Introduction to getCRUCLdata

The getCRUCLdata package provides functions that automate importing CRU CL v. 2.0 climatology data into R, facilitate the calculation of minimum temperature and maximum temperature, and formats the data into a tidy data frame as a tibble::tibble() object or a list() of raster::stack() objects for use in an R session.

CRU CL v. 2.0 data are a gridded climatology of 1961-1990 monthly means released in 2002 and cover all land areas (excluding Antarctica) at 10 arcminutes (0.1666667 degree) resolution. For more information see the description of the data provided by the University of East Anglia Climate Research Unit (CRU), https://crudata.uea.ac.uk/cru/data/hrg/tmc/readme.txt.

Changes to original CRU CL v. 2.0 data

This package automatically converts elevation values from kilometres to metres.

This package crops all spatial outputs to an extent of ymin = -60, ymax = 85, xmin = -180, xmax = 180. Note that the original wind data include land area for parts of Antarctica.

Using getCRUCLdata

Logical arguments are used to specify the climatology elements to retrieve and parse. The create_CRU_*() functions require an additional parameter, dsn to be provided that states where the files are locally stored. The arguments for selecting the climatology elements for importing are:

pre Logical. Fetch precipitation (millimetres/month) from server and return in the data? Defaults to FALSE.
pre_cv Logical. Fetch cv of precipitation (percent) from server and return in the data? Defaults to FALSE.
rd0 Logical. Fetch wet-days (number days with >0.1 millimetres rain per month) and return in the data? Defaults to FALSE.
dtr Logical. Fetch mean diurnal temperature range (degrees Celsius) and return it in the data? Defaults to FALSE.
tmp Logical. Fetch temperature (degrees Celsius) and return it in the data? Defaults to FALSE.
tmn Logical. Calculate minimum temperature values (degrees Celsius) and return it in the data? Defaults to FALSE.
tmx Logical. Calculate maximum temperature (degrees Celsius) and return it in the data? Defaults to FALSE.
reh Logical. Fetch relative humidity and return it in the data? Defaults to FALSE.
sunp Logical. Fetch sunshine, percent of maximum possible (percent of day length) and return it in data? Defaults to FALSE.
frs Logical. Fetch ground-frost records (number of days with ground- frost per month) and return it in data? Defaults to FALSE.
wnd Logical. Fetch 10m wind speed (metres/second) and return it in the data? Defaults to FALSE.
elv Logical. Fetch elevation (and convert to metres from kilometres) and return it in the data? Defaults to FALSE.
dsn For create_CRU_stack() and create_CRU_df() only. Local file path where CRU CL v. 2.0 .dat.gz files are located.

Creating tidy data frames for use in R

The get_CRU_df() function automates the download process and creates tidy data frames as a tibble::tibble() of the CRU CL v. 2.0 climatology elements.

library(getCRUCLdata)

CRU_data <- get_CRU_df(pre = TRUE,
                       pre_cv = TRUE,
                       rd0 = TRUE,
                       tmp = TRUE,
                       dtr = TRUE,
                       reh = TRUE,
                       tmn = TRUE,
                       tmx = TRUE,
                       sunp = TRUE,
                       frs = TRUE,
                       wnd = TRUE,
                       elv = TRUE)

CRU_data

Perhaps you only need one or two elements, it is easy to create a tidy data frame of mean temperature only.

t <- get_CRU_df(tmp = TRUE)

Plotting data from the tidy dataframe

Now that we have the data, we can plot it easily using ggplot2 and the viridis package for the colour scale.

if (!require("ggplot2")) {
  install.packages(ggplot2)
}
if (!require("viridis")) {
  install.packages("viridis")
}

library(ggplot2)
library(viridis)

Now that the required packages are installed and loaded, we can generate a figure of temperatures using ggplot2 to map the 12 months.

ggplot(data = t, aes(x = lon, y = lat)) +
  geom_raster(aes(fill = tmp)) +
  scale_fill_viridis(option = "inferno") +
  coord_quickmap() +
  ggtitle("Global Mean Monthly Temperatures 1961-1990") +
  facet_wrap(~ month, nrow = 4)

Plot of global mean monthly temperatures 1961-1990

We can also generate a violin plot of the same data to visualise how the temperatures change throughout the year.

ggplot(data = t, aes(x = month, y = tmp)) +
  geom_violin() +
  ylab("Temperature (˚C)") +
  labs(title = "Global Monthly Mean Land Surface Temperatures From 1960-1991",
       subtitle = "Excludes Antarctica")

Violin plot of global mean temperatures 1961-1990

Saving the tidy `data.frame` as a CSV (comma separated values file) locally

Save the resulting tidy data.frame to local disk as a comma separated (CSV) file to local disk, using data.table's fwrite().

fwrite(x = t, file = "~/CRU_tmp.csv")

Creating raster stacks for use in R and saving for use in another GIS

For working with spatial data, getCRUCLdata provides a function that create lists of raster stacks of the data.

The get_CRU_stack() functions provide similar functionality to get_CRU_df(), but rather than returning a tidy data frame, it returns a list of raster::stack() objects for use in an R session.

The get_CRU_stack() function automates the download process and creates a raster::stack() object of the CRU CL v. 2.0 climatology elements. Illustrated here is creating a raster::stack() of all CRU CL v. 2.0 climatology elements available.

