Influence of El Niño-Southern Oscillation (ENSO) Events on the Coastline of Central California

Please download to get full document.

View again

All materials on our website are shared by users. If you have any questions about copyright issues, please report us to resolve them. We are always happy to assist you.
 0
 
  Influence of El Niño-Southern Oscillation (ENSO) Events on the Coastline of Central California
Share
Transcript
  ABSTRACTSignificant sea-cliff erosion and storm dam-age occurred along the central coast of Cali-fornia during the 1982–1983 and 1997–1998El Niño winters. This generated interest amongscientists and land-use planners in how his-toric El Niño–Southern Oscillation (ENSO)winters have affected the coastal climate of central California. A relative ENSO intensityindex based on oceanographic and meteoro-logic data defines the timing and magnitude of ENSO events over the past century. The indexsuggests that five higher intensity (relative val-ues 4–6) and 17 lower intensity (relative values1–3) ENSO events took place between1910and 1995. The ENSO intensity index corre-lates with fluctuations in the time series of cy-clone activity,precipitation,detrended sealevel,wave height,sea-surface temperature,and sea-level barometric pressure. Wave height,sea level,and precipitation,which are the pri-mary external forcing parameters in sea-cliff erosion,increase nonlinearly with increasingrelative ENSO event intensity. The number of storms that caused coastal erosion or stormdamage and the historic occurrence of large-scale sea-cliff erosion along the central coastalso increase nonlinearly with increasing rela-tive event intensity. These correlations and thefrequency distribution of relative ENSO eventintensities indicate that moderate- to high-in-tensity ENSO events cause the most sea-cliff erosion and shoreline recession over the courseof a century. INTRODUCTION California’s shoreline is characterized by coastalmountains,sea cliffs,and small pocket beaches.Sea-cliff erosion along the California coast is per-manent and irreversible because eroded bluff ma-terial is lost to the littoral system during landwardmigration of the sea cliff. In spite of the fact thatthere are hundreds of kilometers of rocky,cliffedcoastline along central California and increasingpressure to develop the shoreline (Griggs,1995),there are few studies focused on understandingthe processes of sea-cliff erosion and the evolu-tion of the coastline. This may be because it isdifficult to evaluate the numerous variables thatcontribute to sea-cliff erosion (Shih and Komar,1994). Sunamura (1992) identified two main cat-egories:the factors inherent to the cliff materialthat resist erosion and the external forces that actto erode the sea cliff. The inherent properties in-clude lithology,structure,intact rock strength,permeability,and porosity,as well as cliff height;attributes that tend to be relatively constant overshort to intermediate time scales (Sunamura,1992). The external forces include biologic andseismic activity,which can reduce the resistiveforces of sea cliffs (Kuhn and Shepard,1983;Griggs and Savoy,1985; Plant and Griggs,1990);however,fluctuations in the oceanographic andatmospheric climate occur more frequently andare therefore more important to the evolution of sea cliffs over short to intermediate time scales(Griggs and Johnson,1979; Kuhn and Shepard,1983; Sunamura,1992).Most coastal scientists who have studied theerosion of rocky coasts conclude that the major-ity of sea-cliff erosion occurs during infrequent,energetic storm events (Griggs and Johnson,1979;Emery and Kuhn,1980; Kuhn and Shepard,1983;Sunamura,1992; Shih and Komar,1994). Two of the most recent storm seasons,during which sub-stantial sea-cliff erosion occurred along centralCalifornia,were the 1982–1983 and 1997–1998winters (Griggs and Johnson,1983; Storlazzi andGriggs,1998; USGS/UCSC/NASA/NOAA Col-laborative Research Group,1998); both of theseasons coincided with severe El Niño–SouthernOscillation (ENSO) climatic anomalies. Eventhough the impact of ENSO events on productiv-ity in Peruvian coastal waters has been docu-mented for more than four centuries,there hasnot been a thorough investigation of how thecoastal climate of central California has histori-cally been affected by these large-scale climaticfluctuations.The goals of this project are to:(1)create a rel-ative intensity ENSO event time series applicableto geomorphic studies along the coastline of cen-tral California for the time period