adc provides functions to calculate discharge-based metrics that are useful in water quality concentration and flux regression equations.
Install adc from CRAN:
The development version is available on https://txwri.r-universe.dev/adc and can be installed with:
install.packages('adc', repos = c(txwri = 'https://txwri.r-universe.dev'))
Flow anomalies represent how different the current discharge period is (current day, current week, current month, etc.) from previous periods (previous week, previous month, period of record, etc.).
library(adc) ## example code is lavaca and includes dates and mean daily flow data(lavaca) x <- fa(lavaca$Flow, dates = lavaca$Date, T_1 = "1 month", T_2 = "1 year", clean_up = TRUE, transform = "log10") plot(lavaca$Date, x, type = "l", xlab = "Date", ylab = "Anomaly [unitless]")
The packages also includes functions to calculate an exponentially weighted discounted flow, base-flow, and rate of change for mean daily streamflow. Functions generally work well using the dplyr::mutate()
function to facilitate tidy workflows.
RetroSearch is an open source project built by @garambo | Open a GitHub Issue
Search and Browse the WWW like it's 1997 | Search results from DuckDuckGo
HTML:
3.2
| Encoding:
UTF-8
| Version:
0.7.4