skip to main content

Attention:

The NSF Public Access Repository (NSF-PAR) system and access will be unavailable from 11:00 PM ET on Thursday, May 23 until 2:00 AM ET on Friday, May 24 due to maintenance. We apologize for the inconvenience.


Title: Using Information Flow for Whole System Understanding From Component Dynamics
Abstract

Complex systems that exhibit emergent behaviors arise as a result of nonlinear interdependencies among multiple components. Characterizing how such whole system dynamics are sustained through multivariate interaction remains an open question. In this study, we propose an information flow‐based framework to investigate how the present state of any component arises as a result of the past interactions among interdependent variables, which is termed as causal history. Using a partitioning time lag, we divide this into immediate and distant causal history components and then characterize the information flow‐based interactions within these as self‐ and cross‐feedbacks. Such a partition allows us to characterize the information flow from the two feedbacks in both histories by using partial information decomposition as unique, synergistic, or redundant interactions. We employ this casual history analysis approach to investigate the information flows in a short‐memory coupled logistic model and a long‐memory observed stream chemistry dynamics. While the dynamics of the short‐memory system are mainly maintained by its recent historical states, the current state of each stream solute is sustained by self‐feedback‐dominated recent dynamics and cross‐dependency‐dominated earlier dynamics. The analysis suggests that the observed 1/fsignature of each solute is a result of the interactions with other variables in the stream. Based on high‐density data streams, the approach developed here for investigating multivariate evolutionary dynamics provides an effective way to understand how components of dynamical system interact to create emergent whole system behavioral patterns such as long‐memory dependency.

 
more » « less
NSF-PAR ID:
10452676
Author(s) / Creator(s):
 ;  
Publisher / Repository:
DOI PREFIX: 10.1029
Date Published:
Journal Name:
Water Resources Research
Volume:
55
Issue:
11
ISSN:
0043-1397
Page Range / eLocation ID:
p. 8305-8329
Format(s):
Medium: X
Sponsoring Org:
National Science Foundation
More Like this
  1. At the biosphere–atmosphere interface, nonlinear interdependencies among components of an ecohydrological complex system can be inferred using multivariate high frequency time series observations. Information flow among these interacting variables allows us to represent the causal dependencies in the form of a directed acyclic graph (DAG). We use high frequency multivariate data at 10 Hz from an eddy covariance instrument located at 25 m above agricultural land in the Midwestern US to quantify the evolutionary dynamics of this complex system using a sequence of DAGs by examining the structural dependency of information flow and the associated functional response. We investigate whether functional differences correspond to structural differences or if there are no functional variations despite the structural differences. We base our analysis on the hypothesis that causal dependencies are instigated through information flow, and the resulting interactions sustain the dynamics and its functionality. To test our hypothesis, we build upon causal structure analysis in the companion paper to characterize the information flow in similarly clustered DAGs from 3-min non-overlapping contiguous windows in the observational data. We characterize functionality as the nature of interactions as discerned through redundant, unique, and synergistic components of information flow. Through this analysis, we find that in turbulence at the biosphere–atmosphere interface, the variables that control the dynamic character of the atmosphere as well as the thermodynamics are driven by non-local conditions, while the scalar transport associated with CO2 and H2O is mainly driven by short-term local conditions. 
    more » « less
  2. Abstract

    As drought and wildfire frequency increase across the western United States, our ability to predict how water resources will respond to these disturbances depends on our understanding of the feedbacks that maintain watershed function and streamflow. Previous studies of non‐perennial headwater streams have ranked drivers of low‐flow conditions; however, there is a limited understanding of the interactions between these drivers and the processes through which these interactions affect streamflow. Here, we use stream water level, soil moisture, sap flow, and vapor pressure deficit data to investigate ecohydrological interactions along a mountainous headwater stream. Correlation and cross‐correlation analyses of these variables show that ecohydrological interactions are (a) nonlinear and (b) interconnected, suggesting that analyses assuming linearity and independence of each driver are inadequate for quantifying these interactions. To account for these issues and investigate causal linkages, we use convergent cross‐mapping (CCM) to characterize the feedbacks that influence non‐perennial streamflow. CCM is a nonlinear, dynamic method that has only recently been applied to hydrologic systems. CCM results reveal that atmospheric losses associated with local sap flow and vapor pressure deficit are driving changes in soil moisture and streamflow (p < 0.01) and that atmospheric losses influence stream water more directly than shallow soil moisture. These results also demonstrate that riparian processes continue to affect subsurface flows in the channel corridor even after stream drying. This study proposes a nonlinear framework for quantifying the ecohydrologic interactions that may determine how headwater streams respond to disturbance.

