoo, these figures may vary. 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 2.4).
Table 2.4 Threshold levels for rainfall related flooding in Gan Threshold level Impact (daily rainfall)
16 50mm Puddles on road, flooding in low houses, occasional 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. 230+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.
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.
2.2.3 Wind storms and cyclones
Maldives being located within the equatorial region of the Indian Ocean is generally free from cyclonic activity (Figure 2.5). There have only been a few cyclonic strength depressions that have tracked through the Maldives, all which occurred in the northern and north central regions. According to the hazard risk assessment report (UNDP, 2006) Gan falls within the second least hazardous zone for cyclone related hazards and has a maximum predicted cyclonic wind speeds of 56 Kts (see figure below). There are no such records for the southern region, although a number of gale force winds have been recorded due to low depressions in the region. Winds exceeding 35 knots (gale to strong gale winds) were reported as individual events in Kadhoo annually between 2002 and 2006, all caused by known low pressure systems near Maldives rather than the monsoon (DoM, 2005). The maximum wind speed in Kadhoo during this period was approximately 46 kts.
17 Kulhudhufushi Fonadhoo
Thulusdhoo
Kudahuvadhoo Vilufushi Gan
probable maximum Hazard Zones cyclone wind speed (kts) Villingili 5 96.8 Thinadhoo 4 84.2 3 69.6 2 55.9 Hithadhoo Feydhoo 1 0.0
Figure 2.5 Cyclone hazard zones of the Maldives as defined by UNDP (2006).
Historic records for Gan have indicated that near gale force winds (see Table 2.5) 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 vegetation and backyard crops. Gan does have lush vegetation dominated by larger trees species, which acts to minimise the direct exposure of properties.
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.
The threshold levels for damage are predicted based on interviews with locals and housing structural assessments provided by risk assessment report (UNDP, 2006), as summarized in Table 2.6.
18 Table 2.5 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 2.6 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 vegetation 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 Gan is geographically located to be a very high tsunami hazard zone. The tsunami of December 2004 had devastated a number of islands in the eastern rim of Laamu atoll along with parts of Gan. According to the official estimates, 50% of the island was flooded during this event. Field surveys and aerial photographs immediately after the event revealed that approximately 70% of the island was flooded. Flood waters travelled approximately 1km inland in the northern half while much of the southern end was flushed from east to west. Hence, all the 3
19 settlements were flooded. The settlements of Mathimaradhoo and Mukurimagu along with the ‘industrial zone’ were entirely flooded. Flooding in Thundi settlement was limited to 30%. The significantly high exposure of Mathimaradhoo and Mukurimagu are due to the close proximity to the oceanward coastline.
There were extensive damage to properties in Mathimaradhoo and Mukurumagu and a significant percentage of the population of the island lost much of their livelihood due to the damage to crops and businesses. The tsunami run-up height at the eastern shoreline of the island was reported to be approximately 4m above MSL reducing to 0.3m inland. 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. Tsunami induced tide level within the lagoon predicted using the tide data from the nearest tide station at Hulhule’ shows why the island was not flooded from its lagoonward side (Fig 2.6 and Fig 2.7). It is evident that the tide level within the atoll lagoon did not rise above the elevation of the island.
5 N P3 P2 P1 4 Predicted Tsunami Flooding Decay Curve Tsunami Induced tide level 3 at the nearest tide station (December 2004 tsunami) 2
1
0 Height rel MSL relHeight(m) MSL 200 400 600 800 1000 1200 1400 1600m -1
-2 Distance from oceanward shoreline -3
Fig 2.6. Maximum water level caused by tsunami of December 2004 plotted across the island profile of Gan near Thundi settlement evidently showing the reason why the island did not get flooded from the lagoonward side. Graph also shows the logarithmically decaying flood water level.
20
Fig 2.7. Water level recordings from the tide gauge at Hulhule’ indicating the wave height of tsunami 2004 (source: University of Hawai’i Sea Level Centre, http://ilikai.soest.hawaii.edu/uhslc/iot1d/male1.html)
Comparatively higher exposure of Laamu Atoll may be partially due to the refraction of the wave caused by the Indian Ocean bathymetry as it travelled westwards Maldives (Ali, 2005). The Indian Ocean bathymetry (Fig 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 became 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.
21
Fig 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 Gan is located is 3.2 – 4.5m (UNDP, 2006). Examination of the flooding that will be caused by a wave run-up of 4.5m for the island of Gan indicates that such a magnitude wave will flood the entire island from coast to coast. The first 150-200m from the shoreline will be a severely destructive zone (Fig 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. It also 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. This could cause flooding of the island from the lagoonward side of the island, if the water level rises above the height of the island. The maximum tsunami wave induced water level height predicted for the atoll lagoon near Gan is 1.7m. This could flood the island of Gan not
22 just from the oceanward side of the island but also from the lagoonward side and the entire island will be flooded.
N 5 Threshold level for P3 P2 Extent of most flooding for severe P1 4 destructive zone strucutral damage Tsunami Induced tide level 3
2
1 0 Height rel MSL (m) Heightrel MSL 200 400 600 800 1000 1200 1400 1600m -1 Distance from oceanward shoreline -2 Predicted -3 Tsunami Flooding Decay Curve
Fig 2.9. Tsunami induced tide level within the atoll lagoon Gan will flood the entire island. Graph also shows the flooding decay curve and maximum impact zone for the maximum predicted tsunami at Gan.
2.2.5 Earthquakes
There hasn’t been any major earthquake related incident recorded in the history of Gan 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 Laamu Atoll is geographically located in the highest seismic hazard zone rated 2 out of 5, based on the entire country. According to the report the rate of decay of peak ground acceleration (PGA) for the zone 2 in which Gan is located has a value less than 0.05 for a 475 years return period (see table below). PGA values provided in the report have been converted to Modified Mercalli Intensity (MMI) scale (see column ‘MMI’ in table 3.9 table below). 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
23 attempt has been made to individually model the exposure of Gan Island as time was limited for such a detailed assessment. Instead, the findings of UNDP (2006) were used.