from 1910 to1995; (2)identify the influence of ENSO eventson the external forcing parameters that cause sea-cliff erosion; (3)investigate the impact of ENSOevents on coastal cliff erosion and storm damagealong the shoreline of central California; (4)de-fine what relationships exist between the magni-tude of variation in the forcing parameters,thenumber of damaging coastal storms or occur-rences of sea-cliff erosion,and the relative inten-sity of ENSO events; and (5)define what roleENSO events may play on the evolution of thecentral coast of California over short to interme-diate time scales. OVERVIEW OF ENSO EVENTS ENSO events represent one of the two extremestates of the quasiperiodic fluctuation of the large-scale atmospheric circulation systems across thePacific and Indian Oceans known as the SouthernOscillation. During non-ENSO times when theDarwin,Australia-Tahiti atmospheric pressureanomaly or Southern Oscillation is positive,a re-gion of high atmospheric pressure dominates theeastern equatorial Pacific while the western equa-torial Pacific is characterized by a region of lowatmospheric pressure. This pressure differencedrives the strong easterly equatorial and south-easterly trade winds commonly observed in thelow latitudes of the Pacific. These winds,blowingoffshore along the west coast of the Americas,cause deep,nutrient-rich,cold water to upwell inthe eastern Pacific and push the warm equatorialsurface waters westward. This warm water be-comes superelevated in the western Pacific,cre-ating an overall west to east downward slope of the sea surface and therefore a pressure gradientacross the equatorial Pacific (Wyrtki,1975).236 Influence of El Niño–Southern Oscillation (ENSO) events on the evolutionof central California’s shoreline Curt D. Storlazzi*  Department of Earth Sciences and Institute of Marine Sciences,University of California, Gary B. Griggs Santa Cruz,California 95064-1077  GSA Bulletin ; February 2000; v. 112; no. 2; p. 236–249; 9 figures; 2 tables. } *E-mail:manta@earthsci.ucsc.edu.  With the onset of an ENSO event,an approxi-mate reversal occurs in the Southern Oscillationand in the large-scale atmospheric circulationpatterns along the equatorial Pacific. This rever-sal is marked by the decay of the prevailing east-erly winds and the concurrent development of aseries of prolonged westerly wind events over thewarm pool in the western Pacific. These windsperturb the upper ocean and excite the eastwardpropagation of large-scale waves in the thermo-cline that start to suppress the upwelling in theeastern Pacific (Deser and Wallace,1987; Websterand Palmer,1997). This reversal in wind directionalso allows the potential energy of the sloping seasurface to be released,further inducing the warmwaters to propagate eastward along the equator.This disturbance,similar to an equatorially trappedinternal Kelvin wave,superelevates the local seasurface at the equator and moves across the Pa-cific as a wave-like bulge in sea level (Wyrtki,1975). The Coriolis force not only confines thisbulge to low latitudes but also retards its dissipa-tion by expansion into higher latitudes (Komar,1986). When the superelevated warm water bulgepropagating eastward along the equator collideswith South America,it splits into two portions thatadvance north and south to higher latitudes ascoastally trapped internal Kelvin waves (Enfieldand Allen,1980). These two propagating shelf waves are pinned to the coast by the inclinationof the shelf and slope,increasing sea level alongthe coastline while retarding dissipation to theopposing eastern boundary currents of the Pacificbasin (Wyrtki,1975; Komar,1986). This eastwardshift of the warm water bulge moves the center of organized cumulonimbus development,which isthe principal mechanism for exchanging heat be-tween the ocean’s surface and the atmosphere,into the eastern Pacific. This causes atmosphericcirculation,which is sensitive to shifts in orga-nized cumulonimbus convection,to be perturbed(Climate Diagnostics Center,1997).Although the major vertical convection anom-alies are confined to low latitudes,the effects onthe circulation of mass and energy in the atmo-sphere extend to middle and high latitudes. Dur-ing a typical non-ENSO winter,a region of highpressure is centered over the Gulf of Alaska andAleutian Islands while southeastern Alaska andwestern Canada are dominated by a region of lowpressure,driving the northwesterly winds andwaves common along the west coast of NorthAmerica (Griggs and Johnson,1983; Dingler