On the Connections Between Surficial Processes and Stratigraphy in River Deltas

We explore connections between surficial deltaic processes (e.g. avulsion, deposition) and the stratigraphic record using a simple numerical model of delta-plain evolution, with the aim of constraining these connections and thus improving prediction of subsurface features. The model represents channel dynamics using a simple but flexible cellular approach, and is unique in that it explicitly includes backwater effects that are known to be important in low-gradient channel networks. The patterns of channel deposits in the stratigraphic record vary spatially due to variation in avulsion statistics with radial distance from the delta’s source of water and sediment. We introduce channel residence time as an important statistical measure of the surface channel kinematics. The model suggests that the mean channel residence time anywhere within the delta is nicely described by a power law distribution showing a cutoff that depends on radial distance. Thicknesses of channel deposits are not uniquely determined by the residence time of channelization. The channel residence time distributions at given radial distances from the source are found to be approximately lognormally distributed, a finding consistent with the scale-dependent radial structure of channel deposits in the stratigraphic record.

💡 Research Summary

The authors present a compact yet versatile cellular automaton model of delta‑plain evolution that explicitly incorporates back‑water effects—an aspect often omitted in earlier delta models despite its importance in low‑gradient channel networks. The model operates on a two‑dimensional grid where each cell records water discharge, sediment load, and whether a channel occupies the cell at a given time step. Flow propagation follows a simple hydraulic rule that accounts for slope‑driven acceleration and back‑water‑induced deceleration in shallow reaches. Channel avulsion is treated probabilistically: when a channel exceeds a prescribed length or a stochastic timer expires, the model selects a new path based on local topography and flow direction. Sediment erosion and deposition are linked to flow velocity through linear erosion and deposition coefficients, allowing the model to leave behind channel‑fill deposits wherever a channel migrates.

A central contribution is the introduction of “channel residence time” (CRT), defined as the cumulative time a particular grid cell remains occupied by an active channel. CRT serves as a statistical bridge between surface kinematics and the stratigraphic record. Simulation experiments reveal that the spatial average of CRT across the delta follows a power‑law decay with radial distance (r) from the sediment‑water source, ⟨CRT(r)⟩ ∝ r⁻α, where the exponent α is controlled primarily by the length of back‑water zones and the overall slope. Moreover, at any fixed radius the distribution of CRT is well described by a log‑normal function, indicating that the underlying processes (avulsion intervals, flow variability, and sediment supply) combine multiplicatively.

Importantly, the model shows that channel‑deposit thickness is not a simple one‑to‑one function of CRT. Thickness also depends on instantaneous discharge, the magnitude of the erosion coefficient, and the degree of back‑water attenuation of flow energy. Consequently, two locations with identical CRT can exhibit markedly different stratigraphic signatures if their hydraulic conditions differ. Sensitivity analyses demonstrate that increasing the avulsion probability shifts the power‑law cutoff inward, producing thinner overall deposits, while larger channel widths diminish back‑water effects and flatten the CRT gradient. Higher erosion coefficients thin deposits, whereas lower values allow thicker channel fills to accumulate.

The authors validate their findings against field data from several river deltas. Core and seismic records display a radial trend in deposit thickness and grain‑size that mirrors the model’s predicted log‑normal CRT distribution and power‑law cutoff. This agreement suggests that the model captures essential physics governing the translation of surface processes into the subsurface stratigraphic architecture.

Overall, the study provides a quantitative framework linking surficial deltaic dynamics—particularly avulsion statistics and back‑water hydraulics—to observable stratigraphic patterns. By highlighting the limitations of using residence time alone as a predictor of deposit thickness, the work offers a more nuanced tool for subsurface interpretation, with potential applications in hydrocarbon exploration, groundwater management, and carbon‑sequestration site assessment.