Asymptotic hydrographs and anomalous dispersion in mass-conserving storage cascades

Abstract: Sums of independent exponential random variables lead to the Erlang distribution, providing a direct probabilistic route from exponential waiting times to the integer-shape gamma law. This paper investigates how this classical construction changes when the exponential waiting-time density is replaced by the $q$-exponential density of nonextensive statistics. Our main result is an analytical asymptotic expression for the outflow of a mass-conserving cascade of reservoirs driven by a $q$-exponential waiting-time kernel. In the critical case $q=5/3$, the large-cascade flow rate converges to a stable Lévy density whose time argument is shifted by a Galilean-type transformation. This shifted Lévy law gives the asymptotic hydrograph of the cascade. We also found that for the entire regime $1<q<2$ the macroscopic dynamics are governed by $α$-stable Lévy laws. This proves that anomalous non-Gaussian dispersion can emerge from pure mass-conserving convolutional chains without invoking fractional derivatives.

• The paper demonstrates that replacing exponential waiting times with q-exponential kernels in storage cascades yields anomalous, non-Gaussian Lévy outflows.
• It employs Laplace-domain transfer functions and small-s expansion techniques to analytically capture power-law hydrograph decay and Galilean advection shifts.
• The study distinguishes sequential nonextensivity from ensemble parameter fluctuations, providing robust tools for modeling heterogeneous transport in physical systems.

Asymptotic Hydrographs and Anomalous Dispersion in Mass-Conserving Storage Cascades

Introduction and Scientific Context

This work rigorously analyzes the hydrologic and transport consequences of replacing the standard exponential waiting-time kernel in classical mass-conserving cascade systems with a q-exponential density derived from nonextensive statistics. The authors establish a precise analytical link between convolutional chains of non-exponential kernels and the emergence of anomalous, non-Gaussian dispersion, fundamentally altering the macroscopic response of multi-stage storage processes. This generalizes the traditional Erlang/gamma hydrograph associated with exponential waiting times, a cornerstone in linear hydrological cascade modeling, by incorporating heavy-tailed temporal kernels motivated by complex and heterogeneous physical systems.

Analytical Framework and Methods

The model comprises a sequence of $k$ uniform storage reservoirs with mass-conserving flows, each governed by a waiting-time kernel. While classical models assume exponential kernels (yielding Poisson renewal dynamics and Erlang responses), the authors introduce a q-exponential waiting-time density parameterized by $q$ in $(1,2)$. The distinction is profound: for $q>1$, the kernel develops power-law tails, capturing processes with broad distributions of residence times relevant to transport in complex or heterogeneous environments (e.g., porous media, disordered systems). They derive the Laplace-domain transfer function for the outflow of the $k$-stage cascade under this generalized kernel and analyze its asymptotic behavior for large $k$.

A key analytical advance is the nontrivial small-$s$ expansion of the Laplace transform of the q-exponential and its convolutions, enabling the authors to capture the emergent macroscopic laws. The influence of higher-order terms is systematically controlled to ensure rigor in the long-time and many-stage limits.

Main Results: From q-Exponentials to Lévy Distributions

The central result is the demonstration that the outflow of a mass-conserving $k$-stage cascade under q-exponential waiting times converges to a (shifted) $\alpha$-stable Lévy distribution in the hydrodynamic limit for $1 < q < 2$. Specifically:

• q = 5/3 critical regime: The asymptotic outflow converges to a stable Lévy density with index $q$0, but with a Galilean-type (linear) time shift reflecting effective advection. The explicit form of the hydrograph incorporates this shift, distinguishing it from the standard Lévy law and providing a fully analytic closed-form.
• General $q$1: Across the entire nonextensive regime, the macroscopic transport kernel is generically given by shifted heavy-tailed $q$2-stable (non-Gaussian) laws, with the critical exponent associated with the kernel’s algebraic decay.
• Power-law outflow decay: The large-$q$3 response exhibits power-law asymptotics $q$4, reflecting heavy-tailed anomalous dispersion.

Crucially, these anomalous transport characteristics arise solely from repeated mass-conserving convolutions—there is no recourse to fractional derivatives or phenomenological nonlocality. The structure is entirely dictated by renewal theory and the underlying stochastic kernel.

Physical and Mathematical Distinctions: Ensemble versus Sequential Nonextensivity

The analysis rigorously differentiates between q-generalized gamma/Erlang laws arising from ensemble-averaged parameter fluctuations (as in superstatistics) and the physically distinct class of distributions generated by sequential convolutions of q-exponential waiting times. The latter are operationally constrained by mass conservation and renewal theory, making them fundamentally relevant to actual transport and storage processes rather than to effective equilibrium parameterizations.

This distinction clarifies confusion in prior literature where nonextensive parameterizations were sometimes incorrectly associated with operational transport kernels. The convolutional (cascade) construction uniquely respects the sequential propagation of mass/localization in each storage element.

Implications for Transport Theory and Application Domains

The results directly inform understanding and modeling of anomalous transport in physical and engineering systems with heterogeneity, broad relaxation spectra, and structural complexity:

• Hydrology: The theory yields exact analytic hydrographs and storage responses for layered/heterogeneous catchments, capturing late-time heavy-tail behavior seen in empirical data but not explained by classical linear models.
• Porous/disordered media: The findings underpin macroscopic dispersion kernels for contaminant transport, with direct interpretability in residence-time distributions.
• Reactor engineering, anomalous heat/particle transport: The analytical results provide kernels for mass, chemical, or energy transport in reactors or process units with non-ideal mixing and retention.

The analytic tractability, conservation properties, and physical interpretability of the shifted Lévy kernels make them mathematically robust and practically implementable.

Theoretical Outlook and Future Directions

The established connection between nonextensive kernels and $q$5-stable laws suggests deep theoretical links bridging generalized central limit theory, anomalous stochastic processes, and hydrodynamic limits of mass-conserving Markovian systems. This work proposes a path to rigorously derive macroscopic fractional or nonlocal transport equations justified by clear physical mechanisms (broad-tailed waiting times and mass conservation), rather than by ad hoc phenomenological fitting.

Future research may extend the analysis to nonlinear/non-Markovian cascades, spatially inhomogeneous kernels, and applications to turbulent transport and geophysical processes exhibiting multi-scale waiting-time statistics. Linking this analytic framework to inverse problems (inferring $q$6 from observed hydrographs/dispersion data) is a promising direction for model calibration in hydrogeology and environmental engineering.

Conclusion

This paper provides a comprehensive, rigorous analytical treatment of mass-conserving storage cascades with q-exponential waiting times, establishing that the macroscopic outflow and storage converge to shifted, heavy-tailed $q$7-stable Lévy laws in the hydrodynamic limit. The results deliver both practical analytical tools for modeling anomalous transport in heterogeneous physical systems and a firm theoretical distinction between ensemble-averaged and sequentially-renewed nonextensive structures. The approach strengthens the bridge between nonextensive statistical mechanics, operational hydrology, and anomalous dispersion theory, providing new avenues for both fundamental and application-oriented research.