etal.,1985) as shown in Figure1A. In strong ENSOwinters,however,the region of high pressure ad-vances eastward into north-central Canada and isreplaced by a region of anomalously low pressure(Seymour et al.,1984; Climate Diagnostics Center,1997). This causes the jet stream to intensify andsplit. One section tracks more eastward than dur-ing non-ENSO time,while the other branchswings south over the Hawaiian Islands beforepropagating northeast across California (Fig.1B).This diversion of the jet stream causes the trajec-tories of cyclones and interanticyclonic systems(fronts) to be redirected. These cyclonic and anti-cyclonic motions control the daily weather fluc-tuations at higher latitudes,and their diversion al-ters the srcin,frequency,and strength of stormsacross the northern Pacific and western NorthAmerica (Seymour et al.,1984; Climate Diag-nostics Center,1997). STUDY AREA This study focuses on the coastline of centralCalifornia from Bodega Bay (~70km north of San Francisco) in the north to Point Conception(~95km south of San Luis Obispo) in the south.This 790km section of shoreline is sparselypopulated,except in the vicinity of San FranciscoBay and Monterey Bay. The coast is dominatedby the Coastal Ranges,which are composed of Pa-leozoic metamorphic rocks,Mesozoic igneousrocks,and Cenozoic sedimentary rocks. Thesemountains are structurally controlled by thenorthwest-trending tectonics of the San Andreasfault system and are drained by a number of small,steep perennial streams and a few larger rivers,which are the primary sources of coarse-grainedsediment to the littoral environment (Best andGriggs,1991). The mouths of many of thesestreams were inundated during the Holocenetransgression,forming low-gradient flood plains,coastal lagoons,and marshes in their lowerreaches,many of which are backed by dune fields(Griggs and Savoy,1985; Dingler etal.,1985).The coastline of central California is charac-terized by steep,as much as 100m high,activelyeroding coastal bluffs often incised into upliftedmarine terraces and commonly fronted by low,wave-cut shore platforms,or very small pocketbeaches. These sea cliffs are interrupted at irreg-ular intervals by larger pocket beaches that format the mouths of coastal streams and by infre-quent continuous beaches in sheltered bays. Sea-cliff erosion,with long-term rates ranging from0 to >30cm/yr,is episodic and locally variable(Griggs and Savoy,1985). This erosion typicallyoccurs during the infrequent combination of high tides and extreme storm waves (Griggs andJohnson,1979).The offshore wave climate can be characterizedby three dominant modes:the Northern Hemi-sphere swell,the Southern Hemisphere swell,andlocal wind-driven seas. The Northern Hemisphereswell is typically generated by cyclones in thenorth Pacific off the Aleutian Islands during thewinter months (November–March) and can attaindeep-water wave heights exceeding 8m (Na-tional Marine Consultants,1970). The SouthernHemisphere swell is generated by storms off New Zealand,Indonesia,or Central and SouthAmerica during summer months and,althoughthey generally produce smaller waves than theNorthern Hemisphere swell,they often have verylong periods (20+ s). The local swells typicallydevelop rapidly when low-pressure systems tracknear central California in the winter months orwhen strong sea breezes are generated during thespring and summer (Griggs and Johnson,1979;Dingler et al.,1985). Storms with deep-waterwave heights in excess of 5m occur five times ayear on average (National Marine Consultants,1970; Dingler et al.,1985). TEMPORALLY VARIABLE CLIFF EROSION FACTORSWaves Wave energy,which is proportional to thesquare of wave height,is commonly regarded as adominant physical process leading to coastal ero-sion and sea-cliff retreat along rocky coastlines(Sunamura,1992; Shih and Komar,1994; Griggsand Trenhaile,1995). Hydraulic action,includingcompressional,shear,and tensional forces,is ex-erted on sea cliffs during wave impact (Barnes,1956; Sunamura,1977). When sediment or debrisare available,waves can also exert mechanical ac-tion through abrasion and impact (Sunamura,1992). Together,hydraulic and mechanical forcingmay quarry the sea cliff by prying apart jointedrocks (Baker,1958; Emery and Kuhn,1980).Large waves also facilitate coastal bluff ero-sion by removing protective beach sediment andallowing the waves to directly attack the cliff toe.This is done by increasing sediment suspension,set-up,and offshore flow as wave heights andperiods increase (Holman and Sallenger,1985).Increased set-up also elevates beach water-tablelevels,further facilitating beach erosion,as dis-cussed in the next section. Griggs and Johnson(1983),Seymour et al. (1984),Seymour (1998),and Storlazzi and Griggs (1998) discussed therole that large waves may have played on thecoastal erosion that occurred along Californiaduring the 1982–1983 and 1997–1998 intenseENSO events. Studies by Griggs and Johnson(1979) and Dingler et al. (1985) documented therole of wave action in coastal cliff erosion alongcentral California over longer time periods. Sea-Surface Elevation Higher than normal sea-surface elevationsplayed a major role in the damage and erosionthat occurred during the 1982–1983 and 1997–1998 ENSO events (Griggs and Johnson,1983; EL NIÑO–SOUTHERN OSCILLATION,CENTRAL CALIFORNIA SHORELINE Geological Society of America Bulletin,February 2000237  Flick and Cayan,1984; Komar,1986; Flick,1998;Storlazzi and Griggs,1998). Abnormally high sealevels cause flooding in low-lying areas,elevatethe level of wave attack relative to the cliff toe,andreduce the amount of wave energy lost to bottomfriction during shoaling by increasing the relativewater depth (Carter and Guy,1988; Sunamura,1992; Griggs and Trenhaile,1995). Higher thannormal sea levels also tend to elevate beach watertables,raising pore pressures and thus increasingsediment mobility,enhancing the beach’s suscep-tibility to both subaerial and subaqueous erosion(Bryant,1983; Clarke and Eliot,1988; Mossaetal.,1992). This reduces the effectiveness of thebeach as a buffer,and therefore makes the seacliffs more vulnerable to direct wave attack. Precipitation and Ground Water Although most of the terrestrial sediment sup-plied to the coastline of central California is de-livered by rivers and streams during large dis-charge events,high precipitation generally tendsto enhance coastal erosion along cliffed shore-lines. Precipitation and runoff tend to elevate thelocal sea surface in lagoons and estuaries whileeroding beaches backed by lagoons or sloughs asthe swollen coastal streams breach their barrierspits. The large volume of sediment and debrissupplied to the surf zone by the steep streams thatdrain the Coast Ranges may accelerate sea-cliff erosion through abrasion and impact forces(Griggs and Johnson,1983; USGS/UCSC/NASA/ NOAA Collaborative Research Group,1998). Epi-sodes of heavy precipitation also tend to raise theground-water levels of coastal bluffs,increasingtheir loading and pore-fluid pressures. Increasedpiezometric pressures along joint surfaces reducethe frictional resistance and effective normalstresses in the bluff material; in conjunction withthe increased weight of the bluff due to satura-tion,this may initiate slope failure (Turner,1981;Griggs and Johnson,1983; Kuhn and Shepard,1983). In addition,ground water can promote theweathering and solution of cementing material,altering the cohesive and frictional properties of the material,thus reducing the strength of the seacliff (Griggs and Johnson,1979; Turner,1981). DATA ANALYSIS AND RESULTSHistorical Record of ENSO Events During the past 20yr,our understanding of thedriving mechanisms behind and precursors of ENSO events has made significant progress. Byexamining the occurrence of diverse biologic,at-mospheric,terrestrial,and oceanographic phe-nomena from South America,Quinn et al. (1987)developed a history of ENSO events and their rel-ative intensities back to the1500s. More recently,higher resolution indices using more definitivebut shorter records have been developed. Deserand Wallace (1987) and the National ClimateData Center (1997) compiled records of sea-sur-face temperature anomalies from offshore PuertoChicama,Peru,and southern California,respec-tively. A Southern Oscillation index derived fromsea-level barometric pressure anomalies at Dar-win,Australia,and Tahiti has been generated bythe Pacific ENSO Applications Center (1997),and Wolter and Timlin (1997) devised a multi-variate index derived from sea-surface tempera-ture,wind stress,barometric pressure,and outgo-ing long-wave radiation anomalies. Due to thedifferent parameters,locations,and methods uti-lized by the various researchers,there are someminor discrepancies in the occurrence,timing,and magnitudes of ENSO events during the pastnine decades (Fig.3).The comprehensive data set of Quinn et al.