Working with spatial and climate data from GSODR and GSODRdata
Tomislav Hengl and Adam H Sparks
Introduction
The GSOD data allow for users to use spatial methods with it. Using various R packages such as sf, you can save data as a GeoPackage file which can be read by most GIS software packages or in R using R’s GIS capabilities with other contributed packages as well.
Following is an example of how you might download and save GSOD annual data for a given country, Philippines in this example, and convert it into a KML file for viewing in GoogleEarth. The second portion uses the same GeoPackage file to import the data back into R and combine the GSOD data with CHELSA data from the GSODRdata package available from GitHub and plot the station temperatures for daily GSOD, average monthly GSOD and CHELSA temperatures (1979-2013).
Example - Download and Plot Data for a Single Country
This example will demonstrate how to download data for Philippines for year 2010 and generate a spatial, year summary file, PHL-2010.gpkg, in the user’s home directory and link it with climate data from the GSODRdata package.
Download data for Philippines for year 2010 and generate a spatial, year summary file, PHL-2010.gpkg, in the user’s home directory.
library(GSODR)
PHL <- get_GSOD(years = 2010, country = "Philippines", max_missing = 5)
PHL <- st_as_sf(x = PHL,
coords = c("LON", "LAT"),
crs = "+proj=longlat +datum=WGS84")
write_sf(PHL, "~/PHL-2010.gpkg")Example - Use the GeoPackage File to Create a KML File
Using the GeoPackage that was just created, you can now create a KML file to open with Google Earth and visualise the data spatially and over time.
library(rgdal)
library(spacetime)
library(plotKML)
layers <- ogrListLayers(dsn = path.expand("~/PHL-2010.gpkg"))
pnts <- readOGR(dsn = path.expand("~/PHL-2010.gpkg"), layers[1])
# Plot results in Google Earth as a spacetime object:
pnts$DATE = as.Date(paste(pnts$YEAR, pnts$MONTH, pnts$DAY, sep = "-"))
row.names(pnts) <- paste("point", 1:nrow(pnts), sep = "")
tmp_ST <- STIDF(sp = as(pnts, "SpatialPoints"),
time = pnts$DATE - 0.5,
data = pnts@data[, c("TEMP", "STNID")],
endTime = pnts$DATE + 0.5)
shape = "http://maps.google.com/mapfiles/kml/pal2/icon18.png"
kml(tmp_ST, dtime = 24 * 3600, colour = TEMP, shape = shape, labels = TEMP,
file.name = "Temperatures_PHL_2010-2010.kml", folder.name = "TEMP")
system("zip -m Temperatures_PHL_2010-2010.kmz Temperatures_PHL_2010-2010.kml")Example - Compare the GSOD Weather Data From the Philippines With Climate
Data Provided by the GSODRdata Package.
The the GSODRdata package provides climate data from four climate data sets that have been formatted for easy use with the GSODR package. Following is a description how to install the GSODRdata package and use it to visualise the weather data for the Philippines in 2010 against CHELSA data.
CHELSA (Climatologies at high resolution for the earth’s land surface areas) are climate data at 30 arc seconds for the earth land surface areas.
Description of CHELSA data from CHELSA website
CHELSA is a high resolution (30 arc sec) climate data set for the earth land surface areas currently under development in coorporation [sic] with the Department of Geography of the University of Hamburg (Prof. Dr. Jürgen Böhner, Dr. Olaf Conrad, Tobias Kawohl), the Swiss Federal Institute for Forest, Snow and Landscape Research WSL (Prof. Dr. Niklaus Zimmermann), the University of Zurich (Dr. Dirk N. Karger, Dr. Michael Kessler), and the University of Göttingen (Prof. Dr. Holger Kreft). It includes monthly mean temperature and precipitation patterns for the time period 1979-2013. CHELSA is based on a quasi-mechanistical statistical downscaling of the ERA interim global circulation model (http://www.ecmwf.int/en/research/climate-reanalysis/era-interim) with a GPCC bias correction (https://www.dwd.de/EN/ourservices/gpcc/gpcc.html) and is freely available in the download section.
See http://chelsa-climate.org for more information on these data.
# install.packages("devtools")
devtools::install_github("adamhsparks/GSODRdata")Now that the extra data have been installed, take a look at the CHELSA data that are one of the data sets included in the GSODRdata package.
library(GSODRdata)
str(CHELSA)
#> 'data.frame': 23927 obs. of 46 variables:
#> $ STNID : chr "008268-99999" "010014-99999" "010015-99999" "010882-99999" ...
#> $ CHELSA_bio10_1979-2013_V1_1: num 30 14.3 12.5 12.6 15.2 16.3 13.9 12.7 13.4 12.9 ...
#> $ CHELSA_bio11_1979-2013_V1_1: num 5.8 1 -3.2 -7 -0.2 -2.2 -3 -6.9 -2.1 -1.2 ...
#> $ CHELSA_bio1_1979-2013_V1_1 : num 18.3 7.1 4.2 2.3 7 6.6 4.8 2.4 5 5.2 ...
