VRRIRTIONS IN SEDIMENT RUDCET END MORPNOLOCT RT ORECON INLETa NORTN CSROLINS Richard tk. Stephenson Bast Carolina University James N. McCluskey University of ffisconsin Esu Claire Gambling with the Shore. Proceedings oftire Ninth Annual Conterence oi The Caaiml Socrery. Octoher te- t y,<99m Atlantic City. tyJ. Copyrighr hy The Coartal Society 1995. Many sci<ntists as wvl I as the Na- tional Park Service are on record as bi ing oppas<.d < o < hv Uti'gun In- ler. jvcty project as proposed by the Corps al Engin<.ers. 'I'hi pro- ject would disrupt thi irolagy a the area. Land in thv tap< Hatter- as Rational Seashore has rvcently been approved for construction pur- posess by Congress. B<.caus« t.his proj< ct is vxr.rvmvly contiavvrsial, it seems appropriate to study th<' ramilications surrounding thi pro- ject as much as passiblv, bi lore funding becomes available, The rvsearch here attempts co de- fine thv nature and mapnit.udv af geomorphic change in chv vicinity of Uri gon inlet. Rates of mar- phomi c r ic changv and si ditm nt budg- ets arv pri s<ntvd as bas« I in«dara as thvy arv rv I Jtvil to < lii dynamic stabi I i ty af th< in li t aiid t.hv ad- jacvnc coast.line. Historical Chanivs Uri'gait iiili'c is l!al < at I ii«bJI < i<.'f i sf and systviii si PJr,it ink <indi« I s I and I o t hv nort h rind I'i a I s Iand t.o t hv south. 'I'hi i it i< < I. important 2f13 lor navigat ion sincv it is che only inlv<. at. <h« prvsi.nc time bvtwven <Jpi <<i nry and Hat tvras In 1 i c . Stat» and local of i i cia is view the st.abilizatiun ol th< inlet vssvn- t iii I to <h< economy ol thi area, «specially lor chv local fishing lndusrry at Wanchvse Harbor on Roan- oke Island. Early charts af the scudy a rea show sev< ra I inlets in che vici nity of Uregon lnlvt Figure 1!. Because of thv inaccuracies of these early charts it. is difficulr. to deterraine which mighc b» thv pri.decessor of the pres«nc inlet. The existing inlet opened during chv hurricane of Septi.mber sixth and seventh in IB46, t<n IB43 chart shcws an inlet in t.hi area, but some distance saurh oi <hv presinc location. The rlosing ol koanukv lnl»c, ch» one Hi r W t I < i'r k i I »i yh suppasvd I y usi'd in I '>84, pramp< i d pol i <ic ians of th» p«riod to pram<su <o reopen the j n I i < I t i' I i'«. »d, '!'I« i iilvt was ri lat i vi. I S undi sturb- i d i'sir'I>< I or oct assi< <J I dr»dy iny, u<t< i I Iufr4 whvii Honiivr Br idg» was
Transcript
VRRIRTIONS IN SEDIMENT RUDCET END MORPNOLOCTRT ORECON INLETa NORTN CSROLINS
Richard tk. Stephenson
Bast Carolina University
James N. McCluskey
University of ffisconsin � Esu Claire
Gambling with the Shore. Proceedings of tire Ninth AnnualConterence oi The Caaiml Socrery. Octoher te- t y, <99mAtlantic City. tyJ. Copyrighr hy The Coartal Society 1995.
Many sci<ntists as wvl I as the Na-tional Park Service are on recordas bi ing oppas<.d < o < hv Uti'gun In-ler. jvcty project as proposed bythe Corps al Engin<.ers. 'I'hi pro-ject would disrupt thi irolagy a the area. Land in thv tap< Hatter-as Rational Seashore has rvcentlybeen approved for construction pur-posess by Congress. B<.caus« t.hisproj< ct is vxr.rvmvly contiavvrsial,it seems appropriate to study th<'ramilications surrounding thi pro-ject as much as passiblv, bi lorefunding becomes available,
The rvsearch here attempts co de-fine thv nature and mapnit.udv afgeomorphic change in chv vicinityof Uri gon inlet. Rates of mar-phomi c r ic changv and si ditm nt budg-ets arv pri s< ntvd as bas« I in«daraas thvy arv rv I Jtvil to < lii dynamicstabi I i ty af th< in li t aiid t.hv ad-jacvnc coast.line.
Historical Chan ivsUri'gait iiili'c is l!al < at I ii«bJI < i <.'fi sf and systviii si PJr,it ink <indi« I s�I and I o t hv nort h rind I'i a I s I and t.ot hv south. 'I'hi i it i< < I. important
2f13
lor navigat ion sincv it is che onlyinlv<. at. <h« prvsi.nc time bvtwven<Jpi <<i nry and Hat tvras In 1 i c .Stat» and local of i i cia is view thest.abilizatiun ol th< inlet vssvn-t iii I to <h< economy ol thi area,«specially lor chv local fishinglndusrry at Wanchvse Harbor on Roan-oke Island.
Early charts af the scudy a rea showsev< ra I inlets in che vici nity ofUregon lnlvt Figure 1!. Becauseof thv inaccuracies of these earlycharts it. is difficulr. to deterrainewhich mighc b» thv pri.decessor ofthe pres«nc inlet. The existinginlet opened during chv hurricaneof Septi.mber sixth and seventh inIB46, t<n IB43 chart shcws an inletin t.hi area, but some distancesaurh oi <hv presinc location. Therlosing ol koanukv lnl»c, ch» oneHi r W t I < i'r k i I » i yh suppasvd I y usi'din I '>84, pramp< i d pol i < ic ians ofth» p«riod to pram<su <o reopen thej n I i < I t i' I i'«. »d,
'!'I« i iilvt was ri lat i vi. I S undi sturb-i d i'sir'I>< I or oct assi<<J I dr»dy iny,u<t< i I Iufr4 whvii Honiivr Br idg» was
$5 illiiinl iS!$!6l i !
'l. i, !9$ i$!i!pi!pa
DL< li5$5 Si I6<V I '!LS l!D $6!i!iiifiilRDK<LNW I m$91! 95i i9 ~ 51LDGDIN!if $9
IPL il<fl IISSL i$5! i!
<LOU! NIL<ISSRKN !IN< DRUN!
204
compl<.ted. Whilv it has bven hy-pothesized that. the bridging hascantributvd to inlet migration,narrowing, and shoa ling, it has notbein substantiated. Morc tharlikely, it has been thc deer!.ase instorm fequency and magnifude sincethe hurricanes of the 1930's andthe Ash Wednesday storm of 1962.
Figure 1. Inlet locations of theOuter Banks of North Carolina,Source: Birkemeier, 1981.
f!ata Sources and Anal svsFor this inves<.igation, !.h» shore-line in r.he vicinity of Or»gon ln-lvt is divided into three sections,which are: I! the updrift oceanbeach on southern Bodiv Island, 2!t.hv downdril! a»can beach nn north-ern Pva Island, and �! Or!gon In-let. The Oregon Inlvt svc!ian canbe further Subdivided In! o I br< vcomponents, which ar<.: I! t.hv seashoals or vbb tidal delta, �! thcsound shoals or flno5f I.idal delta,and �! t.hv inlvi gargv or Ihannef,
An inspection of historic maps andchari.s bvr.wevn 1848 and 1982 showsf hat ihl inl!t has bvr.n Inigratingsouthward at. a raf.e of ninety feetpcr yvar I'igurv 2!. Concomitantwit.h c.his southerly migration, isth< change in inlet. throat width Figure 3!. The width of the inletthroat in 1848, 1915, and 1949 rangeed between four tenths and one halfof a mile Figure 4!. The occur-ance of the Ash Wednesday storm re-sult.ed in a widening of the inlet.ta 1.6 miles Figure 3!. ln 1982,calculations showed that the inletthroat. had been Iiarrowing since 1974at a rat» of thrvv hundrf'd and ill iy-six feet pvr y<ar in order to at-tainn its present. widr.h ol aboutthree tent.hs of a milv, Since theopvniny, of I.hc ilil<'r. in 1846, con-siderable change has occurred withrespect. <o the approximate locationof the existing inlvt and r.he pro-posed jet.ties Figurc 6!.
Figurc 2. Inlet in 1848 with pre-scni gorgv ~bown b!twfvn triangles. Af !L r liuke Un i vers i r y vol . Lab. ! .
205
Figure 3. Inlet. in 1915 wi th pre-sent gorge shown between r rtangles. At tcr Duke University Geol. Lab. !.
I'igurc 4. Inlet in 1949 wit h prt-sent. gurge shown between r,ri angle s. At I < r Duke Univcrs it y *'< o I . l.ah. ! .
Figure 5, Inlet. in 1962 wi th pre-sent gorge shown between triangles. Aiter Duke University Geo I . Lab !-
Figurc 6. Inlc in 1974 wtth thc! orat. ion ol the present gorgts theproposed >.t.ty system and thc Bon-ner Hridp. IAtt~ r Uukt Univt rsttyI,'co 1,
20
IO
206
The main channi.l of the inlet hasalso been migrating southwardthrough time and it has also tendedto become shallower since the mid1960's I'igure 7!. Betwe en 1937and 1957, the depch ot r.he mainchannel c:hanged from twenr.y feet cothirty-rwo feet respect.ively. And
Figure 7. In ! er. i:rossect ions.Source: Dolan and Glassen, I 973 ~
when the inlet ai rained its greatestcross-sectional area in 1965, themain channel of the inlec wasthirty-one feet deep. The seeminglygreater depth and cross-sectionalarea recorded in 1957 and in 1965are thought co be related to thelarge number of hut'ricanes that oc-curred in the area from 1954through 1959, and the impact. ot theAsh Wednesday storm ot 1962. Inthe late 1970's, the bridge pilingsnear Pea Is!and were being under-mined by t.he main channel and forc-ed the bridye to be closed tempo-rarily. The channel now must bedredged almost continually co main-rain a depth and location requiredfor navigation purposes.
The period between the Ash Wednes-day storm and the present has sr.enrelatively !ittle storm accivicy.There has bein a rrarked dc.crease inthe frequency and magnicudi uf se-vere northeastcrs and a major hur-ricane has not made landfall with-in i.he loca!icy ot the in!or. dur-ing rhis pc riod. 'I'his quie set nr.pi riod has been marked by a high
accret.ion rate in both the ebb t.i-da! and flood tidal de! tas. Chang-es in the rnorphomer ry of the soundand the sea shoa I s were inadi by re-cording depch measuremencs along aseries of transects, It was toundr hat rhe mean depth of the ebb de I-ta decreased trom �.6 feet in 1961ro 13.7 teer in 1972 Figure 8!,
MEAN DEPTH OF EBB TIDAL DELTA ft!
I 96! !972 1982
Figure 8. Mean depth of ebb tidaldelta, in feet, for 1961, 1972 and1982.
While between 1972 and 1982 themean depth of c.he ebb delta de-creased an additional two feet, to11.7 feet, This rate of sedimencaccumulation represencs an annualincrease of 0.85 cubic yards ofsed imi. ni per square yard of area or805,513 total cubic yards per yearover the area of the ebb c ida I de I-ta Figure 9!, The flood de!ca has
MEAN DEPTH OF FLOOD TIDAL DELTA ft.!
01961 !972 !982
F'igure 9. Mean depth of flood tidaldelta, in feet., for 1961, 1972 and1982.
also been decreasing in depth. Be-tween I'961 and 1972, the meandepth ot the tlood delta decreasedtrom six fret to 4,8 feet, Whilebetween 1972 and 1982, the meandept.h decreased an additonal footto a depth of 3.8 feet. This rateoE sediment accumulation marks anannual inr rease of 0.3 1 cubic yardsof sediment per squat'e yard of areaor 613,098 total cubic yards peryear over the flood tidal delta.The total accretion for both theebb and flood ti.dal delr.as is cal-culated to be 1,418,611 cubic yardsper year.
