hydrotoolbox

Overview

hydrotookbox contains objects, methods and functions to efficiently manage hydro-meteorological data sets. The goal is to provide easy to use functionality to read, plot, manipulate and process hydro-meteorological data records and also for pre and post processing data for hydrological modeling. The package has special features for Argentina and Chile data-sets (SNHI, CR2, IANIGLA, AIC and DGI).

The goal is to provide general objects and methods to be used with any hydro-meteorological data set for:

• condense all the information that a station records in a single object.
• easily plot (dynamic and static graphs) the series for data analysis, reporting and publishing.
• modify the original series in order to remove all kind of mistrustful periods without loosing the original data.
• pre and post process series for hydrological modeling.

For example, creating an hydromet_station object and storing the series just takes a few lines:

library(hydrotoolbox)

path <- system.file('extdata', package = 'hydrotoolbox')

# we first build the station object 
guido <- 
  hm_create() %>%
  hm_build(bureau = 'mnemos', path = path, 
           file_name = 'mnemos_guido.xlsx', 
           slot_name = c('qd', 'evap', 'tair',
                         'tmax', 'tmin', 'wspd'), 
           by = c('day', 'day', '6 hour', 
                  'day', 'day', '6 hour'), 
           out_name = c('q(m3/s)', 'evap(mm)', 'tair',
                        'tmax', 'tmin', 'wspd(km/hr)') )

# show all the loaded data
guido %>%
  hm_show()

now let’s day that we want a report of missing data and basic statistics of the daily discharge and to make a dynamic plot to analyze the series

# we want to know something about mean daily streamflow discharge
guido %>% 
  hm_report(slot_name = 'qd', col_name = 'q(m3/s)')
  
  
# plot discharge
guido %>% 
  hm_plot(slot_name = 'qd', col_name = list( 'q(m3/s)' ), 
          interactive = TRUE, 
          line_color = 'dodgerblue',
          y_lab = 'Q(m3/s)'   ) 

we realize that we need to smooth the series and remove some mistrustful periods

# smooth the series 
guido <- 
   guido %>% 
       hm_mutate(slot_name = 'qd', FUN = mov_avg, 
              k = 3, out_name = 'q_smooth') 

# remove mistrustful measurements 
guido <- 
  guido %>%
    hm_mutate(slot_name = 'qd', FUN = set_value,
              col_name = 'q_smooth', out_name = 'q_set',
              value = rep(NA_real_, 2),
              from = c('1965-08-09', '1974-06-26'),
              to = c('1965-08-25', '1974-07-04') )

now we plot the three versions so see the effect of the proposed modifications. Then we export the object, save the table and the plot.

guido %>% 
  hm_plot(slot_name = 'qd', 
          col_name = list( c('q(m3/s)', 'q_set') ), 
          interactive = TRUE, 
          line_color = c('dodgerblue', 'forestgreen'),
          y_lab = 'Q(m3/s)',
          legend_lab = c('original', 'set')) 
          
# save the hydromet object
saveRDS(object = guido, path = "my_path/guido.rds")

# extract the table and save it
guido_table <- 
  guido %>% 
  hm_get(slot_name = "qd")

wirte.csv(x = guido_table,
          file = "my_path/guido_qd.csv",
          row.names = FALSE)

# create a ggplot2 object and save it
gg_guido <- 
  guido %>% 
  hm_plot(slot_name = 'qd', 
          col_name = list( c('q(m3/s)', 'q_set') ), 
          interactive = FALSE, 
          line_color = c('dodgerblue', 'forestgreen'),
          y_lab = 'Q(m3/s)', x_lab = "",
          legend_lab = c('original', 'set')) 
          
ggsave(filename = "guido_qd.png", 
       plot = gg_guido,
       path = "my_path",
       width = 5, 
       height = 3,
       units = "cm")

We are also interested on monthly mean discharge, so we need to aggregate the daily discharge

# aggregate and relocate
guido <- 
  guido %>%
  hm_agg(slot_name = 'qd', col_name = 'q_set', 
         fun = 'mean', 
         period = 'monthly', 
         out_name = 'q_mean', 
         relocate = 'qm' )

Try to plot the mean monthly discharge…

Installation

To install it, use:

install.packages("hydrotoolbox")

Note

This package is an improved version of hydroToolkit. After using it in a big project, I realize that classes could be reduced and that many aspects of the arguments should be rewritten to make the package more user-friendly. For this new package (hydrotoolbox) I also decided to follow the tidyverse style, so people used to the tidyverse syntax should feel comfortable on using hydrotoolbox.

In summary hydroToolkit will be achieved soon. If you were using it, I strongly recommend you to move to hydrotoolbox.