Package 'AHGestimation'

Properly estimate AHG values

Description

Properly estimate AHG values

Usage

ahg_estimate(
  df,
  allowance = 0.05,
  gen = 192,
  pop = 200,
  cprob = 0.4,
  mprob = 0.4,
  times = 1,
  scale = 1.5,
  full_fitting = FALSE,
  verbose = FALSE
)

Arguments

df	hydraulic data.frame with columns named (Q, V, TW, Y). Q and at least one other are required.
allowance	allowed deviation from continuity
gen	Number of generations to breed.
pop	Size of population
cprob	Crossover probability
mprob	Mutation probability
times	how many times (seeds) should nsga2 be run
scale	should a scale factor be applied to data pre NSGA-2 fitting
full_fitting	should all fits be returned?
verbose	should messages be emitted?

Value

list

See Also

Other AHG: best_optimal(), calc_nsga(), compute_ahg(), min_max(), mismash()

---

Report best optimal

Description

Report best optimal

Usage

best_optimal(best, check, verbose = TRUE)

Arguments

best	best performing method (character string)
check	values to check against
verbose	should messages be emitted

Value

vector

See Also

Other AHG: ahg_estimate(), calc_nsga(), compute_ahg(), min_max(), mismash()

---

Calculate NSGA2 AHG

Description

Calculate NSGA2 AHG

Usage

calc_nsga(
  df,
  allowance = 0.05,
  r,
  scale = 2,
  gen = 96,
  pop = 500,
  cprob = 0.8,
  mprob = 0.05,
  times = 1
)

Arguments

df	hydraulic data.frame
allowance	allowable deviation from continuity
r	fit list
scale	should a scale factor be applied to data pre NSGA-2 fitting
gen	Number of generations to breed.
pop	Size of population
cprob	Crossover probability
mprob	Mutation probability
times	how many times (seeds) should nsga2 be run

Value

data.frame

See Also

Other AHG: ahg_estimate(), best_optimal(), compute_ahg(), min_max(), mismash()

---

Approximate AHG relationships

Description

Approximate AHG relationships using both OLS and NLS methods

Usage

compute_ahg(Q, P, type = "relation")

Arguments

Q	a stream flow time series
P	a corresponding time series of a second hydraulic variable
type	relationship being tested

Value

data.frame

See Also

Other AHG: ahg_estimate(), best_optimal(), calc_nsga(), min_max(), mismash()

---

Calculate the slope of 3D linestring

Description

Given a sf object with 'XYZ' coordinates, return a vector of numeric values representing the average slope of each linestring in the sf data frame input.

The default calculates the slope using 'slope_weighted()'. You can also use 'slope_mean()' or any other function that takes the same inputs as these functions.

Usage

compute_channel_slope(path, fun = slope_weighted, directed = FALSE)

Arguments

path	an XYZ LINESTRING representing the path of travel
fun	The slope function to calculate per element, 'slope_weighted' is the default.
directed	Should the value be directed? 'FALSE' by default. If 'TRUE' the result will be negative when it represents a downslope (when the end point is lower than the start point).

Value

A vector of slopes associated with each linear element The value is a proportion representing the change in elevation for a given change in horizontal distance.

See Also

Other hydraulics: compute_hydraulic_params(), compute_n(), cross_section(), extract_thalweg(), slope_matrix()

---

Approximate channel coeffiecent

Description

Approximate the hydraulic values from AHG fit

Usage

compute_hydraulic_params(fit)

Arguments

fit	output of ahg_estimate

Value

numeric

See Also

Other hydraulics: compute_channel_slope(), compute_n(), cross_section(), extract_thalweg(), slope_matrix()

---

Approximate Roughness

Description

Approximate median roughness using Manning Equation

Usage

compute_n(df, S = 0.02)

Arguments

df	a data.frame with at least Y and V.
S	reach scale longitudinal slope (m/m). Default mean of the nhdplusV2

Value

numeric

See Also

Other hydraulics: compute_channel_slope(), compute_hydraulic_params(), cross_section(), extract_thalweg(), slope_matrix()

---

Approximate channel shape

Description

Get a list of points from x axis of a cross section and max depth and produce depth values for those points based on channel shape

Usage

cross_section(r, TW = 30, Ymax = 2, n = 30)

Arguments

r	The corresponding Dingman's r coefficient
TW	width of the channel at bankfull
Ymax	maximum depth of the channel at bankfull
n	the number of points to construct in the XS

Value

depth values every 1m along the cross section

See Also

Other hydraulics: compute_channel_slope(), compute_hydraulic_params(), compute_n(), extract_thalweg(), slope_matrix()

---

Implements filtering by date

Description

Data is filtered when it is beyond a specified year threshold (e.g. 5 years old). The relative date is based on the newest observation in the data set. Optionally, the maximum flow (Q) record can be retained.