This accumulation of sediment inr.he inlet can be observed in thetot.al area of the shoals exposed atlow water in bot.h the ebb and Eloodt.idal deltas Figure 10!. Between1961 and 1972, exposed shoals werenot observed in the ebb delta.While in 1982, ten acres of shoalswere exposed. Twenty-four acres ofshoals were exposed at low water onthe tlood delta in 1961. This in-creased to fifty-five acres in 1972,and to seventy-nine acres in 1982.However, it must be realized thatsome of this accumulation could bespoil from dredging activity,
SHOALS EXPOSEDAT LOW WATER acresj
IOO
75
DELTA50
25
0
196I I 972 1982
Figure 10. Acres of shoa!s exposedat low water for 1961, 1972 and1982,
An L xaminat ion was made of the up-dr<It and downdriit ocean beaches
Bodie island and Pea island re-spectively in order to determine ifany s>gniticanr dilterences existin the beach/duse complexes Fig-ure ll!. Hara for the analyseswere derived from two engineeringsurveys dating from 1937 and 1976.A tour mile reach of beach was ex-amined on either side of the inlet.Within each section, twenty pro- i!es werc used. The profiles arelocared approximately one thousand eer apart and originate behind aEoredune ridge baseline and ex-tend to mean high wat.er. Bot.h theupdritt and downdrift beaches werefound to be receding with a risingsea level. The maximum amount ofrecession was found to be twenty-Eivc feet per year, and occurred onthe updritt ser tion nt ur i,'oqu<naBeach. The max imum rate oi rt-cession on the downdri t. beach wastwenty eet per year and occurredin th« proximity ol the NationalWildlife Service Headquartersbui!ding on Pea Island. Addition-ally, an increase in dune height inexcess oE ten Eeet has been observ-ed since 1937. This is the resultof the construction and maintenanceof a barrier dune line by theNational Park Service.
Volumetric changes were calculatedfor the updritt and downdriftbeaches Figure 12!. It was foundthat the updri t beaches were erod-ing at a rate of 114,000 cubic yardsannually, while the downdrift beach-es were accreting at a rate of4,000 cubic yards annually. These igures represent. a net loss of5.39 cubic yards o[ sediment perlinear foot ot beach per year forthe updrift beaches and a net gainof 5.87 cubic yards per linear footper year for the downdrift beaches.These changes in the storage char-acteristics o the beach/dune com-plex repr.-ants a variety ot coast-al proces-... which relate tochanges ~ - aches, capes and in-lets.
207
~ I LE ~ 0tT
I XO
m
Dag
II Dltlvl Kl ~ ill vn ls wlAI l IIIB
e125,000
- I2$,000
208
Figure 11. Shore! ine change in thevicinity of Oregon Inlet. Source:Sharica, 1978.
VOLUMETRIC CHANGES IN THE
BEACH/DUNE COMPLEX yds /yr!
I:i Bur«, Vo! um«t.ric chang< s inth< b~ ach/dune cumplt x, tn cubicyard., pi r yi ar.
Using data b«cw««n 1972 and 1978from the Corps pic.r at Duck, NorthCarolina Figure 13!, the averagemonthly and annual predicted ratesof long»hore transport can be de-termined based on the method recom-mended in the Shore Prorection !fan-ual CBRC, 1973!. Along the north-ern portion of the Outer Banks then«r longshore r.ransporc is to thesouth at a rate of 1,068,004 cubicyards p«r y«ar, The sedicnent thar.is transported onshore is a signif-icant. factor in che present infill-ing at. 0 rc.gon lnlec. But a totalof 114,000 cubtc yards of rhe long-shor« transport. is derived from theupdrift b«aches within four milesof thc inlec. 0F th« rota! 1,418,�6]f cubic yards that is depositedannually <n the ebb and liood tidaldeltas, 1,182,004 cubic yard» or a-bout f!!. ! pc rct nt i» d«riv«d from
500600
400 SDO
400300
X200
300
200o
I 00o
o aoI 00
cu 0
o C:-I DO
o U CJo
-I 00
-200
-200-300
-300Jnn. tele
-400OL I nip c Oe cMnc. npc Mnr uune Bu!e nun Sept
I lgur» I I. l u!a rc I ranspc>rl nl.l!uck, N. .. by mantle I ni I'l77 I a19/a. Sour»»: ill rkc lm i»r, I '!III .
209
longshore cransparc. The rernain-ing 236,607 cubic yards of s»dementIs d»clued froln »ilh»r sound orollshore source.s. An »xacc. deter-mination cannot b» mad» on c.h» a-mount. of sediment that. »scap»s c heinl»r.'s sink annually b»cause c.hedowndrift longshore rransport. rateis nor. known, However, it is esti-rnenced t.har. 124,000 c.ubic yards arebeing deposiced annually on thefour mile section of the downdriftbeach an Pea island. This amountis essentia/ly equivalent to theamount being removed ft'om r.he up-drift. bear.hes on Bodie Island.
ConclusionsMost barri»r island inl»ts are in astate. ol dynamic stability and r.endto close wit.h t ime. An inlet ' s de-mise can be related t.o the movementof sediment and all t.hat c.his im-plies, including the closing ofother inlets that are less favor-ably located. At Oregon Inlet withits migracian shoaling and narrow-ing, wave action is decreasing as isthe tidal prism which cends to les-sen the integrated flux of energy,
This situation does nac allow a sig-nificant amounr. of sand L.a move.For an inlet to remain open for along p»rind of time IL requires themaximization of th» t ida I prism ascompar»<l Lo the litloral drift. Thelarg»r and more irr»gular Lhe driftand che snlaI I»r the tidal flow thegreac»r Ch» possibtlicy of che in-let shoaling and closing, Also, in-let e I I ic i»ncy is equal to the in-verse re l ac ianship between the in-let's cross-sectional area and thesound area. In th» case of Oregoninlet rhe cross-sectional area hasd»crc,as»d since 1962 whil» we as-s uln» L I1» satlncl at»a has rema ined r hesame. ln 1965, whi I » t.he cross-sc c t iona I area was larger than in1937, the 197 7 c ross-sect iona I areawas considerably less than in 1937or 1965. lt is suggested here thatthe 1962 Ash Wednesday storm scouredthe inlet., but since that r. ime, de-rreasing storm magnitude and fre-quency has allowed t.he inlet ta be-com» small»r.
Th« lnlc t ' s morpho loyy, which con-sists ot th« sea and suund shoalsand the inlet gorge, se«mingly haschanged with the cyclic nature ofstorms. The inlt t's morphology alsoresponds to changes in t.he longshoredrift and the tidal prism, Whendrift is minimized and the prism ismaximized there is commonly morethan a single dominant channelthrough the inler.. Oregon Inlet,now has a single channel through thesea shoal which is dredged to main-tain its location and a navigatabledepth.
Any attempt to simplify the complar-ity of the inlet processes by jetty-ing or even dredging, is tantamountto alrering the tilt of the earth' saxis' and expecting nothing to hap-pen. It is strongly suggesr.ed herethat more research is needed to de-cidee on a long term managemenr. po li-cy such as jettying; and that theshort term dredging should be con-tinued until jettying or hope fullyother management alternatives aremore intensively studied.
k«li r«nc«sliirkem«i«r, W.A,, «t al., IVII,Users Cuid« ro CIRC Field R«s«archI'ac ility, Nisc. Report gl � 7, U.S.Army, Corps oi Enginr«rs, CoastalEngin«t ring R«sear< h C«nter, Ft.Bt'Ivotr, Va
:ERC, 1973, Shore Protection Nanual;U.S. Army, Corps of Engineers,Coastal Engineering Research Center,Ft. Belvoir, Va.
Dolan, R. and R. Class«n, 1973, Ore-gon Inlet, North Carolina-A Historyoi' Coastal Change, SoutheasternGeographer, Vol. Rill, No. 1.
Stephenson, R.A, and N.F. Johnson,1977, An Analysis of the BarrierIsland Changes on the Ourer Banks,North Carolina, Cape HatterasNational Seashore, National ParkService, Nanr.eo, N.C.
210
NISCONCEIUEN CAUBES OF CLIFF ANB BEACH EROSION IN ROSSLARE NAT,SOUTER&ST IRELAND: A CAUTIONAEF TALE FOR COASTAL ENCINEERS
Ne SRORELINE NAB&ORES
Julian Or f o rd
Department of GeographyThe Queen s University at Belfast
Belfast, Northern Ireland, BT7 1BN
Bill Carter
School of Environmental StudiesThe University of Ulster
Coleraine, Northern Ireland, BT52 3SA
lntroduc t ion
Study Area
I h<~ nort I hh <<1 92* xt<imt
211
Ge<nbline With the Shere. <<OCee<tintft Of the hlinth AnnvefCenference of The Cent<et SOciety, t>etcher 14-17, taaaAttentfc City, NJ. Cof>yn'yht by The Coettet Society tasa
C~rpholog ists should he wa>.yof making hasty judgements <>ncoast eros ion prohlet»s respecially as many of their viewsare accepted with alacrity hycoastal residents and managers.'lhis cautionary study ftescriheshow a ma jor misconcept ion as tothe orig in of a severe coasterosion pro >lem in southeastIreland has spawned a series ofill-founded protection measures.In a country where the totalannual central governinen tappropriation for mastprotection s< heines >s less thanS2~ such lack of ur>der~tending rs a serious matter.
>>t>sslare Bay is in t.he <nxt remesoutheast corner of Ireland Fiq.I !. Whi'ie the east-fac>ng hay >swithin the semi � enclose<i IrishSea, the shoreline plan owes m<>rcto control exercised hy northerlymoving Atlantic swel , refr'actedaround the promrix nt. ht a< !l an<3:.r
0reenore Wint and as suchRnsslare Pay appears to he a goodexample of a crenellate hay Bi lvester 1970! moving towardss<xt>e fi>rm of equi 1 ihrium hetweenplan form and wave action *srecxxJni sed wor ld wide hy theconcept of loq-spiral hays.Rosslare beach extends for 15 km,frcm Greenore lt>int to theestuary mouth spit at theentrance of Wexford Harhour. The2-3 km of southern coast of thehay comprises Low beaches anderr>ding <Jlacial till cliffs up to10 m in hei<Jht. Near Wsslarevillage the cliffs r!ive waylaterally to spit sediments with
s ingle Iow sand dune r i dge,tronted hy, at Iow tide, a 20-50m wi<ie heach. A shore-parsi le 1bar <x <.epics the inner i>earshorezone. The shore 1 ine terminates>n a narr<iw si>nd spit, stretc.h>n;!out on r ll<' souther n s i<is of'ltexf'>r< I Hnrtx>ut, a 5 > k<n2estuary, partly <naintained hycl >sc >at:]e 10-350 cctn<><'s ! f ro<nthe H>ver Blaney,
Harbour is a southerly di rectedspit Raven Spit! suppl ied withsediment fran the rapid!y eroclingfluvio-glacial sand/<! rave! c 1 i f f salong the north County wexfordcoast. Johnston ! 984! adv<ocatesthat the southerly longshoredrift potential in the vicinityof the Raven Spit is four to fivetimes that of the northerlylongshore drift potential alongRosslare Bay. 'Ihe intersectionof the two drift systems isunclear � yet the presence of asubstantial ebb � tide delta at theestuary mouth s<xfgests that spitdynamics must be related tolongshore sediment supply andtidal hydraulic interactions.
