Search

One of the most destructive tropical cyclones ever to strike the U.S., Hurricane Katrina made landfall along the Mississippi coast on 29th August 2005. The Mississippi-Alabama (MS-AL) barrier islands were subjected to storm breaching, area reduction, and vegetation loss caused by a number of parameters including salt spray, saltwater flooding, mechanical damage (e.g., ablation of bark from tree trunks), removal of plants and their soil substrate by scouring, burial under sand, and a 10-month, post-storm period of low rainfall. Repeated acquisitions of remotely-sensed data served as an essential tool in quantifying vegetated and total land area before and after the storm, and post-storm ecological community type and topographic elevation. Vegetated land area continued to decline on some islands in the first year following the storm. However, by November 2007, only 2.2 years after the storm, total vegetated land area had recovered to 72, 96, 77, 93, and 82%, and total subaerial land area to 97, 94, 33, 100, and 104%, of pre-Katrina values on Cat, W. Ship, E. Ship, Horn, and Petit Bois islands by natural re-growth and sediment accretion, respectively. Comparing ecological community-type maps that were developed from field and remotely-sensed data with LiDAR-derived digital elevation models determined that year 2010 ecological community type changed distinctively at the decimeter scale as mean surface elevation ranged from 0.1 m to 1.2 m. Storm-related changes in ecological community type included subtidal to supratidal sand flat, low marsh to wet or dry herbland, and woodland to wet herbland/shrubland.

A large body of chronologically well-constrained and detailed Quaternary data accumulated in the last decades from the coastal-shelf region of the northern Gulf of Mexico. Consequently, a preliminary synthesis of major aspects of the late Quaternary barrier and mainland coastal development on the northern Gulf of Mexico is timely. Several major factors account for the striking differences between barrier size and development styles in the three northern Gulf coast regions. Barriers are nearly continuous in the NW where they dominate by the greatest dimensions and continuity. Sand supply from shelf and mainland sources, continued at variable rates throughout six stages of the Last Eustatic Cycle (LEC). Differences in antecedent topography, in fluvial runoff volumes and relative sea level rise, including rapid overstepping-flooding events were critical in the regional distinctions. Leaving fewer and smaller relict fluvial delta behind, drowned shore and nearshore landforms on the shelf, only two large fluvial systems impacted the NE. The NW shelf and nearshore received abundant sediment supply from the Mississippi and lesser streams; indirectly by ravinement erosion of relict landforms and landward-directed cross-shelf sediment transfer. Sands originated mainly in relict deltas, fluvial, tidal channel, and inlet fill, as well as from submerged shore ridge remnants on the continental shelf. Shelf-margin deltas represented major secondary sources. Barrier formation categories and their respective importance in specific areas represent critical aspects of coastal development. The authorship of various formation concepts (e.g., Penck, Gilbert, McGee, Haage, Ganong, and Keilhack) often is still miscredited or remains unrecognized in the literature. Recognition of the stratigraphic and sedimentological diagnostic characteristics of the basic genetic barrier categories plays a key role in testing the validity of barrier evolution models. Closely-knit process-form relationships prevail between hydrodynamic (sea level change, storm, tidal current, overwash), sedimentary (shoal and island aggradation, rollover, and eolian accumulation) processes, and the resulting landform morphology. The most detailed accounts that deal with the development of the Alabama-Louisiana and Apalachicola barrier chains are particularly instructive. Almost all modern Gulf barriers with identifiable genetic background formed by shoal aggradation between 5.5 and 2.0 ka, during a marked deceleration in sea level rise. Concurrently, a sizable mainland strand plain complex extended the NE coastal plain. Tide- and storm-driven sediment transport to the large Mobile Bay ebb-delta, the focal role of composite east Dauphin Island in establishing the MS-LA island chain and the island-blocking role of Mississippi Delta advance were prime factors in a complex barrier history. Migration, paradoxically, may also occur by transgressive rollover concurrently with regressive landward progradation of island strand plains. Stream discharge and sand reworked from relict lithosomes on the shelf were insufficient in compensating for wave- and tide-related sediment loss. Gulf levels stayed below the present throughout the mid- and late Holocene.