Table 2.7 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 2.8 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 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
2 Based on KATZFEY, J. J. & MCINNES, K. L. (1996) GCM simulation of eastern Australian cutoff lows. Journal of Climate , 2337-2355.
24 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 Gan by 2100 will be nearly 0.4m above the present day MTL .
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. Although no specific studies have been done for Laamu Atoll, the findings from Addu Atoll could be used as a guide to predicted changes.
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 southern Atolls .
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 Gan will be
25 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
Fig 2.10 Graph showing the rate of increase of averaged annual mean precipitation in tropical south Asia (Adger et al., 2004)
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 Gan 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 Gan and surrounding region is summarised below in Table 2.9. It should be cautioned that the values are estimates based on most recent available literature on Maldives 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 are climatological consequences.
26 Table 2.9 Summary of climate change related parameters for various hazards. Element Predicted Predicted change (overall rise) Possible impacts on rate of Hazards in Gan 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 degree of related flooding. Yr 2100: +8.3 Yr 2100: +16.7 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 Gan Island (Tables 2.10-12).
27 Table 2.10 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 > 3.0m 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)
Table 2.11 Slow onset flooding hazards (medium term scenario – year 2050)
1.0.1. H 1.0.2. Impact thresholds 1.0.3. Probability of Occurrence azard
1.0.4. 1.0.5. L 1.0.6. M 1.0.7. S 1.0.8. L 1.0.9. M 1.0.10 . S ow oderate evere ow oderate evere
1.0.11. S 1.0.12. < 1.0.14. > 1.0.15. > 1.0.16. M 1.0.17. V 1.0.18. V LR: Tidal 2.0m 2.0m 3.0m oderate ery Low ery Low Flooding 1.0.13. 1.0.19. S 1.0.20. < 1.0.22. > 1.0.23. > 1.0.24. V 1.0.25. M 1.0.26. L LR: Swell 2.0m 2.0m 3.0m ery high oderate ow Waves 1.0.21. 1.0.27. S 1.0.28. < 1.0.29. > 1.0.30. > 1.0.31. V 1.0.32. M 1.0.33. L LR: 60mm 60mm 175mm ery High oderate ow Heavy Rainfall
28
Table 2.12 Other rapid onset events
Hazard Max Impact th resholds Probability of Occurrence Prediction
Low Moderate Severe Low Moderate Severe
Wind storm NA <30 > 30 knts > Very High Moderate knts 45Knts High Earthquake I < IV > IV > VI Very Unlikely none Low (MMI value 3)
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.
2.4.1 Swell waves and SW monsoon high Waves
Swell waves higher than 3.0m on reef flat are predicted to reach the eastern coastline of the island. These waves may penetrate 200 to 700m inland (Figure 2.11). The wave height on coastline is on average estimated to be 1.0 m and with rapid decline as it moves inland. The runoff on to the island is facilitated by the low topography towards the centre and due to the absence of high coastal ridges along much of the coastline (sees physical environment section).
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 20-50m from the coastline.
3 Refer to earthquake section above
29 Amongst the individual settlements, the Thundi settlement is relatively protected from impacts of swell waves due to the presence of a 1000m wide uninhabited land on eastern side. The settlements of Mathimaradhoo, Mukurimagu and the industrial zone remain exposed due to their location close to the eastern coastline. Structures within the first 100m of the coastline along Mathimaradhoo and Mukurimagu are particularly exposed to severe intensity should an event in the severe category strike. Moreover the entire footprint of these two settlements is exposed to varied intensities from a severe category event.
Thundi
Mathimaradhoo
Industrial Zone
Hazard Zoning Map Swell Waves and Udha
Mukurimagu 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)
0 500 1,000 metres
Figure 2.11 Hazard zoning map for swell waves and southwest monsoon high seas.
30 2.4.2 Tsunamis
When a severe threshold of tsunami hazard (>3.0m on reef flat) is considered, the entire island predicted to be effected (Figure 2.12). If the waves reach beyond 4.0m on reef flat the entire island is highly likely to be flooded due the prevalent tide levels. High intensity waves will flush through the island from the eastern side while tide related surges will occur within the atoll, flooding the western coastline. The intensity of flood waters will be highest 200-250m from the shoreline. Intensity could also be high up to 500m inland owing to the downward slope existing along the length of the island, 300m from the eastern coastline. Impact beyond 500m 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 original tsunami wave height, but the areas marked as low intensity is predicted to have proportionally lower heights compared to the coastline.
31 Thundi
Mathimaradhoo
Industrial Zone
Hazard Zoning Map Tsunami
Mukurimagu Intensity Index
Low 1 2 3 4 5 High
Contour lines represent intensity index based on a severe event scenario (+3.0m to +0.5m on land)
0 500 1,000 metres
Figure 2.12 Hazard zoning map for tsunami flooding.
The settlements of Mathimaradhoo and Mukurimagu will receive high intensity waves due its proximity to coastline. The entire settlement foot prints of these two settlements are considered a hazard zone for severe category tsunamis. The settlement of Thundi is protected from the direct impact of the waves although parts of the settlement could experience strong wave flushing form the north east. Thundi is predicted to be mostly exposed to flooding caused by rise in tsunami related. Impacts will therefore be lower than the settlements on the eastern coastline.
32 2.4.3 Heavy Rainfall
Heavy rainfall above the severe threshold is expected to flood parts of all three settlements (Figure 2.13). The areas predicted for severe intensity are the wetland areas and the topographic lows on the eastern half of the island. These areas act as drainage basins for the surrounding higher areas and due the large size of the island the ‘catchment area’ is considerable for surface runoff during heavy rainfall.