'Ihe spring tidal range is 1,6 min Ross lare bay and 1, 5 m inF4.xford Harbour estuary! . Medianwave heights are < lm frcx«allonshore directions, withsoutheast storm wave~ heing thehighest Orford et al 1983!.
History of Shoreline Erosion andProtection
The entire coast of Rosslare Bayis suffering from erosion.Figure 2 provides a su<Tn<ary ofthe available data based on mapand airphoto analysis, The localview of this erosion, based onthe perception of a resiclentsaction group at Rosslare Strand,is that it was initiated by theconstruction of a railway-f.erryjetty Ireland to !«ales! n~orth ofGreenore R>int in the 1880s atthe position now identifiecl asRosslare Harbour. Tt is thoughtthat the jetty impeded thenorthez ly movement of longshoredrift, so starving the beaches inRosslare Bay and in the processthreatening the centre ot Countyhexf<ord's coastal tourism tradeat Rosslare Strand. Alm<ost allshore protecti<on initiatives havebeen based on tht.< impeded ciriftpr<mis», rxrtw<thstanding thegeneral lack of evidence
In 1924-25 the d ista 1 1200 mstretch of Rossla re Spit werebreached hy severe southeasterlystorm washover thus isolating thecoastguard station at RosslareFort Fig. 3! . 'Ihe spit wasf<nally beheaded in the late192Cs and over the next 50 yearsthe spit terminus retreated about2.5 km to its present position.This initial beheading seems tohave precipitated a long-runningbattle over coastal erosionbetween the local residents, thelocal council F4.xford CountyCouncil!, the national government the Board of Marks! and therailway company Corais ImpairF.'irann! who still maintain a raillink from Fublin to RosslareHarbour. f<< nunber of piecemealcompromise coastal defencesolutions have been undertakensince the 193Cs, including theconstruction of tither andconcrete hulkheads, timbergroynes, timber palisactes andslatted snow fences and rockarmouring . Many of thesesolut ion s have been imposed overshort stretches of shoreline byindividuals wishing to protecttheir own property most notablyby private hotels and by a golfcourse on the Spit. Tn placeslandfill and old cars have beendumped over the cliff to serve asprotection. Between 1964-72attempts to nourish RosslareStrand with sediment from thedeposits impeded updri f t ofRoss la re Harbour, hel ped to slowerosion at. Rosslare, but disputesover sour<oes of. new sediment andlegal action cut t he amount ofnourishment to negligible ! evelsbetween 1971 and 1982. In 1983the railway company CIE had toprovide 105 x 103 m3 of offshoredredge spoil in order to fulfil arequ i remen t for beachnour i shmen t . The sed imen t wasplaced on the beach at Rosslarestrand, hut unfortunately, itsf'ine size meant that tt. wasrapidly removeci offshore gtvinqno lastin<! protection to the
212
beach.
Fur ther sources of sediment fornourishment have yet to heidentified, but impeded driftreserves south of RosslareHarbour are now virtuallyexhausted with little furtheraccretion taking place at a ratesufficient to replenish theamount heing used for nourishmentbetween 1964-71.
1he major coastal problran forIreland in general and Rosslare,specifically, is that there hasnever been sufficient legislationor finance to allow a unifiedcoastal management plan to bedeveloped by which recognition ofgeomorphic processes and resultscould underwrite a coherent andacceptable coastal defence policy.
~rphol ical Assessment
Ccxrrparison of early maps BritishAdmiralty Hydrographic Chartssee Fig. 4! suggests that thepre-jetty erosion was at least ofthe same order, if not greater insane places Fig. 2!, thanpost-jetty erosion OrdnanceSurvey 1:10660!. Most of thecoast is still eroding at between0.6 m and 2.0 m per year thoughthe rates are slowing down due tothe combined impact of coastalprotection measures. 'Ihe tillcliffs show an irregular spatialretreat pattern, ccmrprisingshallow rotational slides,spall ing and loca] ised mudsl idesand slumps. Most > 79%! of thecliff material is of silt or claysized particles, which appear tobe transported seawards insuspension. Very little cliffdebris remains on the beach.Sane sand sized material movesalongshore to the north, hut thevolrxrre is only in the order nf10~103 m3 per year sohnston,1984! and is insufficient toprovide beach protection. Mostof this sand probably movesalongshore in the nearshore bar
system outs i de the groyne f ield,Wwndrift of the resort ofR>sslare Strand, the dunes aretrirrmed frequently hy high tides,and show no signs of any recentforedune accumulation. Fbr mostof its length the dune is onlyone ridge wide and is close tocol lapse hy undermining innumerous places
while construction of theWsslare Harbour jetty did resultin accretion of a small shadowforeland, its openwork character !880-!978! is exceedinglyunlikely to have caused majordisruption to any originaldcminant longshore sedimenttransport. W the basis ofextracted sediment for beachnourishment purposes, the averagepre-jetty drift rate pastRoss 1 are Harbour was 5 x103/m!/vr, 'Ihis is ccrrwrensuratewith the current rate of sedimentnow being eroded in Rosslare Bayand entering the beach c. 6 x103/m3/yr. Ibis suggests thatthe building of the harbour hasnot materially affected thenortherly longshore supply toR>sslare Spit. 'Ihe recentrebuilding of the Harbour �978!with a replacement impermeablesea wall may, however, augerfurther longshore supplydepletion. Inspection of map andfield evidence casts considerabledoubt on the importance of thecoast east of Rosslare as a pastmajor sediment source forsupplying long term beach volumesthat would have protectedRosslare Strand and moreimportantly built Rosslare Spit.Map analysis shows no long r:ermshoreline changes, while thecliffs have well-vegetated,stable profiles at angles ICP-XPlower than the eroding cliffswi thin Mssl are Bay. I4e wouldargue that the development of acrene! late hay at Ross!are, westof the present Harbour position,necessitated similar erosionrates now, as experienced prior
213
to jetty construction! . Thesediment volumes required tobuild and maintain Rosslare Spitare not likely to he suppliedsolely fran erosion of theRosslare-Carnsore Mint tills.The higher erosion ratesexperienced at Rosslare Strandmust reflect the changes in sizeand position of Rosslare Spitrather than solely on the longerterm changes engender' hythe dynamics associated with thecrenellate bay developnent ofRosslare Bay. Given the lossrate frcm the Spit of 38 7 x103/rn3/yr over the period1940-1980, it is clear thatalternative sources of sediment,apart f ran the minor erodingclif f element, have to beconsidered .
The sanewhat inconclusive natureof the jetty evidence promptedconsideration of the dynamics ofk!nxford estuary as a potentialfactor in the Rosslare erosionprobl~. In the rnid-nineteenthcentury the shallower intertidal'sloblands' within the estuarywere extensively reclaimed Furlong, 1970! through theconstruction of low banks, acrossthe north and south arms of theestuary Fig. 5!, followed bypumping. Within a decade�845-1855! these reclamationactivities halved the high waterarea of the estuary, althoughonly about 10% of the tidal prismvolume was lost. The main effectof these schemes has been toalter the volume of tidalinflow/outflow and hence tidalhydraulics and, as a consequence,the stability of the tidalinlet. It is postulated that theloss of the shallow estuarineareas, including sane salt marsh,would have radically altered theebb tide regime, producing ashorter, more concerted flow.Additionally the discharge fromthe River Slaney would not havebeen so readiLy intercepted andretained hy the estuary, so
reinforcing the ebb tidedanination. The sedimentologicalconsequences of this may hetraced from the sequence of maps Fig. 5! taken over the last 140years. 'Ihe shift frcm f leod toehb bias is reflected in a rapidflushing of the fled material,prcgrading the ebb delta. 'Ihiscaused two further changes; one,the bypassing efficency ofsediment fran north Raven Spit!to south Rosslare Spit! wasimpaired, so starving thesouthern beaches RosslareStrand! and causing theprogressive withering of the spitin an apparent updrift directionfran the distal terminus Fig.3!, and two, the burgeoning ebbdelta and change~ in thenearshore of Rosslare Bay priorto jetty construction Fig . 4!resulted in a major perturbationof the inshore wave climate Fig.6!. 'Ihe main change of thisbeing the comparative shelterinqof the Rosslare Strand fromnortheasterly storms allowing thenortherly drift of material toaccelerate. These two processestogether led to the onset ofaccelerated erosion along thesouthern shore. The lack ofsediment moving south round theebb tide rlelta meant a majorreduction in the sand accretionto Rosslare Spit's distal endhence the high rates of erosion.
Conclusion
'Ihe main point of this study isobvious it is dangerous to leapto conclusions as to the cause ofcoastal erosion prohlrans. TheRosslare study exemplifies thiswell. 'Ihe erosion problems ofthe Ray could be related to anycombination of three factors.Cne, that the shoreline isretreating through naturalwasting caused hy, perhaps,sea � level r ise or changingstorminess or by the continuingmovement towards sane form ofwave climate � shoreline
214
Acknowledgements
References
Furlong,of landHarbour .53 � 76.
tong -termsed imp nt
215
equilibria associated with thecrenellate bay morphology.that the reclamation of WexfordHarbour in the 1850s set off achain reaction manifest today inshoreline recession aroundRosslare Bay. Ihree, that theconstruction of a jetty in 1880near Rosslare R>int was solelyresponsible for Rosslare Strand'serosional problems. Option threeseems the least plausible, giventhe existing map and field data,although it has been widely citedas the sole cause, and mostcoastal defense initatives havebeen linked to it. At most thechanges imposed in the system bythe presence of the jetty areonly cosmetic in affect. Optionone <natural erosion! is hard toassess due to a lack ofinformation, but sealevelstability over the last 200 yearsat least means that shorelinedisplacement due to verticalsealevel rise is highlyunlikely. Likewise no evidenceexists for changing storminessover the last two centuries, Asfor natural movement tocrenellate bay equilibrium � thisseems unlikely given the dramaticchange in sediment supply and thecatastrophic demise of the spit.Option two, associating estuarinereclamation with later coastalchange is at least, the mostintriguing, and from the evidencepresented here, albeitcircumstantial, the most likely.%here are a number of other sitesin Ireland where a scmewhatsimilar pattern of cause andeffect can be noted, but as yetthey have gone uninvestigated.'Ihis is a further reflection onthe inexperience of coastalengineers' in Ireland who fail torecognise the interlocking natureof natural environmental systemswhich appear at first glance tobe unrelated.
Lessons for managmnent are easyto accept but hard to implement.lesslare Bay is now so festooned
with coastal engineeringstruct ures of varying designs,origins and ages that it wouldnot be easy to restart on a newbasis. Ihe chances offacilitating estuary bypassingare slim, given the costsinvolved, while construction ofnew defenses would do 1 it tie tohelp an already starved beach.the likely prognosis for RosslareStrand is continued erosionunless a major beach nourishmentscheme with all its additionalramifications related tomaintenance can be undertaken.Given that this kind of operationhas never been successfullyundertaken in Ireland and thatthe cost of the exercise would beconsidered by Government agenciesto be disproportionate tn thevalue of Wsslare Strand, thenthe future for the site looksbleak. 'Ihis scenario does nothowever prevent local propertyinvestment being welcomed byRosslare Strand residents who seeits encouragement as a means offorcing Local and CentralCovernments agencies to undertakeeven more ineffective remedialcoastal protection measures!
We shou!d 1ike to thank MrFbrd bk xford Co. Fngineer! forhis assistance in allowing usaccess to W.C.C. information.Interpretations of the data arPstrictly ours alone and do notnecessarily reflect W.C.C.views . We should also like tothank Cueen's University and theWyal Society for assistance inthe preparation of this paper.
