cantempr

An R package to access long-term homogenized temperature data for 338 Canadian climate stations (for some stations data is available as far back as 1841). These data come from Vincent et al. (2012), and are available in station-specific text files from Environment and Climate Change Canada (see file 'Monthly mean of daily mean temperature').

Data citation: Vincent, L. A., X. L. Wang, E. J. Milewska, H. Wan, F. Yang, and V. Swail (2012) A second generation of homogenized Canadian monthly surface air temperature for climate trend analysis. Journal of Geophysical Research 117(D18110). https://doi.org/10.1029/2012JD017859

Installation

Install from GitHub with:

# first install package 'remotes' if necessary (a dependency of devtools)
# install.packages("remotes") 
remotes::install_github("patrickbarks/cantempr")

Usage

library(cantempr)

Fetch data

Temperature data can be fetched using the cantemp_fetch() function. Temperature data is available at monthly, seasonal, and annual intervals (selected with argument interval).

temp_annual <- cantemp_fetch(interval = "annual") # fetch annual data

Let's take a peek.

head(temp_annual)
#>   prov        station   stnid year interval temp flag
#> 1   BC SHAWNIGAN LAKE 1017230 1914   Annual  9.7    a
#> 2   BC SHAWNIGAN LAKE 1017230 1915   Annual  9.9    a
#> 3   BC SHAWNIGAN LAKE 1017230 1916   Annual  7.8    a
#> 4   BC SHAWNIGAN LAKE 1017230 1917   Annual  8.6    a
#> 5   BC SHAWNIGAN LAKE 1017230 1918   Annual   NA    M
#> 6   BC SHAWNIGAN LAKE 1017230 1919   Annual  8.7    a

We can also fetch station-specific metadata (including lat/lon/elevation, and details on the temporal extent of the temperature data) using the function cantemp_meta().

stn_metadata <- cantemp_meta()

Barcode plot

The cantempr package includes a function to produce 'temperature barcode plots' (see examples here). Here's an example barcode plot using a time-series of mean annual temperatures in Toronto, Ontario.

library(ggplot2)

# subset annual temperature data to Toronto
temp_to <- subset(temp_annual, station == "TORONTO")

# create barcode plot
cantemp_barcode(temp_to, x_breaks = seq(1850, 2010, 20)) +
  ggtitle(paste("Mean annual temperatures in Toronto since", min(temp_to$year)))

Seasonal temperature trends

Here's a more traditional plot of long-term temperature trends, by season, in Hay River, Northwest Territories.

# subset to seasonal temperature data for Hay River, Northwest Territories
temp_seasonal <- cantemp_fetch(interval = "seasonal")
temp_hr <- subset(temp_seasonal, station == "HAY RIVER" & !is.na(temp))

# plot
ggplot(temp_hr, aes(year, temp)) +
  geom_line() +
  geom_smooth(method = "loess") +
  scale_x_continuous(breaks = seq(1900, 2020, 20)) +
  scale_y_continuous(labels = function(x) paste0(x, "\u00B0C")) +
  facet_wrap(~ interval, ncol = 2, scales = "free_y") +
  ggtitle(paste("Mean seasonal temperatures in Hay River, NWT, since", min(temp_hr$year))) +
  theme_minimal() + theme(axis.title = element_blank())

Contributions

All contributions are welcome. Please note that this project is released with a Contributor Code of Conduct. By participating in this project you agree to abide by its terms.