In Thundi settlement the high intensity zone is expected to be located in the southern and western areas, especially where the new resettlement project is being undertaken. The natural depression and occasional wetlands patches in this area performs a drainage function for this section of the island. Similarly, the central wetland area close to the Mathimaradhoo settlement is also expected to lead to flooding on the western parts of the settlement. Similar topographic lows and old wetland areas in the Mukurimagu settlement will also cause moderate to severe flooding.
33 Thundi
Mathimaradhoo
Industrial Zone
Hazard Zoning Map Rainfall Mukurimagu Intensity Index
Low 1 2 3 4 5 High Contour lines represent intensity index based on a severe event scenario (+3.0m to +0.5m on land) Note: White areas represent areas with no data
0 500 1,000 metres 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. The white areas represent areas with no field surveys due to poor accessibility. 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 receives higher intensity due to the predominant westerly direction of abnormally strong
34 winds. The entire island has been assigned an intensity index of 4 for strong winds during a severe event.
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 200m from the oceanward coastline and the northern wetland areas are predicted 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.
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
0 500 1,000 metres
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 atleast 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 Gan 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 Gan.
References
ALI, S. (2005) December 26 2004 Tsunami Impact Assessment and a Tsunami Risk Assessment of the Maldives. School of Civil Engineering and the Envrionment. Southampton, United Kingdom, University of Southampton, . BINNIE BLACK & VEATCH (2000) Enviromental / 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 unsually strong swell, tidal waves hit Maldives Islands [sic]. Male', Maldives, Department of Meterorology. 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. 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 Gan was assessed using three island profiles (see Figure 3.1). Given below are the general findings from this assessment.
The island is generally low lying with an average elevation of +0.9 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 1 m at median tide).
As characteristic of large islands, considerable variations in topography were observed in Gan. The island does not have high elevations comparable to other large islands in Maldives but there are substantial tracts of low areas developed during island formation. These include a major wetland area in the northern half of the island. As can be seen from general trends in low areas shown in Figure 3.5, a low depression extends almost along the length of the island. It has to be noted that the map shows predicted low areas, based on topographic profiles (see Figure 3.2-4) and general field observations. The roads used for the assessment, except profile 3 (Figure 2.4) had been modified during road maintenance. To accommodate this limitation, additional field assessments were undertaken in surrounding unmodified areas during level surveys.
The low areas dictate the drainage system of the island. During heavy rainfall, low areas are regularly flooded. Although much of these areas are located outside the settlements, there are small patches of low areas that fall within the settlements. Hence, parts of Thundi, Mathimaradhoo and Mukurumagu settlements experience occasional rainfall related flooding. In general, larger islands are less exposed to the impacts of ocean induced flooding due their width and the presence of buffer land. However, drainage system prevalent on Gan and the general low elevation of the island appears to negate this advantage. At present the low ridges on the oceanward side and the general gradient towards the centre can cause flood waters around 2.0 m above MSL to flood half of the island. The tsunami of 2004 appeared to prove this trend.
The coastal ridges of Gan are quite low averaging 1.5 m and increasing in height southwards. It is interesting to note that the coastline facing northeast was the lowest
40 while the coastline facing southwest was highest. This pattern tends suggest that the impacts of NE monsoon may not be that prominent and that there is an alternate wave energy source effecting the southwest coastline. Observations by Nasser (2003) in a Maldives wide wave energy study tends to suggest that wave energy approaching from southeast Indian Ocean has a higher wave power and could effect reefs facing southeast. The generally low elevations along the oceanward coastline also tend to suggest that the overall wave energy is low. In any case the low ridges expose much of Gan to the effects of ocean induced flooding.
10214000°N
Thundi P1
Mathimaradhoo
P2 10212000°N Indutrial Zone
10210000°N P3 Mukurimagu
TOPOGRAPHIC SURVEY PROFILES
0 500 1,000 metres 3000°E3380 3 000°E336 000°E 4000 3
Figure 3.1 Topographic survey locations
41
P3 generally End of P2 P1 Oceanward Reclaimed area low elevation Ridge system in the area generally low Ridge Elevation (+1.1m) (+1.5m) Low created Main +0.6m Harbour by water Elevation road quaywall +0.8m (close to runoff wetland area)
1m
0 Lagoonward Side Approximate Mean Sea Level Oceanward side
Note: Profile modified due to road maintenance activities
0 200 400 600 800 1000 1200 1400 1600
Figure 3.2 Topographic profile P1
P3 P2 Hospital Oceanward Ridge Close Elevation Main Road Elevation (+1.6m) +0.9m Elevation to Wetland Lagoonward area +0.5m Ridge +0.75m (+0.8m)
Lagoonward Side Approximate Mean Sea Level Oceanward side
Note: Profile modified due to road maintenance activities
0 200 400 600 800 1000 1200 1400
Figure 3.3 Topographic profile P2
42 Oceanward Ridge N P3 (+1.8m) P2 Low area prevalent in Main Road Elevation Llagoonward much of +0.7m Ridge Elevation Dhandimagu (+0.8m) +0.5m Island
1m
0
Lagoonward Side Approximate Mean Sea Level Oceanward side
0 50 100 150 200 250 300 350 400 450
Figure 3.4 Topographic profile P3
Low Areas
0 500 1,000 metres
43 Figure 3.5 Topographic Lows
Vegetation
Recent vegetation Very low & sparse Vegetation
10214000°N Thundi
Wetland areas
Areas cleared for tsunami housing reconstruction Mathimaradhoo Fairly depleted coastal vegetation 10212000°N
Indutrial Zone
Strong coastal vegetation strip (60m) Strong coastal vegetation
Strong coastal vegetation
10210000°N Dense vegetation & Mukurimagu larger trees Sand mined areas No coastal vegetation General Vegetation cover VEGETATION Dense forest area DISTRIBUTION
0 500 1,000 metres
Figure 3.6 Vegetation Distribution
44 The overall vegetation cover in Gan Island is very high compared to other inhabited islands, primarily due its large size (Figure 3.6). The settlement footprints of the three main settlements and industrial zone cover just 15% of the total land area. Vegetation cover within the settlements is low and is mainly restricted to medium sized backyard fruit trees.