(1987) incorporated a number of parameters(e.g.,storms,flooding,sea-level changes),andbecause they rated the intensity of each ENSOevent in terms of the occurrence or fluctuation inthese parameters,their classification was utilizedas the foundation of our ENSO intensity index.Because Quinn et al. (1987) did not evaluate therelative intensity during non-ENSO and La Niñatimes,the ENSO intensities during these periodswere determined by evaluating the fluctuationsin the standardized Deser and Wallace (1987),National Climate Data Center (1997),PacificENSO Applications Center (1997),and Wolterand Timlin (1997) data sets relative to the Quinnet al. (1987) series.Owing to the higher resolution and precisionof the newer data sets,the relative intensities sug-gested by Quinn et al. (1987) were modified for anumber of ENSO events. The standardized timeseries were scaled to the maximum relative Quinnetal. (1987) index value,and the geometric meanof these rescaled indices and the Quinn et al. val-ues was computed to develop our modified rela-tive ENSO intensity index. Of note is the fact thatthe National Climate Data Center (1997) data setwas weighed only half as much as the other re-scaled data owing to the record’s acquisition from STORLAZZI AND GRIGGS 238Geological Society of America Bulletin,February 2000 120 W180150 W150 E90 W ooooo 90 N o     2    0      N      o 2     0     N       o      5     5      N      o 5     5      N       o LH LH = High pressure= Low pressure= Storm track HHL 120 W180150 W150 E90 W ooooo 90 N o     2    0      N      o 2     0     N       o      5     5      N      o 5     5      N       o Figure 1. Location of high and low barometric pressure regions in the northern Pacific andNorth America along with general storm tracks during (A)a typical non-ENSO or La Niña win-ter,and (B)an intense ENSO winter. Modified after Seymour et al. (1984) and the Climate Diag-nostics Center (1997). AB  offshore southern California,where it could affectour correlation with central California coastalphenomena by imposing local bias.These modifications resulted in a relative ENSOintensity index series that includes 9 single-yearand 12 multiyear ENSO events during the periodfrom 1910 to 1995 (Fig.3). In terms of relativeENSO intensity,the time series developed in-cludes 5 higher intensity (intensity values of 4–6)and 17 lower intensity (intensity values of 1–3)ENSO events. During the early part of the twen-tieth century to the 1940s,an intense ENSO eventoccurred on average once every decade-and-a-half. The period from the 1940s through 1970was marked by a relatively more benign climatewith no higher intensity ENSO events and onlyfour events greater than a relative magnitude of 2occurred; this time span corresponded with a pe-riod of intense development along much of theCalifornia coast (Kuhn and Shepard,1983; Griggs,1995). The past three decades have seen the re-turn to more frequent higher intensity events,similar to the early part of the twentieth century.On average,a high-intensity event transpires onceevery 17yr; however,excluding the 31yr intervalbetween 1941 and 1972,during which no higherintensity ENSO events occurred,this average israised to once every 12.3yr. Lower intensityENSO events transpire once every 2.6yr. For ourrevised intensity index,the normal return intervalfor ENSO events of all intensities is ~2.1yr. Wave Height and Oceanographic Data Seymour et al. (1984) utilized hindcast infor-mation derived from pressure field data for lat35ºN to compile evidence of the correlation be-tween large wave events along California and theQuinn et al. (1987) ENSO time series from Perufor the period between 1900 and 1984. This hind-cast series was reevaluated against our revisedENSO index,and the correlation between largewaves and ENSO events was found to be statisti-cally significant (Table1). Although this correla-tion provides evidence to support the influence of ENSO events on the central coast,we supple-mented the Seymour et al. (1984) data with deep-water wave measurements recorded between1980 and 1995 by three National Data Buoy Cen-ter (1997) buoys off central California and oneCoastal Data Information Program (1997) buoylocated off the Farallon Islands (Fig2).All four of the buoys recorded maximum waveheights during the1982–1983 ENSO event greaterthan 7.6m; these heights exceeded one standarddeviation from the mean maximum yearly waveheights for all of the buoys during the 16yr pe-riod of observation (Table2). Buoys 46012 and23 recorded 8.7m and 7.6m waves,respectively,that exceeded two standard deviations from themean. Buoys 46011 and 23 also recorded waveheights (>6.7m) exceeding one