| |
|
| Submitter's Name |
Nathan Reaver |
| Session Name |
Poster Session - Stream/River Dynamics |
| Poster Number |
39 |
| |
| Author(s) |
Nathan Reaver, Environmental Engineering Sciences Department, University of Florida (Presenting Author) |
| |
David Kaplan,
Environmental Engineering Sciences Department, University of Florida |
| |
Robert Hensley, School of Forest Resources and Conservation, University of Florida |
| |
Matthew Cohen, School of Forest Resources and Conservation, University of Florida |
| |
|
| |
Quantifying and Predicting Three-Dimensional Heterogeneity in Transient Storage Using Roving Profiling |
| |
|
| |
Hydraulic transport is an important component of nutrient spiraling in streams. Quantifying conservative solute transport is a prerequisite for understanding the cycling and fate of reactive solutes, such as nutrients. Numerous studies have modeled solute transport within streams using the one-dimensional advection, dispersion and storage (ADS) equation calibrated to experimental data from tracer experiments. However, there are limitations to the information about in-stream transient storage that can be derived from calibrated ADS model parameters. Transient storage (TS) in the ADS model is most often modeled as a single process, and calibrated model parameters are “lumped” values that are the best-fit representation of multiple real-world TS processes. In this study, we developed a roving profiling method to assess and predict spatial heterogeneity of in-stream TS. We performed five tracer experiments on three spring-fed rivers in Florida (USA) using Rhodamine WT. For each experiment, stationary fluorometers were deployed to measure breakthrough curves for multiple reaches within the river. Teams of roving samplers moved along the rivers measuring tracer concentrations at various locations and depths within the reaches. A Bayesian statistical method was used to calibrate the ADS model to the stationary breakthrough curves, resulting in probability distributions for both the advective and TS zone. Rover samples were assigned a probability of being from either the advective or TS zone by comparing measured concentrations to the probability distributions of concentrations in the ADS advective and TS zones. A regression model was used to predict the probability of any in-stream position being located within the advective versus TS zone based on spatiotemporal predictors (time, river position, depth, and distance from bank) and eco-geomorphological feature (eddies, woody debris, benthic depressions, and aquatic vegetation). Results confirm that TS is spatially variable as a function of spatiotemporal and eco-geomorphological features. |
| |
