Paleodischarge reconstruction using oxbow lake sediments complicated by shifting hydrological connectivity
Introduction
Riverine flooding is among the most costly and deadly natural hazards globally that are expected to get worse with climate change (e.g., Davenport et al., 2021; Dottori et al., 2018; Jongman et al., 2012; Rojas et al., 2013; Wallemacq et al., 2018). The intensity of heavy precipitation has increased and is projected to continue due to anthropogenic global warming (e.g., Allen and Ingram, 2002; Easterling et al., 2017; Min et al., 2011). However, the correlation between riverine flooding and anthropogenic climate change is unclear (e.g., Dickinson et al., 2019; Mallakpour and Villarini, 2016), particularly at the river-basin scale (Dankers et al., 2014), adding significant uncertainty to flood hazard projection. There are several challenges in correlating changes of instrumental riverine flooding with anthropogenic climate change. First, instrumental records can be affected by land-use changes and river engineering (e.g., Munoz et al., 2018; Villarini and Smith, 2010). Second, most instrumental records are relatively short and not useful for separating the effects of anthropogenic climate change and long-term natural climate variability. Third, instrumental records may not capture the most extreme riverine flooding. Therefore, paleoflood hydrology emerged to use geomorphological (e.g., Baker, 1977, 2008; Knox, 1985; Kochel and Baker, 1982; Shi et al., 1987) and alluvial sedimentary proxies (Fuller et al., 2018; Ishii et al., 2017; Jones et al., 2010; Korponai et al., 2016; Leigh, 2018; Munoz et al., 2015, 2018; Peng et al., 2019, 2020, 2020; Toonen et al., 2017, 2020, 2020; Wolfe et al., 2006) to reconstruct long-term pre-instrument records to investigate climatic and anthropogenic controls on riverine flooding.
Alluvial sedimentation is episodic (Aalto et al., 2003; Shen et al., 2015) and may not scale with flood magnitude (Gomez et al., 1995; Lecce et al., 2004; Magilligan et al., 1998; Sambrook Smith et al., 2010). However, recent studies demonstrate that continuous sedimentary records in some alluvial niches may be used for quantitative high-resolution paleoflood reconstruction on centennial to millennial timescales (Toonen et al., 2020). The coarseness of historic infill sediments in oxbow lakes was shown to correlate with instrumental discharge (e.g., Munoz et al., 2018; Toonen et al., 2015), forming the basis to retrodict paleodischarge beyond instrumental records. Natural oxbow lakes are relict channels cut off from rivers due to flooding and channel migration (Constantine et al., 2010; Fisk, 1947; Hooke, 1995; Rowland et al., 2009; Toonen et al., 2012). These lakes are naturally dammed by plug bars at the cutoff location and receive coarser sediments from higher-stage floodwaters that entrain and transport more coarse sediments in suspension overtopping the plug-bar barrier. Therefore, extreme floods are indicated by silt to sand laminae in lake infill sequences (Munoz et al., 2018; Toonen et al., 2015). Plug-bar growth and river migration can change the hydrological connectivity between oxbow lakes and rivers, which may cause distinct shifts in the sedimentary architecture of the infill (Toonen et al., 2012). Furthermore, climate (Knox, 1993) and land-use changes (Syvitski et al., 2005) can change the sediment load of rivers, which may affect sedimentation in an oxbow lake. Therefore, the reliability of paleodischarge reconstruction depends on whether the sediment coarseness-discharge correlation established using the most recent infill sediments can be applied to older sediments of different infilling stages (Toonen et al., 2015, 2020).
To investigate this issue, we compared sediment coarseness of a ~60 yr old oxbow lake of the Waccamaw River in South Carolina, USA (Fig. 1) with peak annual discharge at a nearby United States Geological Survey (USGS) river gauge station (USGS 02110500). The lake infill shows distinct lithologic variations and can be divided into two sections. The instrumental discharge corresponding to the upper section was used to establish criteria to identify flood-event deposits and define the sediment coarseness-discharge correlation. The correlation was then used to retrodict discharges of event deposits in the lower section. The retrodicted discharges were compared with the instrumental data to test the paleodischarge reconstruction.
