Systematic karst pool erosion within the conglomerate platforms ofMoorea, French Polynesia

Christine A. WaljeskiDepartment of Earth and Planetary Science

University of California, Berkeley, CA, 94720waljeski@eps.berkeley.edu

ABSTRACT. Karst morphology appears as a prominent erosional feature on carbonate platforms fronting the Motu ofMoorea, French Polynesia. Conglomerate platforms on Motu Temae Motu Tiahura, and Motu Ahi show varying levelsof karstic influence. The horizontal surface topography of these platforms and the presence of hydrologic and lithologiclayering present an environment in which karst pools may develop. Interaction between the platform with both freshwater from rain and seawater from the adjacent lagoon leads to the formation of pool and basin structures which erodevertically as well as horizontally through the platform and in some cases out to the lagoon. Two distinct evolutionarysystems are recognized within these pools with stages of development defined for each system. Throughout thedevelopment of these pools, defining characteristics including surface textures, influx systems, drainage systems andphysical morphology change and evolve. Differences in both small and large-scale structural elements over the area ofthe platform dictate where and how pools develop. The formation and evolution of these karst pools does in some caseseffect the morphology of the platform on a larger scale. In addition to acting as an erosional mechanism for theplatform, karst pools can also act as depositional environments for new sediment and material.

1. IntroductionKarst erosion is an influential process in

tropical regions with typical rates of erosioncalculated at 45kg/km2/year (Drew, 1985). Withsuch high karst erosion rates noted in the tropics,one expects exposed limestone in these areas toexhibit well known karst morphologies. Theconglomerate platforms which occur along thebarrier reef faces of the Motu of Moorea, FrenchPolynesia provide an environment prone to rapidlithification and solution of the calciumcarbonate material they are made of. Karst typemorphology is visible on these conglomerateplatforms including karst pools similar to karsticbasins documented in the Bahamas (Bourrouilh-Je Jan, 1998). The conglomerate platformspresent on the five motu of Moorea, FrenchPolynesia have been suggested to be deposited asa result of large storms and cyclones. (Murphy,1992) As a storm or detrital deposit thesestructures contain clasts ranging in size from finesand to large coral boulders which creates aunique environment for erosional anddepositional processes.

This study examines the surface karstfeatures of the conglomerate platforms on theMotu of Moorea to whether there is a systematicevolution to the erosion of these platforms.Further investigation is then made into the natureof the features present and the dominantenvironmental processes that influence the karst

erosion. Bourrouilh-Je Jan (1998) describes thekarstification of shallow water carbonateplatforms in the Bahamas and the Tuamotu,finding that lithification and solution occursimultaneously in these environments with bothfreshwater and sea water adding to thedevelopment of these karst systems. Theplatforms involved in this study are partiallyemergent and partially submerged. Cementationof subaerial and submarine carbonates differaccording to Matthews (1969). This study alsoexamines whether structural differences betweensubaerial and submarine cemented environmentsinfluence erosional evolution. Alternatively anull hypothesis is that karst features on theconglomerate platforms evolve in the samemanner and by the same physical and chemicalprocesses at all locations along the platformsregardless of structure on a large or small scale.

2. Methods

2.1 Study SitesAll field work for this study was

conducted along the conglomerate platforms onthe barrier reef sides of Motu Temae, MotuTiahura and Motu Ahi all of which are locatedon Moorea, French Polynesia (17°30’ S,149°50’W). Field experiments were performedprimarily at Point Aroa (17°28.402’S, 149°46.419’W) on the northwestern end of Motu

Temae with observations for comparison made atMotu Tiahura and Motu Ahi. All experimentsand observations were carried out duringOctober and November 2003.

a.

b.Figure 1 . a. Map of French Polynesia, b. Mapof Moorea depicting field sites. * depicts primaryarea for field experiments.

The study area at Aroa Point representsonly a small section of the large conglomerateplatform present along the front of motu Temae.This field site includes the area 100m west of thelight at point Aroa all the way to the eastern mostend of the platform.