Majority of the vegetation cover on the island comprise of a wide ranging medium to low species. There are patches of larger trees distributed across the island. Much of the larger trees have been reported as cleared for forestry. The most prominent patch of large trees (coconut palms) is located south of Thundi settlement. At present a large proportion of this patch has been cleared for tsunami resettlement.
The coastal vegetation around the island is dense and well established. Coastal vegetation around the settlements has been largely depleted, however. Coastal vegetation in more than 75% of the coastline around the settlement of Mukurimagu has been cleared while over 50% of Mathimaradhoo Island has been cleared. On average the coastal vegetation belt is 30m wide within the settlement areas in Mathimaradhoo and over 80 m wide around the industrial development zone.
Coastal vegetation in the northwest and the northeast corner of the island is relatively young and sparsely distributed. The tsunami of 2004 seemed to have effected the coastal vegetation, but has since recovered by 2007.
Ground Water and Soil
Gan Island has a substantial layer of fresh water (MPND, 2005). Water lens depth varies across the island based on topography. Generally the water table could be reached with less than 1m at median tide in all areas. This could decrease to 0.5 m during spring high tides or more during heavy rainfall, especially in low lying areas.
Gan’s ground water is generally in good conditions and no traces of contamination were reported (MPND, 2005). There were no shortages of potable water in the past due to the good quality of ground water and availability of rainfall reserves.
The soil conditions were not assessed across the island due to time limitation. Gan is reported to have some of the most fertile soils in the atoll (MPND, 2005).
45 3.1.2 Coastal Environment
Sediment Mov ement Island naturally expanding northward. Process af f ected Low coastal ridges due to harbour construction
Severe Erosion %%
Thundi NE monsoon Low areas wind and storm generated Swell wav es SW monsoon wind generated wav es & Sediment f lows residual swell blocked by harbour Mathimaradhoo Wetland wav es %% Low coastal ridges Low Areas
Large sediment Indutrial Zone Shallow budget reef f lat
Sea grass Long distance swell wav es originating f rom South Indian Ocean Sediment and current Sediment f low f low changes seasonally changes seasonally
Low Areas Two seperate islands naturally merged Mukurimagu
Severe Erosion
Ov ergrowth of COASTAL AND TERRESTRIAL seagrass & Seagrass FEATURES v ery low currents
Erosion
Currentl f lows altered due to reclamation 0 500 1,000 metres
Figure 3.7 Coastal Features
Figure 3.7 summarises the coastal characteristics of Gan Island. The coastline is predominantly affected by monsoon wind driven waves and long distance swell waves originating from south east Indian Ocean. On the ocean ward side, wave activity is more prominent during the north east monsoon. On the lagoon wardside the impact of
46 monsoon wind generated waves are controlled by the closed nature of the reef, but nonetheless creates enough conditions to generate waves within the atoll. Impacts from waves originating within the atoll are expected to be limited.
The oceanward coastline does not resemble a high energy coastline, based on the dominant geomorphic features. Furthermore, the coastline facing north and northeast was observed to have less exposure to strong wave energy than the stretch of coastline facing southwest. The net result is a low coastline with the no prominent ridge systems. The exceptions are the southern half of the oceanward coastline which increases in height, although still lower than other larger islands like Seenu Atoll Hithadhoo or Gnaviyani Fuvahmulah.
The lengthy coastline along the oceanward and lagoonward sides has meant that the effects of longshore drift are prominent. Sediment was observed to be transported along long distances and was seen to vary across the two monsoon seasons. The lagoonward coastline has a larger sand budget and tends to move more quantities. This extent of transport may have been limited due to the presence of solid structures and dredged areas on both ends of the island.
Gan was observed to be expanding northwards especially on its northeast and northwest corners. This process has been hampered over the last 4 years due to the harbour development in Thundi settlement. Large areas are now seen to be eroding from the north possibly due to sand supplies being restricted during the southwest monsoon.
Coastal erosion has been reported as a major issue by all three settlements. These events appear to be seasonal or long-term cyclic changes occurring to the coastline. The trends and patterns in erosion are difficult to forecast due to the recent introduction of coastal modifications. Over the past 40 years, 3 ha of land have been eroded from the island, especially the northern end of the island. During the same period over 2.5 ha of new land have been added. Hence the net loss is insignificant, considering the size of the island. However, coastal erosion is predicted to become a major environmental issue, atleast in the medium-term, due to irreversible changes being brought to the coastal environment by human activities.
47 3.1.3 Marine environment
General Reef Conditions
General historical changes to reef conditions were assessed anecdotally, through interviews with a number of fishermen. The general agreement amongst the interviewees was that the quality of reef areas on both sides of the island has declined considerably over the past 50 years. Much of these changes were reported to be in the form of excess sedimentation in the lagoonward side and general coral decline on the oceanward side. The construction of causeways may have played a role in this reduction of quality.
Sea grass overgrowth is a major problem on both the lagoonward and oceanward lagoon. The case of the lagoonward side is more prominent due to the slow currents and coastal modifications blocking sediment transport. The condition in the southwest corner of the island has deteriorated over the past few years with foul smell becoming a major issue.