Section snippets
Study area
The Waccamaw River starts from Lake Waccamaw in North Carolina. It meanders south-southwest across the outer Atlantic Coastal Plain for about 230 river km before discharging into Winyah Bay in South Carolina (Fig. 1). The Waccamaw River watershed is located on the southern side of the Cape Fear Arch that is part of a structural high of the crystalline basement rock known as the Carolina Platform (Grow and Sheridan, 1988). The watershed is underlain primarily by the Cretaceous Pee Dee Formation
Field and laboratory methods
A 1.24 m long core, CL2-1 (33.9039 °N, 78.7216 °W), was taken from the oxbow lake using a piston corer in March 2018. The core was first scanned at 0.625 mm intervals down-core using a GE Optima CT scanner at Conway Medical Center in Conway, South Carolina, and then split into two halves. One half was archived, and the other half was subsampled for grain-size analysis at 1 cm intervals down-core for the 70 cm of lake infill. The sediments below 70 cm are medium to coarse sand channel lag
Lithology and grain-size analysis
The infill sediments are dark to brown-gray mud to sandy mud and rich in leaf and woody debris. The CT image shows a predominantly dark gray color with intervening light gray bands, corresponding to the muddy matrix and sand laminae, respectively (Fig. 4a). Sand laminae are abundant throughout the core and become more frequent and thicker below ~40 cm (Fig. 4b).
The grain-size distributions of the infill sediments show multiple modes at 10–20 μm, ~100 μm, and 200–400 μm, respectively (Fig. 3a).
Discussion
The infill of CL2 at the location of CL2-1 consists of sand laminae interbedded in organic-rich mud. It shows a fining upward trend, along with an upward decrease in density, thickness, and coarseness of the sand laminae (Fig. 4b). Superimposed on this trend, D90 and its variability show abrupt decreases above ~40 cm, which is in sharp contrast with the instrumental discharge record that shows more floods exceeding the 5-year threshold with larger magnitudes corresponding to the upper 40 cm of
Conclusions
We studied the infill sediments of an oxbow lake of a small river in the US Atlantic Coastal Plain. We found that sedimentation occurred in multiple stages that are controlled by plug-bar development. In addition, the lake-infill sedimentary architecture resembles those of continental-scale rivers, suggesting that natural infilling processes in cutoff lakes may be scale-independent.
Comparing infill sedimentary architecture with instrumental discharge indicates that local morphology applies the
CRediT author statement
Zhixiong Shen: conceptualization, methodology, formal analysis, data curation, visualization, supervision, and writing – review & editing; Molly Aeschliman: investigation, formal analysis, writing – original draft; Nicholas Conway: investigation, formal analysis, writing – review & editing.
Data availability
Grain-size data of this research are publicly available under Creative Commons Attribution 4.0 International (CC BY 4.0) license in Coastal Carolina University's Digital Common data repository (Shen, 2021).
Declaration of competing interest
The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.
Acknowledgements
This work was supported by the Professional Enhancement Grant to Z. Shen, and the Marine Science Undergraduate Research Fellowship to M. Aeschliman by Coastal Carolina University. We thank K. Washington for assistance in the laboratory, and J. Tompkins from the Conway Medical Center for providing CT scan services. The paper was improved with comments by four anonymous reviewers.
References (57)
Paleoflood hydrology: origin, progress, prospects
Geomorphology
(2008)- et al.
An end-member algorithm for deciphering modern detrital processes from lake sediments of Lake Donggi Cona, NE Tibetan Plateau, China
Sediment. Geol.
(2012) - et al.
An analysis and comparison of observed Pleistocene South Carolina (USA) shoreline elevations with predicted elevations derived from Marine Oxygen Isotope Stages
Quat. Res.
(2014) - et al.
Storm-generated Holocene and historical floods in the Manawatu river, New Zealand
Geomorphology
(2018) River channel adjustment to meander cutoffs on the River Bollin and River Dane, northwest England
Geomorphology
(1995)- et al.
Middle to late Holocene flood activity estimated from loss on ignition of peat in the Ishikari lowland, northern Japan
Global Planet. Change
(2017) - et al.
Global exposure to river and coastal flooding: long term trends and changes
Global Environ. Change
(2012) Responses of floods to Holocene climatic change in the upper Mississippi valley
Quat. Res.
(1985)Vertical accretion sand proxies of gaged floods along the upper Little Tennessee river, Blue ridge Mountains, USA
Sediment. Geol.
(2018)- et al.
Investigating the relationship between the frequency of flooding over the central United States and large-scale climate
Adv. Water Resour.
