In mkoohafkan/RAStestR: Functions for testing RAS outputs

Introduction

The R package RAStestR provides a library of functions for reading, manipulating and visualizing HEC-RAS output data. Specifically, it accesses the HDF5 data file associated with an HEC-RAS "plan". This vignette provides a brief introduction to the basic functionality of RAStestR.

An example project

Outputs from a simple quasi-unsteady sediment transport simulation are included with RAStestR. Basic information about the plan can be displayed with list_plan_info.

library(RAStestR)
ras.file = system.file("sample-data/SampleQuasiUnsteady.hdf", package = "RAStestR")
list_plan_info(ras.file)

RAStestR provides functions for accessing data on model geometry and simulation outputs. Simulation outputs are read into R via two basic functions: read_standard() and read_sediment(). Most hydraulic data such as flow, water surface elevation, shear stress, etc. are contained in single tables which are read in using read_standard(). Some outputs from sediment transport models reported by grain class in individual tables; these tables can be read in and automatically grouped using read_sediment(). In the example below, the standard table "Shear Stress" and the sediment table "Mass Bed Change Cum" are read into R.

shear.stress = read_standard(ras.file, "Shear Stress")
vol.change.cum = read_sediment(ras.file, "Vol Bed Change Cum")

You can also choose to extract data for only a subset of output times, river stations, or grain classes. RAStestR provides listing functions to retrieve possible values.

list_output_times(ras.file)
list_stations(ras.file)
list_grain_classes(ras.file)

read_standard and read_sediment return wide-format tables which match the row-column structure of the HDF file. extracting subsets of the data is more convenient with wide tables as rows correspond to output times and columns correspond to river stations. However, many operations in R are more convenient to use with long-format tables. The to_longtable function collapses the cross-section columns in a wide table automatically. The to_widetable() function can convert a long table back into the original wide format. The R package dplyr provides a rich environment for manipulating long-format tables.

# wide table format (first five columns)
head(shear.stress[1:5])
# long table format
shear.stress.long = to_longtable(shear.stress, "ShearStress")
head(shear.stress.long)
# convert from long-format to wide-format
#to_widetable(shear.stress.long, "Station", "ShearStress")

read_standard, read_sediment and to_longtable read the river stations, time stamps and grain classes as characters. This is inconvenient for e.g. plotting results over time or along a reach. The function reformat_fields() will convert the time stamps to POSIXct datetime format, and river stations to numeric format.

shear.stress.long.fmt = reformat_fields(shear.stress.long)

mkoohafkan/RAStestR documentation built on July 14, 2019, 11:41 p.m.