MotuTiahura(17°29.179’S,149°46.419’W) is located on the northeastern corner ofMoorea. The conglomerate platform runs along

the length of the motu on the barrier reef sideand ranges in width from about 3 to 25m.

Motu Ahi is located on the westerncoast of Moorea just offshore from the village ofAfareaitu. The conglomerate platform runs alongthe western side of the motu and wraps aroundfollowing the trend of the barrier reef. Theplatform width at Motu Ahi ranges from lessthan 1m to 40 m.

2.2 Field WorkThe conglomerate platform at Temae

was observed at times corresponding to differenttidal situations including morning low tide(6am), midday high tide (12pm), and eveninglow tide (6pm). This observation was performedto grasp the type of water movement and influxthat affects different parts of the conglomerateplatform. In an attempt to simulate storm likeconditions and the influence of large stormevents on the platform, observations were madeduring a period of large swell (October 9, 2003)and during two periods of rain.

In surveying the larger part of thewestern end of the platform the Western most200 meters of the platform were chosen for usein field experiments due to their intensekarstification. All pools were recorded in termsof distance from the lagoon platform interface,relative size ,the composition and texture of thewall and bottom cement, the composition andtexture of the surrounding cement, whether theplatform location containing the pool was beingundercut by a current, connections to other poolsor the lagoon, observable intake and drainagesystems, and the presence or absence of loosematerial.

Upon analyzing all pool data collectedthe pools were separated into subcategoricalsystems based on common characteristics.Each pool system was then further separated intodevelopmental stages beginning with the un-pooled surface cement and ending pool thatcould signify the end of that system. Stages weredefined by their various cement textures, thedrainage and influx systems visible and anyunique morphologies that appeared characteristicof that particular stage.

In order to determine the effect of tideson different systems and stages, stages weretested for drainage and influx at both high andlow tide. To test drainage, flouresein dye wasdiluted in seawater and then added to water inthe pool. The path of the dye functioned to trackthe path of water entering and exiting the pool.

Moorea

*MotuTemae (PointAroa)

Motu Tiahura

Motu Ahi

In order to better understand thedistribution of different stage pools along theplatform, a count of pool stages was made overthe whole study site at Point Aroa. Starting at thefurthest island structure west of the Point Aroalight, the length of the platform was walked andpools of each stage in each system were tallied.Sampling

Hand samples were taken from keyplatform layers and cement types using a rockhammer and chisel. These samples were thenobserved with a hand lens and in some casesexamined in thin section with a petrographicmicroscope.

3. Results

3.1 IntroductionThe fieldwork performed at the

Point Aroa field site resulted in thecharacterization of two distinct karst poolsystems. These systems are referred to as thenear lagoon system and the inland system namedprimarily for the dominant location of each. Byobserving each pool system, characteristic stagescould then be defined to describe thedevelopment of each pool system. Thefollowing systems were described fromobservation made at Point Aroa. Karstic featuresat Motu Ahi and Motu Tiahura were alsoobserved to compare features with those presentat Point Aroa.

3.2 Near Lagoon SystemThe near-lagoon system is defined as

occurring near the edges of the conglomerateplatform in contact with the lagoon. This area isaffected continually by rising and falling tides,wave action, and longshore current. Interactionbetween the platform edge and the lagoon alsocreate a layering system between the vadose andphreatic zones (figure 2) which is characteristicof this pool system. The near lagoon systemcontains 7 distinct developmental stages.

3.2.1 GeologyThe area of the field site dominated by

the near lagoon system is largely horizontal withfew large boulders cemented on top. Towards thewestern end, the platform breaks into island typestructures. The island structures as well as thearea connected to the main platform is layeredwith what appears as an erosional gap just belowthe high tide line (figure 2). This gap separatesthe vadose and phreatic layers. The upper vadoselayer contains poorly sorted clasts of coral heads

and other calcareous material. Closest to theedge of the platform it is difficult to distinguishbetween clasts due to the extreme pitting. In thephreatic layer, algae and bioerosion preventclasts from being distinguishable.