     
    more » « less
  3. Abstract

    The concept of causal interactions between components is an integral part of hydrology and Earth system sciences. Modelers, decision makers, scientists, and other water resources stakeholders all utilize some notion of cause‐and‐effect to understand processes, make decisions, and infer how systems react to change. However, there are different perspectives on the meaning of causality in complex systems and, further, different frameworks and methodologies with which to detect causal interactions. We propose here that information theory (IT) provides a compelling framework for the detection of causality and discuss approaches for several levels of analyses that capture interactions that range from pairwise to multivariate in nature. We illustrate these types of analyses with an example based on weather station time series variables, in which variables may interact pairwise or jointly and on short to long time scales. In general, many unsolved or even unanticipated questions in the hydrologic sciences could benefit from this perspective.

     
    more » « less
  4. Abstract

    Climate change is increasing the frequency, severity, and extent of wildfires and drought in many parts of the world, with numerous repercussions for the physical, chemical, and biological characteristics of streams. However, information on how these perturbations affect top predators and their impacts on lower trophic levels in streams is limited.

    The top aquatic predator in southern California streams is nativeOncorhynchus mykiss, the endangered southern California steelhead trout (trout). To examine relationships among the distribution of trout, environmental factors, and stream invertebrate resources and assemblages, we sampled pools in 25 stream reaches that differed in the presence (nine reaches) or absence (16 reaches) of trout over 12 years, including eight reaches where trout were extirpated during the study period by drought or post‐fire flood disturbances.

    Trout were present in deep pools with high water and habitat quality. Invertebrate communities in trout pools were dominated by a variety of medium‐sized collector–gatherer and shredder invertebrate taxa with non‐seasonal life cycles, whereas tadpoles and large, predatory invertebrates (Odonata, Coleoptera, Hemiptera [OCH]), often with atmospheric breather traits, were more abundant in troutless than trout pools.

    Structural equation modelling of the algal‐based food web indicated a trophic cascade from trout to predatory invertebrates to collector–gatherer taxa and weaker direct negative trout effects on grazers; however, both grazers and collector–gatherers also were positively related to macroalgal biomass. Structural equation modelling also suggested that bottom‐up interactions and abiotic factors drove the detritus‐based food web, with shredder abundance being positively related to leaf litter (coarse particulate organic matter) levels, which, in turn, were positively related to canopy cover and negatively related to flow. These results emphasise the context dependency of trout effects on prey communities and of the relative importance of top‐down versus bottom‐up interactions on food webs, contingent on environmental conditions (flow, light, nutrients, disturbances) and the abundances and traits of component taxa.

    Invertebrate assemblage structure changed from a trout to a troutless configuration within a year or two after trout were lost owing to post‐fire scouring flows or drought. Increases in OCH abundance after trout were lost were much more variable after drought than after fire. The reappearance of trout in one stream resulted in quick, severe reductions in OCH abundance.

    These results indicate that climate‐change induced disturbances can result in the extirpation of a top predator, with cascading repercussions for stream communities and food webs. This study also emphasises the importance of preserving or restoring refuge habitats, such as deep, shaded, perennial, cool stream pools with high habitat and water quality, to prevent the extirpation of sensitive species and preserve native biodiversity during a time of climate change.

     
    more » « less
  5. null (Ed.)
    Autonomous active, elastic filaments that interact with each other to achieve cooperation and synchrony underlie many critical functions in biology. The mechanisms underlying this collective response and the essential ingredients for stable synchronization remain a mystery. Inspired by how these biological entities integrate elasticity with molecular motor activity to generate sustained oscillations, a number of synthetic active filament systems have been developed that mimic oscillations of these biological active filaments. Here, we describe the collective dynamics and stable spatiotemporal patterns that emerge in such biomimetic multi-filament arrays, under conditions where steric interactions may impact or dominate the collective dynamics. To focus on the role of steric interactions, we study the system using Brownian dynamics, without considering long-ranged hydrodynamic interactions. The simulations treat each filament as a connected chain of self-propelling colloids. We demonstrate that short-range steric inter-filament interactions and filament roughness are sufficient – even in the absence of inter-filament hydrodynamic interactions – to generate a rich variety of collective spatiotemporal oscillatory, traveling and static patterns. We first analyze the collective dynamics of two- and three-filament clusters and identify parameter ranges in which steric interactions lead to synchronized oscillations and strongly occluded states. Generalizing these results to large one-dimensional arrays, we find rich emergent behaviors, including traveling metachronal waves, and modulated wavetrains that are controlled by the interplay between the array geometry, filament activity, and filament elasticity. Interestingly, the existence of metachronal waves is non-monotonic with respect to the inter-filament spacing. We also find that the degree of filament roughness significantly affects the dynamics – specifically, filament roughness generates a locking-mechanism that transforms traveling wave patterns into statically stuck and jammed configurations. Taken together, simulations suggest that short-ranged steric inter-filament interactions could combine with complementary hydrodynamic interactions to control the development and regulation of oscillatory collective patterns. Furthermore, roughness and steric interactions may be critical to the development of jammed spatially periodic states; a spatiotemporal feature not observed in purely hydrodynamically interacting systems. 
    more » « less