Based on numerous publications and field work on the northern Gulf of Mexico coast, U.S. and other coastal areas, the present objective is to meet the need for an improved system of the definition and classification of barriers and their components. This topic includes morphological and functional interrelationships between beach ridges, barriers, and similar landforms. The term “beach ridge” was applied to a plethora of sub- and supratidal ridge types. In contrast, the presently proposed definition restricts the term to relict berm and foredune ridges, composed of sand or gravel and stabilized after isolation from shore zone of active processes. Barriers sustain paralic hydrology, sediment, biotic and nutrient condition in the coastal realm. In combination with adjacent lagoons and river deltas, these landforms represent physical barriers that lessen marine and related influences in brackish paralic-coastal environments. Benefiting from historic evolution of concepts and nomenclature, the task also involves the correction of misattributions rooted in early publications. Island, spit, and mainland barriers thus maintain lower salinities, reduce wave erosion, protect nutrient-rich paralic deposits and mitigate storm-induced inundation. “Fetch-limited lagoonal barrier islands” are considered valid barriers only in lagoons where they face inlet-transmitted full-marine conditions. “Mainland barriers” of several subcategories often are important sediment archives of past sea levels, hydroclimate, climate and tectonic-isostatic changes. They consist mostly of ridgeplains, at the start firmly attached or later migrated and welded to the mainland shore. In addition to local hydrodynamic and topographic influences, formation conditions may reflect overall paleogeographic conditions, including fluvial impact. In progradational (regressive) island settings, barriers consist mainly of strandplains that laterally may alternate with narrow spit-like sectors. They include only single, marine, respectively, lagoon-side shore ridges, with secondary dune fields sandwiched between them. Field observations and drillcore studies of microfossil- and sediment-based depositional facies suggest that most barrier islands originated by the often underestimated shoal-aggradation mode. Two additional genetic models are recognizable; a fourth remains hypothetical. Remotely-sensed images indicate that post-hurricane regeneration of the largest deltaic barrier chain in the Gulf of Mexico stalled in recent years. Late Pleistocene barriers significantly influenced present coastal landforms in two extensive regions. Dramatic changes in past decades indicate that sea-level rise and stormier future hydroclimate may likely accelerate the degradation of deltaic and non-deltaic barrier islands.

Delineation of the boundary between the Last Interglacial (LIG) and the last (Wisconsinan) Glacial Stage in North America represents a critical, yet unresolved issue. Subdivisions of the late Pleistocene are based on oxygen isotope, ice cover, and pollen stratigraphic data. Boundaries defined by isotope chronology hinge on complex interrelationships between δ18O in foraminifer tests, ice volumes stored on land, and coeval sea-level position. In the absence of adequate pollen-stratigraphic documentation, Pleistocene subdivision boundaries were harder to establish in North America than in Europe. Time-transgressive pollen zones revealed increased lengths of the climatically-floristically defined LIG from the European subarctic to the Mediterranean. Conflicting definitions of “Sangamon,” as representing only the last interglacial of minimum ice cover and higher temperatures or broadly defined, “sensu lato,” also incorporating early part of the Last (Wisconsinan) Glacial Stage persist in the North American literature. The exclusively interglacial age of the Sangamon Geosol, originally used in dating the Sangamonian Stage proved untenable. Designation of an “Eowisconsinan” interval corresponding to Susbtages MIS 5d-a also lacks merit. Despite climate- and vegetation-related discrepancies, pollen- and coastal deposit-based comparisons between Europe and North America during MIS 5 and the Holocene are useful in establishing the climate history of the North American Sangamonian and subsequent early Wisconsinan substages. An overarching MIS 5 cooling trend represented by scattered subarctic and high-mountain ice accumulation events followed the MIS 5e Eemian–Sangamonian temperature peak. Adoption of the general European practice that asymmetrically splits MIS 5 into a short MIS 5e interglacial and a long early Wisconsinan Glacial (MIS 5d-a) interval is preferred in North America as well. Subdivisions in the normalized δ18O curve that serve as the chronological framework and the wealth of European pollen data support this approach. While multiple pre-Sangamon Pleistocene marine-paralic intervals do occur on the NW Gulf coast, all pre-Sangamon Pleistocene marine and brackish-inshore deposits had been removed by erosion in the NE coastal plain. A single inshore-nearshore marine sediment and highstand interval is well-documented in this region. The LIG highstand sequence correlates with varied Eemian marine and paralic MIS 5e deposits encountered along northern and western European, Siberian, and additional shores. Apart from reliably dated Sangamonian S Florida coral reefs, identification and dating of LIG highstand deposits remain highly problematical in SE Atlantic shore terraces.