Figure 2. Boundary between vadose andphreatic layers. Field notebook (20cm height)shown for scale. Porosities in the vadose layerare 25-60% while those in the phreatic layer are10-30%. (Nunn, 1994)

Figure 3. northwestern edge of platform fromPoint Aroa. Flagstone surface texture on top ofplatform. Island structures in the distance arecharacteristic of near-lagoon system.

3.2.2 PetrologyIn thin section the samples taken from

the near lagoon setting show a distinctionbetween material in the vadose layer and

Vadose layer

Phreatic layerErosional gap

material in the phreatic layer. Sample TA01represents the uppermost cemented layer in thenear lagoon system. In thin section individualcalcite clasts are surrounded by finer grainedcement and large pores. There is no evidence ofsecondary crystallization. Sample TA05represents the area of the platform just below thevadose-phreatic boundary. This sample is takenfrom just underneath the erosional gap thatoccurs at this boundary. In thin section one cansee evidence of secondary crystallization withfiner grained crystals of calcite rimming theconglomerate fragments.

3.2.3 Near- lagoon Stages

Stage 1Stage 1 in the near lagoon system is

defined by the texture and relief of theconglomerate platform surface. Pooldevelopment has not yet started at this point butthe surface cement is characteristic. The surfaceof the platform is rough and pitted with highrelief on a small scale and algae covering a largepercentage of the surface.Individual pits appearon the mm-cm scale

Figure 4 . Stage 1 of near-lagoon system. Algalcolonies can be seen encrusting some of thesurface.

Stage 2At stage 2 (figure 5a) of the near lagoon

system, pool development becomes visible. Poolshape at stage 2 is still dictated by thesurrounding clast morphology. Pools of stage 2are low points in naturally occurring topography.Pools wall are the same in texture as the stage 1starting cement. Surfaces are still highly pittedand visibly porous. At this stage influx sourcewaters are rain or wave action. Drainage of the

pool during this stage can be observed to occurvertically through the pitted porous cement of thepool walls and bottom. When such pools aresituated above undercut ledges drainage is seento occur straight through the upper layer and intothe lagoon.

Stage 3Stage 3 (figure 5b) pools in the near-

lagoon system show a contrast between bottomcement texture and texture of the cement ofsurrounding material. Pools are shallow butbottom cement is much smoother with lesspitting and less visible porosity. The influxsystem for this stage is splash from wave actionas well as water from rain. Drainage of stage 3pools is vertical through the bottom cementalthough extremely shallow stage 3 pools may beflushed out by successive waves or mayexperience evaporation.

Stage 4Pools at stage 4 (figure 5c) show

physical erosion through the top vadose layer ofthe platform. The bottom of the pool has at thispoint begun to break through to the erosional gapthat occurs between the vadose and phreaticlayers of the near lagoon platform environment.Breakthrough to this erosional gap beteen layersoffers a new pathway for both influx anddrainage. Influx is derived from wave splash,rainwater, and now during high tide water mayenter along the boundary between the vadose andphreatic layers. Drainage also takes advantage ofthis new pathway. Water may also drain viaporous flow through the pool walls.

Stage 5At stage 5 (figure 5d), pools have

similar physical morphology as that of stage 4.Pool wall texture is similar to that of stage 4 butthe pool bottom fully reveals the boundary tothe phreatic layer. Influx is derived from rain andwave action but flow along the erosional gap atthe phreatic boundary has increased and the poolmay be filled or drained along the boundarylayer during tidal fluctuations. At Stage 5 thedrainage pathway along the boundary layer mayerode enough to create a window like structureon the lagoon side of the pool at this layer.

a.

b.

c.

d.

e.

f.

Phreatic Zone

Vadose Zone

Sea Level

Vadose Zone

Vadose Zone

Vadose Zone

Vadose Zone Vadose Zone

Phreatic Zone

Phreatic Zone

Phreatic Zone

Phreatic Zone

Sea Level

Sea Level

Sea Level

Sea Level

Erosional gap atboundary

Windowstructure

Influx andoverflowchannel

Figure 5. Erosional evolution of near lagoon system including stages 2-7 shown in cross-section.a. stage 2 pool, b.stage 3 pool, c. stage 4 pool. Erosional gap at boundary between vadose andphreatic zones is shown. d. stage 5 with window structure at boundary layer. e. stage6 showingwindow structure and dominant inflow and overflow channel. F. stage 8 pool.