3.1.4 Modifications to Natural Environment
Coastal Modifications
As noted earlier, much of the coastal modifications have been undertaken on the eastern shoreline of the island (Figure 3.8). Below is a summary of major modifications.
• The two islands of Gan and Maandhoo have been joined together through land reclamation. Water flow through the lagoon pass between the islands has been halted permanently. There appears to be major localised changes to the coastal sediment transport and erosion patterns around the region, following the reclamation activity.
• A harbour was developed in the eastern coastline of Thundi settlement in 2006. This included dredging activities and construction of solid structures perpendicular to the shoreline, blocking sediment flow along the eastern coastline. The area north of the harbour is highly mobile as the island was continually growing in the region. Currently the sediment supply to the area has been restricted and is very likely to cause erosion in the long term.
48 • There two areas with abandoned harbour work: east of Mukurimagu and north Thundi Settlement. These areas have dredged harbour basins close to shoreline and remnants of disposed dredge material. In Mukurimagu, along with the dredged area, there is a 20m long pile of dredge material located perpendicular to the shoreline. In thundi settlement the, dredge material no longer exists and the dredged area has become considerably shallow, possibly during the tsunami of 2004. These modifications cause disruptions to sediment flow and have implication for localised erosion and accretion, and possibly to the sediment flow regime around the island.
• Illegal sand mining is a major problem facing Gan Island. During the field visit, small scale commercial mining was observed around the island. The islanders reported that these activities were common and that the authorities were unable to monitor these activities due limitations in regulation enforcement. Continued sand mining especially from the oceanward coastline will facilitate severe erosion if the net production of settlement exceeds the rate of natural erosion and sand mining.
• Breakwaters have been constructed to mitigate coastal erosion in Mukurimagu and Thundi settlement. The structures in Mukurimagu settlements are constructed using sand-cement bags and coral pieces. It was constructed to protect an historical site which is now within a few meters of the shoreline.
• The harbour constructed in Maandhoo Island comprise of 4m deep dredge area and a breakwater perpendicular to the shoreline, extending to the eastern reef edge. Since the reclamation of land between Gan and Maandhoo Island, the coastline of the two islands is now merged. The presence of the Maandhoo harbour has essentially blocked all sediment flows along region leading excess sedimentation on the southern side and severe erosion on the northern side. The excess growth of seagrass can also be partly attributed to the reduction in current flows due to harbour construction.
49 Abondoned Harbour Area naturally works joined in the past Opened up during the tsunami illegal sand mining 10214000°N Thundi Dredged harbour and coastal protection
Reclaimed area Large scale sand mining Mathimaradhoo from existing land close to Solid structures beach areas prevent sediment Gan 10212000°N transport Indutrial Zone Piled Jetty
Sand mining from beach Reclamation of Abondoned Harbour wetland areas works Coastal vegetation 10210000°N cleared partly by human Sand Pier activities and during tsunami Mukurimagu
coastal protection to prevent erosion Reclaimed area Two islands joined COASTAL AND TERRESTRIAL Habour and MODIFICATIONS perpendicular breakwater to shoreline - Bocks sediment flow
Maandhoo Fisheries 0 500 1,000 Complex metres 3000°E3380 3 000°E336 Kadhoo 0°E 4000 3 Airport Figure 3.8 Coastal modifications around Gan
Terrestrial Modifications
• The overall terrestrial environment of the island is in relatively good condition, but that of settlement areas has been considerably modified.
50 • The vegetation within the settlement areas are limited to large ‘shade’ and backyard fruit trees.
• Much of the coastal vegetation on the island is intact, but that of settlement areas have been highly modified. Coastal vegetation cover is highest in the Thundi and Mathimaradhoo, and lowest in Mukurimagu settlement with just a few meters of vegetation. While Mathimaradhoo settlement has a strong coastal vegetation system, it has been considerably modified due to clearing, road development and sand mining.
• Land reclamation of wetland areas without considering the elevations and impacts on drainage systems has caused such areas to flood during heavy rainfall. This is most prominent in Thundi and Mukurimagu Settlements.
• Large scale sand mining, using heavy machinery, has been undertaken close to the oceanward shoreline near the Mathimaradhoo Settlement. Much these mining activities were undertaken during the road development activities for the main road and the new industrial zone. Most of these areas are within 10-30m if the coastline and involved complete clearing of vegetation. In some parts near Mathimaradhoo, the width of the vegetation is barely 5m and has the potential to be breached in the event of abnormal erosion. Since the mined areas are almost to the low tide level, such a breach may cause permanent land loss.
3.3 Environmental mitigation against historical hazard events.
3.3.1 Natural Adaptation
Gan Island has signs natural adaptation to varying climatic conditions in the past. The adjustment of ridges, coastal processes and drainage patterns are evident from initial assessments and require further empirical assessments to understand the adaptation processes. The defensive mechanisms established for the storms, especially in the southern half of oceanward shoreline, are critical for minimising impacts of sea induced hazards as well. It has to be noted that the extent of natural defence systems in Gan are small compared to other large islands in Maldives such as Seenu Hithadhoo, Haa Dhaalu Kulhudhuffushi or Fuvahmulah. It may be due to the lack of major natural hazards, especially ocean induced natural hazards, on the island.
51
3.3.1 Human Adaptation
Gan has very few modifications undertaken to directly prevent natural hazards. The main activities include construction of breakwaters to prevent erosion in the historical site near Mathimaradhoo settlement and breakwaters to protect the harbour in Thundi settlement. More adaptation activities have been undertaken on land to prevent rainfall related flooding including rising of roads and houses to prevent flooding. The lack of both natural and human adaptation measures could be taken as crude indicator of the historical exposure of the island to natural hazards, specifically climate related hazards.