Basic differences between non-deltaic regressive and deltaic transgressive barrier islands reflect major contrasts in geological settings and sediment sources. Two island groups on the N. Gulf of Mexico provide unique perspectives of genetic and geomorphic contrasts applicable in a worldwide context. The near-extinction of the deltaic transgressive Chandeleur barriers and reduction of the sturdier prograded Mississippi–Alabama (MS–AL) chain are related to differences in sediment sources, storm, and anthropogenic impact. 160 years of documentary evidence points to contrasting geological settings, development history, sediment sources, and island morphology as responsible for different island erodibility and life spans. The non-deltaic chain received larger volumes of coarser, less erodible medium sand from the NE Gulf coast. Onshore sand flux from reworked delta deposits received from the retreating delta shoreface initiated the fragile, thin, and isolated transgressive Chandeleur islands. Fine-grained sand from unconsolidated muds of abandoned Mississippi-St. Bernard delta lobes maintained two distinct transgressive barrier island categories. In the absence of quantitative data on cross-shore transport, discrepancies between estimated littoral drift volumes and sand reserves for nourishment remain unexplained. Medium-sandy MS–AL barriers have resisted storm events far better than delta barriers. However, even the former chain did undergo 26 to 53% area reduction since 1848. Anthropogenic intervention stymied island growth. Emerging intertidal berm-basins formed on sandy shoal platforms in storm-eliminated sectors have contributed to partial island recovery. Delta attrition by wave erosion, tectonic, and compactional subsidence had accelerated delta lobe and barrier island decay. Intensive storm erosion culminating in and following Hurricane Katrina came close to eradicate the highly vulnerable Chandeleur barrier chain. Lacking adequate nourishment, after devastating cyclones only small islands reemerge and persist temporarily from the shoal belt. A four-stage barrier evolution model, globally applicable to transgressive deltaic barriers, is based on documented changes in late Holocene Mississippi sub-deltas.

The designation “coastal barrier”, related terms and their varied interpretations, and critical concepts in coastal plain and nearshore geology, underwent numerous reincarnations over the years. This designation was first applied to shore-parallel sand and gravel-built ocean-front islands and spits that separate open marine environments from paralic inshore environments that they shelter. A more recently recognized third category includes mainland strandplain, chenier, and dunefield barriers, either as original components of mainland coasts or subsequently welded to the coastal plain. A parallel but closely related terminology, developed mostly in Australia and Brazil, identifies several specific process categories recognizable in the three basic barrier groups. In the absence of lagoons and tidal wetlands, shore ridges that front and are contiguous with the mainland coastal plain have been mistakenly assigned to the barrier island category. Greatly diminished wave energy and consequently minor littoral drift volumes account for the relative rarity and much smaller dimensions of prograded landforms in estuarine islands and bay shore beaches in paralic-inshore basins. Due to wave and related littoral drift processes, however diminished in comparison with open marine shores, morphological similarities between aggraded landforms of open marine through estuarine settings are to be expected. However, the multitude of proposed “fetch-limited”, diminished wave energy estuarine islands, characterized by diminutive narrow beaches, sand spits, even occasional small dunes, and miniature strandplains, lack the essential functional attributes intrinsic to true barrier islands. While barrier islands may form even on relatively low wave energy and/or sediment-deficient marine shores, the term is inappropriate for islands located in low-energy paralic basins, including estuaries, or in coastal plains where frontal shore ridges are not backed by lagoons. Landforms erroneously ascribed in a variety of facies settings as barrier islands or islands frequently are mainland-attached beach ridge plains or beach-fringed marshlands instead. Narrow marshland beaches and mainland strandplain barriers are segmented into short islands by intervening narrow tidal creeks, fluvial distributary channels, or storm cuts that separate them. Improved fundamental and functional definitions of barrier islands, their comparison with the many genetic categories of islands and mainland shore beaches found in estuarine basins and wetlands should lead to a better understanding of paralic island evolution processes on the global scale.