Usage

date_filter(df, years, keep_max = FALSE)

Arguments

df	a data.frame with at least a date and Q field.
years	the number of allowed history
keep_max	Should the largest flow record be kept, even if older then "years"

Value

data.frame

See Also

Other filters: mad_filter(), nls_filter(), qva_filter(), significance_check()

---

Extract Thalweg From a data.frame of cross sections, a classified thalweg can be extracted as the connected LINESTRING

Description

Extract Thalweg From a data.frame of cross sections, a classified thalweg can be extracted as the connected LINESTRING

Usage

extract_thalweg(xs, crs = 5070)

Arguments

xs	a data.frame containing cross sectional data. Required columns are hf_id, cs_id, X, Y, Z
crs	the CRS of the XY coordinates

Value

XYZ LINESTRING object

See Also

Other hydraulics: compute_channel_slope(), compute_hydraulic_params(), compute_n(), cross_section(), slope_matrix()

---

Implements filtering by median absolute deviation

Description

An iterative outlier detection procedure is run based on to the linear regression residuals. Values of log-transformed TW, V, and Y residuals falling outside a specified median absolute deviation (MAD) envelope are excluded. Regression coefficients were recalculated and the outlier detection procedure was reapplied until no outliers are detected. This method was identified in HyG

Usage

mad_filter(df, envelope = 3)

Arguments

df	a data.frame with at least a Q and one other AHG field (Y. TW, V).
envelope	MAD envelope

Value

data.frame

See Also

Other filters: date_filter(), nls_filter(), qva_filter(), significance_check()

---

Find thresholds for coefficient and exponent limits.

Description

Find thresholds for coefficient and exponent limits.

Usage

min_max(df, scale = 2)

Arguments

df	hydraulic data.frame
scale	Scale by set factor. This limits the exponent at coefficients to the range of (1/s) * nls; s * nls

Value

list

See Also

Other AHG: ahg_estimate(), best_optimal(), calc_nsga(), compute_ahg(), mismash()

---

Compute all combos!

Description

Compute all combos!

Usage

mismash(v, V, TW, Y, Q, r, allowance)

Arguments

v	values
V	Velocity time series
TW	Top width time series
Y	Depth time series
Q	Discharge time series
r	rrr TODO
allowance	Allowable deviation from continuity

Value

list

See Also

Other AHG: ahg_estimate(), best_optimal(), calc_nsga(), compute_ahg(), min_max()

---

Implements NLS filtering

Description

An NLS fit provides the best relation by relation fit. For each provided relationship, an NLS fit is computed and used to estimate the predicted V,TW,Y for a given Q. If the actual value is outside the specified allowance it is removed.

Usage

nls_filter(df, allowance = 0.5)

Arguments

df	a data.frame with at least a Q and one other AHG field (Y. TW, V).
allowance	how much deviation from observed should be allowed (default = .5)

Value

data.frame

See Also

Other filters: date_filter(), mad_filter(), qva_filter(), significance_check()

---

Normalized Root Mean Square Error

Description

Normalized root mean square error (NRMSE) between sim and obs, with treatment of missing values

Usage

nrmse(sim, obs)

Arguments

sim	numeric vector simulated values
obs	numeric vector observed values

Value

numeric

See Also

Other evaluation: pbias()

---

Sample gage data Manual measurements made at NWIS site 01096500 Q_cms is a mandatory argument and at least one of TW_m, V_ms, or Y_m.

Description

Sample gage data Manual measurements made at NWIS site 01096500 Q_cms is a mandatory argument and at least one of TW_m, V_ms, or Y_m.

Usage

nwis

Format

A data frame with 245 rows and 6 columns:

siteID

NWIS ID

date

date of measurement

Q_cms

Steamflow (cubic meters per second)

Y_m

Depth (meters)

V_ms

Velocity (meters per second)

TW_m

Top width (meters)

---

Percent Bias

Description

Percent Bias between sim and obs, with treatment of missing values.

Usage

pbias(sim, obs)

Arguments

sim	numeric vector simulated values
obs	numeric vector observed values

Value

numeric

See Also

Other evaluation: nrmse()

---

Implements filtering by continuity

Description

The function tests if the measured Q is outside of the expected range based on the product of measured velocity, top-width, and depth (e.g. Q≠vA)

Usage

qva_filter(df, allowance = 0.05)

Arguments

df	a data.frame with a Q, Y, TW, V and field.
allowance	how much deviation from equality should be allowed (default = .05)

Value

data.frame

See Also

Other filters: date_filter(), mad_filter(), nls_filter(), significance_check()

---

Implements significance check

Description

The relationship between all supplied log transformed variables are computed. If the p-value of any of these is less then the supplied p-value an error message is emitted.

Usage

significance_check(df, pvalue = 0.05)

Arguments

df	a data.frame with at least a Q and one other AHG field (Y. TW, V).
pvalue	Significant p-value (default = .05)

Value

data.frame

See Also

Other filters: date_filter(), mad_filter(), nls_filter(), qva_filter()

---

Calculate the gradient of line segments from a 3D matrix of coordinates

Description

Calculate the gradient of line segments from a 3D matrix of coordinates

Usage

slope_matrix(mat, lonlat = TRUE)

slope_weighted(mat, lonlat = TRUE, directed = FALSE)

slope_mean(mat, lonlat = TRUE, directed = FALSE)

Arguments

mat	Matrix containing coordinates and elevations. The matrix should have three columns: X, Y, and Z. In data with geographic coordinates, Z values are assumed to be in meters. In data with projected coordinates, Z values are assumed to have the same units as the X and Y coordinates.
lonlat	Are the elements provided in longitude/latitude coordinates? By default, value is from the CRS of the routes ('sf::st_is_longlat(...)').

Value

A vector of slopes associated with each LINE element The output value is a proportion representing the change in elevation for a given change in horizontal distance.

See Also

Other hydraulics: compute_channel_slope(), compute_hydraulic_params(), compute_n(), cross_section(), extract_thalweg()

---