3.4 Environmental vulnerabilities to natural hazards
3.4.1 Natural Vulnerabilities
• The low elevation of the island is the main natural vulnerability of Gan. Similar to the overall low topography of the island, the ridges around the island are low compared to the other large islands assessed under this project. This can cause surges and waves 1.5m above MSL to overtop the ridges and flood.
• The island is located in a high tsunami impact zone due to the ocean floor topography off the Laamu atoll eastern rim. (Shifaz, 2004).
• It’s location on the eastern rim generally exposes the island to flooding events and tsunamis (UNDP, 2005)
• As is the case with large islands studied in this project, Gan has substantial variations in topography. These variations have exposed all three settlements to occasional rainfall induced flooding. Amongst the three settlements, Mukurimagu and Thundi reported higher frequency of rainfall flooding. Profiles taken across the Mukurimagu settlement confirms the presence of major low area with less than 0.5m from the water table. Similar cases were observed during the assessment in Thundi settlement.
• The low areas with in the island, especially on the western side of the island appear to facilitate inundation during flooding events most likely due to the downward gradient. During the 2004 tsunami it was reported that flood waters reached 800m inland. Analysis of topographic variation against flood extent
52 revealed the substantial effects of a low area on the flood extents. These patterns were observed along two main east-west oriented roads of the island and around Mathimaradhoo Settlement which lies adjacent to a wetland area. Hence, the topographic variation may play a major role in exposing Gan to both rainfall induced and ocean induced flooding events.
• The North-south orientation of the island along with low elevation exposes the majority of the island’s eastern shoreline to ocean induced flooding hazards.
• The narrow width of Gan around Mukurimagu settlement coupled with low elevation and low ridges exposes the settlement to severe flooding hazards. The impact of tsunami was most heavily felt on Mukurimagu.
• Reef width appears to play an important role increasing or decreasing the impacts of ocean induced wave activity. The proximity of Gan Island coastline to reef edge may increase the exposure of the island to certain sea induced Hazards. Implications of the existing distance needs to be studied further to establish a concrete relationship.
3.4.2 Human induced vulnerabilities
• The main human impacts on natural hazards have largely been a result of present land use.
o In the island of Mathimaradhoo and Mukurimagu the settlement is located close to the eastern coastline. Coastal vegetation clearing has been undertaken to accommodate this expansion. In Mathimaradhoo, the extent of clearing has been controlled, but the coastal vegetation of much of Mukurimagu is all but removed. It was reported that a large part of coastal vegetation was affected during the tsunami. However, it is more likely that the root cause of coastal vegetation loss was due to human clearing, especially the undergrowth.
o Large areas of the densely vegetated land have been cleared for agriculture and forestry. Such clearings close to the ocean ward coastline facilitates to increase extent of flooding. A large proportion of the agricultural crops located close to the eastern coastline were lost during the tsunami.
53 • Large areas of land has been cleared and mined for sand close to the eastern coastline. Sand mining was carried out to obtain material for road development during the establishment of industrial zone development. Since then the main roads have also been maintained using similar sand mining methods. The areas mined are left as large holes, in some areas with barely 5m of coastal vegetation separating the mined area and high tide line. These mining activities seriously undermine the coastal defensive system of the mined area. If the narrow stretch of vegetation separating the sea is breached by wave action, it is highly likely that the mined areas would get flooded during high tide. It was also reported by the islanders that the mined areas acted as a drainage zone to mitigate the impact of the tsunami. While it is likely that the localised effect of tsunami might have been slightly reduced, the long term effects of such mining activities could be disastrous in terms of exposure to natural hazards.
• Sand mining has also been undertaken in the beach areas for construction activities of Gan. There have been reports that sand mining from beach areas are undertaken at a commercial scale especially following the numerous construction activities on the island. Sand mining from inhabited island beaches have been banned, and continued mining would lead to serious implications in the future especially in terms of coastal erosion.
• Coastal modification activities around the island have caused localised exposure to natural hazards. In the island of Mukurimagu, coastal structures have been built in response to coastal erosion around a historical site. On the eastern side of the same settlement harbour works were started and abandoned leaving a dredged area and improperly disposed dredge material. Such activities are often associated with undesired changes to the surrounding coastal environment, most often coastal erosion.
• Past continuous road maintenance activities on the island to mitigate rainfall flooding has caused the road to be raised higher than the surrounding housing plots. As a result, in some area of the island, especially in Mukurimagu settlement, the houses have become the drainage area for the roads causing flooding in unlevelled houses during heavy rainfall.
3.5 Environmental assets to hazard mitigation
54 • Gan is the largest island in Maldives and its size is considered the biggest asset against ocean induced flooding events. However, the advantage of size is available primarily to the Thundi settlement as it is located on the western coastline along widest area of the island.
• Generally strong coastal vegetation exists right around the island, especially on the ocean ward side. The only exception in the Mukurimagu ward where the coastal vegetation strip is narrow and degraded.
• The vegetation within the island is amongst the densest that can be found in any inhabited island. Much of the island is therefore protected from the effects of strong winds. However, there is a noticeable lack of larger trees in the island.
• The coastal processes along the western coastline of the island appear to be functioning well without much human intervention. The blockage of water and sediment flow between Gan and Maandhoo due to land reclamation may have had considerable effect on the southern half of Gan but is expected to have minimal impact on the rest of the coastline.
• Although a number of modifications have been undertaken on he western side of the island, the fact that the coastal processes on the eastern side of the island operates independently allows the natural adaptive functions of the island to natural hazards remain intact.
• There is a well established drainage system, dominated by wetland areas in the east and west restricting the impact of rainfall related flooding to these areas.