N, �970! 'the historyreclamation in Wexford
J. Wbxf ord Soc. I 3,
,Iohnston...h. �984!sediment =,uppl y,
216
transport and shoreline evolutionon 'open' and 'closed' cellularcoasts: Co. wexford and CoR>nega1, Ireland. Unpublished D,�ril. 'Thesis, The New Universityof Ulster, Coleraine, 349 pp.
Orford, J.D., Carter, R.W.G. andJohnston, T.W. �983!Discussion of particle sizegrading on a shingle beach.Earth Sci. Pirblin!, 5, 247-249.
Silvester, R. [1970! Growth ofcrenulate shaped bays toequilibrirjrr J. Waterways andHarbour Dis., Proc. Pm. Soc. Civ.
4', 275-287,
Fig. l Location of RosslareHarbour, Co. Wexford
Fig. 2 ~tal units of RosslareBay with rrean annual shore-line erosion rates andcoastal protection rreasures.Note that data for periodsA and B come from Hydro-graphic and Topographicsurveys respectively
Fig. 4
217
Erg. 3 Changes in the positionof Rosslare Spit �925�1983! based on map andair � photo analysis. Notethe beheading and retreatof the spit terminus,
Changes in the bathym tryof Rosslare Bay during theperrod �845 � 1883!, priorto any affects of thefirst jetty at RosslareBarbour. Definitions;Erosion= the 1883 contouris landward of the 1845positicn. Severe erosion.
where the 1883 -20' or -12' con-tour is landward of the 1845 -12'or -6 ' contour. Deposition;where the 1883 contour rs seawardof the 1845 position. Note thedeposr ties on the northern flankof the eb-:ide delta and theerosion bc.rn on the southern flankof the cbo-' 'de delta, and rn Ross-!are Bay ~ ' '.882.
Fig. 5 Changes w the bethyrretryof Wexford Harbour and theebb-tide delta �845-1958!and the extent of estuaryreclamation �847-1854!.
218
Fig. 6. The alteration in long-shore wave p3wer for twr7wave direction due tobathy77etriC ChangeS inRosslare Bay prior toRosslare Harbour con-struction. Ho: off-shore wave heightTO wave peliod, 2:d2rection of waveapproach. Note thereduction of southerlydrift in the lee of Ross-lare Bay for both wavedirections by l883,
i,"* 'OI *01
0
l000
10
9200
HO I OSm
7: 0.70
2 OOH100
Ii + -I 2883IO01
0 00 I Z
92 ~100
1000
0 Z 0 Z OlZ0
1000 HO 1,2710
7 02 ~
2 "0"
2883
100
10
219
Gernhhhg with the Shore. Procsedingt of the Ninth Annoe!Conierence ot The Coectel Society. Octoher i&lg t98eAtlentic city, htt. copyright hy The cototel soaety t98a
LAND USE CONTROLS TO REDUCE COASTALEROSION DAMAGE IN WISCONSIN
Professor Douglas A . Yanggen
Agt icultural Economics DepartmentUniversity of Wisconsin
Hadison, Wisconsin
Introduction
There are three general types of erosion-hazard situations found alongthe cnast of the Great Lakes. The first involves erosion of beaches.Beaches are periodically being worn away in some areas whileaccumulating in other areas, Structures placed too close to eroaingbeaches will be subject to flooding and undermining by waves. Asecond related type of shore erosion involves sand dunes. Dunesprotect against wave attack and floading. D'sturbance of dunes andremoval of vegetatinn will subject them to wind and wave erosion andresult in a loss of their protective functian . Bluff retreat is thethird type of erosion hazard. It is the most prevalent and mostserious problem in Wisconsin and is the major focus of this paper,
Significant erosion is taking place along almost one-half ofWisconsin's 620 miles of mainland shoreline on Lake Hichigan and LakeSuperior. On Lake Michigan, short-term recession rates of 3-15 feetper year have been recorded along sandplains and 2 � 6 feet per yearalong high blufflines. On Lake Superior, recession rates of 2-5 feetper year are cammon along blufflines and rates in excess of 10 feetper year have been recorded around bays. As the shoreline recedes,houses and other structures are damaged or destroyed. Damages of over$16 million occurred during the 1972-76 high water period and willincrease as erosion continues and additional development takes placein erosion hazard areas.
Causes of Shore Erosion
Wave erosionWind driven waves are the primary erosive force or. Wisconsin's LakeMichigan and Lake Superior coasts. Beaches are continuously changing
221
as waves add or remove materials. During storms, high steep wavesrcmove beach materials and carry them lakeward. In periods betweenstorms, small waves tend to carry material shoreward and build up thebeaches. Waves that approach the shore at an angle create a"longshore current" that runs parallel to the shore. Longshorecurrents move materials along the shoreline in a process called"littoral drift" that can help replenish beaches. If this littoraldrift is blocked, however, rapid erosion may occur.
The extent to which waves and currents erode shorelines depends upon avariety of factors. Among these factors are: storm direction andintensity; wind strength and duration; the configuration of the lakebottom; structures which affect littoral drift; and lake levels.Higher lake levels caused by increased rainfall in the drainage basinor storm surges allow waves to attack the shoreline farther back thanusual, greatly accelerating erosion. Lake Michigan varies between ahistoric high of 580.2 feet and a low of 575.4 feet. Lake Superi. orlevels range from 598.2 feet to 602,0 feet. In addition, a stormsurge on the open coast can raise water levels several feet during thestorm. In some bays the storm surge can be even greater, reachingfour to six feet on occasion. property owners may build too close tothe shore during periods of Iow lake levels because they believe theshoreline is stable. When higher lake levels occur the protectivebeaches which normally dissipate the force of the waves are againsubmerged and waves directly attack the shoreland. Shorelandsdowndrift from structures which trap the movement of sand, such asgroins, piers and off-shore breakwaters, are subject to particularlysevere erosion.
Bluff retreatWave attack undercuts or steepens bluffs causing unstable slopeconditions. Even if wave-induced erosion could be completely halted,bluff recession would continue as the force of gravity acts to movematerial on the unstable slope to a lower position. Over a longperiod of time, the slope would reach a stable angle of repose wherethe stresses acting to move material down the slope and the resistanceof the materials on the bluff to these stresses are in balance. Thetexture size and shape! of the material.s comprising the bluff and theground water pressures determine the resistance to shearing. Theshear stress of the materials in the bluff is primarily influenced bythe bluff height and the angle of the slope. Water entering the soiland frost action can further weaken the bluff. Localized ground waterconditions may greatly reduce slope stability. When water percclatingthrough the soil reaches an impermeable layer, the water may pushmaterial out from the bluff face causing collapse of the overlyingsoil. Building development and other activities which add weight orwater to the bluff also increase its instability.
Host bluff retreat occurs as slope failure, that is, periodic landslides. The most common forms of land slides are slumping andtranslational sliding. Bluffs composed of silt and sand tend toslough off in shallow layers in the form of translational slides.This occurs when materials moves down the sloping bluff face along asingle slide surface in a layer several inches to one or two feetthick. Along some parts of the shorelines this typo of bluff failureis probably the most important. However, it is impossible to quantify
222
the effects of translational sliding without detailed measuremen orto predict its occurrence in more than a general sense. Slumps tendto occur in more cohesive materials where a large intact mass of thebluff slides downward in a rotational motion. Slumping usually takesplace fairly rapidly and the movement of one slump block can remove upto 50 or 100 feet of bluff top. This type of landslide could resultin both loss of like and considerable property damage Edil andVallejo, 1976! .
The absence of vegetation on the face of a bluff indicates an activestate of erosion. Bluffs with an unvegctated face are subject tosurface erosion from rain and runoff. Development which remcvesvegetation, creates impervious surfaces, and increases runoffaccelerates this erosion. Ravine corrosion is common along manybluffs. Stormwater drainage from intermittent streams, road ditches,storm drains and other sources of concentrated runoff causes ravinesto form. Over time, these ravines may enlarge to threaten structuresplaced too close to them. Vegetation on the top of the bluff incombination with storm water controls such as earthen harms mayintercept and divert surface runoff preventing it from eroding thebluff. In some cases, the amount of material removed by surfaceerosion may exceed the amount removed by slumps and slides.
Wave erosion, landslides and surface erosion often occur incombination at a particular site. In these cases, attempts to dealwith erosion which do not take into account all three of these factorswill be unsuccessful. Thus, vegetation and storm water controls willnot control bluff recession that is due to sliding and slumping unlessthe slope has been stabilized. The slope cannot be stabilized unlessthe toe of the bluff has been protected against further wave attack.
Bluff recession is influenced by many factors. These factors do notoccur in a uniform or continuous process over the short term; e.g.,wave attack which is related to fluctuating lake levels. After eachincrement of slope failure the slope temporarily stabilizes untilthese forces once again decrease slope stability and precipitateanother slope failure. A slope may appear stable and become heavilyvegetated for a period of years. However, active recession during thelife span of a typical building is inevitable for many of the coastalbluffs.
The Structural A roach To Dama e ReductionThere are two basic approaches to reducing erosion damages: thestructural method and the land use cont~ol method. The structuralmethod relies on slowing down the erosion process by constructingdevices to protect against wave attack or to stabilize the bluff. Theland use control method focuses on adjusting land use to the erosionhazard by setting structures back a safe distance and controllingrunoff.
Ccmmon structural methods to reduce wave attack are to armor the shoreby rip-rap revetments or bulkheads and to build protective beaches bydevices such as groins or nearshore breakwaters. Devices to stabilize
223
bluffs including reshaping the bluff to s stable angle andconstructing terraces with retaining walls. Structural methods aremost appropriate for protecting existing development against erosion.They may also be necessary for protecting sewage treatment plans,ports and other high value new developments which have to be locatedin the immediate shoreland area. The setback approach is preferablelargely because of the limitations of structural attempts from both aprivate and public point of view. Among these limitations are: �!Attempts to adequately protect against waves may not be feasible froman engineering point of view, e .g., effective protection mayrequire stabilization of a coastal reach which is longer than the sitein question. �! Structural measures are usually too costly inrelation to the value of the land proposed to be protected. Theeffective life of a structure is reflected in its construction cost.Improperly- designed> installed or maintained protective measures willfail and are a waste of money. �! Structural measures may haveadverse off-site effects. Grains may cause accelerated erosion bystarving down drift beaches. Shore armorment may deflect waves whicherode adjoining property. �! The form of shore protection mostcommonly used by individual property owners is loose dumping of stoneor concrete rubble . This practice affords only short term protection.Besides destroying the natural beauty of the shoreline, this materialoften ends up on the bed of the lake, impairing the public rights innavigable waters.
The Land Use Control A roach to Dame e Red~ctionThe land use control approach focuses on the safe location ofdevelopment in shoreland areas. It adjusts land use to the erosionhazard by the appropriate setback of buildings and other vulnerableuses in erosion-prone areas. Local zoning ordinances and subdivisionregulations can require that new development be placed landward oferosion hazard setbacks. Establishing a safe setback involves severalsteps: a! identifying the areas subject to erosion; b! determiningshore recession rates; c! selecting the length of time during whichregulated uses are to be protected from recession; and d! in the caseof bluffs, the additional step of estimating the stable slope angle.
There are two basic approaches to determining erosion hazard � thesite specific method and the reach method, The site specific methodrequires a geotechnical engineering analysis at each site at the timedevelopment is proposed. This method may require a report analyzingamong other things: a! wave-induced erosion based upon recessionrates and wave energy calculations; b! geologic conditions includingthe soils at the site and their properties and stabililty; and c!groundwater and surface water conditions. While the site specificapproach may be technically accurate, it is too costly and timeconsuming for all but the most expensive development.