Before its western sector was stranded and/or buried ca. 4.0–3.8 ka BP (3.9–3.7 ka 14C), the Mississippi–Alabama chain of regressive barrier islands extended well into present southeastern Louisiana. Westward-directed net littoral drift, ebb-deltas, and microtidal inlet bypassing were instrumental in the formation of elongated, narrow, sandy barrier platform sectors on which these islands, mostly of strandplain topography, have originally emerged. The development of sizable subtidal–intertidal berm basins, ringed by swash and foreshore berm ridges that emerged after storms, then filled by storm-mobilized sand, has aided posthurricane recovery. These processes are linked to discrete stages in aggradational barrier genesis. Increasingly frequent and destructive cyclones reduced island areas to laterally extensive subtidal barrier platform intervals. Enhanced overwash across lengthened platform sectors reduced drift volumes and consequently island progradation. Deepened ship channels facilitated san...

High Resolution Heavy Mineral Analysis (HRHMA) of late Pleistocene terrace samples, their Tertiary source rocks, and modern river sediments provided an effective tool for reconstructing sediment provenance and mapping heavy mineral provinces in southwest Louisiana. Each province, linked to a discrete source region, represents Pleistocene fluvial channel belts within which depositional activity was controlled by periods of climate, sediment supply, and sea level changes. Four coastal heavy mineral provinces have been identified. The Northern Province (NP), drained by the lower reaches of the Sabine and Calcasieu Rivers underlies level mid- and late Pleistocene coastal terrace surfaces and is distinguished by high-grade metamorphic assemblages (kyanite, staurolite, sillimanite) and abundant zircon, probably of Ouachita Mts. derivation. Transporting eroded Cretaceous, Tertiary, and Pleistocene coastal plain deposits, the modern Calcasieu and Sabine River sands in west-central and southwest Louisiana and east Texas, display identical heavy mineral composition to that of the NP. Level Late Pleistocene coastal terrace areas in the east represent the Red River Province (RRP) with dominant epidote, tourmaline, garnet, and zircon. Its mineralogy is influenced significantly by Paleozoic–Mesozoic sedimentary units that frame the drainage basin upstream. Modern Red River sands differ in their spectra both from Red River Pleistocene coastal terrace and valley terrace deposits, interpreted by temporal fluctuations in sediment supply initiating a variable contribution of detritus from different sources. Tributaries that drain formations with high concentrations of high-grade metamorphic minerals also affected Red River valley Pleistocene terrace deposits in west-central Louisiana, enriching them in kyanite and staurolite. The Mississippi Province (MP) occupies the eastern-southeastern area of the low, flat, gently seaward-sloping Prairie coastal terrace. Whereas modern Mississippi alluvium is dominated by hornblende, pyroxenes, and epidote, as the result of post-depositional dissolution, pyroxenes are rare in the MP. The Mixed Suite Province (MSP) reflects MP, RRP, and to a lesser degree, NP signatures and forms the Prairie fluvial coastal plain surface closer to the Texas state line. Raw data of the principal heavy minerals were used for statistical analysis. Statistical parameters proved consistent with mineralogy-derived reconstruction of sediment provenance and provinciality of heavy mineral suites, thus providing an independent and objective support to data interpretation. Optical and thermal luminescence dating at other Gulf locations [Otvos, E.G. (2005). Numerical chronology of Pleistocene coastal plain and valley development; extensive aggradation during glacial low sea levels. Quaternary Internat., 135 91–113.] supports the pre-Sangamon ages of the Intermediate Pleistocene terraces in the NP area. Sangamon (135–116 ka), Eowisconsin (114–76 ka), and Wisconsin (74–36 ka) dates characterize the four provinces in the low, level northern Gulf Prairie coastal plain. Refuting earlier assumptions that coastal plain aggradation occurred only during marine highstand phases, thermal and optical luminescence dates indicated that, despite the low Eowisconsin and Wisconsin eustatic sea levels of several preglacial and glacial stages and substages, coastal plain alluviation, paradoxically, recurred between 106 and 35 ka BP. An interesting outcome of our heavy mineral study is the recognition and dating of a previously undocumented, rare ash-fall event that originated in Caribbean andesitic volcanoes. It was identified by the presence of a volcanogenic heavy mineral suite, composed of pristine euhedral clinopyroxene, sphene, zircon, apatite, and hexagonal biotite. Unaffected by fluvial reworking, this suite was recovered from a MP sample, dated ca. 86 ka BP.

Based on more than 60 newly acquired luminescence ages, field information, and sediment granulometry data, a comprehensive chronostratigraphic framework is presented for Pleistocene alluvial coastal plain and valley terraces and coastal barrier trends on the northern Gulf of Mexico between Texas and NW Florida. Luminescence ages 216–188 ka from the second youngest coastal terrace coincide with the highstand during the second youngest Pleistocene Marine Isotope Stage (MIS 7a). TL ages from the oldest Prairie alluvium and OSL-ages from the Gulfport barriers; 124–116 ka, are consistent with the Sangamon interglacial substage, MIS 5e. Forest assemblages of the last two Pleistocene highstands are very similar to the present hard pine-dominated arboreal flora, established under warm-temperate, humid conditions. Ages of the youngest, Prairie–Beaumont coastal terrace overlap with the Eowisconsin (MIS 5d–5a) and Wisconsin (MIS 4 and 3) stages, associated with a sea-level range of −80 to −30 m. Coastal plain aggradation, not limited to interglacial highstands occurred during the much longer preglacial and glacial low sea-level stages. The Prairie–Beaumont coastal plain is a collage of seamlessly merged surfaces, aggraded between ca.135 and 30 ka in several alluviation stages. Periodically dry conditions, associated with enhanced slope erosion and consequent increase in sheetwash, colluvial, and fluvial sediment flux may have induced substantial aggradation at significant distances inland from the coeval shorelines during depressed preglacial and glacial sea-levels. This balance between erosion and sediment delivery may explain the diminished control of lower Eowisconsin and Wisconsin sea-levels (base-levels) on coastal entrenchment and aggradation. The combined effects of alluvial aggradation and subsequent uplift, modified by surface erosion produced the present coastal plain topography. An early post-Sangamon phase of deep entrenchment involved the Amite, Sabine, Neches, and the Pearl Valleys in Eowisconsin and early Wisconsin times. Except for the universal impact of the deep Last Glacial Maximum (LGM) entrenchment, the alluviation–incision cycles that produced two to four sets of terraces of aggradational and strath origin did not occur in all valleys synchronously and with identical effects. In certain valleys terrace aggradation coincided with the LGM record lowstand and the following late Wisconsin transgressive deglacial hemicycle. In contrast with intensive valley filling, inferred for the MIS 6 glacial interval, Holocene backfilling is far from complete in most late glacial entrenched valleys.

A bottomland flora that prevailed between ∼9900 and 6000 cal yr B.P. in a North Carolina stream valley may not reflect a regionally much wetter Atlantic climate, coeval with record drought in the Great Plains region and assumed dry Gulf coastal conditions. Such conditions were inferred for 6000 ± 1000 yr ago when the Bermuda High may have consistently occupied summer positions far to the NE. Arid episodes coeval with the Little River local wet interval are known from eolian sediments and pollen spectra in the Atlantic and the Gulf coastal plain. For multiple reasons, the regional extent, intensity, and duration of coastal aridity and alternating wet phases and the Bermuda High positions are not yet adequately constrained. The climate and edaphic causes for the steadily growing predominance of southern pines over hardwoods, achieved between ∼8900 and 4200 cal yr B.P. at different sites at different times are similarly still unresolved. New data from Shelby Lake, AL, reconfirms that no credible field or other proxy evidence exists for a previously postulated “catastrophic Gulf hurricane phase” in the late Holocene.