3.6 Predicted environmental impacts from natural hazards
The natural environment of Gan and islands Maldives archipelago in general are appear to be resilient to most natural hazards. The impacts on island environments 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 impacts from major events such as tsunamis. The recovery of the Gan Island environment, especially vegetation, ground water and geomorphologic features following the tsunami is a good example 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
55 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 Gan may continue to naturally adapt to rising sea level, especially with a relatively unmodified eastern coastline. There are two scenarios for geological impacts on Gan. First, if the sea level continues to rise as projected and the coral reef system keep up with the rising sea level 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 Gan (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 Gan stands to undergo substantial change (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)).
Gan 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).
56 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 Gan.
Hazard Scenario Probability Potential Major Environmental Impacts at Location Tsunami (maximum scenario) 4.5m Low • Widespread 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 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 in the industrial zone such as diesel and chemicals are leaked. • Salinisation of ground water lens to a considerable period of time causing ground water shortage. If the rainwater collection facilities are destroyed, potable water shortage would be critical in Mathimaradhoo and Mukurimagu Settlement. • Widespread damage to crops (short-term) • Widespread damage to coastal protection infrastructure • 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 • Moderate –high damage to coastal protection infrastructure • Minor geomorphologic changes in the northern oceanward shoreline and lagoon 1.32m (2.30m Very Low • Moderate damage to coastal vegetation storm tide) • Long term or permanent damage to selected inland vegetation especially common backyard species such as mango and breadfruit trees. • Salt water intrusion into wetland areas and island water lens causing loss of some flora and fauna.
57 Hazard Scenario Probability Potential Major Environmental Impacts at Location • Contamination of ground water if the sewerage system is damaged or if liquid contaminants in the industrial zone such as diesel and chemicals are leaked. • Salinisation of ground water lens to a considerable period of time causing ground water shortage. If the rainwater collection facilities are destroyed, potable water shortage would be critical in Mathimaradhoo and Mukurimagu Settlement. • Loss of crops • Moderate –high damage to coastal protection infrastructure • Minor-moderate geomorphologic changes in the oceanward shoreline and lagoon • 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. • Minor damage to open field crops 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. • Widespread damage to open field crops • Minor changes to coastal ridges Heavy rainfall 187mm Moderate • Minor to moderate flooding in low areas, including roads and houses. 240mm Very Low • Widespread flooding but restricted to low areas of the island. 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
58 Hazard Scenario Probability Potential Major Environmental Impacts at Location areas
3.7 Findings and Recommendations for safe island development plan
The following findings and recommendations are based on the assessment of existing Physical development plan of November 2005.
• In spite of its size Gan doesn’t have high natural ridges to protect it from major flooding events. Perhaps ridges are absent due to the lack of major storm events around Gan. In order to mitigate the effects of future flooding around settlement areas alternate solutions may be required. It is recommended that the present solution of 2.4m high revetment be reconsidered on Gan Island as hard engineered structures along a functioning coastline would have implications for the future. Instead, alternative means of raising the existing ridges must be sought based on appropriate site specific studies.
• Due to the low elevation and topographic variations on the island, consideration needs to be given for potential flood zone in future settlement planning. At present the low areas have mainly been allocated to agriculture or are preserved as wetland areas. However, the planned expansion of Mathimaradhoo settlement towards the north may expose them to flood hazards. If the settlement expansion is unavoidable, appropriate flood prevention measures may be required to protect the settlement.
• The proposed width of EPZ zone may need to be reviewed due the lack of a natural high ridge system on the island.
• Much of the new housing being developed in Thundi for tsunami resettlement appears to be located in a low area. The reclamations done under the project appears to have done little to raise the area. There are implications for rainfall induced flooding as this specific site has been reported to be a flooding zone during heavy rainfall. Further analysis needs to be undertaken to assess the topography of the area and mitigate potential flooding hazards.
• Due to its large size, the island relies on a functioning drainage system to reduce rainfall induced flooding around the island. The proposed development activities on the island should carefully consider the implications on the drainage system
59 and mitigate potential hazards using artificial systems, wherever development is unavoidable.
3.8 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 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 topographic variations 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.
60 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.
o Detailed GIS basemaps for the assessment islands.
o Coastal change, flood risk and climate change risk modeling 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
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.
Ministry of Planning and National Development (MPND) (2005). Infrastructure Development for Poverty Alleviation, Volume II - L.Gan. Male', Maldives, Ministry of Planning and National Development.
61 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.
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.
62 4. Structural vulnerability and impacts
Located on the eastern rim of the atoll, L. Gan is predominantly exposed to ocean-originated flooding, tsunami and wave/surge floods. Both hazard extents can cover more than half of the island.
4.1 House vulnerability
More than 50 houses are identified as vulnerable on L. Gan Island, which accounts for about 15% of the total number of existing houses on the island. Houses with extremely poorly physical conditions account for less than 10% and are mainly located in the Mukurimagu division of the island.
4.1.1 Vulnerability type
The house vulnerability of L. Gan is dominantly attributed to the weak physical structure and poor protection of the houses. As shown in Fig. 4.1 , 64% of the vulnerable houses are weak in their physical structure and around 36% poor in their protection. In contrast, few houses are found their plinths lower than the adjacent road surface.
4.1.2 Vulnerable houses
Accordingly, the vulnerable houses identified can be divided into three major groups: weak buildings, weak buildings with poor protection, and normal buildings with poor protection. As shown in Fig. 4.2 , more than 50% of the vulnerable houses are weak in their physical structure only; 40% poor in protection only; and about 10% are weak in both their physical structure and boundary wall.
63 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 Type of house vulnerability.
L. Gan WB
WBPP 0%0% WBLE
36% WBPPLE
53% PP LE 0%2% 9% PPLE
Fig. 4.2 Distribution of vulnerable houses.
