Download as a PDF

Coastal Sediments '07

© 2007 ASCE

WINTER STORM AND TROPICAL CYCLONE IMPACTS ON THE SHORTTERM EVOLUTION OF BEACHES AND BARRIERS ALONG THE NORTHEASTERN GULF OF MEXICO Gregory W. Stone1, Baozhu Liu2, and Felix Jose3 1. Coastal Studies Institute, Department of Oceanography and Coastal Sciences, Louisiana State University, 218A Howe-Russell Geoscience Complex, Baton Rouge, LA 70803, USA. [email protected] 2. Coastal Studies Institute, Louisiana State University, 218A Howe-Russell Geoscience Complex, Baton Rouge, LA 70803, USA. [email protected] 3. Coastal Studies Institute, Louisiana State University, 218A Howe-Russell Geoscience Complex, Baton Rouge, LA 70803, USA. [email protected] Abstract: Here we present data indicating the complexity and highly

variable response of barrier islands and beaches to the impacts of tropical cyclones and winter storms along the northern Gulf of Mexico. Data indicate that (1) barrier islands can conserve mass during catastrophic hurricanes; (2) less severe hurricanes and tropical storms can promote rapid dune aggradation and contribute sediment to the entire barrier system; (3) cold fronts play a critical role in the post-storm adjustment of the barrier by deflating the subaerial portion of the overwash terrace and eroding its marginal lobe along the bayside beach through locally generated, high frequency, steep waves; and (4) barrier systems along the northern Gulf do not necessarily enter an immediate post-storm recovery phase, although nested in sediment-rich nearshore environments. The fluid mud environment off west Louisiana coast plays a significant role in damping wave energy associated with tropical cyclones. However, no significant surge attenuation appears apparent. It is anticipated that these findings will have important implications for the longer-term evolution of coastal systems in the northern Gulf of Mexico.

1


© 2007 ASCE

INTRODUCTION

Recent data suggest that the North-central Gulf of Mexico coast has undergone an increase in the number of tropical cyclone landfalls over the past decade. Louisiana State University and the USGS have monitored the Florida panhandle and Alabama coast since the mid 1990’s using airborne LIDAR and field surveys. The resultant data sets provide a unique time series capturing morphological change and poststorm adjustment due to three powerful events (Opal, 1995, Ivan, 2004 and Katrina, 2005), weaker hurricanes and numerous tropical storms. It is the objective of this paper to identify, quantify and elucidate storm responses and post-storm adjustment along a heterogeneous and geologically complex coastal region. The concept of “barrier mass conservation” is further examined. Study sites along the Florida panhandle (A on Fig. 1), Mississippi, (B) and south-central Louisiana (C, D and E) are used to investigate the effects of geological variability on storm response. The significance of locally generated, high frequency waves in estuaries/bays during winter storms is also investigated and interpreted within the context of short-term coastal evolution.

D

B

A

E C

Fig. 1. MODIS image of the northern Gulf of Mexico showing locations of beaches/barriers monitored during the study program discussed in text.

GEOLOGICAL VARIABILITY

In an otherwise tectonically stable region of the northeastern Gulf of Mexico, the coast appears to be storm dominated and undergoing a reduction in sediment volume in the sub-aerial barrier unit, a phenomenon that appears attributable to storm dominance (barrier islands off Mississippi coast are the typical example, location B on Figure 1). 2


© 2007 ASCE

Along the Louisiana coast, barrier islands are routinely breached and/or stripped of their coarser (fine sand) fraction of sediment that overlays an organic, silty-muddy, cohesive core (location C on Figure 1). While the sand veneer is typically infused into the finer-grained bay sediments, the organic and fine-grained core material is exposed due to turbulence associated with storm waves. This response was noted after Hurricane Andrew (1992) (Stone and Finkl, 1995), Tropical Storm Isidore (2002) and Hurricane Lili (2002) (Stone et al., 2003; Allison et al., 2005). Larger chains that constitute those barriers west of the modern delta were not significantly impacted by either Hurricane Katrina or Rita (2005) (Stone et al., 2005; Guidroz et al., 2006; 2007). Similar to during Hurricane Ivan (2004), the Chaundeleur Islands (location D on Figure 1) were severely breached during Hurricane Katrina (2005).

Fig. 2. Chandeleur Islands, Louisiana, before hurricane Katrina, on 13 October 2004 (left) and after, 16 September 2005. Images courtesy of USGS.

