DRIVERS OF SHORELINE CHANGES IN THE NORTHERN TUAMOTU REEF ISLANDS,

Virginie DUVAT, UMR LIENSs 7266, University of la Rochelle-CNRS, & Bernard SALVAT, USR 3278, EPHE CNRS UPVD, Perpignan-Moorea, France

13 TH INTERNATIONAL CORAL REEF SYMPOSIUM – 19-24 June 2016, Honolulu, Hawai’i 1. STUDY AREA

Ø 111 reef islands distributed among four located between 14°51’S and 15°30’S Ø Contrasng geomorphic (size, shape, structure) and human features (urbanised vs. uninhabited and unexploited islands) Ø Including the islands concentrang human assets, i.e. major infrastructures, main villages and economic acvies 2 176,273 ha, 241 islands Avatoru urban area, Land area: 66 km2 2,567 inhab. Tourism, fishing, 44,885 ha Agriculture 30 km2 540 inhab. Tourism, fishing

Tuherahera village, Atoll

8 densely populated islands concentrang most human assets Main village (Tuherahera) and nearby islets (14 islands)

3 4,735 ha, 13 islands (2 > 300 ha) South-eastern side of Atoll Land area: 17 km2

272 inhab. Tourism and agriculture

11,735 ha, 20 km2 1,256 inhab. South-western side of Atoll Pearl farming

4 Atoll Date Image Scale 2. MATERIAL AND METHODS type 17-18/10/1966 B/W aerial 1: 10,000 20/10/1981 B/W aerial 1:15,000 Rangiroa 27/03/1984 B/W aerial 1:15,000 Ø Aerial imagery (33- to 52-y me period) 26/09/1998 Color aerial 1:15,000

01/10/2013 Pleiades SI

1962 B/W aerial 1:10,000 1. Georeferencing of aerial photographs, using ground 20/10/1981 B/W aerial 1:5,000 control points extracted from the 2013/2014 high-resoluon Tikehau 27/03/1984 B/W aerial 1:15,000 satellite images (RMS error: 0,26-1,72 m) 26/09/1998 Color aerial 1:15,000 22/08/2014 Pleiades SI / 04/02/1976 B/W aerial 1:5,000 04/02/1976 B/W aerial 1:10,000 Mataiva 27/03/1984 B/W aerial 1:15,000 13/11/1994 Color aerial 1:15,000 01/11/2013 Pleiades SI / 08/03/1969 B/W aerial 1:20,000 21/10/1981 B/W aerial 1:25,000 21/10/1981 B/W aerial 1:8,000 Takaroa 22/10/1984 B/W aerial 1:30,000 01/03/1995 Color aerial 1:15,000 01/03/1995 Color aerial 1:4,000 01/10/2013 Pleiades SI / B/W photographs: Topographic Section of the Town Planning Division of the Government of French Polynesia; Pléiades satellite imagery 2013 and 2014 from the French National Centre for Spatial Studies (CNES) 5 2. Digizaon of the “stability line” (SL) + base of the beach (BB)

SL: seaward edge of the vegetaon or of the stabilised beach ridge in non-vegetated areas along natural coastline; seaward edge of human-built structures along modified coastline

Vegetaon line Base of the beach

Stability line

B/W photographs: Topographic Section of the Town Planning Division of the Government of French Polynesia; Pléiades satellite imagery 2013 and 2014 from the French National Centre for Spatial Studies (CNES) 6 3. Total shoreline posion error (image resoluon + georeferencing + shoreline digizaon): <4 m (±4 m = stable)

4. Generaon of shoreline change stascs using DSAS extension (NSM + EPR)

7 Ø Coral reef data collected since the beginning of the 1980s 1986 1. Detailed descripons of the impacts of the 1983 TCs on the Laboute reef outer slopes (% of coral destrucon, processes at work), based on

the comparison of pre- and post-cyclone data -Vivien and

2. Coral reef monitoring data collected since the establishment of Harmelin permanent staons in the early 1990s -Vivien 1985, and Harmelin 1985, Laboute

from

Modified

8 Ø Informaon on the impacts of the 1983 TCs on shoreline posion and island morphology

9 Ø Generaon of data on human pressures (coastal developments, engineered structures, mining, dredging in reef flats, etc.), based on image analysis and fieldwork

Reclamaon Sand mining in Dredging dunes

MATAIVA ATOLL TAKAROA ATOLL Uninhabited south-eastern Capital island part of the atoll

10 3. RESULTS CHANGES IN ATOLL AND REEF ISLAND LAND AREA

- No significant change in the total net land area of the study atolls (-1.12 to +1.66%)

- 78 (77 %) islands exhibited no significant change in area, while 15 islands (15 %) increased and 8 islands (8 %) decreased in size

