Complex sedimentary processes in large coastal embayments and their potential for coastal morphological and paleo tropical cyclone studies: A case study from Choctawhatchee Bay Western Florida, U.S.A

Highlights

• This article provides a high-resolution record of the evolution of Choctawhatchee Bay Florida, USA during the Holocene.

• The article discusses the spatial variability of different sedimentary processes within the Choctawhatchee Bay.

• It evaluates the suitability of large back-barrier coastal systems in hurricane-prone areas for paleo hurricane studies.

• The article gives evidence for the hurricane variability in the area during the last 8 kyrs

Abstract

Storminess and sea-level can both have a significant impact on landforms in cyclone-prone coastal regions, although much of our understanding comes from short-timescale modern observations. This study aims to understand the variability of sediment transport and deposition in the Choctawhatchee Bay/Santa Rosa Island in the northern Gulf of Mexico, establishing the dominant sediment transport processes and morphological response of the barrier system to long-term variations in storminess and rising sea-levels.

Here, we study the spatial and temporal changes in physicochemical properties of the sedimentary record of Choctawhatchee Bay to examine the character and fidelity of records of storm impacts spanning the Holocene. Proxies for marine and terrestrial conditions in the cores situated closer to the present barrier (proximal) show that sedimentation in coastal areas and marine influence of the bay during the last ~8000 yrs. were mainly determined by barrier response to the Holocene transgression and changes in storminess. In contrast, sedimentation close to the landward shore was governed by terrigenous input. The correlation of grain size and terrigenous proxies with regional hurricane records indicates that hinterland erosion by the rainfall during hurricane events is likely the dominant terrigenous sediment transport mechanism in areas close to the landward shore of the bay. These results suggest that sediment archives in large coastal deposition environments are equally suitable for sea level and cyclone modulated coastal morphological studies and paleo tropical cyclone studies, depending on the location, selected with an understanding of sedimentation processes in the vicinity.

Introduction

Tropical cyclones (TC) in the Indo-Pacific and hurricanes in the Atlantic pose a growing threat because of the increasing population and wealth in TC-prone areas (Pielke et al., 2008). Therefore, researchers are interested in determining the mechanisms governing TC frequency and intensity changes. Instrumental records span time intervals far too short to appropriately assess risk and understand the climatic forcing responsible for TC activity changes. In contrast, sediment archives preserved in various coastal depositional settings, such as coastal bays, lagoons, marshes, sinkholes, and blue holes on carbonate platforms, offer unique insights into the variability of storm activity over 1000's years. Knowledge of past storm patterns can help identify the forcing factors that control storminess, giving insights into potential future variability, which is a valuable asset for coastal management.

Site selection for paleo-TC studies is challenging because coastal systems are also undergoing complex changes in response to sea-level rise. Variations in storminess can also alter the sensitivity of a given site to storm impacts and influence the preservation potential of storm-induced deposits. For example, sediment supply and transport pathways may change, altering the character of storm-induced sediment transport and deposition (Sallenger Jr, 2000; Otvos, 2011). Storm surges are erosive events that can erode beach, dune, and barrier bars and back-barrier lowlands (Reimnitz and Maurer, 1979; Morton and Barras, 2011). During high-energy events, erosion can even remove portions of the sedimentary record (Morton, 2002; Eisemann et al., 2018). About 48 cm of sediment was eroded by hurricane Harvey which made landfall in 2017, in the San Jacinto Estuary (Du et al., 2019).

The well-developed Holocene sea-level and TC records of the northern Gulf of Mexico make Choctawhatchee Bay on the panhandle of Florida an ideal laboratory (Brandon et al., 2013; Lane et al., 2011; Rodysill et al., 2020 Milliken et al., 2008; Donnelly and Giosan, 2008to examine the effects of different coastal processes on the sediment record. Here, we studied the spatial and temporal changes in physicochemical properties of the sedimentary record of Choctawhatchee Bay (Fig. 1) to examine the character and fidelity of records of storm impacts spanning the Holocene.