The reach method uses generalized formulas to estimate the erosionhazard . Nuch of the information needed is available from studies madethrough the Wisconsin Coastal Zone Nanagement Program. Erosion HazardArea Napa at a scale of 1 inch equals 2,000 feet delineate areas witherosion potential. These maps also show short-term recession rates
224
�966-1975! and long-term recession rates at selected intervals. Ingeneral, it is preferable to use the long-term rate as a measure ofrecession. In speaking of the variation over time in average retreatrates, a technical paper of the Corps of Engineers notes: "Thus, thenn h hi * n nh n n hh h n i nn rntend to decrease with time. The variance of these estimates wouldalso tend to decrease thus, the precisions increase! jn directproportion between the number of years between surveyshu Hands,1979!.
Kstablishin a Recession Rate SetbackA recession rate setback distance can be established by multiplyingthe average annual recession rate by the assigned design life of thestructure to be protected e.g. ~ 30 years, 50 years or I.OO years for aresidence!. The selection of the appropriate regulatory time spanduring which buildings are to be protected from recession is adecision to bc made by local policy makers in Wisconsin. The State ofMichigan requires permanent structures to be set back the distance ofthe 30 year recession rate, but recommends that a greater setback isdesirable Michigan Department of Natural Resources, 1973!. TheProvince of Ontario measures the 100 year recession rate and thestable slope angle Ministry of Natural Resources Ontario, 1978!.
A 50 year rate appears to bc a reasonable minimum figure, since itapproximates the useful life to a typical residence. To illustrate,assuming a 50 year design life and a long term recession rate of 2feet per year; regulated structures would have to be set back 100fcct.from the ordinary high watermark. The recession rates shown inthe Technical Report Appendices and Krosion Nasard Maps sre consideredas a general guide for determining the recession rate in a given ares.In areas with highly variable recession rates or where structures haveaccelerated erosion, it may be necessary to make additional studies orto determine the recession rate at the particular site whendevelopment is proposed.
Kstsblishin a Stable Slo e SetbackAssuming for a moment that no further wave-induced erosion takesplace, it is also necessary to determine the additional setbackrequired to locate buildings outside unstable slope areas. Theultimate angle of repose of a stable slope reflects the angle ofinternal friction of various materials has been documented byengineering analysis. Kven though actual bluff failure at aparticular site depends upon local variations in the soil profile,groundwater conditions, vegetative cover, surface drainage and otherfactors, the stable angle of repose of various classes of materialscan provide a reasonable rule of thumb to estimate slope stability,Thus knowing th height of a bluff, its slope angle, and thepredominant material of which it is comprised takes into account somekey site-specific factors.
A generalired stable elope angle of 2Q feet horirontsl distance to 1foot vertical distance �1.8' hss been selected for most lake Michiganerosion hssard areas. This figure is based upon studies of relative
225
slope stability of bluffs along Lake Michigan which took into accountstratigraphy, parent materials, bluff height and slope angle Misconsin Coastal Management Program 1977! . In addition, on LakeMichigan slopes of approximately 21,8 degrees �$: 1!, naturalvegetation occurs and that vegetation can effectively control manymass wasting processes. The predominantly clayey soils on LakeSuperior tend to be less stable. A generalized stable slope angle ofthree feet horizontal distance to one foot vertical distance �8.4 !has been suggested for regulatory purposes in these areas.
Structures, such as residences, that would be damaged by slope failurecan be protected by requiring them to be located outside of unstableslope areas. A stable slope setback for a 50 foot high bluff would beestablished as folio~a: An angle of 21.8' �Q feet horizontaldistance to 1 foot vertical distance! is measured from the ordinaryhigh watermark. The point at which this angle intersects the bluff isthe edge of the stable slope. This means that the stable slopesetback would be 2.5 stable elope angle! x 50 feet bluff height! of125 feet from the ordinary high watermark.
Determinin The Frosion Hazard Setback for ZoninThese computations of recession rate and stable slope angle can beused to establish an erosion hazard setback in a zoning ordinance.Within this setback line high value structures which would be severelydamaged by erosion or activities which would accelerate erosion can beregulated. Usinp our previous examples, in an ordinance that requireda 50 year period for protection against recession the erosion hazardsetback would be 100 feet from the ordinary high watermark for a beacharea with a 2 fnot per year recession rate. Assume there is anotherarea with the same recession rate but which also has a 50 foot highbluff. Here the erosion hazard setback would be the stable slopesetback �0 ft. x 2.5 ft.! 125 feet plus the recession rate setbackof 100 feet or a 225 foot erosion hazard setback line. The erosionhazard setback can be modified if the landowner provides technical dataproving that a different recession rate is warranted, slope conditionsare more stable than assumed, or that tbe erosion hazard, althoughcorrectly estimated, can be mitigated by structural protection.
Ad ustin Land Use to the Erosion HazardThe damages that vill result from shoreline erosion depend upon boththe severity of the erosion hazard and the type of land use that willbe affected. As the shoreline continues to erode, the land willeventually be lost but the major portion of the damage comes fromdestruction of structures on the land. Open space land uses such asagriculture, forestry and parks is the most appropriate land use inmany erosion hazard areas, other things being equal. However, somefacilities such as marines, water intakes, sewage treatment plants,ports, and certain industries may require a location in the immediateshoreline area, For these shoreline dependent uses careful siting toavoid high hazard erosion areas and well designed erosion mitigationmzasures are important to avoid unnecessary damage. In the main,these uses are ones for which it may be economically feasible toprovide effective structural protection, An investigator of shoreline
226
erosion in Southeastern Wisconsin, commenting on structural erosionprotective measures, notes "For the most part, the successfulstructures observed were built either by units of government or, ta alesser extent, by industry. These structures are massive, wellengineered and constructed, and probably much too expensive to bejustified for even the most valuable residential properties" Hadley,p. 27! .
Since regulations generally apply only to new development, theeffectiveness of regulations depends upon the existing land usedevelopment and ownership patterns. These patterns vary widely butmay be characterized by the following general categories; I! ruralareas where the land consists of large tracts of open space use insingle ownership, e.g. ~ farms and forests; �! rural areas where theland has been divided into smaller tracts through subdivision plate orsale of individual lots but is not yet developed or only partiallydeveloped; �! suburban areas where the land has been substantiallydeveloped along the immediate shoreline and development consists ofinfilling, i.e., construction on the undeveloped shoreline lots; �!developed areas where the first tier of lots has been largely builtupon and development is occuring within the second tier of lots withinan area still subject to erosion; and �! urban areas where almost theentire shoreline is developed in depth. In general, regulations havetheir best potential in relatively undeveloped areas.
Relocation and Structural I'rotection for Devel o cd AreasLots already occupied by buildings are largely beyond the scope ofregulations. The only appropriate regulatory provisions are thosedesigned to control activities which would accelerate erosion or whichcontrol the expansion of structures subject to damage. The owner ofan existing structure subject to substantial erosio~ damage has twobasic options: �! attempt to mitigate the erosion hazard byprotecting against wave erosion and stabilizing the slope ar �!relocate the structure. Permanent relocation outside the erosionhazard area could mean moving the structure to the rear of the sameparcel if the lot is af sufficient depth, or moving it to a differentlot not subject to erosion . "Relocation is an alternative chat cannotbe overemphasized, Erosion is a natural geologic process that isextremely difficult to stop. The alternative to build shoreprotection or to relocate must be weighed against the consequence offailure. Depending upon the type of structure yau mipht consider, itmay cost the same to relocate as it would to build shore protection.Should a protective structure fail, then your investment in thestructure is lost and your home or cottage still in danger" U.S. ArmyCorps of Engineered' North Central Division. p . 14! .
The following informatian on relocation costs is based upon 1979 dataas presented in the state shore erosion plan. A number of factorsaffect the cost of relocation. "They include lot depth, theavailability of new building sites, esse of site access, buildingconfiguration and size, amount of subfloor access, number of publirutility disconnections, and the availability of experienced movers.Because relocation is typically only considered during emergency
periods, the amount of land lakeward of a building is a criticalfactor. Between 15 to 20 feet of clearance is normally required forsafe operation of equipment. Moving costs of a small cabin orcottage, medium size ranch style housel' and large mansion can beexpected to range between $3,000-$4,000, $7,000-$9,000, and$30,000-$40,000, respectively. These < osts do not include sitepreparation costs at the new location" Springman and Born, 1979,p.87! .
In cases of individual hardship where lots are too shallow to permitconstruction meeting the erosion hazard setback, it may sometimes berea onable to permit s moveable structure such as a mobile home orresidence designed so that it can be readily relocated . Allowing suchstructures within the setback line should be done only on a case bycase hasis after a careful investigation of the particular situation.Appropriate conditions should be attached to development permission inthese instances.
Land Use Controls for Undevelo ed AreasOn lots of adequate depth the most satisfactory approach is toproperly locate the structure in the first place. This means thatstructures should be set back a safe distance from erosion hazardareas. The setback should be based upon a consideration of therecession rate ~ in all cases, and a stable s3ope angle in the case oferodible bluffs.
Zoning ordinances and subdivision regulations are important tools thatlocal government csn use to require that new land uses take erosionhazard into account. Subdivision regulations and zoning complementeach other. Zoning focuses primarily on the uses of land, thedimension of lots, and the location of structures on the lot. Lotdimensions are important to ensure the lot is deep enough to permitstructures to be safely located behind the required erosion hazardsetback line, Zoning can also control grading' filling, vegetativeremoval, installation of protective devices and other activities thatmay accelerate erosion. These activities can be made conditional usesto require that they be undertaken in a manner that avoids adverseeffects.
Subdivision regulations focus on the process of dividing larger tractsof land into lots for purposes of sale or building. For undevelopedareas which have not been divided into lots, subdivision regulationshave particular promise. The larger size of the parcel involved makesit more likely that economically feasible engineering solutions can befound to storm water managements grading and filling and erosionprotection measures. Subdividers can be required to designate erosionhazard areas on the plat ~ and restrict this area to park or open spacefor the use of the residents of the subdivision.
~Sam le Erosion Hazard ProvisionsSample erosion hazard provisions designed to supplement local zoningordinances and subdivision or regulations have been prepared for'Risconsin's local governments Yanggen, 198 l!. In general, these
228
provisions regulate erosion prone lands by: a! establishing snerosion hazard setback line; b! restricting or prohibiting uses whichare vulnerable to erosion damage or which may impair public rights innavigable waters; c! requiring special review of erosion protectiondevices to ensure that they are properly designed and installed and donot have substantial adverse environmental impacts; and d! regulatingland disturbance, vegetation removal, runoff and other activitieswhich may increase erosion. Flexibility in the zoning is achievedthxough conditional use provisions which allow landowners to provethat: slope conditions are more stable than assumed; a slowerrecession rate is warranted; or that erosion hazards can be mitigatedby structural protection or placing a moveable building on the site.After public notice and a bearing on a conditional use, the localZoning Agency may grant development permission, deny it, or allowdevelopment to take place subject to specified conditions.
Subdivision regulations are an important tool for reducing erosiondamages and/or protecting potential purchasers of erosion-prone lands.The subdivision regulations contain provisions that allow localgovernment to: a! require each lot to have adequate area to meet theerosior hazard setbacks; b! prohibit the subdivision of lands subjectto serious erosion unless the hazards are overcome; c! require thedesignation oi erosion hazard setbacks on the plat; d! limit the useof hazardous lands through deed restrictions or dedication of the landto the public; c! review proposals for stormwater drainage, gradingand similar activities which may accelerate erosion to ensure they areundertaken in a manner compatible with conditions on the site; f!require the subdivider to install reasonably necessary publicimprovements, including erosion control measures or provide a suretythat the improvements will be installed; and g! require that ero ionprotection measures are maintained by a properly constituted privateagency with assessment powers.
