An R binding package for calling Google Earth Engine API from within R. Several functions have been implemented to make simple the connection with the R spatial ecosystem.
• Installation • Hello World • How does rgee work? • Guides • Contributing • Citation • Credits
Google Earth Engine is a cloud-based platform that allows users to have an easy access to a petabyte-scale archive of remote sensing data and run geospatial analysis on Google's infrastructure. Currently, Google offers support only for Python and JavaScript. rgee
will fill the gap starting to provide support to R!. Below you will find the comparison between the syntax of rgee
and the two Google-supported client libraries.
JS (Code Editor) | Python | R |
---|---|---|
var db = 'CGIAR/SRTM90_V4'
var image = ee.Image(db)
print(image.bandNames())
#> 'elevation' |
import ee
ee.Initialize()
db = 'CGIAR/SRTM90_V4'
image = ee.Image(db)
image.bandNames().getInfo()
#> [u'elevation'] |
library(rgee)
ee_Initialize()
db <- 'CGIAR/SRTM90_V4'
image <- ee$Image(db)
image$bandNames()$getInfo()
#> [1] "elevation" |
Quite similar, isn't it?. However, there are additional smaller changes should consider when using Google Earth Engine with R. Please check the consideration section before you start coding!
Install from CRAN with:
install.packages("rgee")
Install the development versions from github with
library(remotes)
install_github("r-spatial/rgee")
Additionally, rgee
depends on the Python packages: numpy and ee. To install them, users can follow any of these three methods:
- Use ee_install (Highly recommend for users with no experience with Python environments)
rgee::ee_install()
- Use ee_install_set_pyenv (Recommend for users with experience with Python environments)
rgee::ee_install_set_pyenv(
py_path = "/home/csaybar/.virtualenvs/rgee/bin/python", # Change it for your own Python PATH
py_env = "rgee" # Change it for your own Python ENV
)
Take into account that the Python PATH you set must have installed the Earth Engine Python API and numpy. The use of miniconda/anaconda is mandatory for Windows users, Linux and MacOS users could also use virtualenv. See reticulate documentation for more details.
Other option, only possible for MacOS and Linux, is just set the Python PATH:
rgee::ee_install_set_pyenv(
py_path = "/usr/bin/python3",
py_env = NULL
)
However, rgee::ee_install_upgrade and reticulate::py_install will not work until you set a Python ENV.
- Use the Python PATH setting support that offer Rstudio v.1.4 >. See this tutorial.
After install Python dependencies
(and Restart R!!), you might use the function below for checking the status of rgee.
ee_check() # Check non-R dependencies
library(reticulate)
library(rgee)
# 1. Initialize the Python Environment
ee_Initialize()
# 2. Install geemap in the same Python ENV that use rgee
py_install("geemap")
gm <- import("geemap")
Upgrade the earthengine-api
library(rgee)
ee_Initialize()
ee_install_upgrade()
- All
rgee
functions have the prefix ee_. Auto-completion is your friend :). - Full access to the Earth Engine API with the prefix ee$....
- Authenticate and Initialize the Earth Engine R API with ee_Initialize. It is necessary once by session!.
rgee
is "pipe-friendly", we re-exports %>%, butrgee
does not require its use.
1. Compute the trend of night-time lights (JS version)
Authenticate and Initialize the Earth Engine R API.
library(rgee)
ee_Initialize()
Adds a band containing image date as years since 1991.
createTimeBand <-function(img) {
year <- ee$Date(img$get('system:time_start'))$get('year')$subtract(1991L)
ee$Image(year)$byte()$addBands(img)
}
Map the time band creation helper over the night-time lights collection.
collection <- ee$
ImageCollection('NOAA/DMSP-OLS/NIGHTTIME_LIGHTS')$
select('stable_lights')$
map(createTimeBand)
Compute a linear fit over the series of values at each pixel, visualizing the y-intercept in green, and positive/negative slopes as red/blue.
col_reduce <- collection$reduce(ee$Reducer$linearFit())
col_reduce <- col_reduce$addBands(
col_reduce$select('scale'))
ee_print(col_reduce)
Create a interactive visualization!
Map$setCenter(9.08203, 47.39835, 3)
Map$addLayer(
eeObject = col_reduce,
visParams = list(
bands = c("scale", "offset", "scale"),
min = 0,
max = c(0.18, 20, -0.18)
),
name = "stable lights trend"
)
Install and load tidyverse
and sf
R package, after that, initialize the Earth Engine R API.
library(tidyverse)
library(rgee)
library(sf)
ee_Initialize()
Read the nc
shapefile.
nc <- st_read(system.file("shape/nc.shp", package = "sf"), quiet = TRUE)
Map each image from 2001 to extract the monthly precipitation (Pr) from the Terraclimate dataset
terraclimate <- ee$ImageCollection("IDAHO_EPSCOR/TERRACLIMATE") %>%
ee$ImageCollection$filterDate("2001-01-01", "2002-01-01") %>%
ee$ImageCollection$map(function(x) x$select("pr")) %>% # Select only precipitation bands
ee$ImageCollection$toBands() %>% # from imagecollection to image
ee$Image$rename(sprintf("%02d",1:12)) # rename the bands of an image
Extract monthly precipitation values from the Terraclimate ImageCollection through ee_extract
. ee_extract
works similar to raster::extract
, you just need to define: the ImageCollection object (x), the geometry (y), and a function to summarize the values (fun).