Choctawhatchee Bay, FL is a drowned river valley system currently separated from the Gulf of Mexico by the Okaloosa peninsula and Santa Rosa Island (30.49^oN, −86.58^oW and 30.39^oN, −86.10^oW; Fig. 1). The bay opens to the Gulf of Mexico through “East Pass” (Destin Pass), which today is a nearly 500 m wide inlet at the east end of the Okaloosa barrier fronting the bay.

According to data from Pensacola, FL, the area receives the highest rainfall during the summer (Jun-Sep) and winter (Jan-Mar) (NOAA). During most fall and winter, rainfall occurs in the region mainly associated with frontal systems from the northwestern United States. During most spring and summer, rainfall occurs mainly by convective processes and tropical storms (Baigorria et al., 2007). The Choctawhatchee River discharges into the bay about 37 km east of East Pass. Monthly discharge values at Bruce, FL (USGS 02366500) (Fig. 1) show that the Choctawhatchee River discharge is high during December – April, with the highest discharge in March. Annual discharge data for the period 1931–2017 show that annual discharge varied between 311 and 57 m³/s. The tidal range within Choctawhatchee Bay is averaging 0.15 m (Ruth and Handley, 2006). Choctawhatchee Bay connects to Pensacola Bay by a narrow back-barrier channel formed behind Santa Rosa Island. According to Otvos (1985), over half of Santa Rosa Island is underlain by relict Pleistocene Gulfport formation made of barrier sand. Fort Walton Beach on the eastern portion of Santa Rosa Island is composed of Holocene sandy sediment. In contrast, the Okaloosa peninsula to the east of Destin Pass is largely the Pleistocene Gulfport formation.

While some investigators have argued for a complex Holocene sea-level history for the Gulf of Mexico, with a series of high and low stands (e.g., Tanner et al., 1992; Morton et al., 2000), dated sea-level indicators demonstrate that the region experienced a largely monotonic increase in sea level over the last 10,000 years (Milliken et al., 2008; Donnelly and Giosan, 2008) (Fig. 2). In general, the relative sea-level in the Gulf of Mexico increased about 20 m over the Holocene at a gradually decreasing rate. For example, the overall rate of rising over the last few millennia was only about 0.5 mm/year, compared to approximately 4 mm/year in the early Holocene (Milliken et al., 2008). According to the local tide gauge (Station 8,729,840, Pensacola, FL), the current sea-level rise rate over the last century is approximately 2.3 mm/year.

(https://tidesandcurrents.noaa.gov/sltrends/sltrends_station.shtml?id=8729840).

The barrier complexes of the Gulf of Mexico formed during the Holocene as post-glacial sea-level rise slowed down after around 5000 yrs. BP (e.g., Otvos Jr, 1970; Otvos, 1982; Otvos, 1985; Rodriguez et al., 2004; Rodriguez and Meyer, 2006; Törnqvist et al., 2004). Over the last 150 years, sea-level rise rates have increased dramatically in response to climate warming (e.g., Kemp et al., 2011), and sea-level rise is likely to accelerate as warming continues (Kopp et al., 2016). Moore et al. (2010) reveal that substrate composition, followed in rank order by substrate slope, sea-level rise rate, and sediment supply rate, are the most critical factors determining barrier island response to sea-level rise. Simple morphodynamic models suggest width and height drowning are the key modes of barrier failure in response to accelerated sea-level rise (Ashton and Lorenzo-Trueba, 2018). Because of the recent increase in the rate of sea-level rise, barriers may be more susceptible to breaching and overtopping during storm events today and into the future than they were 150 years ago. Studying sedimentological and geochemical changes in the proximal and distal sites to the barrier would reveal the barrier response to the sea level and its environmental impact on the bay, providing important clues about potential future changes in the system. Better constraints on the character of the evolution of back-barrier bay systems will provide important context for interpreting potential storm reconstructions from these environments.

In the Gulf of Mexico, the most dominant types of severe storms are TCs. In contrast to the slowly decelerating rates of long-term sea-level rise, hurricane activity in the northeastern Gulf of Mexico appears to have changed significantly over the latter half of the Holocene (Fig. 2). Reconstructions of event beds likely related to hurricanes from coastal ponds and embayments in the northeastern Gulf of Mexico indicate significant centennial-scale variability within much of the last 4500 years, experiencing more frequent intense hurricane strikes than were suffered historically (Brandon et al., 2013; Lane et al., 2011; Rodysill et al., 2020; Fig. 2). In particular, the intervals between 4500 and 2300 years ago and 1500 and 700 years ago were much more active in terms of intense hurricane activity in the northeastern Gulf of Mexico than in the last 700 years (Lane et al., 2011; Brandon et al., 2013).

Over the last few decades, TC storm surges overwashed the western part of Santa Rosa Island and deposited sand sheets in Santa Rosa Sound. Several TCs significantly impacted the eastern Santa Rosa barrier island over the last few decades. Hurricane Dennis, which made landfall in 2005 as a Category-3 storm and created over 3 m surge at eastern Santa Rosa barrier; Hurricane Ivan, which made landfall in 2004 as a Category −3 hurricane and created 3–4 m storm surge; Hurricane Opal, which made landfall in 1995 as a Category −4 hurricane and created about 3 m surge (National Hurricane Center, 2012) are some of them. During Hurricane Ivan, extensive inundation and overwash occurred within 100 km from the storm center at landfall. Significant beach and dune erosion occurred as far as 300 km east of the storm center (Wang et al., 2006). The island morphology changed from a discontinuous foredune backed by hummocky back-barrier dunes and maritime forest (at the cuspate headlands) to washover terraces at the headlands and washover corridors between headlands (Houser, 2008).

Modern as well as Paleoclimate records for mid-late Holocene show tight coupling between the strength of El Ninὸ-Southern Oscillation (ENSO) and winter rainfall intensity (Donders et al., 2005; Cronin et al., 2002). Records show establishing modern ENSO frequency between 7000 and 5000 yrs. BP and increasing the ENSO intensification after 3500 yrs. BP (Donders et al., 2005). While the future of hurricane activity in the Gulf of Mexico is uncertain, basic theory (e.g. Emanuel, 1987), statistical relationships between large-scale climate parameters and TC activity(e.g., Mann et al., 2009), and downscaling global models (e.g., Emanuel, 2013; Knutson et al., 2007; Knutson et al., 2015) indicate that the intensity of TCs may increase as earth's climate continues to warm from increased concentrations of greenhouse gases in the atmosphere. Some studies have concluded that we already see an increase in the frequency of the most intense storms (Elsner et al., 2008; Kossin et al., 2013; Webster et al., 2005) and overall North Atlantic TC activity has increased over the last several decades. However, these conclusions have been challenged due to the brevity and inconsistency of the instrumental record (Landsea et al., 2006). Alternatively, Sobel et al. (2016) suggest that detectable TC activity trends will only emerge if greenhouse gas forcing continues to outpace anthropogenic aerosol-related cooling over the coming decades. Many modeling studies point toward increasing TC intensity over the coming century, but the significant regional variance is likely with some locations seeing an increase in the intensity of TCs and others seeing little change or even decrease inactivity. The modeling of Knutson et al. (2015) points toward more Category 4 and 5 TCs in the Gulf of Mexico by 2100 CE.

Detailed records of past hurricane activity are essential for understanding hurricane variability on multidecadal to centennial time scales. The main objective of this study is to understand the complex sedimentary processes in back-barrier depositional environments and fundamentally assess their potential for hurricane reconstructions using Choctawhatchee Bay, Florida, an ideal site to study such processes in a large hurricane-prone back-barrier bay.