West of the Atchafalaya delta (location E on Figure 1) a predictable response occurred along the Chenier plain and the coast farther west during Hurricanes Lili (2002) and Rita (2005). Fine sand was eroded from the upper shoreface and beach and deposited on the adjacent marsh surface as a well defined veneer which widened considerably to between 100-150 m west towards the point of landfall near the Louisiana-Texas border. Guidroz et al., (2006; 2007) reported extensive damage to the coastal marsh vegetation, even up to 25 km inland, due to super elevated storm surge associated with Hurricane Rita. Overwash deposits and evidence of excessive 3


© 2007 ASCE

salt water intrusion were observed for the Cameron and Vermilion Parishes, (Figure 3) southwestern Louisiana. However, the aerial observation study also reported initial post-storm beach recovery, within a matter of two weeks from the day of landfall. Significant contrasts in post-storm adjustment are evident over decadal time-scales on contrasting the sediment rich Florida-Alabama coast and the barriers of South-central Louisiana. Longer-term monitoring along the Florida coast implies post-storm adjustment can take up to 10 years for recovery of approximately 75% of the sediment budget after major events (e.g., Hurricanes Frederic, 1979 and Opal, 1995) (Stone et al., 1999). Although post-storm adjustment of the Louisiana coast is

Fig. 3. Holly Beach, Louisiana, before Hurricane Rita, on 16 June 2001 (upper) and after, 28 September 2005. Images courtesy of USGS, Coastal Hazard Team.

considerably slower, infusions of sand from updrift and offshore sources are apparent. Nevertheless, the multi-decadal increasing trend in storminess is causing a distinct down-turn in the sediment budget.

4


© 2007 ASCE

FIELD DATA

As a case study, topographic and bathymetric surveys have been conducted for a 6 km stretch along the western half of Santa Rosa Island (location A, Figure 1) over a 10 year period. Altogether 30 surveys were conducted since February 1996 and pre and post storm island morphology associated with all major storm events were monitored during this decade long field study program. The details of the transects and the methodology adopted are given in Stone et al., (2004). The morphological evolution of the island in terms of change in volume of sediments is presented in Figure 4. For uniformity, shoreline location was defined as NGVD zero. To study the different responses of the various sub-environments, the cross-island profile was divided into four sections as follows: Gulf beach (foredune to -1.5 m NGVD), dune (foredune to bay beach dune), bay beach (dune to Mean Low Water), and bayside platform (MLW- -1.5 m NGVD). STORM IMPACTS ON MORPHOLOGY

Stone et al., (2004) described in detail the post-storm morphological adjustment of the island after Hurricane Opal (1995). The study revealed that approximately 90% of the sediment eroded from the foredune-beach-nearshore environment was deposited as expansive overwash deposits on the interior and bayside flanks of the barrier. The barrier increased by an average of approximately 40 m in width along the bay side suggesting that the system conserved mass during Hurricane Opal. As a consequence of this mass conservation response, the Gulf side shoreline of the island retreated an almost equal distance. During the ensuing 2 years this newly deposited wedge of material along the adjacent bay was significantly reworked by northerly waves during the post-frontal phase of cold fronts which resulted in a significant shoreline recession (Stone et al., 2004). The response of north facing beaches/nearshore to these high-frequency steep waves generating in the bay during winter time are discussed in more detail later in this paper. The time series volume change data for the study area over the period February 1996 – August 2005 is given in Figure 4 and Table 1. These data include the signatures of 7 storms and 10 cold front seasons, which contained over 300 winter storms. Even though the impact of Tropical Storm Josephine resulted in a net loss of sediments from the system, the barrier immediately regained sediment during the period November 1996 and July 1997. Due to the sustained southerly winds during Hurricane Danny a significant amount of sediment migrated from the nearshore and was re-desposited on the berm and overwash platform. The increased vegetation density during the summer also helped the re-deposition of sediment. Prior to

5


© 2007 ASCE

Fig. 4. Volume change for each of the barrier sub-environments from 2/96-8/05 at the Santa Rosa Island site, west Florida.

landfall of Hurricane Georges, the barrier went through a phase of sustained accretion, particularly along the gulf beach. The impact of this storm resulted in the migration of considerable volumes of sediment, eroded from the berm, across the

6


© 2007 ASCE

Table 1. Sediment Volumes of the Entire Study Area and each Sub Environment: (calculated by BMAP to -1.5 m NGVD) Survey

Entire Area 6 3 (10 m )

Gulf Beach 6 3 (10 m )