Phreatic Zone

a.

b.

c.

d.

e.

f.

Figure 6. Photographic documentation of stages2-7 of the near lagoon system at Point Aroa. a.Stage 2 b. stage 3 c.stage 4 d.stage 5 e.stage 6 f. stage7

Windowstructure

Influx/overflow channel

Windowstructure

Stage 6Pools at stage 6 (figure 5e) often have

developed window structures at the vadosephreatic boundary layer. In addition a dominantinflux channel may be developed showing thepreferred direction for influx and drainage duringperiods when the pool overflows. Influx duringstage 6 comes from rain, wave action or waterflowing along the erosional gap or windowstructure during high tide. During low tidedominant drainage is along the vadose-phreaticboundary while at high tide when the erosionalgap may be submerged, drainage may be bothalong the boundary as well as through the poolwalls. During storms or large swells, drainageand influx tends to occur along the newly formeddominant influx channel located near the poolsurface.

Stage 7Stage 7 (figure 5f) is the final stage of

the near-lagoon pool. At this stage the pool wallclosest to the lagoon has been completely erodedaway. The pool is now open to the lagoon. Thebottom of the pool may or may not have erodeddown below the low tide level, meaning it maybe left dry at low tide. Influx and drainage duringthis stage are highly dependent on tidalfluctuations as there is no longer a barrierbetween the pool and the lagoon.

3.2.4 Pool AbundancesWithin the Point Aroa field site at motu

Temae a total of 253 pools belonging to the nearlagoon system were counted. The distributionbetween the various stages is shown in table 1.Stage 1 is not included in this table as itrepresents the starting surface cement of thelagoon system.

Stage# Number ofpool

Percent oftotal pools

2 102 40.33 101 39.94 33 13.05 6 2.46 5 2.07 6 2.4Table 1. Abundances of near lagoon stages 2-7in the Point Aroa field site.

3.3Inland SystemThe inland pool system dominates areas

of the platform that are removed from theplatform lagoon interface. This area is notaffected by tidal influences on a daily scale.Ocean water influx occurs only from high tidescombined with storm conditions such as largewaves and swells which are sufficient to deliverlagoon water onto the platform. This system isbroken down into 5 characteristic stages.

3.3.1 Inland-system GeologyThe inland system, system B occurs in

areas of the platform where the conglomerateappears highly cemented at the upper mostlayers. The uppermost 20cm of the platformsurface is often made up of a flagstone or shinglestructures. This uppermost layer appears wellcemented both with the naked eye and in handsample.Individual clasts are visible but theyoccur with very little relief. Below the flagstonecap layer the conglomerate material becomesmore easily recognizable as individual clasts.These clasts are poorly sorted and while towardsthe lagoon the large clasts are well-cemented, onthe motu side of the platform this lower layer ispoorly consolidated and barely cemented.

3.3.2 Inland Stages

Stage 1Stage 1 of the inland surface refers to

the initial surface texture and composition of theplatform within the areas dominated by thissystem. The surface ranges from extremely flatand smooth with little pitting, to areas that arecomposed of a layer of large flagstone likestructures that may demonstrate some pitting.

Stage 2The first appearance of pooling in the

inland system occurs as topographic lows in thesurface cement with channels from the flagstonejoints leading to a central depression. Influxduring this stage is due to rain and waves largeenough to reach far back on the platform.Drainage of this stage occurs vertically throughthe pool bottom cement or along the jointchannels that may lead to topographically lowerpools.

a.

b.

c.

d.

e.