229
Edil, 'l'uncer B. and Vslle]o, Luis E.; Shoreline Erosion and Landslidesin the Great Lakes, Wisconsin Sea Grant Program Report Nc . 15,University of Wisconsin, Madison, 1976.
Medley, David; Shoreline Erosion in Southeastern Wisconsin, WisconrinGeological and Natural History Survey, Special Report Number 5 ~1976.
Hands, Edward B. "Changes in Rates of Shore Retreat, Lake Michigan,l967-76." Technical Paper No. 79-4, U.S, Army Corps ofEngineers.
Michigan Department of Natural Resources; A Plan for Michi an'sShorelands, Lansing, August 1973.
Ministry of Natural Resources, Province of Ontario and Fnvironment Canada;A Guide for the Use of Canada/Ontario Great Lakes Flood andErosion Prone Area Ma in ~ Burlington, Ontario, March 1978.
Springman, Roger and Born, Stephen; Wisconsin's Shore Frosion Plan:An A raisel of 0 tions and Strate ies, Wisconsin CoastalManagment Program, Madison, 1979.
United States Army Corps of Engineers, North Central Division; "HelpYourself", Chicago.
Yanggen, Douglas A,; Re ulations to Reduce Coastal Erosion Losses,Wisconsin Coastal Management Program, 1981.
230
Gambling wrrh rhe Shore. proceeoinm of the fvinrh AnnueiConference of The Control gocfery. October 14-17, 1984.Arlentrc city, haf. copyright by' The coertel Society 198K
AMALTSIS OF BRIGANTINE NATIONAL WILDLIFE REFUGEAREA USING TRENATIC NAPPER TN! SENSORT DATA
Kam Rachardson
Nsnager, Landsat Remote Sensing LabGraduate School of Oceanography, University of Rhode Island
Kingston, Rhode Island
Niels West
Associate Professor, Dept. of Geography and Narine AffairsUniversity of Rhode Island
Kingston, Rhode Island
Satellite remote sensing using the new Landsat Senaor � theThematic Mapper TM! � may provide coastal zone cover assessment in atimely and cost effectfve manner. Thematic Napper TM!, data �9S2! apart of the Brigantine National Wildlife Refuge Arear' New Jersey, hasbeen obtained from the National Marine Fisheries Service, NortheastFisheries Center, Sandy Beck. The nb!ective is tO Claaaify the TMfnformatfon and develop images which can be used to identify existingland cover which may greatly enhance coastal zone management. All sevenbands from the TM were used to develop the unsupervised classificationof the reflectance values of the associated land cover, The applicationof the TM with the increased spatial resolution was the primary reasonfor the study. Additional incentives were to apply information from theblue, mid-infrared, and thermal bands which are unique to the TM sensor.The processing was done using the ELAS software package written byNASA's National Space Technology Laboratory, Earth Resources Laboratory ERL!, Bay St. Louis ~ Mississippi.
New Jersey'S Coastal Management Plan was approved in two phases,the first, covering the state's Atlantic counties including Nonmouth,Ocean and Cape May, was passed in 197S. Since then' the state has beenimplementing its coastal plan which includes inventorying its extensivecoastal wetlands. One Of New Jersey's most diversified wetlandsincludes Brigantine National Wildlife Area BNWA! which fs part of theEdwin B. Forsythe National wildlife Refuge. This area encompasses bothmodified and natural wetlands. The paper reports on the development ofan unsupervised classification for a part of the extensive wetlandscovering the southern portion of BNWA which is located in Atlanticcounty, New Jersey, approximately ten miles north of Atlantic city. The
231
study area is bounded on the Forth by Oyster Creek Road and is locatedentirely within the Few Jersey Coastal Area Regulatory Area CAFRA!.The BFwR northern boundary is the rapidly growing suburban communitiesof Oceanville and of Smithville Fig 1!.
Enny state coastal plans have been hampered by the absence ofadequate information to develop a fully integrated coastal zone plan.Consequently, ihe availability of comprehensive and reliable land coverinformation will enhance not only the interpretation of the image, butalso facilitate the implementation of the management plan. Since 1972 'the Landsat Kultispectral scanner Rss! has provided regional plannersand resource managers with increasingly reliable and cost effective landcover surveillance. Fany studies, too numerous to quote, have beenundertaken in terrestrial settings, yet coastal and nearshoreapplications have been relatively few except for Klemas and hisstudents. Representative studies dealing with wetlands and biomasssurveillance include early inventories of marshes Klemas et.al., 1979!,analysis of stressed wetlands Klemas et.al,, 1983!, biomass assessments Bartlett et.al., 1977 and Bartlett and Klemas 1979a, 1979b, 1981, andButera, 1983!. Other authors have used Landsat remote sensingtechniques to facilitate fisheries Brucks, 1979!.
One of the principal problems with the MSS has been its relativelycoarse resolution which is approximately one acre. While sufficient formany regional applications, the Landsat HSS system may not be detailedenough for land cover applications in the coastal wetlands whereindividual vegetative stands may be much smaller.
The three new bands are: blue visible light, a mid-infrared, and athermal infrared band. Another ma!or advance of the Landsat IV systemis its capability to communicate directly with the Earth. Data transferfrom the spacecraft to the processing center at Goddard Space FlightCenter is accomplished by the Tracking and Data Relay Satellite System TDRSS!. Previously, the Landsat systems relied on an onboard tape
232
The wetland environment by its very nature is an extremelydifficult area to survey from the ground, in part because of itsinaccesibility Richardson, 1984! and in part due to the absence ofadequate ground truth information. To date, the Landsat system is theonly satellite sensor that is even close to having the required surfaceresolution to study the wetland land cover which is often characterizedby highly diversified vegetative cover. This paper discusses the use ofthe Tg sensor, which has a surface resolution of 30 meters, or 0.25acres' as opposed to the more conventional Hultispectral scanner Rss!.The TH relies on the same technology as the Nss, and recordselectromagnetic radiation from seven different bands, also referred toas channels, as follows:
QC
U
Z'
0I�
Z
I�Z
U
233
recorder for data storage until the time the satellite over-passed anearth receiving station. These onboard tape recorders had twooperational constraints: a limited lifespan' and a built in time delayin data transmission. The Landsat 1V provides near real time datareception and a greater orbital life expectancy.
The data used for this analysis was obtained from a small reductionof the TM scene 40150-15084 Row 14 ' Path 33, dated December 13, 1982.The larger scene is centered over Dover, Delaware and measuresapproximately 100 by 100 miles and covers an area approximitely 10 F 000square meters. The classification used for this paper was based on asubset of the Dover image measuring approximately 12 miles by 12 milesand applied to the center portion of the BWNRA, including the twoimpoundments immediately to the east of the administration building andextending across the bay and barrier island where it includes the northwesterly portion oi the borough of Brigantine Fig 1!.
Digital data storage requirements for a whole scene are veryextensive totaling 290 F 858,000 bytes, but ihe working subset comprised16,900 ' 096 bytes. The TN data used for this study is consideredEngineering Data by NOAA, which does not guarantee for its accuracy.
The Landsat data was processed using the software package developedby Earth Resourses Laboratory Applications software ELAs!, written bythe Earth Resources Laboratory of NASA's National Space TechnologyLaboratory, Bay st. Louis, Mississippi Junkin, 1980!. The softwarepackage is made up of more than one hundred modules, each of which isresponsible for a specific type of processing. The software is userinteractive and enables the analy ist to develop and manipulate avariety of color image displays. The ELAS sOftware operates under acontro1 file environment, which "keeps track" of a11 processing resultsfrom each module used in the data processing. This information isstored for the user in the control file for subsequent reuse and update,All values stored in the control file by the user can be used by otherELAS modules either as data input or intermediate information which maybe required to perform subsequent asks.
The data window for the BNMR include lines 1 to 480 and elements5550 to 6062 resulting in a window consisting of 512 by 480 pixels. Thewindow was chosen by ingesting the entire scene for channel one throughseven, and then selecting the lines and elements which most closelycoincided with the physical boundaries of the southern portion of theBNwR. The computer compatible tapes were read using the module thatreads the tape and re-formats the data into a disk file for directaccess by the other software modules. Each channel of TM data is storedon a single tape, and all seven channels were read in for this area ofinterest. The specific window parameters are input to the module andthen run to read the tapes. The output disk file is formatted by ELASand the data is read into it for later processing,
There are two ways in which land cover information may be obtainedfrom a Landsat scene, supervised and unsupervised. The supervisedclassification requires apriori knowledge of the area to be analyzed.
234
In these instances, the analysist knows the geographical location ofspecific land cover types and the associated reflectance values. Thesesignals are then being used to classify all other pixels in the scenewith similar reflectance signals. The unsupervised classificationrefers to a procedure by which a classification scheme is developedusing a search routine, which passes a three line by three elementwindow through the data from all seven channels to compute homogeneoustraining field statistics. The statistical data is stored in a subfileof the control file for later use by other ELAS modules. The spectral
signatures were defined for thiS classification by the lower nouna ot0.1 standard deviation and a standard deviation upper bound of 1.0.Furthermore, the mean value times the coefficient of variation valuecould not exceed five percent.
The statistical output Table 1! represents the unsupervised landcover classes that were clustered from this window of data. The finalstatistical preparation consists of those merged statistics that havethe smallest scaled distance between any pair greater than the minimumscaled distance of 3.0. The mean values for each channel electromagnetic band! are then plotted and the resulting curve is thespectral signature which characterizes each class within the data. Thissignature data is used fn the classification to identify ground covertype by pattern recognition or clustering through the use of maximumlikelihood technique. According to Bayes' rule, the pixel is assigned aclass value in accordance with the probability of fts occurrence withinthat class. The mean vector and covariance matrix from the trainingsite statistics computed by the search routine previously discussed!was used to determine the best fit assignment for the classification.Each input pixel is assigned a class from the statistic into which thatpixel fits the best. A disk file containing ihe classified pixels isthen written for image display and analysis.
The unsupervised classification produced thirty-five land and waterclasses Table I!, twenty-three of which were identified based oninformation obtained from the oceanville and Brigantfne Inlet 7 I/2minute quadrangle series. In addition, interpretation was enhanced byvegetative transect made available by the US Pish and Wildlife ServiCe.
significant man-made modifications have impacted the BgwR andenvirons. The Intracostal waterway Iow! divides the image in two withseveral old spoil islands clearly in evidence, including Black Point andshoal Island. The most dominant features are the two impoundmentslocated fn the west central portion of the image. The westernmost fsrepresentative of a fresh water marsh into which the Doughty creekdrains. A ditched and controlled saltwater marsh environment is locatedimmediately east of the freshwater impoundment providing an environmentwhich differs from both the diked freshwater marsh and the uncontrolledbay environment surrounding both impoundments. Finally, a particallydeveloped barrier island separates the estuarine system to the west fromthe open ocean.
A total of thiry-five individual classes resulted from theunsupervised classification, twenty-three of which were associated withvegetative covers. For illustration purposes. three examples of blackand white "density slicing" are presented in Figs. 2-4. Theclassification of the combined reflectance values for each of the sevenTK bands have been converted tO a gray scale rendition in Fig. 2. Thespecific class associated wfth a particular land cover is identified bysubstituting white for that gray scale level for the class. Thus, inFfg. 2 ~ white is substituting a land cover class which on the grayscale has been depicted as dark gray. In Fig. 3, white has been 'slicedthrough' a land cover which approaches black and which is identified asshallow bay water and associated drainage channels. These areas havelittle or no emergent vegetation. It will be noted that not all thewater has been included in this class. On the ocean side of the barrierbeach the shore drops off rapidily identifying shallow water within a
236
Figure 2
Figure 3
237
Figure 4
short distance of the shoreline. Conversely, deeper water extends ashort distance into the bay through the Brigantine Xnlet. Finally, thevery shallow water located within and immeditely surrounding the twoimpoundments have nct been included the this class.
