1. Testing models for river avulsion style with remote sensing and numerical simulations
NSF, Geomorphology and Landuse Dynamics Program (NSF 1911321)
One of the most dramatic, yet rarely witnessed, events is when a river avulses (or moves) to a new spot in the adjacent floodplain. This process is important for river systems because it dictates sediment delivery to floodplains and major flooding hazards. Despite this, once an avulsion occurs, we cannot predict how this process works. This is because avulsions occur infrequently (every 1000 years) and the lack of observations stifle model development.
In this proposal, we seek to advance the science of river avulsions by using the remote sensing record and numerical simulations to test models for how they work. Based on preliminary data, our hypothesis is that avulsion style changes moving downstream—near the mountain front avulsions tend to reoccupy older channels, while those farther away tend to create their own channels. This arises because sediment size decreases downstream, and finer sediment is more easily transported overbank into the floodplain, filling abandoned channels, and forcing avulsions to create new channels.
We will test this hypothesis by:
- detecting avulsions in modern foreland basins by using big data remote sensing tools, e.g., Google Earth Engine, to search through Landsat data,
- quantifying avulsion tendency to reoccupy or create their own channel with our fingerprinting algorithm that isolates the area disturbed by the avulsion, and
- building a cellular avulsion model that includes downstream fining and overbank sedimentation.
We will use the model to test the connection between downstream fining and avulsion behavior.
Our understanding of river avulsions has been dominated by a few well-documented modern examples, and a plethora of preserved deposits in the rock record. The modern examples are limited in number and the deposits in the rock record contain limited information. The intellectual merit of this project is its transformational approach to use the modern remote sensing record to detect recent avulsions and from those observations develop new models for how avulsion work.
We propose a two-pronged approach that uses big data remote sensing tools, e.g., Google Earth Engine, to develop innovative new methods for avulsion detection, and then create updated simulation models of river avulsion to test controls on the avulsion process. The new remote sensing methods for avulsion detection can be used for general river change detection and will be shared with the community.
River movement is an engaging, dynamic process that can excite school-aged children about earth science. Even though Earth Science is engaging, a pipeline problem exists, and students lose interest in Earth Science by the time they enter high school. Our broader impact activity seeks to assess if a curriculum based on realistic, three-dimensional learning increases likelihood in pursuing Earth Science. We will design a science curriculum that uses Google Earth Engine to make observations about earth’s rivers and connects those observations to theory. To measure if this kind of learning experience increases interest, we will collaborate with Dr. Meredith Park Rogers (IU School of Education) and the Saturday Science Quest program at Indiana University, which she also directs.
In collaboration with Dr. Park Rogers, we will develop, teach, and assess student’s engagement in a 5-week program called When Rivers Move. Working with children in upper elementary through middle school grades, we will engage students in activities to illustrate how observations from Google Earth Engine can be used to construct models of river movement. We will assess how effective this curriculum is by measuring the impact on student motivation and enthusiasm through documenting the students’ participation in the weekly activities and conducting pre and post-interviews.
2. Understanding deltas from the lens of their channel networks
NSF, Geomorphology and Landuse Dynamics Program (NSF 1812019)
River deltas are complex ecogeomorphologic systems exhibiting a large variability in their morphology and in the configurations of the channels that tile their deltaic surface. These channels dictate how water, sediment, and nutrients are spread over the delta top, and thus are key for delta self-maintenance and resilience to external perturbations. Yet, while tributary channel networks have been exhaustively studied over the past 50 years, the study of Delta Channel Networks (DCNs) is still in its infancy.
In this proposal, we seek to undertake a comprehensive study of deltas through the lens of their delta channel networks. Our hypothesis is that DCNs encode information about the processes that created them and this hypothesis will be tested via quantitative analysis of an extensive data base of DCNs extracted from satellite images of 60 field deltas of diverse forcings, geologies, hydroclimatic environments and levels of anthropogenic influence, as well as numerically simulated deltas using Delft3D under controlled conditions, using metrics that rigorously capture DCN topology and flux organization.
The ultimate goal is to:
- relate DCN patterns to the underlying physical processes that create them
- propose a DCN-based delta classification framework
- explore principles behind delta self-organization and explain deviations from optimality, and
- advance our understanding of delta response to change.
Deltaic surfaces are tiled by channels that join and split in multiple and complex ways as they distribute water, sediment and nutrients from the apex to the delta top and to the shoreline. Despite significant advances in studying delta processes and the morphology and dynamics of deltas as affected by first order controls of river, wave and tide dominance, the study of deltas from the point of view of the topology and flux organization of their complex delta channel networks (DCNs) has not been adequately explored to date. Here we propose a DCN-based study of deltas via a three pronged approach of theory (graph theoretical approaches), numerical modeling (analysis of the results of numerically simulated deltas under a range of controlled conditions), and analysis of an exhaustive number of field deltas spanning diverse physical environments.
Innovative new methods for automatic extraction of DCNs from satellite imagery will be developed and a global set of DCNs will be produced and shared with the community. We will use state-of-the-art physically-based models to develop for the first time a comprehensive database of numerically simulated deltas where the main physical controls are systematically explored. At the heart of the proposed research, and inspired by the complex structures of DCNs, network theory concepts will be introduced and further advanced for development and analysis of a suite of sufficient metrics towards a reduced dimensionality space for DCN-based delta comparison and process inference.
Our approach draws on many disciplines -information theory, graph theory, engineering, and earth science- and as such it provides a setting within which learning modules can be developed for students working across fields to enhance a quantitative approach to analysis of natural and coupled human natural systems. We will develop, collaboratively between the University of California Irvine (UCI) and Indiana University (IU), a 2-week module that focuses on delta connectivity through laboratory experiments, numerical modeling, and theory as part of the Summer Institute for Earth Surface Dynamics (SIESD) sponsored by NSF which brings 50 graduate students/early career researchers together with senior scientists from academia and federal agencies.
We will also develop a short course on "Delta Connectivity" and offer it at the Community Surface Dynamics modeling system annual meeting. As part of a newly funded National Research Training (NRT) program at UCI, called Ridge to Reef R2R, we will aim to attract new students to the wonder of landscapes and their channel networks. UCI is a Hispanic Minority Serving Institution and we will engage in research 3-4 undergraduates of diverse background per year and attract them to pursue STEM careers, capitalizing also in the highly successful California Alliance for Minority Participation (CAMP) and the Diverse Educational Community and Doctoral Experience (DECADE) programs at UCI.