Figure 7. Photo-documentation of stages 1-5 ofthe inland system at Point Aroa. a. stage 1showing flagstone texture. b. stage 2 c.stage 3showing joint channels d. stage 4 e. stage 5

Joint channels

Central pool

Stage 3 At stage 3 the central pool begins to

deepen and associated joint channels leading intothe central pool begin to widen. Influx duringthis stage is due to large waves during high tides,large swells or storms and rainwater. Drainageduring this stage is along joint channels when thepool is overfilled either by heavy rains or largewaves but when the level reaches below theoverflow point, the dominant drainage is verticalthrough the bottom cement. Evaporation is notedin some pools of this stage.

Stage 4At stage 4 in the inland system, pools

have eroded vertically through the flagstone caplayer and the pool bottom begins to reveal a newlithologic layer. This new bottom layer is madeup of distinguishable clasts in finer cement.Influx for this stage can come from rainwater orlarge waves and swells. In the case of waterentering from the lagoon, the developed jointchannels between nearby flagstone piecesprovide directed pathways. Drainage occursvertically through the new bottom texture or insome cases under the flagstone layer.

Stage 5Pools in stage 5 of development exhibit

pool bottoms that have fully eroded through theflagstone cap layer and begun to erode down intothe underlying poorly consolidated layer. Bothpool walls, and bottom in this stage aredominantly made of poorly sortedunconsolidated clasts. Individual clasts of detritalcoral are visible and easily distinguishable.Cementing is often poor and porosity may below enough to be visible to the naked eye. Influxof water comes from rainwater and large wavesor storm swells. Drainage during this stageoccurs through the pool walls and bottom. Poolsin stage 5 often collect small pieces ofunconsolidated material.

3.3.3 AbundancesIn the Point Aroa field site at Motu

Temae a total of 115 pools were found in theinland system. The number of pools found ofeach stage of the inland system is shown in table2. Stage 1 abundance does not appear as itrepresents the starting surface cement and doesnot yet show pool erosion.

Stage # Number ofpools

% of totalpools found

2 54 47.03 37 32.14 19 16.55 5 4.3Table 2. Abundances of inland stages 2-5 in thePoint Aroa field site.

3.4 Motu TiahuraThe conglomerate platform at Motu

Tiahura demonstrates a similar visible layeringto that of the field site at Point Aroa. The surfaceof the platform at Motu Tiahura showed moreoccurrence of the flagstone texture and wasdotted with many large coral boulder cementedon top of the horizontal platform. The lateralends of the platform were broken into islandstructures larger than those at Point Aroa. Manypools were filled in with sand and other detritalmaterial. Late stage near-lagoon system pools arecommon creating a platform edge that is dottedwith island structures and small peninsulas.

3.5 Motu AhiThe conglomerate platform at Motu Ahi

appears erosionally different than that at PointAroa. Undercutting along the lagoon edge isextreme. There is a small back lagoon that occursbetween the platform and the main motu withonly small areas where connections are subaerial.Karst pooling at Motu Ahi is not common asidefrom small early stage pools on the surface.Other erosional features appear dominant on thisplatform.

4. Discussion

4.1Evolution of near-lagoon systemStage 1 of the near lagoon system

exhibits the most extreme pitting of all thecements seen in this study. This stage isoccurring on the top most layer of the nearlagoon platform environment. The algal layer ontop of this cement may be adding to thekarstification of the surface on a microscopicscale through a phenomenon referred to asphytokarst (Patterson and Sweeting, 1983). Theformation of phytokarst can happen on a veryshort time frame, in as little as four years(Jamesand Choquette, 1984). This is an area where newclasts originating from the lagoon will bedeposited and may undergo simultaneouslithification and solution (Bourrouilh-Je Jan,

1998). With the possible rapid formation ofphytokarst and cementation of newly depositedclasts, the texture of stage1 is likely tocontinuously erode and lithify in a rough mannerwith no time for any smoothing of cement. Asthis stage is entirely in the vadose layer, anywater infiltrating the surface from above willflow down due to gravity and porosity andtherefore it will act to constantly dissolvecalcium carbonate from the top most layer andcarry it farther down in the platform to bereprecipitated or carried into the lagoon waters(Nunn, 1994).