- The smallest islands recorded the highest rates of change (±14.64% per decade)

- 7 out of the 8 islands that decreased in area are very small (<3 ha); and the islets that disappeared were ny (<0.6 ha)

- All of the islands (16) larger than 50 ha were stable in area, including 5 capital islands out of 6

Reef islands are robust landforms 11 CHANGES IN SHORELINE AND ISLAND POSITION

- Ocean shorelines are more dynamic than shorelines: 33 islands (31 %) showed changes in ocean shoreline posion – 19 advance/14 retreat- vs. 11 Lagoon islands (10%) showed changes in lagoon shoreline posion – 5 advance/6 retreat

- 90 islands (85 %) were stable in posion while 16 islands (15 %) migrated

Lagoonward migraon - Migrang islands: 10 islands <5 ha; 3 islands are 560 ha) are the most stable in posion

- Lagoonward migraon: 5 islands out of 6 are located on the southern and Lagoon south-eastern margins of Tikehau, Mataiva and Takaroa atolls, suggesng that the trade winds contribute to island migraon

- Oceanward migraon: 9 islands out of 10 are located on the south-western side of Takaroa Atoll Oceanward migraon SW Takaroa Atoll Most reef islands are stable in posion 12 DRIVERS OF SHORELINE CHANGE

- Sea-level rise - Storms: TCs + distant source swells - Coral reefs - Human disturbances

Becker et al., 2012

Northern Tuamotu: 2.5 ± 0.5 mm > Majuro, Tarawa, but < Funafu 13

(9/03/83 - 6hUTC) na 0 50km 100km e e a V v (11/04/83 - 18hUTC) 1983 TCs Re

8 - 9m (Reva) 10m STABILITY LINE (Veena) 7- 8m (Veena) a am 15°S Mataiva Tikehau Or Ta TAKAROA 2 TC category 4m (23/02/83 - 18hUTC) (SS scale) (Orama) 2 4m Rangiroa (Veena)

2 Ve li 9m 2 (Orama) 3 5m Stability line (Veli) 3 Ta 3 15°.50'S (01/02/98 Makateate 3 12hUTC)

149W 148°W Position toutes les 6 heures Vitesse de Direction et hauteur de et vitesse des vents moyens 1 déplacement 2 la houle significative 3 4m 60 à 70 noeuds < 10km/h (Veena)

70 à 80 noeuds 10 à 20km/h Principaux villages 4

80 à 90 noeuds 20 à 30km/h < 100 habitants 100 à 700 hab. 90 à 100 noeuds >30km/h > 700 habitants

Impacts of the 1983 TCs Reva (March) and Veena (April) on Mataiva Atoll 14

(9/03/83 - 6hUTC) na 0 50km 100km e e a V v Ahe Re (11/04/83 - 18hUTC)

Impacts of the 1983 TCs Orama (February) and Veena (April) on 8 - 9mRangiroa Atoll (Reva) 10m (Veena) 7- 8m (Veena) a Study area am 15°S Mataiva Tikehau Or 4m (23/02/83 - 18hUTC) (Orama) 4m Rangiroa (Veena)

Ve li 9m (Orama) 5m (Veli) 0 1 km Islands Average changes in the Average changes in the posion of the Stability posion of the Base of the Makateate 15°.50'S Line (m) Beach(01/02/98 (m) 1 Erosion (-10.85) Stable to accreon (3.61)12hUTC) 2 Erosion (-8.25) Stable (1.15) 3 Erosion (-8.06) Stable (2.38) 4 Erosion (-12.49)149W Stable (-0.83) 148°W 7 Erosion (-7.18) Accreon (+9.30) Position8 Stable (-2.42) toutes les 6 heuresAccreon (+8.05) Vitesse de Direction et hauteur de etChanges in shoreline posion on the ocean coast vitesse des vents moyens 1 déplacement 2 la houle significative 3 Marked retreat of the Stability Line: erosional impact of the cyclonic waves 4m 60 à 70 noeuds < 10km/h 15 (Veena)

70 à 80 noeuds 10 à 20km/h Principaux villages 4

80 à 90 noeuds 20 à 30km/h < 100 habitants 100 à 700 hab. 90 à 100 noeuds >30km/h > 700 habitants