The last illustration Figure 4! shows the extensive emergentmarshland vegetation consisting largely of @gyring BnlgBg and ~~gllgrnf~ which characterizes much of the bay and surrounds theseveral natural and manmade islands.
Because of high printing costs associated with the publishing evenblach and white pictures only three of the original 35 land coverclasses have been discussed in this paper. A more efficient way toassess land covers requires a supervised classification. This procedureexpects geographical land cover information which reflectance range isused to identify areas with similar reflectance values. This procedureis more labor intensive and requires the identification of of severalsites for each of the land cover classes to be included in the analysis.
While the application use of the TK sensor is still in its infancyand largely confined to terrestrial sites, there is considerableevidence that this system may be particularly useful in assessing landcover changes in the coastal zone including both the terrestrial andmarine portions of this important environment.
238
Bartlett, D.S. et.al!, "Variability of Wetlands Reflectance andits Effects on Automatic Categorization of Satellite Imagery,"
Amer. Sac. Photogram. Annual Meeting, Washington ~D.C., 1977
Bartlett' D.S. 6 Klemas, V. ~ "Assessment of Tidal Wetlands Habitatand praductivity," ~MRR 13th Int. symp. on Remate sensingof Env., Ann Arbors MI, April' 1979
Bartlett, D.S. 6 Klemas, V.. "Quantitative Assessment of EmergentBiomass and Species Composition in Tidal 'Wetlands Using RemoteSensing," ji~hW� Workshop on Wetlands and EstuarineProcesses and water Quality Modelling, llew orleans, LA, June.1979
Klemas, v. et.al.!, "Inventory of Delaware's wetlands," QlRl lgcggl�979!, 433-439
Klemas, V. et.al.! ~ "Remote Sensing of Saltg Marsh Biomass andStress Detection," AdK BRBSR RRS �983!,2,8,219-229
Richardson, K.A., "Wetlands Classification Using Landsat ThematicMapper Data Unsupervised Classification Approach", ~i~dTenth International Symposium Machine Processing af RemotelySensed Data, Purdue University/LARS, June 12-14, 1984.
239
Gambling svirh the Shore. Proceedings of rhe bhhrh AnnualConference of The Coesnil Society. October 14.17, 1984.Arlenric City, NJ. Copyrighr by The Coastal Saciery 198a
The protei.t.ion of beaches or more accur-ately, thv m;iintinanii. iil' l.he high waterlinc at an exist.ing lucat.ton can beef lect ed in a variety of ways. In urbanareas such as resorts, a rigid mass can-c rei.e si rue cure may be built ta ensurer lie permanent e uf the adjacent struct-ures.
I"'it et iiil s Used in Mire-Enclosed ki -Ra
llvdern gabion mattresuggesrs, fabr icaredusually mild steel asteel and pl est ics hgabion originated wifilled with stone orhave been documenteduries agu I! and towire came into use ored years ago.
At the ache-r end of the scale, beachesmay be replensshed by rlie duicping ofsand either from inland or fram dredgingoffshore sandbdnks which may in turnaffect the wave climaci.,
fresh far Caastal StructuresThe use of wire enclosed rtp � rap, gabionsgives a structure that is neither rigidnor comple rely free to adjusr ro anylevel but does, with careful design,af fer an opportunity to stabs 1 i.ze thelieach at a 1iisiced raiigc of leva�'ls.
Although other types ran be used, thispaper wi11 refer only ta mesh fabricaredinta a liexagana 1 shape with a doubletwist betwei n ineshes. This, because ofthe appi'araiice, has also been designatedas "I'r i pie Twist". The wire used, mildstci 1 cvnforii.ing ta AST."1 A641A andFederal Speci t scat ion QQ-461-!l. is zinc-coaced galvanized! prior to weaving
Structures builr. wir.h tliis r.ype ofmaterial may include bulkheiids or wiil 1 s,revet tsi rits arid groins or d combiiiaf ivn
241
F'lexible beach Prot.ectton
sses are, as the titlefrom wire. This is
lthaugh high tensileave been used but thech wickerwork basketswtth sari. Usesback ro China cent�Italy where steel
nly abouc one hund-
Thus ensuring that che roating is uniform. For use in a marine environmentzinc-coating is not by itself sufficientprotection for the wire vhich thereforereceives a coating of PVC extruded ontoit prior to the weaving proress, Thiscoating is formulated to resist salt orbrackish water, the effects of ultra-violet radiation and frequent vettingand drying.
Rock for Use in Gabion Structures
As with rock for all applications farerosion control, whether as rip-rap orfor use enclosed in vire cages, thematerial must be both hard and durablecapable of resisting veathering, abras-ion and exposure to wave action. Incercain special cases softer marerialmay be used as is the case with a numberof strurtures on the Barbadian coastlinewhere only coral is readily available.In such cases the structures, usuallylow walls and/or groins, have been builtfor hateliers and resort developers viththe full realization that periodicallythe lids will have to be opened andadditional material inserted.
This process is only practicable forsmall walls and revetments and couldnot be employed in larger projectswhich vould be vulnerable should poorquality fill be used.
Although a fairly vide range af sizesmay be used for rip-rap, that used forwire-enclosed rip-rap is much smallerfrom, say, four inches to rvelve inchesand for use in a coastal environmentvhere the srructure is exposed to waveaction an even smaller range is notonly desirable but essential. Largermaterial, over eight inches in anydimension, must not appear in thesurface layers and, ideally, should beavoided altogether. This assists inmaximizing the number of voids in vhichvave energy may be dissipated vhilekeeping the void content as low aspossible.
Additional Naterials for use vith Wire-Enclosed Ri -Ra
Althaugh the stone used is, as explained
above, fairly small there is often aneed far a filter layer between the sailand the protective armour. This may bea layer of gravel, typically a six inchthickness for I> inch material ar a goodquality geotextile, or a combinatior ofboth, This compares with the usualrequirement for two layers under rip-rapto protect against the same vave climatewhere the rip-rap is of such size as tonecessitate intermediate filter layers.
Geotextiles
When geotextiles are used, considerablecare must be taken in the selection ofthe type to be specified, having regat'dto the sail type, the method af constr-uction i.e. the fabric must resist impactby falling rock, and the angle at whichit is to be placed,
The Design of Wire-Enrlosed Rip-Rap orGabian Structures
At present there is not a large volumeof published data on the use of gabionsin coastal structures. Reseach vascarried out in New South Wales Brown,1979! on revetments, Same work vascarried out there also in making compar-isons betveen types of mesh Kabaila, 1979!
Currently research is under vay in Crenobleprance by Sogreah who are undertaking amajor study on mattress revetments underwave ar,tion.
The structures may be classified intovarious types for whirh design data canbe considered,
Vertical Walls or Bulkheads
The use of wite-enclosed rip-rap, gabions.for verrical veils is, of course, quitecomman both on dry land and along rivers.Generally, designs are as for any massgravity vali having due regard to the freedraining quality of the structure.
Howevers vertical gabion walls are notusually recommended along the shorelineunless situated abave normal high waterlevels when they may function basicallyr.o control the erosion af cliffs by
242
wave action. of a toe wall.
Revetments
Bd � ! Sr-I! cot 0
Groins
It is as sloping revetments that gabionsfind many uses in stabilizing the shore-line. With slopes of one to three, maxi-mum thin revetments may be used to protectthe base of dunes, to protect pipelinecrossings and to provide pedestrianaccesses to the beach.
The design of such t'evetments was, aspreviously mentioned. investigated byBrown, in 1979 who came up with thefollowing formula for thickness of thegsbions mattress. placed on slopes steep-er than 3.5:l.
When Hd is design wave heightSr is specific gravity of rock fillV is fractional percentage of voids
betweer filling rock.
For usual quarry stone I-V!S -1! = Ir
This, can, however give misleadingresults in certain conditions. Considera sloping revetment at 1 to 3 to resistwaves of maximum height 6 feet.
Assume substituting in the above formula�! we get:
TM= 60.66 A 8 inches.3xlx3
While a lining of this t'hickness mightwell be sufficient if the water leveldoes not vary greatly between high andlow water mark, it would give rise todifficulties if such a layer were toterminar.e, as is often the case, with anapron level or set flush into the beachat the toe. There would, with six footwaves, be some risk of the mattress beinglifted up onto itself and thereforespecial attention would be required atdesign stage to avoid this possibility.Possible solutions would be to carry therevetment into the beach well below thelevel to which it might scour; ro increasethe thickness of the lower section of therevetment by 50K or more or the provision
Indeed, 8 inches does not allow morethan two layers of stone and prudencewould dictate an increase to, perhaps,twelve inches for structures receivingsix foot waves during a few storms eachyear,
Table 1. Table of recosszended thick-
nesses for use as a general guide only!See following page
Notes. I! Not recosssended at thisslope
�! Would require specialtreatment. at the toe
�! Filter layer essential
This table, based on the work of Brown,will become invalid when research bySogreah is complete and is published bythe sponsors, Officine Naccaferri SpAof Bologna, Italy.
In the meantime, it cannot be emphasizedstrongly enough that it is a guide onlyand, as in all coastal works, localconditions must be taken into account atall stages of design. As with all wire-enclosed rip-rap structures, but evenmore vital with coastal work, the meshwhen installation is complete must betight so that the stone will not moveunder wave action,
Flexible groins may be used to retainbeach material to stabilize beach levelsprovided ir is realized that the sand soretained will not be available to re-plenish another area. This simple factls sometimes overlooked in groin pro-jects where massive structures may sointerrupc the natural litteral driftthat materials are swept into deeperwater and are lost to the beach systemcompletely.
For this and other reasons, gabion groinare usually designed with their inshoreends no more than three feet above beachlevel and the offshore end half of thatfigure. The cross section is such as towithstand the wave action at thelocation ar which they are built,
244
Gabion grains may readily be raised inheight or can have units removed inorder to bring them to the desiredprofile relative ta the beach, animportant advantage vhere beaches arein transition.
Sparing is not readily calculated anda careful study made of local conditionscarried out prior ro the use of anygroin system. Gabians have, on occas-ion been used as temporary structuresin order to research the likely resultsof groins at specific locatians.
'ther Gabion Structures
Alang the beach flexible gabions mayalso be used as slipvays, as smalljetties and far offshore breakwaters.In the first case, slipways, gabionsmay be placed flush with the beach ifthe natural slope is acceptable, andthen surfaced dawn ta a foot or twabelow low water with a suitablematerial leaving the balance of theslipvay just in gabions,
Jetties may also be built either ascontinuous structures ar as piersbridged by, say, timber decking.If continuous consideration must begiven to the effect of the structureon the littoral drift.
Offshore breakwaters may be builtparallel or at an angle ta the shore-line. Their function is to interruptthe flov of beach material being de-posited on. or withdrawn fram thebeach thus building up the beach behindeach structure. Some very successfuluses have been made of this techniquebut it is not suitable for universaluse.
Sitin Considerations
One of the majar considerations indeciding vhether to use wire-enclosedrip-rap is that of the nature of thebeach material itself. If the beach isaf large gravel which is thravn greatdistances by heavy wave actian duringstarms then gabions used alone vill have
Conclusions
ln areas af natural beauty revetments maypresent an unwelcome intrusion and in suchcases the use af gabion mattresses pre-sents a further advantage. The smallinterstices readily fill with sand andthe structure attracts volunteer vegetat-ion. In some sites the mattresses be-came lost to viev in the course of ayear or two.