CRU_stack <- get_CRU_stack(pre = TRUE,
                           pre_cv = TRUE,
                           rd0 = TRUE,
                           tmp = TRUE,
                           dtr = TRUE,
                           reh = TRUE,
                           tmn = TRUE,
                           tmx = TRUE,
                           sunp = TRUE,
                           frs = TRUE,
                           wnd = TRUE,
                           elv = TRUE)

The create_CRU_stack() function works in the same way with only one minor difference. You must supply the location of the files on the local disk (dsn) that you wish to import.

t <- create_CRU_stack(tmp = TRUE, dsn = "~/Downloads")

Plotting raster stacks of tmin and tmax

Because the stacks are in a list(), we need to access each element of the list individually to plot them, that's what the [[1]] or [[2]] is, the first or second element of the list. Here using [[1]] we will plot the monthly average minimum temperature for all twelve months.

library(raster)

plot(t[[1]])

Plot of raster layers from minimum temperature stack

To plot only one month from the stack is also possible. Here we plot maximum temperature for July. Note that we use indexing [[2]] as before but append a $jul to the object. This is the name of the layer in the raster::stack(). So, we are telling R to plot the second object in the tmn_tmx list, which is tmx and from that raster stack, plot only the layer for July.

plot(t[[2]]$jul)

Plot of maximum temperatures for July

Saving raster objects to local disk

The raster stack objects can be saved to disk as geotiff files (others are available, see help for raster::writeRaster() and raster::writeFormats() for more options) on the Data directory with a tmn or tmx prefix to the month for a file name.

library(raster)

dir.create(file.path("~/Data"), showWarnings = FALSE)
writeRaster(
  t$tmn,
  filename = file.path("~/Data/tmn_", names(t$tmn)),
  bylayer = TRUE,
  format = "GTiff"
  )

  writeRaster(
  t$tmx,
  filename = file.path("~/Data/tmx_", names(t$tmn)),
  bylayer = TRUE,
  format = "GTiff"
  )

Advanced usage

Caching files for later use

When using the get_CRU_df() or get_CRU_stack() functions, files may be cached in the users' local space for later use (optional) or stored in a temporary directory and deleted when the R session is closed and not saved (this is the default behaviour already illustrated above). Illustrated here, create a tidy data frame of all CRU CL v. 2.0 climatology elements available and cache them to save time in the future. In order to take advantage of the cached data, you must use the get_CRU_df() function again in the future. This functionality is somewhat modelled after the raster::getData() function that will not download files that already exist in the working directory, however in this case the function is portable and it will work for any working directory. That is, if you have cached the data and you use get_CRU_df() again, it will use the cached data no matter what working directory you are in. This functionality will be most useful for writing scripts that may be used several times rather than just once off or if you frequently use the data in multiple analyses the data will not be downloaded again if they have been cached.

Create a list of raster stacks of maximum and minimum temperature. To take advantage of the previously cached files and save time by not downloading files, specify cache = TRUE.

tmn_tmx <- get_CRU_stack(tmn = TRUE,
                         tmx = TRUE,
                         cache = TRUE)

Downloading files outside of R using another FTP program

A second set of functions, create_CRU_df() and create_CRU_stack(), is provided for users that may have connectivity issues or simply wish to use something other than R to download the data files. You may also wish to use these if you want to download the data and specify where it is stored rather than using the cache functionality of get_CRU_df() and get_CRU_stack().

The create_CRU_df() and create_CRU_stack() functions work in the same way as get_CRU_df() and get_CRU_stack() functions with only one major difference. You must supply the location of the files on the local disk (dsn) that you wish to import. That is, the CRU CL v. 2.0 data files must be downloaded prior to the use of these functions using a program external to R, e.g., FileZilla or some other FTP program. In this instance it is recommended to use an FTP client (e.g., FileZilla) to download the files, save locally and use one of these functions to import the data into R and generate your desired object to work with.

t <- create_CRU_df(tmp = TRUE, dsn = "~/Downloads")

CRU CL v. 2.0 reference and abstract

Mark New (1,*), David Lister (2), Mike Hulme (3), Ian Makin (4)
A high-resolution data set of surface climate over global land areas Climate Research, 2000, Vol 21, pg 1-25
(1) School of Geography and the Environment, University of Oxford, Mansfield Road, Oxford OX1 3TB, United Kingdom
(2) Climatic Research Unit, and (3) Tyndall Centre for Climate Change Research, both at School of Environmental Sciences, University of East Anglia, Norwich NR4 7TJ, United Kingdom
(4) International Water Management Institute, PO Box 2075, Colombo, Sri Lanka

ABSTRACT: We describe the construction of a 10-minute latitude/longitude data set of mean monthly surface climate over global land areas, excluding Antarctica. The climatology includes 8 climate elements - precipitation, wet-day frequency, temperature, diurnal temperature range, relative humidity,sunshine duration, ground frost frequency and windspeed - and was interpolated from a data set of station means for the period centred on 1961 to 1990. Precipitation was first defined in terms of the parameters of the Gamma distribution, enabling the calculation of monthly precipitation at any given return period. The data are compared to an earlier data set at 0.5 degrees latitude/longitude resolution and show added value over most regions. The data will have many applications in applied climatology, biogeochemical modelling, hydrology and agricultural meteorology and are available through the School of Geography Oxford (http://www.geog.ox.ac.uk), the International Water Management Institute “World Water and Climate Atlas” (http://www.iwmi.org) and the Climatic Research Unit (http://www.cru.uea.ac.uk).