4.2 Houses at risk
Houses on L. Gan Island are highly exposed to ocean-originated flooding. As shown in Fig. 4.3, more than 340 existing houses are exposed to tsunami flooding, accounting for 90% of the total existing houses; 200 houses are located in the swell wave/surge flood-prone area, making up to 55% of the total houses.
In the tsunami flood-prone area, 44 houses are weak in their physical structure and/or with poor protection, accounting for 13% of the exposed houses. Given a tsunami wave setup of 4 m, around 5 houses may be subjected to a serious
64 damage, which accounts for 1.5% of the exposed houses. The total house damage may result in 2.5% of the population displaced.
In the wave/surge flood-prone area, 32 vulnerable houses are identified, accounting for 8.5% of the total existing houses on the island. Given an inundation depth of 0.5-1.0 m, around 6% of the exposed houses within the wave/surge flood-prone area may be subjected to a slight damage and little population displacement will be expected.
According to the most recently updated land use plan, the establishment of new settlements on the western coast may change the exposure of houses to tsunami inundation significantly. As shown in Fig. 4.3, around 104 proposed plots will be exposed to a tsunami inundation of less than 0.5 m water depth, but no potential damage to these new houses can be expected.
No new plots are exposed to wave/surge floods.
Table 4.1 Houses at risk on L.Gan Island. Exposed Vulnerable Potential Damage Hazard houses houses Serious Moderate Slight Content type # % # % # % # % # % # % TS 336 89.4 44 13.1 5 1.5 29 8.6 10 3.0 292 87.0 W/S 206 54.8 32 8.5 0 0 0 0 13 6.3 174 84.5 RF ------Flood Earthquake 376 100 34 9.0 Wind 376 100 34 9.0 ------Erosion
65 4.3 Critical facilities at risk
Most critical facilities of L. Gan Island, including island offices, schools, mosques, hospital, power houses, communication sites, and waste sites, are exposed to ocean-originated floods (Fig. 4.4). Located in an inundation area of more that 1.5 m water depth, Murikumagu school and a proposed waste water plant are vulnerable and may be subjected to a moderate to serious damage. During the tsunami of 2004, part of the school building, together with its boundary wall, were completely destroyed. In particular, there is still no boundary wall around the school buildings. It may be even affected by a medium intense normal wave/surge flood. Therefore, the Murikumagu school is at high risk.
The exposure of critical facilities to wave/surge floods is extensive, too. Fig. 4.4 (right) shows that 1 power houses, 2 island office, 3 schools, 2 mosques, 2 communication sites, 2 proposed waste water plans, and 2 proposed waste sites are exposed. However, the physical damage caused by a 0.5-1.0 m inundation can be minor. Most critical facilities within this hazard zone are variably content- affected only, depending on their plinth level.
All most facilities of L. Gan are not vulnerable to earthquake, exposed to a PGA of 0.05.
4.4 Functioning impacts
Tsunami floods can lead to an extensive disruption of the functioning of most critical facilities. The degree of disruption varies, depending on the nature of the facilities. As shown in Table 4.3, the functioning of most exposed facilities may be disrupted for a few days. Disruption of school activities and power supply in the Mukurimagu, however, may last a few weeks to a few months, given current social-economic conditions. By the time when the survey of this project was conducted, the Murikumagu school is still left without a boundary wall, directly
66 exposed to ocean-originated floods; the power house has not be recovered to its full capacity.
In contrast, the functioning impact of wave/surge flooding is relatively limited, including a few day disruption of schooling and degradation of sanitation.
Table 4.2 Critical facilities at risk on L. Gan Island. Critical facilities Potential damage/loss Hazard type Monetary Exposed Vulnerable Physical damage value 3 power houses, 1 1 school, Most are content- hospital, 3 island 1 affected. The school office, 4 schools, 3 proposed is subjected to mosques, 3 waste seriously damaged; Tsunami communication sites, water waste water plant a 2 proposed waste plant slight damage water plans, 2 proposed waste sites 1 power houses, 2 None Content-affected Flood island office, 3 schools, 2 mosques, Wave/Surge 2 communication sites, 2 proposed waste water plans, 2 proposed waste sites Rainfall - - - - Earthquake All facilities No No No Wind - - - - Erosion - - - - Note: “-“ means “not applicable”.
67 Table 4.3 Potential functioning disruption matrix Flood Function Earthquake Wind Tsunami Wave/surge Rain fall Administration 1) A few days
Healthcare A day
Housing 2%
Education A few weeks to A few days months (Mukurimagu school) Religion A few days A day
Sanitation 3) localized & months localized & secondary contamination months secondary contamination Water supply
Power supply A few weeks (PH12.4)
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 L. Gan:
• Avoid locating key critical facilities (i.e. waste water plants and disposal sites) in the intense flood-prone area. • Enhance building codes in the ocean-originated flood-prone area on the eastern coast. For example, strong boundary walls (concrete foundation, structured, and strong material) with a proper height are
68 highly recommended. Alternatively, strong buildings with normal boundary wall can also be considered. • Setting up an EPZ with a proper ridge height along the eastern coast can reduce the hazard extent significantly. A buffer zone with a width of 50 m, proposed in the land use plan, seems reasonable. However, no proper EPZ is considered for the coastal segment of the Mukurimagu division yet. In fact, among all the coastal segments of L. Gan, the Mukurimagu coast badly needs an EPZ with a high standard. If the relocation of houses and critical facilities is not possible, a small-scale oceanward land reclamation along the Mukurimagu coast should be considered, which is supposed to be one of the cost-effective options. • Retrofit the power house and school in the Mukurimagu division, if ocean-originated flooding along the Mukurimagu coast is not mitigated.
69
Fig. 4.4 Houses at risk associated with tsunami floods (left) and swell wave/surge floods (right).
70
Fig. 4.5 Critical facilities at risk associated with tsunami floods (left) and swell wave/surge floods (right).
71