(9/03/83 - 6hUTC) na 0 50km 100km e e a V v (11/04/83 - 18hUTC) Ahe Re (9/03/83 - 6hUTC) na 0 50km 100km BASE OF THE BEACH e e a V v (11/04/83 - 18hUTC) Ahe 8 - 9m Re (Reva) Base of the beach 10m 8 - 9m (Veena) 7- 8m (Reva) 10m (Veena) (Veena) 7- 8m (Veena) a a am am 15°S Mataiva 15°S Mataiva Tikehau Or Tikehau Or 4m (23/02/83 - 18hUTC) 4m (23/02/83 - 18hUTC) (Orama) 4m Rangiroa (Orama) (Veena) 4m Rangiroa Ve (Veena) li 9m (Orama) 5m Ve (Veli) li 9m (Orama) 15°.50'S (01/02/98 Makateate 5m 12hUTC) (Veli) 149W 148°W

15°.50'S (01/02/98 Makateate Position toutes les 6 heures Vitesse de Direction et hauteur de 1 2 3 12hUTC) et vitesse des vents moyens déplacement la houle significative 4m 60 à 70 noeuds < 10km/h (Veena)

149W 148°W 70 à 80 noeuds 10 à 20km/h Principaux villages 4

80 à 90 noeuds 20 à 30km/h < 100 habitants Position toutes les 6 heures Vitesse de Direction et hauteur de 100 à 700 hab. 1 2 3 90 à 100 noeuds >30km/h et vitesse des vents moyens déplacement la houle significative > 700 habitants 4m 60 à 70 noeuds < 10km/h (Veena)

70 à 80 noeuds 10 à 20km/h Principaux villages 4

80 à 90 noeuds 20 à 30km/h < 100 habitants 100 à 700 hab. On the ocean side of islands, 90 à 100 noeuds the Base of the Beach mainly exhibited either stability, or >30km/h advance (>30 m in places), indicang that coral reefs (healthy) provide sediments to the > 700 habitants island system 16 1984-1994 (post-cyclone period) shoreline changes 1. The rate of shoreline change remained high over the post-cyclone decade, compared to the values obtained for fair-weather periods that are not consecuve to TCs

2. Over the post-cyclone decade, trends reverse: - The Stability Line advances - The Base of the Beach exhibits either stability, or retreat = reformaon of beach ridges that were washed away by the cyclonic waves (sediment input to the inner part of islands = island upward growth) -Le Jan, 1984 Bourrouilh from

17 Modified CROSS-ANALYSIS OF SHORELINE AND CORAL REEF DATA

1. Storm events make the living coral coverage on the reef outer slopes fall from 50-60 % to 4-5 %, indicang that sediment producon occurs

2. But, the predominance of steep (>45°) outer slopes makes around 75 % of broken corals be lost to islands

Takapoto, March 2016. VD Despite “limited” sediments inputs to islands during TCs, islands were found to be resilient

18 HUMAN INFLUENCE ON CURRENT AND FUTURE ISLAND RESILIENCE

Ø Human acvies currently influence shoreline change on 63 islands represenng 57.5% of the island sample: their influence on shoreline change is not limited to island capitals (6)

Ø Among these 63 “disturbed islands”, 29 are highly-modified and have lost their resilience as a result of human disturbances (enrely fixed shoreline/reclaimed reef flats obstrucng sediment deposion)

19 HUMAN INFLUENCE ON CURRENT AND FUTURE ISLAND RESILIENCE

Ø The fact that most “disturbed islands” were either stable (53%), or expanded (14%) over the study period does not mean that these islands are sll healthy and would be resilient in the future if they were eroded by a TC:

Ø Two arguments:

- Stability in island area is oen due to the compensaon of the retreat of the natural shoreline by an arficial advance of the modified shoreline

- Many islands have experienced accelerated shoreline destabilisaon and beach loss over the last three decades: caused by human disturbances on some islands, but not on others – decrease in sand supply?

20 TIKEHAU ATOLL

Human acvies increasingly influence shoreline change and will probably remain a key driver of island resilience in the next decades. Highly-modified islands (26%) have lost the capacity to respond to TCs by upwards growth, which is concerning in the context of accelerated sea level rise.

hp://lienss.univ-larochelle.fr/storisk WIDESPREAD INFLUENCE OF HUMAN DISTURBANCES ON TAKAROA ATOLL

19 42 “disturbed islands”

23

22 34 WIDESPREAD SAND MINING ON MATAIVA ATOLL… IMPACTING THE SEDIMENT BUDGET OF SOME BEACHES

23 FURTHER STUDIES… + Use remote Develop assessments of TCs sensing impacts on other atolls techniques to ( and Takaroa, with generate similar topographic conclusions) profiles of the reef outer slopes to link sediment inputs to the island system and the reef outer Digize the limit of slopes profile fresh sediment deposits + to determine which DOCUMENT islands and areas are THE IMPACTS concerned by cyclone- OF HUMAN driven upwards growth 24 IMPACTS OF TC ORAMA (1983) ON TAKAPOTO ATOLL ACTIVITIES