~Sblt h
�! Appleton, B. "Li Bing's Masterpiece Stands Testof Time"
Landon May. 1984
�! Brovn. C,T, "Gabion Report. or some factors affectingthe use of Maccaferri Gabions and RenoMattresses for Coastal Revetments.University of New South Wales, WaterResearch Laboratory, Manly Vale NSWAustralia October, 1973
�! Kabaila, A.P. "Structural Study of Tvo Types ofGabion Mesh"
"Asphalt in Hydraulics - MS 12"
245
a limited life. They can be used forsuch locatians but only if they areadditionallv protected by, for example,sand mastic asphalt vhich is compatiblewith gabians,being flexible and part.�icularly suited to a wide range ofhydraulic structures Asphalt inHydraulics: The Asphalt Insti tute,1976
Of course, similar difficulties can ariseif loose rip-rap is placed, say, at thetoe of a gabion revetment subj ect to waveaction and it is usually necessary taallov for site clearance in such cases.Of course, where such rip-rap is sited incalm areas ar outside the normal range ofwave action, it vill not present any suchhazard.
�! Asphalt InstituteNo~ember, 1976
Although the best coastal defenseis provided by nature and ideallynothing should be built that cauldbecome endangered because of etosion,structures are necessary for avariety af reasons.
Wire-enclosed rip-rap farms a use-ful method when neither mass con-crete or pumped sand meets theneeds. It can be attractive and,with care, became buried arith sandand/ar vegetation thus giving anatural appearance vhen this isadvantageous.
Although research ta date islimiced, it is hoped that designdata vill become available in thenear future.
Cernbtbry wttb tbe Strore. frocoertbrtn of the ffbrtb 4nnenrConference of ttre Ctrectel Saorety. Oetobce N-1 y, 1884.4ttenrfn Cfty, SJ. Cofryrtrfbt by rbe Ctrecmt Strorety tattK
SE& LEVEL RISE &RE SECRELII PL&REIN'
Rorbert P. Panty
Center for Coastal and Environmental StudiesRutgers, The State University of Nev Jersey
Rev Rrunswiclr., Rev Jersey 08903
Rates of modern sea level change have drawn the attention ofscientists for many decades. Prior to isotopic analyses the issue wassubject to speculation and intuition. With the advent of techniques toprovide absolute dates of transgressi.on and changes in habitats in theroastal environment, the speculatio« has been reduced and we now have harddata with which to unravel the past t:rends of sea level change.
Of course the past date and interpretations are valuable and becomeeven more so if their trends are projected into future predictions. Notall practitioners of these studies ~ however, believe that the future ratesof change will continue to follow the trends of the past. Recent reportsfrom an EPA- sponsored investigation suggest that sea level rise may beincreasing and amount to from +0.9 m to +2. 13 m by the year 2075 Bertha«d Titus, 19b4! . This is a rather substantial rate of change consideringthe Scholl and Stuiver thesis that sea level has risen but 1.6 m in thepast 3500 years �967! . If the predic.tions have validity, they forecastsignificant impacts on our coastal zone and the many activities that occurin it.
Despite the concern espoused in acier,tific circles about thescenarios attendant to these high rates of sea level change, these seemsto be little evidence of reactiot, irom within the planning or managementcommunity. There may be several reasons offered for this ostrich-likeapproach:
1! past sea level changes have been so minor and over such a longduration that they are beyond the temporal scale usuallypracticed in the planning milieu;
2! the realization nf the tremendous magnitude of. change is beyondthe logical planning process. In this rase the temporal scaleis so foreshortened that planners fee1 totally intimidated byits magnitude and are unable to arcept its effects.
Irdependent of whether sea level will rise at the low ratecalculation, the high rate, or ary rate, it would seem important that anyplanning or management agency consider the possible impacts of this eventon their jurisdiction. Nay the planners and managers excuse themselvesfrom this tcsponsibi! ity by denyirp the facts of sea level rise? Nay theyconsider a raft of alternatives pertezning to the long-range land usedecisions but neglect the possibility that the land surface will no lonperexist in its same dimensions or locations in that identical temporal span?
The following discussion is intended to address some of the factualinformation available concerning barrier island response to sea level riseand to point to elements involved in the planning process. It is meant toapp1y the sea level rise scenarios to a planning context and in so doingprovide the plarners with a basis for long-range land use decisions.
General Situation
One model of barrier ic Land development that is appropriate to NewJersey is that of island migration up the continental shelf as sea levelis rising. This model stresses the washover and inlet sedimentetxonprocesses that transfer sediment from the oceanszde of the island to thebayside Figure 1!. The driving forces for the migration are sea Jevelrise and a negative sediment budgets. The effects are a loss of volumefrom the seaward side oi the island, a retreat of the seaward margin, ar;alongshore transport of sediment to inlet locations and shoal development,and an overland transport to the hayward margins of the i.eland. The papscale of these processes in coastal New Jersey may be interpreteu trom Cdates Figure 2! pretaining to transgressions of the barrier islands andtheir associated habitats. One interpretation is that for each unit ofvertical rise in sea level, the barrier feature and its habitats aredisplaced horizontally in an inland direction on the order of 200 times Figure 3! . It is 1 ikely that this proportion is appropriate only forthose barriers that are a few hundred meters wide. The larger' widerbarriers would not react to sea level changes so readily.
figure 1. Coastal processes active in sediment budget determination andisland migration.
Figure 2. Holocene sea level curve for coastal New Jersey. Source:Daddario �961!; Heyerson �922!; CCES dates collected by Psuty at severallocations.
Scale of Change
One of the classic papers in coastal geomorphology, known as theBruun Rule Eruun 1962!, states that ss sea level rises the beach faceadjusts to the inundatxon by drawing sediment irom the total beach protileto raise the nearshore bottom. The effect of this translocation ofsediment is s farther inland penetration of the new sea level beyond thatcaused by a simple raising of the water level. For purposes of providingan example of the effects of a one meter rise of sea level on coastal NewJersey, the following representation based on existing data is presented:
Assumptionswave base 10 mdistance to wave base 2000 mmaximum island elevation 10 mwidth of island 1000 merosion constant et 0,5 m/yr
Conceptually, the Bruun Rule tollovs this calculstzonHorizontal Zone of Activity
Sea Level Rise xVertical Zone of Activity
Shor~lineDisplacement
2000 8Shoreline Displacement
10m+10m1 m Rise
1 m Rise 100 m Shoreline Displacement
1 m Rise + Shoreline Erosion ~ 150 m Shoreline Retreat/CenturyIf the assumption continues that the 1.0 m rise occurs over 100
years, the affect of the negative sediment budget must also be includedfor that time period. The erosion value is added to the dtsplacementvalue and thus for our example in hew Jersey the seaward margin of ourbarrier island vill have retreated 150 m in one century. The 150 mshoreline retreat value is a combination of the effects of the one meterrise of sea level and the rate of erosion produced by a negative sedimentbudget. Other scenarios can be applied in the same manner. That is, atwo meter rise vill result in a 250 m retreat over a century, and a threemeter' rise vill produce a 350 m retreat.
Th» Bruun Rule gives us cause for concern if the calculations holdand they seem to be supported by empirical analyses in the Great Lakesrelating shoreline displacement with lake level fluctuations lpnds 1984!.11owever, the calculations of island displacement based on C data fromN Jersey point to a 200-times shift. Thus a 1.0 m rise in seal Levelev era pmay produce a 200 m displacement plus 50 m erosion over the perio o acentury, pro uc ngr d cing a total retreat of 250 m. The magnitude of such a
10value can be better appreciated vhen one considers that even the minoryear interval is characterized by a 25 m retreat under this scenario.
250
Figure 3. Battier island displacement as s product of sea level rise inNev Jersey.
Future Barrier Configuration
Given a continuing rise of sea level and a negative sediment budget,there are two barrier island configurations that may evolve Figure 4!.Planners could use these configurations to consider how current and futureuses could remain compatible.
Figure 4. Barrier is!und change scenarios as a product of predicted sealevel rise. A. Migration inlane. B. Browning in place.
Figure 4A represents a low barrier that will be subject to overwashand inlet sediment transport ]eading to barrier island migration as thesea rises. This island will remain low and broad. Applying the abovecalculations by way of illustration, the island will migrate inland by 200m as a result of displacement and will narrow by an additional 50 mbecause of sediment loss. The island will remain very dyramic. with veryfew high elevations and probably a discontinuous dune line. This wouldnot be favorable location for permanent structures.
Figure 4B points to a situation where natural end cultural factorscombine to elevate the island and cause it to remain above the level ofoverwash during the rising sea level. Given our basic assumptions asoutlined earlier, the rise in sea level will remove sediment from theseaward side of the island to compensate for the deeper water in thenearshore zone. The effect of this action will be to narrow the beach andscarp the dunes.
However, the island is being drowned from the bayside as well becausethe general water level is rising. The overall effect is a rapidnarrowing of th~ island. The bayside ot the barrier will probably sufferfaster rates of displacement because the slopes of the lee side arenormally more gentle than the oceanside, Given this scenario, the barrierwill essential!y drown in place, actively losing sediment from theoceanside portion and passively losing area from the hay -iae. Carried toan ultimate, the deeper bayside waters will be sites of larger waves thanpreviously possible and this could lead to additional active sedimentloss, thus adding to the rate oi narrowing of the barrier island.
251
Planning?
There is an assumption that planning involves the consideration of arange of possible scenarios and a range of actions to deal wi.th theseevents. If planning is to accommodate a rapid rise of sea level it seemslikely that the scenario in Fig. 4B will be the de facto plan put intoeffect. A several meter rise iti sea level will all but isolate the dunalridge. However, there will be a limit to the upward growth of the ridgeand eventually it may be subject to washover and inlet processes.Planning for this event not necessarily an eventuality! should be put ineffect now if planning is to retain its meaning. Planning shouldincorporate the variables of:
l. a diminishing spatial base2. a narrowing beach resource3. limited development on a high dune ridge4 . increased inlet instability5. higher wave energy and erosion on bayside6 . total re-evaluation of FEMA insurance zones as the barriers shiit
and clearance above water level is reduced.
Sea level changes are continuing, There is some question about therate of change. However, there is no question about the need to considerthe logical impacts of this rise in our coastal zone regardless of t: hemagnitude of the rise.
Bibliography
Berth, H.C., and James G. Titus, 1984. Greenhouse Effect and Sea LevelRise: A Challen e for this Generation. Van Hostrand, hew York, 325PP ~
Bruun, P., 1962. Sea Level Bise as a Cause of Shore Erosion. Journalof Waterwa and Harbor Division, American Society of Civil Engineers,88: 117-130.
Daddario, J.J,, 1961. A Lagoon Deposit Profile neer Atlantic City, NewJersey . Bulletin of the New Jersey Academy of Science, 6: 7-14.
Hands, E.B. ~ 1984. The Great Lakes as a Test: Model for 1'rofile Res onseto Sca Level Chan e. Miscellaneous Paper CERC 84-14. U.S. ArmyCorps of Engineers, Washington, D.C., 26 pp.
Meyerson, A.l , 1972. Pollen and Psleosalinity Analyses from a HoloceneTidal Marsh Sequence, Cape May County, New Jersey. Marine Geolo12: 335-357.
Scholl, D.W., ard M. Stuiver, 1967. Recer.t Submergence of SouthernFlorida; A Comparison with Adjacent Coasts and other Eustatic Data,Geolo ical Societ of America Bulletin, 78; 437-454.
