SNAPverse R package: snapgrid
This article is originally published at https://blog.snap.uaf.edu
This post introduces the
snapgrid R package, a data package in the SNAPverse R package ecosystem. It contains a number of convenient
raster objects sourced from spatially explicit maps commonly used at SNAP.
snapgrid contains a collection of rasterized maps focused on Alaska and Canada, including vegetation input for the ALFRESCO wildfire model, fire management options map layers, domain template layers for the Alaska “statewide” classic ALFRESCO domain, and 1-km ALFRESCO and 2-km climate templates for the Alaska/western Canada domain. This is the second R package member of the SNAPverse introduced here, following
snappoly, and like
snappoly it is one of the simplest. It provides auxiliary data rather than SNAP data end products.
snapgrid is essentially the rasterized map complement to
This is a very simple introduction to
snapgrid. A subsequent post will provide more examples. That will also be followed by a description of how to integrate maps from
snapgrid in one analysis when you need to combine multiple source maps, but they are in vector and raster formats, respectively.
The following maps are included in the current version of the package:
akcan2km: Alaska/western Canada 2-km downscaled climate domain template layer.
akcan1km: Alaska/western Canada 1-km ALFRESCO domain template layer.
ak1km: “Statewide” classic ALFRESCO domain template layer.
swveg: Vegation map layer for the classic “statewide” ALFRESCO model spatial domain.
swfmo: Statewide fire management options zones map layer.
swratios: Statewide fire management options zone ratios map layer.
swflam: Statewide GBM vegetation-mediated, climate-driven landscape flammability mask map layer.
You can install snapgrid from github with:
# install.packages('devtools') devtools::install_github("leonawicz/snapgrid")
Data sets from the above list can immediately be referred to directly once
snapgrid is loaded. To access documentation on the package from R, use
?snapgrid, or for a data set for example,
library(snapgrid) swveg #> class : RasterLayer #> dimensions : 1450, 1374, 1992300 (nrow, ncol, ncell) #> resolution : 1000, 1000 (x, y) #> extent : -656204.4, 717795.6, 940439.8, 2390440 (xmin, xmax, ymin, ymax) #> coord. ref. : +proj=aea +lat_1=55 +lat_2=65 +lat_0=50 +lon_0=-154 +x_0=0 +y_0=0 +datum=NAD83 +units=m +no_defs +ellps=GRS80 +towgs84=0,0,0 #> data source : in memory #> names : swveg #> values : 0, 7 (min, max)
snapgrid, these auxiliary map layers can easily merge into existing analysis workflows in R just by loading the package and referring to their names, or without even explicitly loading the package if you are explicit with the package namespace instead, e.g., by calling
snapgrid::swveg. The maps are immediately available in your R session to be used for subsetting other data sets, computing region-specific statistics, or even just providing a quick, convenient graphic of the spatial domain you may we working with for documentation purposes.
Like any of the SNAPverse R packages, even if you are not using other R packages and functions that are part of the broader SNAPverse for analyzing your SNAP data set,
snapgrid can be leveraged directly and independently as needed. There is no requirement to couple it in your workflow with other SNAPverse packages that pertain to statistical analysis. The SNAPverse R package ecosystem offers the independence to select only the pieces you need and inject them into you analysis chain where applicable.
Cleaning and curating is kept to a minimum for this package. Included raster layers cover multiple spatial domains, namely Alaska and Alaska/western Canada, but they all share the NAD83 Alaska Albers Equal Area Conic projection in common:
+proj=aea +lat_1=55 +lat_2=65 +lat_0=50 +lon_0=-154 +x_0=0 +y_0=0 +datum=NAD83 +units=m +no_defs +ellps=GRS80 +towgs84=0,0,0
Please visit source website for post related comments.