Feb-96 5.896 1.173 Apr-96 5.877 1.195 May-96 5.863 1.185 Jun-96 5.822 1.141 Jul-96 5.833 1.152 Aug-96 5.85 1.165 Sep-96 5.819 1.127 Tropical Storm Josephine Oct-96 5.664 0.982 Nov-96 5.658 0.98 Dec-96 5.655 0.968 Jan-96 5.732 1.048 Feb-97 5.737 1.048 mar-97 5.771 1.088 Hurricane Danny Jul-97 5.712 1.013 Sep-97 5.635 0.94 Jun-98 5.8 1.058 Tropical Depression Georges Oct-98 5.633 0.861 Jan-99 5.618 0.831 Mar-99 5.624 0.837 Jul-99 5.722 0.953 Nov-99 5.605 0.846 Mar-00 5.614 0.829 Mar-01 5.64 0.796 Feb-02 5.853 0.962 Jul-02 5.996 1.115 Tropical Storm Isidore Oct-02 5.926 0.955 Hurricane Ivan Oct-04 5.94 0.916 May-05 5.811 0.864 Tropical Storm Arlene Jul-05 5.79 0.711 Hurricane Dennis Aug-05 5.804 0.706

Dune (106 m3)

Bay Beach 6 3 (10 m )

Bay Beach & Bayside Platform (106 m3)

3.974 3.945 3.947 3.952 3.95 3.952 3.952

0.295 0.291 0.291 0.292 0.292 0.293 0.293

0.749 0.737 0.731 0.729 0.731 0.733 0.74

3.94 3.938 3.945 3.941 3.947 3.945

0.291 0.29 0.29 0.29 0.29 0.289

0.742 0.74 0.742 0.743 0.742 0.738

3.963 3.961 4.011

0.286 0.287 0.282

0.736 0.734 0.731

4.09 4.095 4.086 4.083 4.084 4.105 4.185 4.149 4.163

0.279 0.28 0.288 0.287 0.285 0.283 0.276 0.267 0.271

0.682 0.692 0.701 0.686 0.675 0.68 0.659 0.742 0.718

4.25

0.273

0.721

3.611 3.665

0.349 0.318

1.413 1.282

3.756

0.329

1.323

3.711

0.344

1.387

island to form a series of overwash terraces. This resulted in a considerable reduction in the volume of sediments in the gulf beach environment. This downward trend in gulf beach volume was reversed after May 2001, six years after landfall of Hurricane Opal. However, Tropical Storm Isidore (September 2002) again reversed that trend 7


© 2007 ASCE

and subsequently, the gulf beach has been undergoing a sustained erosional phase which was exacerbated by the land fall of Hurricane Ivan (2004), Tropical Storm Arlene (2005) and Hurricane Dennis (2005). As per our latest survey in August 2005, the sediment volume along the gulf beach was at its lowest level during the entire span of the study period. The overwash platform, however, has been following a different trend over this decade of monitoring. Beginning May, 1997, two years after landfall of Opal, the overwash platform had been showing a steady increase in volume, mainly at the expense of episodic migration of sediments from the nearshore and beach, driven by sustained southerly waves and storm surge associated with the occasional hurricanes and tropical storms. However, a conspicuous downturn in volume occurred with the landfall of Hurricane Ivan in September 2004. For the track and other meteorological details of the Hurricane Ivan, see Stone et al., (2005). Persistent southerly waves and storm surge associated with landfall of Hurricane Ivan, a category 3 hurricane, resulted in substantial transport of sediment from the dune and overwash platform to the bay beach/nearshore. In addition, the island breached across the study area at a few locations and channels occurred in the backshore area. However, contrary to the gulf side environment, the overwash platform had been gradually gaining in sediment volume since Fall 2004 and underwent erosion during Hurricane Dennis (July 2005). As discussed in Stone et al., (2004), the bay beach is susceptible to frequent cold front generated winter storms and the morphodynamics of this environment are variable. During our survey period, the bay beach sediment volume increased since 2002. Apparent increases in the volume of sediment were observed immediately after landfall of Hurricane Ivan. Sediments reworked from the gulf shore and overwash platforms were redistributed and deposited on the bay beach and the adjacent platform during this event. However, these sediment deposits were reworked by the northerly waves associated with winter cold fronts and a net reduction in volume was observed until the land fall of Tropical Storm Arlene (2005). The trend had been reversed and the net sediment volume in this environment was increasing until our last survey. DISCUSSION Influence of Storms on Morphodynamics

Data indicate the importance of overwash events in supplying sediments to the bayside beaches along Santa Rosa Island, and the importance of cold fronts in rapidly eroding the overwash deposits and causing a significant amount of deflation on the subaerial portion of the barrier. These data have added significantly to our 8