(2016)
An improved method for paleoflood reconstruction and flooding phase identification, applied to the Meuse River in The Netherlands
Global Planet. Change
Paleoflooding reconstruction from Holocene levee deposits in the Lower Meuse valley, The Netherlands
Geomorphology
Climate change and river floods in the European Union: socio-economic consequences and the costs and benefits of adaptation
Global Environ. Change
Investigation and verification of extraordinarily large floods on the Yellow River
J. Hydrol.
Assessment of hydrologic vulnerability to urbanization and climate change in a rapidly changing watershed in the Southeast U.S
Sci. Total Environ.
Lower Rhine historical flood magnitudes of the last 450 years reproduced from grain-size measurements of flood deposits using End Member Modelling
Catena
Muddled or mixed? Inferring palaeoclimate from size distributions of deep-sea clastics
Sediment. Geol.
Episodic sediment accumulation on Amazonian flood plains influenced by El Niño/Southern Oscillation
Nature
Constraints on future changes in climate and the hydrologic cycle
Nature
Stream-channel response to floods, with examples from central Texas
Geol. Soc. Am. Bull.
Overbank sedimentation rates in former channel lakes: characterization and control factors
Sedimentology
Controls on the alluviation of oxbow lakes by bed-material load along the Sacramento River, California
Sedimentology
First look at changes in flood hazard in the inter-sectoral impact model intercomparison project ensemble
P. Natl. Acad. Sci.
Contribution of historical precipitation change to US flood damages
P. Natl. Acad. Sci.
Seasonality of climatic drivers of flood variability in the conterminous United States
Sci. Rep.
Increased human and economic losses from river flooding with anthropogenic warming
Nat. Clim. Change
Precipitation change in the United States
Efficient retention of mud drives land building on the Mississippi Delta plain
Earth Surf. Dynam.
Cited by (9)
Holocene Thermal Maximum paleofloods improve flood frequency analyses in the lower Tennessee River Basin (USA)
2024, Global and Planetary ChangeHydrologic connectivity and morphologic variation of oxbow lakes in a pristine alpine fluvial system
2023, Journal of HydrologyIncorporating alluvial hydrogeomorphic complexities into paleoflood hydrology, magnitude estimation and flood frequency analysis, Tennessee River, Alabama
2022, Journal of HydrologyCitation Excerpt :The paleoflood records used in this research were reconstructed from two sediment cores sampled from a relict, natural levee located on the left bank of the Tennessee River, approximately 10 km from Guntersville, AL (USA). Many paleoflood studies conducted in alluvial settings target low-lying, floodplain environments distanced from the channel (oxbows, meander scars, paleochannels, etc.) (see for example, Toonen et al., 2015; Munoz et al. 2018; Shen et al., 2021). Such locations have some distinct advantages since their positions away from the active channel can mean less erosion through time, and in some cases, a sense of flood magnitude can be ascertained based on the inundated area required for flooding to have occurred at the location.
Palaeoflood level reconstructions in a lowland setting from urban archaeological stratigraphy, Rhine river delta, the Netherlands
2022, CatenaCitation Excerpt :In dynamic lowland settings, the rise of water level with discharge is modest and reconstruction of past morphology is complex (van der Meulen et al., 2020). Hence, alternative methods for palaeoflood reconstruction have been developed for alluvial reaches that focus on relative magnitudes and circumvent the requirement of knowing past flood levels (Toonen et al., 2013; Munoz et al., 2018; Peng et al., 2019; Toonen et al., 2020; Sheng et al., 2021). Still, in light of recent advances in palaeoflood modelling (Dierx et al., 2021; van der Meulen et al., 2021), it is particularly relevant to reconstruct flood levels for alluvial settings, as these are generally characterized by dense population and high flood risk.
A novel binary pipette splitting sediment subsampling method for improving reproducibility in laser-diffraction particle-size analysis
2021, MethodsXCitation Excerpt :Particle-size distribution (PSD) analysis is fundamental to sedimentary geology and has been used for investigating sedimentation processes [e.g. 1–4], paleoclimate [e.g. 5–7], and guiding soil classification [e.g. 8], for instance. Recent development in unmixing PSD [e.g. 9,10] revitalizes PSD analysis for paleoclimatic [e.g. 11,12] and paleohydrological studies [e.g. 13,14,15]. Laser-diffraction analysis has been established as one of the standard methods for PSD measurement because it has high precision, reduces measurement time, and works with a small sample size (<1 g) that enables, for example, ultrahigh-resolution (mm-scale) stratigraphic studies.
Morphometry of Tidal Meander Cutoffs Indicates Similarity to Fluvial Morphodynamics
2024, Geophysical Research Letters