The Holocene history of the SE Louisiana–Mississippi coast and a recently revised regional sea-level curve fail to support the concept of an abrupt Gulf-wide decline from +2.0 m to −1.5 m and regression, with barrier aggradation between 4.1 and 3.9 14C ka BP, triggered by massive onshore sediment transport. A subsequent rapid rise followed by minor oscillations are equally unsubstantiated. Suggestions for an overwhelming role of onshore transport during the assumed regression or otherwise may not be justified. A small hardcrust lens in the continuous subaqueous barrier sequence does not prove major fall in sea-level, regional regression and subaerial exposure. Elevation and lithofacies of Florida shore scarps and radiocarbon-dated beach ridges fail to support sea-level fluctuations. The presented paleogeographic and sea-level models indicate continuous shoal and associated island formation between ca. 5.2 and 4.0 14C ka BP, uninterrupted by sea-level decline. Rapid progradation of a Mississippi River–St. Bernard delta lobe terminated barrier development.

Several new concepts have challenged the traditional view of northern Gulf coast sea-level history in recent years. Claims for higher than present mid and late Holocene sea-levels implied substantial sea-level oscillations. Such high-stands, known globally in low latitudes, have never been credibly documented on the Gulf or on unglaciated Atlantic coasts. Lacking direct indications, numerous lines of indirect proof have been suggested. Proxy evidence was based on items as the reinterpreted geological record of NW Florida lakes, estuaries, and depositional conditions of Texas coast landforms. Pleistocene barrier sectors, barrier islands, relict foredune ridges, washover fans/flood-tidal deltas, and other sedimentary units and their morphology were claimed as highstand indicators. Unrealistically old or young radiocarbon and luminescence dates, the result of diagenetic changes and other contamination have resulted in questionable conclusions. A large body of sediment, stratigraphic and geomorphic d...

A wealth of new data provides a well-constrained chronology of mid- to late Holocene coastal development in the Louisiana–Mississippi borderland that may also be utilized in a globally applicable sedimentation model. Barrier sand and deltaic mud sequences illustrate a process by which potential ground water and hydrocarbon reservoir rocks accumulated unusually rapidly and were preserved. Set against decelerating Holocene sea-level rise and locally variable subsidence rates, the study provides an example of the interplay between an emerging, prograding, and partially stranded barrier island chain, a sizable estuary, and several extensive delta lobes. Utilizing microfossil fauna-based depositional facies information and archaeological data, absolute dates helped to reconstruct the history of the Alabama–Louisiana barrier chain and deltas between ca. 5.7 and ca. 1.5 14C ka BP. Protected by the substantial dune cover that prevented island submersion, the regional eustastic transgression paradoxically was synchronous with significant progradational barrier and deltaic regression. The earliest barrier islands emerged ca. 4.6–4.4 14C ka BP (ca. 5.7–5.0 cal years) when sea level stood at ca. −1.0 to −1.5 m. These islands isolated Mississippi Sound from the greater Gulf of Mexico. The absence of a lagoonal-inshore sediment interval beneath the islands and the 3–15 m thick basal nearshore marine muddy–sandy unit that blankets the Pleistocene surface refutes the transgressive history of barrier initiation. The islands aggraded on a 3.0–16.5 m thick fine sandy shallow-marine regressive platform lithosome that, in turn, overlies a transgressive muddy–sandy nearshore marine lower sediment interval. Avulsion of the Mississippi River had abruptly reduced nearshore salinities by ca. 3.9–3.7 ka 14C BP. Renewed Mississippi delta growth induced rapid aggradation and progradation on the opposite Pearl River delta-mainland shore as well. Gulf influences have rapidly diminished in areas so affected. A new absolute chronology of the Mississippi–St. Bernard delta lobes constrains these events. Delta growth and mainland progradation first isolated, then severely constricted the Lake Borgne embayment. Accompanied by ongoing subsidence, the western barriers were stranded, then buried. Shoaling, related to St. Bernard delta progradation, interfered with westward littoral drift to maintain Cat Island. Archaeology provided important supplementary data for dating environmental changes. Refuting earlier suggestions that Native colonization rapidly followed delta complex formation, the earliest known Indian sites postdated the start of the associated St. Bernard delta lobe by 1.9–3.1 14C ka.

