From the movement of rivers, to the generation of catastrophic landslides and the evolution of entire landscapes, many processes that shape the surface of the earth are characterized by a high degree of variability; variability that is not linked to environmental factors, but to complex internal dynamics. Describing such complexity and variability requires stochastic models that describe processes probabilistically and large datasets to calibrate these models. What can we do when nature presents to us only one of the many possible evolutions of a highly complex system?
In this paper, we describe a framework to calibrate stochastic models of morphodynamic systems with a single time-series of data. By “morphodynamic system” we refer to a system that is characterized by changes in shapes or position of objects. Rivers that are moving back and forth across a floodplain are a great example for a morphodynamic system that is characterized by complex internal dynamics. Here, we demonstrate the framework using an experiment of braided rivers moving in a flume. Yes, this is the same experiment that we used in our last paper to study the average behavior of lateral channel movements (Link). Here, we are interested in the variability.
In simple terms, the framework consists of generating a large number of “synthetic” time-series from a stochastic model. These synthetic time-series will vary depending on the input parameters to the model. We calibrate these parameters by finding model outputs that are statistically equivalent to the data. One of the key aspects of the framework is the choice of statistical tests to compare the data to the model. We propose three statistical tests to compare the behavior of channel movements in model and datasets, but these statistical comparisons are modular and can be adapted or expanded to suit the studied morphodynamic system.
Hoffimann, J., Bufe, A., Caers J. (accepted). Morphodynamic Analysis and Statistical Synthesis of Geomorphic Data: Application to a Flume Experiment. Journal of Geophysical Research: Earth Surface. Journal link.