Stage 2 pools in the near lagoon systemare the product of lower topography. If enoughwater is deposited on top of the platform therewill be a flow toward the low points. Thebottoms of these stage 2 pools still reflects themorphology of the clasts that form it but as thepool is inundated by both fresh water and seawater the dissolution and reprecipitation of thepool bottom will act to gradually form asmoother, less porous bottom cement that ischaracteristic of stage 3 pools. This planed downbottom cement may be due to the shallow natureof these early stage pools. Early stage pools aresubject to influx, flushing and evaporation, all ofwhich may cause rapid dissolution andprecipitation. As stage 3 pools erode verticallydue to the vadose dissolution process describedby Nunn (1994) it is likely that mosttopographically low pools will receive somedrainage from neighboring topographicallyhigher pools and cause them to combine intolarger later stage pools. This may explain whythe abundance of earlier pool stages such asstage 2 and 3 is higher than that of the laterstages 4,5,6, and 7. (Table 1)

As pools evolve between stage 3 andstage 4 the volume increases so they are capableof holding greater amounts of water. Thisincreased volume allows them to erode verticallyand horizontally as the water moves from thepool into the porous surrounding cement. Oncethe pool bottom of stage 3 pools reaches thevadose-phreatic boundary the subsequent stage 4pool will take advantage of the differences inporosity between these layers (Nunn, 1994) anddrain along the erosional gap. This erosional gapprovides a new pathway for both influx anddrainage. A hypothesis is that this pathwaybecomes the favored drainage route for pools ofthis stage which then accounts for the increasederosion along this boundary and the windowstructures often associated with stage 5 pools.This newly formed window structure in stage 5

pools means that both influx and drainage arelikely to occur at this point during tidal changes.As the platform edge continues to erodevertically, a channel is eroded above the windowstructure leading to the morphology of a stage 6pool. As wave action physically erodes the poolwall above the window structure, rainwater willlikely continue to erode the overlying channeluntil the two erosional features meet and the poolis open to the lagoon in its final stage 7 form. Ifthe stage 7 pools occur on a peninsula likestructure or an area of platform surrounded ontwo sides by water then there is the possibilitythat erosion of the back pool wall will result inthe formation of island structures like those seenin figure 8.

4.2 Evolution of the inland systemThe inland system of karst begins with

flagstone surface cement defined as stage 1. Thissurface on a large scale is very flat allowing rainand lagoon water that infiltrate the system todeposit in extensive shallow pools. Waterdirected into these pools by the joint channels ofthe flagstones helps to define the stage 2 pool ofthis system. The early stages of the inlandsystem appear to have surface cements ofrelatively low porosity. This is observed both inthe slow drainage of these pools as well as inhand sample. The presence of this well cementedcap layer may be due to possible higherevaporation rates in these shallow regions withlarge surface area. If these areas are inundatedwith freshwater which will immediately act todissolve the surface layer, and then are subject toevaporation that will oversaturate the watercausing precipitation, then a smoother lessporous surface cement would be expected. Thismay also explain why there is very little relief ofindividual clasts visible in the early stages of thissystem. Beneath this cap layer is a lessconsolidated poorly cemented layer. This porousand poorly consolidated conglomerate layerallows stage 4 and 5 pools to erode vertically assurface water flows with gravity andsubsequently erodes this bottom layer. Thehighly porous nature of this layer may be due toinflux of fresh ground water coming from themotu. This can be thought of as the phreaticfreshwater lens described by James andChoquette (1984). The presence of freshgroundwater flowing out from the motu may alsohelp to explain the erosion of the platform on themotu side. The late stage inland system pools atPoint Aroa very commonly contain rubble-sizedsediment suggesting simultaneous deposition and

erosion making it difficult to create a wellconsolidated lithology.

Motu TiahuraAt the conglomerate platform of Motu

Tiahura most of the surface cement appearedsimilar to that of the inland system at the PointAroa field site. This may be due to the highinflux of small sand sized sediment into thedeveloping pools. Studies in the Bahamasshowed subaerial carbonate sediments to cementrapidly under tropic conditions. (Davis, 1996)Large coral boulders cemented to the top of theplatform are more common at Motu Tiahura thanat Point Aroa suggesting that influx of materialfrom large storms is common here.