While the effects of major hurricanes have been intensively studied, less is known about the impact of the weaker but more frequent tropical cyclones, such as Hurricane Georges (1998). This hurricane, Category 2 at landfall, was non-typical in its effects. While high waves offshore and slow forward speed just before landfall resulted in island degradation, identical to that of Category 5 Hurricane Camille in 1969, the impact on the mainland was quite different. Only approximately 15% of the sand volume eroded by Camille in 1969 was removed from the Harrison County's mainland beach this time. Backshore areas of East Belle Fontaine Beach have prograded by 3–7 m. 20–90 cm vertical aggradation took place at several locations on its 10–45 m wide backshore. The short duration of hurricane-strength winds over the mainland and the availability of compensating sand supplies from adjacent sediment sources in the waning phase of the storm explain the limited extent of mainland shore erosion. Retreating shor...

Luminescence dating of extensive dune fields and associated eolian sandsheets provided a chronology of recently recognized Pleistocene and early Holocene dry climate episodes in the currently humid warm temperate northern–northeastern Gulf of Mexico region. Scattered parabolic dunes and clusters of intersecting parabolic dunes, along with elongated shore-transverse and shore-parallel dunes, developed. These landforms occur in a 390-km-long and 2- to 3-km-wide, semicontinuous belt in southeast Alabama and northwestern Florida. Dune elevations reach ± 22 m. Sangamon coastal barrier sectors were the primary source of the eolian sand. Deflation was coeval with early Wisconsin to mid-Holocene marine low sea-levels and associated distant shorelines. Early Holocene dune dates were synchronous, with indications of a hypsithermal dry interval in southeast Louisiana, the Yucatan, and the south Atlantic seaboard. Overlapping with dry episodes in Yucatan and the High Plains, Texas dunes and Louisiana and Texas prairie mounds, especially in the southwest Texas coast still dominated by dry climate, suggests intervals of early to late Holocene drought. The dates provide the basis for identifying and correlating Wisconsin, early, and late Holocene climate phases between currently semiarid and humid, coastal and interior areas. They contribute to future studies, including interregional paleoclimate modeling. Although Pleistocene coastal eolian deposition coincided with glaciation in the northern interior and with cooler temperatures of a reduced Gulf of Mexico, Holocene aridity phases may have been related to major variations in the position of high-pressure cells, storm tracks, and branches of the jet stream, and even to prolonged La Niña conditions.

Thirty-five sand hills that form six scattered groups rise abruptly from the flat late Pleistocene coastal plain in southeastern Louisiana. New studies confirm their eolian origin. For the first time, several late Wisconsin to early Holocene episodes of arid climate conditions have been recognized and dated in this currently humid warm-temperate subtropical region. Periods of dune formation and reactivation (28,800 to 7900 yr B.P.) were determined by the thermoluminescence method. The onset of the current climate in this Gulf coastal region postdates early Holocene time. The textural and structural homogeneity of the ridge lithosomes, good sorting of their sand fraction, and the dominantly orange hues of the dune sediments contrast with the underlying yellowish–brown to light-brown sandy silts and the well-stratified, occasionally gravelly sands of the underlying alluvial Prairie Formation. Sharply defined, unconformable ridge bases; symmetrical, oval, occasionally parabolic mound shapes; and steep slopes confirm the dune origins. The dominant orientations of ridges and ridge chains clearly reflect paleowind directions. Age comparison with dunes of the lower Mississippi Valley, the northeastern–eastern Gulf of Mexico coast, and south Atlantic coastal areas confirms the existence of at least seasonally dry climate conditions from early Wisconsin to middle Holocene times. The onset of the modern humid-subtropical climate phase in this region thus dates back only to the middle Holocene.