standard de-viation for the 1986–1987 ENSO event whilethe waves observed at buoys 46011 and 46013(>7.8m) exceeded one standard deviation for the1980 and1990–1994 events,respectively. Overall,the more recent offshore wave measurements forthe central coast appear to support the Seymouret al. (1984) and Seymour (1998) conclusion thatENSO events tend to be marked by the presenceof large,damaging waves. Sea-Surface Elevation Data Flick and Cayan (1984) provided a compre-hensive review of the sea-level fluctuations forSan Diego from 1926 to 1984; however,they didnot correlate the anomalies with the history of ENSO events. Detrended records of sea-levelfluctuations from two stations in Central Americaand two stations in South America were examinedalong with records from a station 400km south of the study area in San Diego and one station in SanFrancisco (National Ocean Service,1997). TheCentral and South American stations were utilizedto determine if the fluctuations observed in thesea-level records at North American stations werethe result of the ENSO or some other regionalphenomenon and not just local conditions. If pos-itive sea- level fluctuations were observed in theCentral and South American station records con-currently with positive fluctuations at the NorthAmerican stations,then the fluctuations seen atthe North American stations could confidently beconcluded to be of regional srcin.The1911–1914,1939–1941,1957–1958,1965–1966,1972–1973,and the1982–1983 ENSOevents stand out in the records for all six stationsas fluctuations in maximum annual sea level thatexceeded one or two standard deviations from themean maximum sea levels for the total opera-tional records of the stations. The two Californianstations recorded fluctuations in maximum an-nual sea levels that exceeded two standard devia-tions from the mean maximum sea levels for the1911–1914,1939–1941,1957–1958,and the1982–1983 ENSO events (Fig.4). The San Franciscostation recorded significant positive fluctuations EL NIÑO–SOUTHERN OSCILLATION,CENTRAL CALIFORNIA SHORELINE Geological Society of America Bulletin,February 2000239 BUOYPRECIPITATION STATIONTIDAL GAUGE LEGEND CALIFORNIA * + ++ **** SAN FRANCISCOSANTA CRUZWATSONVILLESALINASSAN LUIS OBISPOSAN DIEGOSANTA MARIAHALF MOON BAYBODEGAFARALLON ISLANDS 0100200 km Pacific Ocean  STUDYAREA  N  #46013#46012#46011#23 124 °  W120 °  W42 °  N39 °  N37 °  N35 °  N33 °  N115 °  W117 °  W120 °  W Figure 2. Coastline of California displaying the location of deep-water buoys,precipitationstations,and the two United States tidal gauges utilized in this study.  240Geological Society of America Bulletin,February 2000 Figure 3. The record of anomalies in different indices used to develop our revised relative ENSO intensity index. The agreement between thedifferent indices clearly increases during the latter half of the twentieth century as the number and quality of instruments used to develop the in-dices increased. TABLE 1.CORRELATION BETWEEN ENSO EVENTS AND PARAMETERSParameterENSO events coinciding with Total years/T-statisticCorrelation significant deviations in the ENSO yearssignificance parameter’s records*level(%)(%)Maximum annual significant wave heightBuoysAll intensity (1–6) ENSO events6116 / 90.171>50Higher intensity (4–6) ENSO events10016 / 24.4290.1Hindcasts † All intensity (1–6) ENSO events9172 / 327.2280.1Higher intensity (4–6) ENSO events10072 / 84.5180.1Maximum annual de-trended sea levelAll intensity (1–6) ENSO events5986 / 413.1071Higher intensity (4–6) ENSO events9086 / 138.2210.1Annual accumulated precipitationAll intensity (1–6) ENSO events2686 / 4010.7830.1Higher intensity (4–6) ENSO events7986 / 1314.2100.1Cyclones impacting CaliforniaAll intensity (1–6) ENSO events3750 / 233.2921Higher intensity (4–6) ENSO events7550 / 515.0570.1Cyclones propagating to within 5°of CaliforniaAll intensity (1–6) ENSO events7950 / 232.8261Higher intensity (4–6) ENSO events10050 / 57.7990.1Erosive or damaging stormsAll intensity (1–6) ENSO events7686 / 414.3010.1Higher intensity (4–6) ENSO events10086 / 138.3090.1*Deviations greater than one standard deviation from the mean. † The hindcast wave data is from Seymour et al.(1984).
Related Search
Similar documents
View more
We Need Your Support
Thank you for visiting our website and your interest in our free products and services. We are nonprofit website to share and download documents. To the running of this website, we need your help to support us.

Thanks to everyone for your continued support.

No, Thanks