© 2007 ASCE

understanding of the longer-term morphological maintenance of barrier islands in the Gulf and the importance of extratropical and winter storms (see Figure 5). Few studies have been conducted on the post-storm adjustment phase of barrier islands. Moreover, those that have been conducted (cf. Sexton and Hayes, 1991; Dingler and Reiss, 1995) are limited to annual topographic surveys, which significantly reduce the resolution of morphological recovery of the barrier system. Profiles are typically conducted during the same month in successive post-storm years, restricting conclusions to the net amount of sediment deposited on the barrier during the poststorm phase (Dingler and Reiss, 1995). Given that >90% of sand comprising largely the subaerial barrier mass at the Florida study site could be accounted for after two severe hurricanes, the implications are highly significant for (a) the source of material during post-storm recovery, and (b) the prevailing concepts that sediment eroded from, in particular, the upper shoreface during storms, is transported offshore to the continental shelf and lost from the barrier system. These issues have not been adequately addressed in the barrier island literature. In addition, these preliminary data have potentially significant bearing on the application of existing models to the upper shoreface-foreshore (see Swift 1976; Swift et al., 1985). Barrier islands along the northern Gulf of Mexico are oriented generally east-west from southern Louisiana to the Florida panhandle (Fig. 1). North-facing beaches on the bay side of these barrier islands are particularly susceptible to wave attack during northerly winds, which generally accompany the passage of cold fronts. Significant erosion has been observed in areas where the adjacent fetch in the sound or bay is relatively long (Stone and Morgan, 1993; Chaney and Stone, 1996; Armbruster et al., 1995; Armbruster, 1997). During winter seasons, three distinct end-member type cold fronts impact the northern Gulf of Mexico (Lewis and Hsu, 1992; Pepper, 2000; Pepper and Stone, 2004): 1) the mid-latitude cyclone; 2) the Arctic surge; and 3) the Gulf cyclone. The extratropical weather systems produce local variations in wind direction, intensity, and duration, and thereby significantly affect wave and sediment transport processes along the back-barrier beach. Fronts cross the northern Gulf approximately 30 times each year and with the exception of tropical storms and hurricanes, are the only known natural mechanism to generate relatively high waves in these bay/sound environments. A common regional weather phenomenon accompanying the passage of cold fronts is the strong post-frontal northerly wind. Due to the sheltering effect of the island, southerly winds are not capable of generating high waves along the backbarrier beach. Influences of the Gulf swell are minimal except in the vicinity of the

9


© 2007 ASCE

tips of the islands and inlets. Strong northerly winds, therefore, constitute the primary driving force for modifying the bayside barrier beach system.

Fig. 5. Time series of volumetric change along the northern Gulf for 3 sub-environments and the entire barrier system 1996-2006 with an applied fourth order polynomial curve to identify apparent trends.

The general morphological characteristics of these back-barrier beaches are fundamentally different from the features developed by storm overwash processes (Schwartz, 1981; Otvos, 1982), which are critical to the origin and supply of sediment to these systems (Stone, 1998). These differences indicate the existence of

10


© 2007 ASCE

a general mechanism/s governing the dynamics and morphological maintenance of the back-barrier beach. Post-Frontal Forcing

It is reasonable to assume that due to the sheltering effect of barrier islands along the northern Gulf (Stone and McBride, 1998), nearshore sediment transport along the back barrier beach is not significant during southerly winds. Where the northerly wind has a long fetch and is capable of generating relatively high waves, water-level setup occurs. Our work to date indicates that the strong northerly wind is the primary energy supply to the back-barrier beach in terms of significant sediment transport and morphological change. We acknowledge the longer-term role of relative sea level rise in these areas but take the position that the critical forcing mechanism/s for mobilizing and transporting sediment are a primary function of wave-current and wind forcing. The area is microtidal and thus, tidal currents play a lesser role in sediment resuspension in shallow water. The frontal event has distinct meteorological and sea state signatures associated with its pre- and post-frontal phases. In Figure 6 we present a time series of wind speed (A), wind direction (B) and significant wave height (C) for a frontal passage in December 2002. The data were obtained from a WAVCIS (www.wavcis.lsu.edu) metocean station located at the 6 m isobath in Mississippi Sound, north of West Ship Island. Winds blow from the south prior to arrival of the front at the site and reach a maximum speed of approximately 7 m/s. Southerly wind speeds decrease as wind direction veers clockwise to the north over a matter of a few hours (B). Wind speeds rapidly increase across the Sound and peak at approximately 13.5 m/s. During the pre-frontal phase, wave heights are low in the Sound,