ee_nc_rain <- ee_extract(x = terraclimate, y = nc["NAME"], sf = FALSE)
Use ggplot2 to generate a beautiful static plot!
ee_nc_rain %>%
pivot_longer(-NAME, names_to = "month", values_to = "pr") %>%
mutate(month, month=gsub("X", "", month)) %>%
ggplot(aes(x = month, y = pr, group = NAME, color = pr)) +
geom_line(alpha = 0.4) +
xlab("Month") +
ylab("Precipitation (mm)") +
theme_minimal()
3. Create an NDVI-animation (JS version)
Install and load sf
, after that, initialize the Earth Engine R API.
library(magick)
library(rgee)
library(sf)
ee_Initialize()
Define the regional bounds of animation frames and a mask to clip the NDVI data by.
mask <- system.file("shp/arequipa.shp", package = "rgee") %>%
st_read(quiet = TRUE) %>%
sf_as_ee()
region <- mask$geometry()$bounds()
Retrieve the MODIS Terra Vegetation Indices 16-Day Global 1km dataset as an ee.ImageCollection
and select the NDVI band.
col <- ee$ImageCollection('MODIS/006/MOD13A2')$select('NDVI')
Group images by composite date
col <- col$map(function(img) {
doy <- ee$Date(img$get('system:time_start'))$getRelative('day', 'year')
img$set('doy', doy)
})
distinctDOY <- col$filterDate('2013-01-01', '2014-01-01')
Define a filter that identifies which images from the complete collection match the DOY from the distinct DOY collection.
filter <- ee$Filter$equals(leftField = 'doy', rightField = 'doy')
Define a join; convert the resulting FeatureCollection to an ImageCollection.
join <- ee$Join$saveAll('doy_matches')
joinCol <- ee$ImageCollection(join$apply(distinctDOY, col, filter))
Apply median reduction among matching DOY collections.
comp <- joinCol$map(function(img) {
doyCol = ee$ImageCollection$fromImages(
img$get('doy_matches')
)
doyCol$reduce(ee$Reducer$median())
})
Define RGB visualization parameters.
visParams = list(
min = 0.0,
max = 9000.0,
bands = "NDVI_median",
palette = c(
'FFFFFF', 'CE7E45', 'DF923D', 'F1B555', 'FCD163', '99B718', '74A901',
'66A000', '529400', '3E8601', '207401', '056201', '004C00', '023B01',
'012E01', '011D01', '011301'
)
)
Create RGB visualization images for use as animation frames.
rgbVis <- comp$map(function(img) {
do.call(img$visualize, visParams) %>%
ee$Image$clip(mask)
})
Define GIF visualization parameters.
gifParams <- list(
region = region,
dimensions = 600,
crs = 'EPSG:3857',
framesPerSecond = 10
)
Use ee_utils_gif_* functions to render the GIF animation and add some texts.
animation <- ee_utils_gif_creator(rgbVis, gifParams, mode = "wb")
animation %>%
ee_utils_gif_annotate(
text = "NDVI: MODIS/006/MOD13A2",
size = 15, color = "white",
location = "+10+10"
) %>%
ee_utils_gif_annotate(
text = dates_modis_mabbr,
size = 30,
location = "+290+350",
color = "white",
font = "arial",
boxcolor = "#000000"
) # -> animation_wtxt
# ee_utils_gif_save(animation_wtxt, path = "raster_as_ee.gif")
rgee
is not a native Earth Engine API like the Javascript or Python client, to do this would be extremely hard, especially considering that the API is in active development. So, how is it possible to run Earth Engine using R? the answer is reticulate. reticulate
is an R package designed to allow a seamless interoperability between R and Python. When an Earth Engine request is created in R, reticulate
will transform this piece into Python. Once the Python code is obtained, the Earth Engine Python API
transform the request to a JSON
format. Finally, the request is received by the Google Earth Engine Platform thanks to a Web REST API. The response will follow the same path.
Created by: - EN and POR: Andres Luiz Lima Costa https://bit.ly/3p1DFm9 - SPA: Antony Barja Ingaruca https://barja8.github.io/
Please note that the rgee
project is released with a Contributor Code of Conduct. By contributing to this project, you agree to abide by its terms.
👍 Thanks for taking the time to contribute! 🎉👍 Please review our Contributing Guide.
Think rgee is useful? Let others discover it, by telling them in person via Twitter or a blog post.
Using rgee for a paper you are writing? Consider citing it
citation("rgee")
To cite rgee in publications use:
C Aybar, Q Wu, L Bautista, R Yali and A Barja (2020) rgee: An R
package for interacting with Google Earth Engine Journal of Open
Source Software URL https://github.com/r-spatial/rgee/.
A BibTeX entry for LaTeX users is
@Article{,
title = {rgee: An R package for interacting with Google Earth Engine},
author = {Cesar Aybar and Quisheng Wu and Lesly Bautista and Roy Yali and Antony Barja},
journal = {Journal of Open Source Software},
year = {2020},
}
We want to offer a special thanks 🙌 👏 to Justin Braaten for his wise and helpful comments in the whole development of rgee. As well, we would like to mention the following third-party R/Python packages for contributing indirectly to the improvement of rgee: