An official website of the United States government
River deltas are dynamic systems whose channels can widen, narrow, migrate, avulse, and bifurcate to form new channel networks through time. With hundreds of millions of people living on these globally ubiquitous systems, it is critically important to understand and predict how delta channel networks will evolve over time. Although much work has been done to understand drivers of channel migration on the individual channel scale, a global-scale analysis of the current state of delta morphological change has not been attempted. In this study, we present a methodology for the automatic extraction of channel migration vectors from remotely sensed imagery by combining deep learning and principles from particle image velocimetry (PIV). This methodology is implemented on 48 river delta systems to create a global dataset of decadal-scale delta channel migration. By comparing delta channel migration distributions with a variety of known external forcings, we find that global patterns of channel migration can largely be reconciled with the level of fluvial forcing acting on the delta, sediment flux magnitude, and frequency of flood events. An understanding of modern rates and patterns of channel migration in river deltas is critical for successfully predicting future changes to delta systems and for informing decision makers striving for deltaic resilience.
more »
« less
Characterization of Deltaic Channel Morphodynamics From Imagery Time Series Using the Channelized Response Variance
River deltas are complex, dynamic systems whose channel networks evolve in response to internal and external forcings. To capture these changes, methods to extract and analyze deltaic morphodynamics automatically using available remotely sensed imagery and experimental observations are needed. Here, we apply a promising method for the automatic extraction of channel presence called RivaMap, on both synthetic and experimental data sets, to investigate the changes experienced by the system in response to five changes in forcings. RivaMap is an automated method to extract nonbinarized channel locations from imagery based on a singularity index that combines the multiscale first and second derivatives of the image intensity to favor the identification of curvilinear features and suppress edges. We quantify how the channelization varies by computing the channelized response variance (CRV), which we define as the variance of each pixel's singularity index response through time. We find that increasing magnitudes of sediment inflow (Qs) and water inflow (Qw) result in corresponding increases in the maximum CRV. We find that increasing the ratio ofQstoQwresults in increased number of channelized areas. We see that adding cohesion to the exposed sediment surface of the experimental delta results in decreased magnitude and decreased number of channelized areas in the CRV. Finally, by observing changes to the CRV over time, we are able to quantify the timescale of internal channel reorganization events as the experimental delta evolves under constant forcings.
more »
« less
- PAR ID:
- 10449143
- Publisher / Repository:
- DOI PREFIX: 10.1029
- Date Published:
- Journal Name:
- Journal of Geophysical Research: Earth Surface
- Volume:
- 124
- Issue:
- 12
- ISSN:
- 2169-9003
- Page Range / eLocation ID:
- p. 3022-3042
- Format(s):
- Medium: X
- Sponsoring Org:
- National Science Foundation
More Like this
-
-
Abstract Coastal deltaic aquifers are vulnerable to degradation from seawater intrusion, geogenic and anthropogenic contamination, and groundwater abstraction. The distribution and transport of contaminants are highly dependent on the subsurface sedimentary architecture, such as the presence of channelized features that preferentially conduct flow. Surface deposition changes in response to sea‐level rise (SLR) and sediment supply, but it remains unclear how these surface changes affect the distribution and transport of groundwater solutes in aquifers. Here, we explore the influence of SLR and sediment supply on aquifer heterogeneity and resulting effects on contaminant transport. We use realizations of subsurface heterogeneity generated by a process‐based numerical model, DeltaRCM, which simulates the evolution of a deltaic aquifer with different input sand fractions and rates of SLR. We simulate groundwater flow and solute transport through these deposits in three contamination scenarios: (a) vertical transport from widespread contamination at the land surface, (b) vertical transport from river water infiltration, and (c) lateral seawater intrusion. The simulations show that the vulnerability of deltaic aquifers to seawater intrusion correlates to sand fraction, while vertical transport of contaminants, such as widespread shallow contamination and river water infiltration, is influenced by channel stacking patterns. This analysis provides new insights into the connection between the depositional system properties and vulnerability to different modes of groundwater contamination. It also illustrates how vulnerability may vary locally within a delta due to depositional differences. Results suggest that groundwater management strategies may be improved by considering surface features, location within the delta, and the external forcings during aquifer deposition.more » « less
-
Abstract Deltaic river networks naturally reorganize as interconnected channels move to redistribute water, sediment, and nutrients across the delta plain. Network change is documented in decades of satellite imagery and laboratory experiments, but our ability to measure and understand channel movements is limited: existing methods are difficult to employ efficiently and struggle to distinguish between gradual movements (channel migration) and abrupt shifts in river course (channel avulsions). Here, we present a method to extract channel migration from plan‐view imagery using particle image velocimetry (PIV). Although originally designed to track particles moving in a fluid, PIV can be adapted to track channels moving on the delta surface, based on input estimates of channel width, migration timescale, and maps of the wet‐dry interface. Results for a delta experiment show that PIV‐derived vector fields accurately capture channel‐bank movements, as compared to manually drawn maps and an independent image‐registration technique. Unlike other methods, PIV targets the process of channel migration, excluding changes associated with channel avulsions and overbank flow. PIV‐derived migration rates from the experiment span an order of magnitude and are reduced under lower sediment supply and during sea‐level rise, supporting recent models. Together, results indicate that PIV offers a fast and reliable way to measure channel migration in river networks, that channel migration rates under non‐cohesive conditions can displace channels a distance comparable to their width in the time needed to aggrade ∼10% of the channel depth, and that migration direction is ∼60% orthogonal to mean flow direction and ∼40% flow‐parallel overall.more » « less
-
Abstract Notwithstanding the large number of studies on bedforms such as dunes and antidunes, predicting equilibrium bedform type and geometry for a given flow regime, sediment supply and caliber remains an open problem. Here, we present results from laboratory experiments specifically designed to study how upper regime bedform type and geometry vary with sediment supply and caliber. Experiments were performed in a sediment feed flume with flow rates varying between 5 and 30 l/s and sand supply rates varying between 0.6 and 20 kg/min. We used both uniform and non‐uniform sands with geometric mean diameters varying between 0.22 and 0.87 mm. Analysis of our data and data available in the literature reveals that the ratio of total (bedload plus suspension) volume transport rate of sediment to water dischargeQs/Qwplays a prime control on upper regime equilibrium beds. Equilibrium bedforms transition from washed out dunes (lower regime) to downstream migrating antidunes (upper regime) forQs/Qwbetween 0.0003 and 0.0007. For values ofQs/Qwgreater than 0.0015, the bedform length increases withQs/Qw. At these high values ofQs/Qw, equilibrium in fine sand is characterized by upstream migrating antidunes, cyclic steps, and significant suspended load. In experiments with coarse sand, equilibrium is characterized by plane bed with bedload transport in sheet flow mode. Standing waves form at the transition between downstream migrating antidunes and upstream migrating bedforms.more » « less
-
Abstract We present a novel quantitative test of a 50‐year‐old hypothesis which asserts that river delta morphology is determined by the balance between river and marine influence. We define three metrics to capture the first‐order morphology of deltas (shoreline roughness, number of distributary channel mouths, and presence/absence of spits), and use a recently developed sediment flux framework to quantify the river‐marine influence. Through analysis of simulated and field deltas we quantitatively demonstrate the relationship between sediment flux balance and delta morphology and show that the flux balance accounts for at least 35% of the variance in the number of distributary channel mouths and 42% of the variance in the shoreline roughness for real‐world and simulated deltas. We identify a tipping point in the flux balance where wave influence halts distributary channel formation and show how this explains morphological transitions in real world deltas.more » « less
