Get LERI data over a region of interest and save a GeoTIFF
Maxwell B. Joseph
2019-09-06
The leri R package provides easy access to the Landscape Evaporative Response Index (LERI) data - an experimental drought monitoring and early warning guidance tool produced by the National Oceanic and Atmospheric Administration.
The LERI product is available from the year 2000 to present at a 1 km spatial resolution over the continental United States, at the following timescales:
- 1, 3, 7, and 12 month
- 8 day accumulated or non-accumulated from April - October
More information on the LERI product is available on the NOAA LERI homepage.
This vignette covers a common use case acquiring data over a region of interest defined by a shapefile, masking the LERI data to that region, and saving GeoTIFF files containing data for the region of interest.
Defining a region of interest
By default, the leri package returns data for the continental United States, southern parks of Canada, and northern parts of Mexico. But, you may only be interested in a region of interest, as defined by a shapefile. Here, you will load a shapefile for the state of North Carolina that is distributed by default with the sf package.
If you are using a different shapefile, replace system.file("shape/nc.shp", package="sf") with its file path, e.g., st_read("path/to/file.shp").
The roi object contains multiple columns of data, and a geometry column that contains spatial information on the region of interest, which in this case consists of multiple counties.
Because you don’t necessarily care about each county, but rather you want the entire state (including all counties) you can use a spatial union to join data from all counties:
Acquiring LERI data
To acquire LERI data, you can use the get_leri() function. You will fetch the 8 day accumulated timescale data for the week of August 13, 2018:
The leri_raster object is a RasterLayer, and you can see information on the spatial extent, resolution, and coordinate reference system by printing the object:
leri_raster
#> class : RasterLayer
#> dimensions : 2834, 6612, 18738408 (nrow, ncol, ncell)
#> resolution : 0.009, 0.009 (x, y)
#> extent : -126, -66.492, 23.994, 49.5 (xmin, xmax, ymin, ymax)
#> crs : +init=epsg:4326 +proj=longlat +datum=WGS84 +no_defs +ellps=WGS84 +towgs84=0,0,0
#> source : memory
#> names : LERI_8day.ac_Apr01.Aug20_2018.nc
#> values : 3.278688, 96.72131 (min, max)
Plot the data with a custom color palette to see what the data look like:
Masking to the region of interest
Now you want to subset or mask the LERI data to the region of interest. First, you need to ensure that the raster data and the polygon for the region of interest have the same coordinate reference system.
Now, graphically verify that they align as expected:
Now, you can crop the LERI data to match extents with the region of interest, then mask the raster set all values outside of the region of interest to NA. Because the raster package requires sp objects, rather than sf objects, you will coerce our roi to a sp object first.
You can plot the masked raster along with the ROI to confirm:
Saving GeoTIFF output
To write a GeoTIFF file of our masked_leri object, you can use writeRaster: