Get daily rainfall from CHIRPS, produced by the Climate Hazards Center (CHC), for points, polygons or an area. The data are read from the CHC server, which is the main source and serves both CHIRPS v3 (default) and CHIRPS v2.
Usage
get_chirps(object, dates, server = "CHC", ...)
# Default S3 method
get_chirps(object, dates, server = "CHC", as.matrix = FALSE, ...)
# S3 method for class 'SpatVector'
get_chirps(object, dates, server = "CHC", as.raster = TRUE, ...)
# S3 method for class 'SpatRaster'
get_chirps(object, dates, server = "CHC", as.raster = TRUE, ...)
# S3 method for class 'SpatExtent'
get_chirps(object, dates, server = "CHC", as.raster = TRUE, ...)
# S3 method for class 'sf'
get_chirps(object, dates, server = "CHC", as.sf = FALSE, ...)
# S3 method for class 'sfc'
get_chirps(object, dates, server = "CHC", as.sf = FALSE, ...)
# S3 method for class 'geojson'
get_chirps(object, dates, server = "CHC", as.geojson = FALSE, ...)Arguments
- object
input, an object of class
data.frame(or any other object that can be coerced todata.frame) with longitude and latitude in the first two columns,SpatVector,SpatRaster,SpatExtent,sf,sfcorgeojson- dates
a character or Date vector "YYYY-MM-DD", with one date for a single day or two dates for the start and end of the period
- server
a character, "CHC" (default and recommended) or "ClimateSERV". See the section on servers
- ...
additional arguments passed to
get_chirps_rawandextract. See details- as.matrix
logical, returns a
matrixwith one row per location and one column per day, named "YYYY-MM-DD"- as.raster
logical, returns an object of class
SpatRaster- as.sf
logical, returns an object of class
sf- as.geojson
logical, returns an object of class
geojson
Value
A data frame of CHIRPS data in the long format, ordered
by id and date, or a matrix, raster, sf or
geojson object as requested:
- id
the index for the rows in
object- date
the date of each value
- lon
the longitude as provided in
object- lat
the latitude as provided in
object- chirps
the CHIRPS value in mm
Details
Additional arguments
version: CHIRPS version, "3.0" (default) or "2.0". CHC retires CHIRPS v2 after the December 2026 release
type: CHIRPS v3 product, "rnl" (default) or "sat". See
get_chirps_raw
resolution: numeric, resolution of CHIRPS v2, 0.05 (default) or 0.25 degrees
operation: for polygons, the function used to summarise the cells of each polygon on each day
| operation | value | |
| max | = | 0 |
| min | = | 1 |
| median | = | 2 |
| sum | = | 4 |
| average | = | 5 (default value) |
Coverage: CHIRPS v3 covers 60°S to 60°N and CHIRPS v2 covers 50°S to 50°N, from 1981 to near present. Final data for a month arrive some weeks after it ends.
Polygons: with sf, SpatVector or geojson
polygons, each day is summarised over the cells that touch the polygon,
with the function set by operation. The lon and lat
columns give the centroid of each polygon.
Messages: each request states what it downloads, here daily
rainfall from CHIRPS, with the period, so it is not mistaken for
temperature from get_chirts. Set
options(chirps.quiet = TRUE) to hide these messages.
Indices: for rainfall indices (dry spells, heavy rain days,
intensity) use climatrends::rainfall(). The matrix returned with
as.matrix = TRUE is its input, see the examples.
Servers
CHC (default) reads the CHIRPS files from https://data.chc.ucsb.edu through terra. It serves CHIRPS v3 and v2, gives the value of the cell that holds each point, and handles many points and long periods in one call. It is the recommended source.
ClimateSERV is kept for backward compatibility. It serves CHIRPS v2 only, averaged over a small buffer around each point, so its values can differ from the CHC values. It sends several requests per location and is slow for more than a few points. ClimateSERV support for CHIRPS may be removed in a future release. Its extra arguments are dist, the buffer distance around each point, nQuadSegs, the number of segments per buffer quadrant, and operation as above.
SpatVector, SpatRaster and SpatExtent input always
uses CHC.
References
Funk, C., Peterson, P., Harrison, L. et al. (2026). Scientific Data, 13,
718.
doi:10.1038/s41597-026-07096-4
Funk C. et al. (2015). Scientific Data, 2, 150066.
doi:10.1038/sdata.2015.66
See also
get_chirps_raw for the rasters,
get_chirts for temperature, and the climatrends
package for rainfall and temperature indices
Examples
if (FALSE) { # interactive()
library("chirps")
library("terra")
dates = c("2017-12-15", "2017-12-31")
lonlat = data.frame(lon = c(-55.0281, -54.9857), lat = c(-2.8094, -2.8756))
# Case 1: return a data.frame in the long format
r1 = get_chirps(lonlat, dates)
# Case 2: a single day, from the CHIRPS v3 product based on IMERG
r2 = get_chirps(lonlat, "2017-12-15", type = "sat")
# Case 3: return a matrix, one column per day
r3 = get_chirps(lonlat, dates, as.matrix = TRUE)
# Case 4: input SpatVector and return raster
f = system.file("ex/lux.shp", package = "terra")
v = vect(f)
r4 = get_chirps(v, dates, as.raster = TRUE)
# Case 5: input SpatExtent and return a raster within the extent
area = ext(c(-66, -64, -6, -4))
r5 = get_chirps(area, dates)
# Case 6: total rainfall per polygon and day
r6 = get_chirps(v, dates, as.raster = FALSE, operation = 4)
# Case 7: rainfall indices with climatrends
if (requireNamespace("climatrends", quietly = TRUE)) {
climatrends::rainfall(r3, day.one = dates[1], span = ncol(r3))
}
}