Beach ridges, frequent components of Quaternary coastal plains, and other coastal landforms, have been cited as indicators of the positions of ancient seashores and associated sea levels. Numerous authors utilized the term beach ridge for active and relict, usually wave-built supratidal and/or intertidal forms. Wind-built ridges have been only occasionally included in the definition. The term was applied also to submerged, landward-shifting, eventually stranded bars. A consistent redefinition of the term is highly desirable. Beach ridges should include all relict strandplain ridges, whether dominated by wave/swash-built or by eolian lithosomes. All active ridge-like shore features, regardless of dimensions, morphology, and origin are excluded. Because of the resistance of coarse-clastic ridges to wave and wind erosion, swash-built gravel or coarse shell (“storm”) ridges may build several meters above the level of high tide. Swash-built high berms, even on pure sandy beaches, exceed the highest tides during episodes of wind-induced, record water levels. Frequently but not always burying underlying low-relief “berm ridges” of berm lithosomes, sequences of relatively steep multiple foredunes are commonly named beach ridge plains. The narrow, subparallel relict foredunes that form these strandplains presently are designated as eolian beach ridges. Beach ridges, thus, are defined as relict, semiparallel, multiple wave- and wind-built landforms that originated in the inter- and supratidal zones. Until separated from the shoreline by progradation, sandy, pebbly or shell-enriched backshore berm ridges behind an active foreshore should not be considered beach ridges. Strandplain progradation is either continuous or, with the inclusion of subtidal (“cat's eye”) ponds, discontinuous. Contrary to claims, transgressive cheniers do not represent “true cheniers” alone; within their overall progradational context, cheniers, a special category of beach ridges bracketed by subtidal–intertidal mudflats, may be transgressive or regressive in character. Landward-driven, transgressive ridges should be designated beach ridges only after they are stabilized on intertidal flats. When recognizable between clearly identifiable intertidal and overlying eolian intervals, the horizontal interface between these lithosomes in beach ridges may help the reconstruction of ancient tide/lake levels. Diagnostic sedimentary textures, structures, and fossils, however, often may be unavailable in the deposits. Along with various types of elevated terraces composed of raised marine deposits and certain coastal landforms of erosional origin that occur worldwide, beach ridges of clearly proven wave-built origin may also serve as indicators of ancient higher-than-present sea levels.

Record high mid and late Holocene relative sea levels, significantly above the current sea level have been widely documented in certain areas of the Australo-Pacific region, Brazil (Lessa et. al. 2000), and parts of Africa and India that were not characterized by tectonic uplift and/or glacio-isostatic rebound. Such elevated sea levels are yet to be reliably identified on our Gulf and Atlantic shores. Radically departing from traditional Gulf of Mexico sea-level curves, Morton et al. (2000) now suggest recurring highstands throughout the entire second half of the Holocene Epoch. Their findings echo earlier ideas in the same vein (see Otvos 1995, 1997, 1999). However, a detailed analysis of several lines of evidence presented by Morton et al. raises serious doubts about these conclusions. Morton et al. accept the position of beach-ridge summits as indicative of higher-than-present elevations. Thus, seaward-declining beach ridge summits on Bolivar Peninsula, Galveston Island, and other coastal areas of Texas in their view reflect late Holocene sea levels that fell from their highest elevations. Although beach ridges are of wave-constructed (berm ridge), foredune, or composite origins (Otvos 2000), Morton et al. fail to distinguish adequately among these categories by sediment and landform analysis. The problem is that under favorable circumstances, wave-built berm ridges may be accurate markers of ancient sea levels or lake levels. Foredunes, on the other hand, aggrade to varying supratidal elevations, rising several meters above the associated high-tide level. For this reason, summit elevations of relict foredunes can not provide precise data on corresponding sea level (Otvos 1995). ### Foredune Ridges Foredune ridges often overlie, mask, and bury wave-built berm ridges (Beal and Shepard 1956). Because sand is usually readily available for wind transport on most Gulf shores, the upper interval in most beach ridges tends to be of eolian origin. Except when wave-built …