#> $ CHELSA_bio12_1979-2013_V1_1: num 214 1889 2209 563 1710 ...
#> $ CHELSA_bio13_1979-2013_V1_1: num 54.6 223.7 254.7 73 200.6 ...
#> $ CHELSA_bio14_1979-2013_V1_1: num 0.1 87.7 108.9 26.7 78.5 ...
#> $ CHELSA_bio15_1979-2013_V1_1: num 104.9 30.3 30.1 30.7 32.1 ...
#> $ CHELSA_bio16_1979-2013_V1_1: num 155 651 753 216 590 ...
#> $ CHELSA_bio17_1979-2013_V1_1: num 0.6 273.5 332.6 85.5 244.7 ...
#> $ CHELSA_bio18_1979-2013_V1_1: num 1.7 433.3 448.3 198.7 330.9 ...
#> $ CHELSA_bio19_1979-2013_V1_1: num 96.9 490.4 621.1 122 494.4 ...
#> $ CHELSA_bio2_1979-2013_V1_1 : num 23 12.5 16.7 20.1 15.6 17.1 16.4 18.6 14.8 13.3 ...
#> $ CHELSA_bio3_1979-2013_V1_1 : num 48.3 44.8 45.3 45.5 45 44.3 43.8 44.6 43.8 43.8 ...
#> $ CHELSA_bio4_1979-2013_V1_1 : num 884 486 592 734 573 ...
#> $ CHELSA_bio5_1979-2013_V1_1 : num 40.5 21.5 22.2 23.6 23.9 25.4 23.1 23 21.8 20.4 ...
#> $ CHELSA_bio6_1979-2013_V1_1 : num -7.2 -6.5 -14.5 -20.7 -10.7 -13.3 -14.3 -18.7 -11.9 -9.9 ...
#> $ CHELSA_bio7_1979-2013_V1_1 : num 47.7 28 36.8 44.3 34.7 38.7 37.4 41.7 33.8 30.3 ...
#> $ CHELSA_bio8_1979-2013_V1_1 : num 10.5 5.7 -1.7 12.5 1.3 8.5 6.8 12.5 7 7.2 ...
#> $ CHELSA_bio9_1979-2013_V1_1 : num 26.7 7.8 8.4 -0.1 12.2 3.8 9.6 0.1 9.8 9.3 ...
#> $ CHELSA_prec_10_1979-2013 : num 3.6 230.6 237.5 48.9 192.8 ...
#> $ CHELSA_prec_11_1979-2013 : num 10.5 219.9 237.8 41.3 189.6 ...
#> $ CHELSA_prec_1_1979-2013 : num 35.8 198.4 227.7 40.3 189.5 ...
#> $ CHELSA_prec_12_1979-2013 : num 25.4 215.5 248.3 40.5 201.5 ...
#> $ CHELSA_prec_1979-2013_land : num 214 1912 2170 559 1720 ...
#> $ CHELSA_prec_2_1979-2013 : num 46.5 153.3 188.4 32.5 153.7 ...
#> $ CHELSA_prec_3_1979-2013 : num 54.8 151.9 176.4 31.8 145 ...
#> $ CHELSA_prec_4_1979-2013 : num 28.3 101.1 114.4 26.6 91.6 ...
#> $ CHELSA_prec_5_1979-2013 : num 7 90.6 107 38.9 79 72.1 75.6 46.7 91.3 85.1 ...
#> $ CHELSA_prec_6_1979-2013 : num 1.2 100.5 110.1 58 84 ...
#> $ CHELSA_prec_7_1979-2013 : num 1 116.7 119.2 69.7 90.1 ...
#> $ CHELSA_prec_8_1979-2013 : num 0.3 165.6 159.6 72.7 121.6 ...
#> $ CHELSA_prec_9_1979-2013 : num 0.1 204 229.6 57.9 181.4 ...
#> $ CHELSA_temp_10_1979-2013 : num 19.2 8 4.3 2.3 7.2 6.7 5.7 2.4 5.6 5.9 ...
#> $ CHELSA_temp_11_1979-2013 : num 13.2 4.5 0.2 -3 3 2.2 1.2 -2.7 1.6 2.2 ...
#> $ CHELSA_temp_1_1979-2013 : num 4.9 1.2 -3.4 -7.2 -0.4 -2.4 -2.5 -7 -1.9 -1 ...
#> $ CHELSA_temp_12_1979-2013 : num 7.7 2.1 -2.7 -6.5 0.3 -1.4 -1.3 -6.3 -0.7 -0.1 ...
#> $ CHELSA_temp_1979-2013_land : num 18.3 7.1 4.1 2.3 6.9 6.5 5.1 2.5 5 5.2 ...
#> $ CHELSA_temp_2_1979-2013 : num 7 0.8 -3.2 -6.6 -0.2 -2.2 -2.8 -6.4 -2.1 -1.2 ...
#> $ CHELSA_temp_3_1979-2013 : num 12.2 2.3 -0.8 -3.3 2.2 0.8 -0.9 -2.7 -0.6 0.1 ...
#> $ CHELSA_temp_4_1979-2013 : num 18.3 5.1 3 1.5 5.8 5.1 3 1.7 2.9 3.1 ...
#> $ CHELSA_temp_5_1979-2013 : num 23.5 9 7.6 6.4 10.1 10.7 7.8 6.6 7.3 7.1 ...
#> $ CHELSA_temp_6_1979-2013 : num 28.2 12 10.9 10.9 13.2 14.7 11.9 11.3 11.1 10.5 ...
#> $ CHELSA_temp_7_1979-2013 : num 28.2 12 10.9 10.9 13.2 14.7 11.9 11.3 11.1 10.5 ...
#> $ CHELSA_temp_8_1979-2013 : num 29.7 14.2 12.2 12.1 14.9 15.8 14 12.2 13.3 12.8 ...
#> $ CHELSA_temp_9_1979-2013 : num 25.1 11.5 8.6 7.7 11.5 11.6 10.2 7.6 9.9 9.8 ...Compare the GSOD weather data from the Philippines with climatic data provided by the GSODR package in the CHELSA data set using dplyr functions to join the CHELSA and GSODR data for plotting.
library(dplyr)
#>
#> Attaching package: 'dplyr'
#> The following objects are masked from 'package:stats':
#>
#> filter, lag
#> The following objects are masked from 'package:base':
#>
#> intersect, setdiff, setequal, union
library(ggplot2)
library(reshape2)
cnames <- paste0("CHELSA_temp_", 1:12, "_1979-2013")
clim_temp <- CHELSA[CHELSA$STNID %in% pnts$STNID,
paste(c("STNID", cnames))]
clim_temp_df <- data.frame(STNID = rep(clim_temp$STNID, 12),
MONTHC = as.vector(sapply(1:12, rep,
times = nrow(clim_temp))),
CHELSA_TEMP = as.vector(unlist(clim_temp[, cnames])))
pnts$MONTHC <- as.numeric(paste(pnts$MONTH))
temp <- left_join(pnts@data, clim_temp_df, by = c("STNID", "MONTHC"))
#> Warning in left_join_impl(x, y, by$x, by$y, suffix$x, suffix$y): joining
#> factors with different levels, coercing to character vector
temp <- temp %>%
group_by(MONTH) %>%
mutate(AVG_DAILY_TEMP = round(mean(TEMP), 1))
df_melt <- na.omit(melt(temp[, c("STNID", "DATE", "CHELSA_TEMP", "TEMP", "AVG_DAILY_TEMP")],
id = c("DATE", "STNID")))
ggplot(df_melt, aes(x = DATE, y = value)) +
geom_point(aes(color = variable), alpha = 0.2) +
scale_x_date(date_labels = "%b") +
ylab("Temperature (C)") +
xlab("Month") +
labs(colour = "") +
scale_color_brewer(palette = "Dark2") +
facet_wrap( ~ STNID)
Comparison of GSOD daily values and average monthly values with CHELSA climate monthly values
Notes
Sources
CHELSA Climate Layers
CHELSA (climatic surfaces at 1 km resolution) is based on a quasi-mechanistical statistical downscaling of the ERA interim global circulation model (Karger et al. 2016). ESA’s CCI-LC cloud probability monthly averages are based on the MODIS snow products (MOD10A2). http://chelsa-climate.org/
Elevation Values
90m hole-filled SRTM digital elevation (Jarvis et al. 2008) was used to identify and correct/remove elevation errors in data for station locations between -60˚ and 60˚ latitude. This applies to cases here where elevation was missing in the reported values as well. In case the station reported an elevation and the DEM does not, the station reported is taken. For stations beyond -60˚ and 60˚ latitude, the values are station reported values in every instance. See https://github.com/ropensci/GSODR/blob/master/data-raw/fetch_isd-history.md for more detail on the correction methods.
WMO Resolution 40. NOAA Policy
Users of these data should take into account the following (from the NCEI website):
“The following data and products may have conditions placed on their international commercial use. They can be used within the U.S. or for non-commercial international activities without restriction. The non-U.S. data cannot be redistributed for commercial purposes. Re-distribution of these data by others must provide this same notification.” WMO Resolution 40. NOAA Policy
References
Jarvis, A., Reuter, H. I., Nelson, A., Guevara, E. (2008) Hole-filled SRTM for the globe Version 4, available from the CGIAR-CSI SRTM 90m Database (http://srtm.csi.cgiar.org)
Karger, D. N., Conrad, O., Bohner, J., Kawohl, T., Kreft, H., Soria-Auza, R. W., et al. (2016) Climatologies at high resolution for the Earth land surface areas. arXiv preprint arXiv:1607.00217. (http://chelsa-climate.org/)
Stachelek, J. (2016) Using the Geopackage Format with R. URL: https://jsta.github.io/2016/07/14/geopackage-r.html