Introduction to the imgw package (EN)

In imgw: Polish Meteorological and Hydrological Data

knitr::opts_chunk$set(
  collapse = TRUE,
  comment = "#>"
)
library(imgw)
library(tidyr)
library(dplyr)
options(scipen=999)

The main goal of the imgw package is to give convenient and programmable access to the Polish database of the archive meteorological and hydrological measurements.

It consists of tools that are:

• Giving more accessible access to the public data stored in the Institute of Meteorology and Water Management (IMGW)
• Downloading the selected data from among eleven different forms of data depending on the interval and the data type
• Providing the description of the parameters in two languages: Polish and English, as well as in the abbreviated form

Functions

The imgw package consists of nine main functions - four for meteorological data, four for hydrological data and one for sounding meteo:

1. Meteorological data:

2. meteo_hourly() - downloading meteorological the data with hourly interval

3. meteo_daily() - downloading meteorological the data with daily interval
4. meteo_monthly() - downloading meteorological the data with monthly interval

5. meteo() - downloading meteorological the data with interval the user want to choose

6. Hydrological data:

7. hydro() - downloading hydrological the data with interval the user want to choose

8. hydro_daily() - downloading hydrological the data with daily interval
9. hydro_monthly() - downloading hyrological the data with monthly interval

10. hydro_annual() - downloading hyrological the data with annual interval

11. Rawinsonde data:

12. meteo_sounding() -- downloading rawinsonde data (+metadata)

Most of the functions mentioned above have similar arguments allowing to choose:

• rank - type of the stations "synop", "climate", "precip" (only meteo functions)
• year - vector of selected years (e.g., 1966:2000)
• status - logical argument TRUE or FALSE; if TRUE the measurement statuses will be erased (only meteo functions)
• coords - logical argument TRUE or FALSE; if TRUE the coordinates are added to the stations
• station - selection of the stations; it can be the ID of stations (numeric) or name of the station (CAPITAL LETTERS (character))
• col_names - format of the columns names; three types of column names are possible: "short" - default, values with shorten names, "full" - full English description, "polish" - original names in the dataset

Database

imgw also has a few additional databases:

• hydro_abberv/meteo_abberv - a dictionary containing all original descriptions of parameters in both languages and the abbreviations

abbev = meteo_abbrev
head(abbev)

• hydro_stations/meteo_stations - datasets of almost all meteorological/hydrological stations containing their ID, latitude, and longitude

station = meteo_stations
head(station)

Application

We will show how to use our package and prepare the data for spatial analysis with the additional help of the dplyr and tidyr packages. Firstly, we download ten years (2001-2010) of monthly hydrological observations for all stations available and automatically add their spatial coordinates.

h = hydro_monthly(year = 2001:2010, coords = TRUE)
head(h)

The idex variable represents id of the extremum, where 1 means minimum, 2 mean, and 3 maximum.^[You can find more information about this in the hydro_abbrev dataset.] Hydrologists often use the maximum value so we will filter the data and select only the station id, hydrological year (hyy), latitude X and longitude Y. Next, we will calculate the mean maximum value of the flow on the stations in each year with dplyr's summarise(), and spread data by year using tidyr's spread() to get the annual means of maximum flow in the consecutive columns.

h2 = h %>%
  filter(idex == 3) %>%
  select(id, station, X, Y, hyy, Q) %>%
  group_by(hyy, id, station, X, Y) %>%
  summarise(srednie_roczne_Q = round(mean(Q, na.rm = TRUE),1)) %>% 
  spread(hyy, srednie_roczne_Q)

library(knitr)
kable(head(h2), caption = "Examplary data frame of hydrological preprocesssing.")

The result represent changing annual maximum average of water flow rate over the decade for all available stations in Poland. We can save it to:

• .csv with: write.csv(result, file = "result.csv", sep = ";",dec = ".", col.names = T, row.names = F). This command saves our result to result.csv where the column's separator is ;, the decimal is ., we are keeping the headers of columns and remove names of rows which are simply numbers of observation.

• .xlsx with: write.xlsx(result, file = "result.xlsx", sheetName = "Poland", append = FALSE) This command saves our result to result.xlsx with the name of the sheet Poland. Argument append=TRUE add the sheet to already existing xlsx file. To save data in .xlsx you have first to install package with command: install.packages("writexl"), and add it: library(writexl).

library(sf)
library(tmap)
library(rnaturalearth)
library(rnaturalearthdata)
world = ne_countries(scale = "medium", returnclass = "sf")

h3 = h2 %>% 
  filter(!is.na(X)) %>% 
  st_as_sf(coords = c("X", "Y"))

tm_shape(h3) + 
  tm_symbols(size = as.character(c(2001:2010)),
             title.size = "Średni przepływ maksymalny") +
  tm_facets(free.scales = FALSE, ncol = 4) + 
  tm_shape(world) + 
  tm_borders(col = "black", lwd = 2) +
  tm_layout(legend.position = c(-1.25, 0.05),
            outer.margins = c(0, 0.05, 0, -0.25),
            panel.labels = as.character(c(2001:2010)))

imgw documentation built on March 26, 2020, 7:37 p.m.