ABSTRACT:

Attached is the Cobble App in Matlab developed by Erin Bray, for calculation of cobble shape parameters as reported in Bray et al "Influence of particle lithology, size, and angularity on rates and products of bedload wear: an experimental study" (In Review).

For the Cobble App to work, install Matlab version 2022b; Add the Image Processing Toolbox; Add the Computer Vision System Toolbox.Photo files must be saved in grayscale (no RGB embedded when saving photo files).
Files of photos can be saved as .tif or .tiff (both should work in the cobble app)
All extraneous white edges/borders or dots in files need to be removed (there were some stray white specks in the background of one of the photo files that was reduced to 75% resolution).
The Photo ID string column such as "P1_A1_N_PT" needs to be consistently formatted, with no extra spaces or extra characters, in both the Excel spreadsheet and in the photo file names, with no changes to the file string name even if you reduce the photo resolution to 75%. To use the merge functionality within the Cobble App, which pairs image-based shape parameters with corresponding handheld measurements of mass, diameter of each particle, the Excel spreadsheet needs to always have the identical number of columns and name of columns.

ABSTRACT:

We developed a spectral river energy balance model (FLUVIAL-EB, Bray et al., 2017) to predict energy fluxes and river temperature along a large lowland regulated river and, more specifically, developed a module within the model with the specific aim of quantifying the response of river temperatures to perturbations in atmospheric variables. For clear, flowing water that is usually present below dams, FLUVIAL-EB couples a spectral radiation balance model with turbulent energy fluxes, bed conduction, and a 1D hydraulic model employed over the longitudinal profile of a lowland river whose water depth and velocity vary with distance downstream. The dynamic component of the model accounts for the feedback between spatial and temporal variability in water temperature and changes in the atmospheric fluxes and conduction into or out from the streambed. The predicted water temperature is used to compute the latent, sensible, net longwave, bed conduction, and advective energy flux at every time step. Absorbed shortwave radiation is computed for every wavelength in the solar spectrum, and then integrated across all wavelengths. The continuous component of the model interpolates between measurements at meteorological stations at discrete times. Because the governing differential equation is instantaneous, the input meteorological variables must be available for any values of x and t. From hourly averages of input data, we calculated instantaneous data by generating a cumulative sum, applying a smoothing spline, and then taking the derivative. Doing so allows for a continuous spatial and temporal field of the entire river based on hourly meteorological data and modeled steady-state hydraulic values under bankfull flow conditions, with temporal resolution up to 30 s and spatial resolution of every 100 m along the river. The compressed file of the model folder ('FLUVIAL-EB_BrayDozier_2023.zip') contains approximately 82 Matlab scripts and .mat files that, together, make up the entire FLUVIAL-EB model; all are compatible in Matlab version 2023b. The primary command line function, used to run model simulations, is titled 'riverExplicitSoln.m'. To run the model for the spatial extent on the San Joaquin River, CA, USA, you must install the following Matlab Toolboxes: Computer Vision, Curve Fitting, Database Toolbox, Image Processing Toolbox, Mapping Toolbox, Signal Processing Toolbox, Statistics and Machine Learning Toolbox, and the Raster Reprojection Toolbox (not on Mathworks, written by Jeff Dozier and attached below). For examples of command lines used in model simulations, see 'runmodel_baseline.m'.