Motu AhiThe conglomerate platform at Motu Ahi

shows far less karst pooling than at the otherplatforms in this study. Undercutting on thelagoon edge of the platform is extremesuggesting that dominant erosion along theplatform edge is from the bottom up perhaps inpart due to the longshore current that runs alongthe barrier reef side of Motu Ahi. Thisundercutting may be responsible for collapse ofplatform edges before late stage pools are able todevelop.

Figure 8. Island structures at northwestern tip ofMotu Temae. Stage 7 pool present in middleisland

ConclusionsStructural and lithologic variations

within a conglomerate platform give rise todistinct systems of development in the formationof karst pools. Stages within systems may havesimilar traits but overall evolution of a stagewithin a particular system differs dependingupon the physical structure of the platform at thatlocation as well as the dominant intake anddrainage systems that can be associated with thestage. The karst pools occurring at Point Aroa onMotu Temae show a systematic evolution thatstarts with surface cement and develops into alarge scale erosional feature. Karst pools in thisarea can be categorized into stages belonging totwo distinct evolutionary systems defined as anear-lagoon system with 7 distinct stages and aninland system with 5 stages. While karst erosionon Motu Tiahura and Motu Ahi share somesimilar features with that of Motu Temae, thedominant erosional mechanism at Motu Ahiappears to be undercutting and collapse of theplatform edge. Karst pooling is a prominentfeature at Motu Tiahura but the development ofthe pools looks to be a combination of the twosystems found at Point Aroa. The majority of thesurface cement at Motu Tiahura is similar to thatin the inland system of Point Aroa yet the near-lagoon layering system is still a prominentinfluence in pool development on Motu Tiahura.

The layering of the platform into thevadose and phreatic zones with an erosional gapin between acts as a major influential structure tothe development of both systems at Point Aroaand Motu Tiahura.

Future research may be interested toinvestigate the chemistry of the karst erosion ofthe conglomerate platform and how changes inrainfall and ocean surface salinity effect thedevelopment of pools at different times of theyear. It is also possible that the other Motu ofMoorea may offer environments that presentfurther systems of development of these karstpools.

AcknowledgementsI would like to thank the professors and

graduate student instructors of IB 158/ ESPM107 2003 for interacting with me in a scientificway I have never experienced before at UCBerkeley. I would like to thank the Gump stationstaff for making my stay and experience inMoorea unforgettable. I would like to thank LetiParnell, Vicky Bertics and Curtis Pehl formotivating me in my fieldwork and finally Iwould like to thank the whole Department of

Earth and Planetary Science for supporting me intaking this opportunity to expand my horizons inthe scientific world. I would also like to thankJohn Williams for his late night company andtechnical support.

ReferencesBourrouilh-Je Jan, F.G., (1998) The role of

high-energy events (hurricanes and/ortsunamis) in the sedimentation, diagenesis andkarst initiation of tropical shallow watercarbonate platforms and atolls. SedimentaryGeology 118:3-36

Davis, J.J. (1996). Rapidity of freshwatercalcite cementation- implications for carbonatediagenesis and sequence stratigraphy.Sedimentary Geology 107 :1-10

Drew, D.(1985). Karst Processes and Landforms.

Macmillan Education Ltd, London, 63pp.James, N.P., Choquette, P.W., (1984). The

Meteoritic Diagenetic Environment.Geoscience Canada 11: no 4

Murphy, F., (1992). The Geomorphology andEvolution of the Motu of Moorea, FrenchPolynesia. University of California, Berkeley,101pp.

Nunn, P.D. (1994). Oceanic Islands. BlackwellPublishers, Oxford 413pp

Paterson, K., Sweeting, M.M. 1986 NewDirections In Karst. Geo Books, 613pp

Steers, J.A., and Stoddart, D.R. (1977). TheOrigin of Fringing Reefs, Barrier Reefs andAtolls. Biology and Geology of Coral Reefs 4:21-105

Viles, H.A., (2001) Scale issues in weatheringstudies. Geomorphology 41:63-72

.