example_data
twinkle includes three example datasets to augment those available in the stars package and elsewhere. These include examples of
-
a small subset of SST satellite data,
-
a
starsobject of the volcano dataset, -
and a tiny toy dataset.
One month of daily
MUR
SST rasters (as one GeoTIFF) from 2014 covering Penobscot Bay, Maine.
The bounding box [west, east, south, north] is
[-69.2, -68.49, 43.78, 44.5]. Dates provided are 2014-06-01 through
2014-06-30. Also provided are daily sst slope and daily cumulative sst
(origin January 1). These are named 20140601-20140630-sst.tif,
20140601-20140630-sst_slope.tif amd 20140601-20140630-sst_cum.tif.
sst <- stars::read_stars(system.file("datasets/20140601-20140630-sst.tif",
package = "twinkle"))
# stars object with 3 dimensions and 1 attribute
# attribute(s):
# 20140601-20140630-sst.tif
# Min. :281.4
# 1st Qu.:283.7
# Median :284.5
# Mean :284.4
# 3rd Qu.:285.2
# Max. :287.6
# NA's :46050
# dimension(s):
# from to offset delta refsys point values x/y
# x 1 71 -69.195 0.01 WGS 84 FALSE NULL [x]
# y 1 72 44.505 -0.01 WGS 84 FALSE NULL [y]
# band 1 30 NA NA NA NA NULL
slope <- stars::read_stars(system.file("datasets/20140601-20140630-sst_slope.tif",
package = "twinkle"))
cum <- stars::read_stars(system.file("datasets/20140601-20140630-sst_cum.tif",
package = "twinkle"))These data are derived from… > JPL MUR MEaSUREs Project. 2015. GHRSST Level 4 MUR Global Foundation Sea Surface Temperature Analysis. Ver. 4.1. PO.DAAC, CA, USA. Dataset accessed [2021-02-08] at https://doi.org/10.5067/GHGMR-4FJ04
100 points within Penobscot Bay, selected at random, stored in
geopackage format. Use
sf::read_sf(system.file("datasets/penbay-points.gpkg", package = 'twinkle'))
to read them in.
pts <- sf::read_sf(system.file("datasets/penbay-points.gpkg", package = 'twinkle'))
# Simple feature collection with 100 features and 5 fields
# geometry type: POINT
# dimension: XY
# bbox: xmin: -69.18 ymin: 43.79 xmax: -68.49 ymax: 44.41
# geographic CRS: WGS 84
# # A tibble: 100 x 6
# index cell col row band geom
# <int> <dbl> <dbl> <dbl> <dbl> <POINT [°]>
# 1 95384 3368 31 48 19 (-68.89 44.03)
# 2 19875 4539 66 64 4 (-68.54 43.87)
# 3 136212 3300 34 47 27 (-68.86 44.04)
# 4 145992 2856 16 41 29 (-69.04 44.1)
# 5 130243 2443 29 35 26 (-68.91 44.16)
# 6 78418 1738 34 25 16 (-68.86 44.26)
# 7 60452 4220 31 60 12 (-68.89 43.91)
# 8 104298 2058 70 29 21 (-68.5 44.22)
# 9 55121 4001 25 57 11 (-68.95 43.94)
# 10 64280 2936 25 42 13 (-68.95 44.09)
# # … with 90 more rowsA polygon within Penobscot Bay stored in
geopackage format. Use
sf::read_sf(system.file("datasets/penbay-polygons.gpkg", package = 'twinkle'))
to them in.
poly <- sf::read_sf(system.file("datasets/penbay-polygons.gpkg", package = 'twinkle'))
# Simple feature collection with 1 feature and 0 fields
# geometry type: POLYGON
# dimension: XY
# bbox: xmin: -69 ymin: 43.9 xmax: -68.6 ymax: 44.2
# geographic CRS: WGS 84
# # A tibble: 1 x 1
# geom
# <POLYGON [°]>
# 1 ((-69 43.9, -68.6 43.9, -68.8 44.2, -69 43.9))par(mfrow = c(1,2))
i <- 1
plot(sst[1,,,i], reset = FALSE, key.pos = NULL,
main = paste("sst", i),
add.geom = list(sf::st_geometry(dplyr::filter(pts, band == i)),
col = "orange", pch = 19))
plot(poly, add = TRUE, border = "pink", col = NA, lwd = 2)
i <- 2
plot(sst[1,,,i], reset = FALSE, key.pos = NULL,
main = paste("sst", i),
add.geom = list(sf::st_geometry(dplyr::filter(pts, band == i)),
col = "olivedrab", pch = 19))
plot(poly, add = TRUE, border = "pink", col = NA, lwd = 2)
par(mfrow = c(1,1))A stars version of this nice
example of the built in
volcano dataset. Points and polygons are also provided but not shown
here. See volcano_points() and volcano_polygon().
First read in the volcano data with a single band, but three different attributes (variables). Note that each attribute is independently scaled.
Note that to plot multiple attributes in one view, there are fewer options for plotting notably applying a legend. The method is to set the
mfrowargument topar()to your desired layout, then call thestarsplot method withreset = FALSE, key.pos = NULL.twinkleincludes a simple wrapper function,mplot(), to manage this for you. The second attribute values are ‘flipped’ to dramatize the value ranges which are scaled separately.
v_a <- volcano_multi(what = "attributes") |>
rlang::set_names(c("volcano_1", "volcano_2", "volcano_3")) |>
dplyr::mutate(volcano_2 = abs(232 - volcano_2))
# stars object with 2 dimensions and 3 attributes
# attribute(s):
# v1 v2 v3
# Min. : 90.28 Min. :111.9 Min. :108.0
# 1st Qu.:103.72 1st Qu.:128.6 1st Qu.:124.0
# Median :119.09 Median :147.6 Median :142.4
# Mean :125.03 Mean :155.0 Mean :149.5
# 3rd Qu.:144.06 3rd Qu.:178.6 3rd Qu.:172.3
# Max. :187.28 Max. :232.2 Max. :224.0
# dimension(s):
# from to offset delta refsys point values x/y
# x 1 87 6478705 10 NZGD49 / New Zealand Map ... NA NULL [x]
# y 1 61 2668015 -10 NZGD49 / New Zealand Map ... NA NULL [y]
mplot(v_a)
But we can reorganize so that instead of three different attributes with single bands that we have three bands for a single attribute. Note the plot scales all bands into the same color scale - something that does not happen when plotting multiple attributes.
v_b <- volcano_multi(what = "bands", threshold = 120)
# stars object with 3 dimensions and 1 attribute
# attribute(s):
# X
# Min. : 92.43
# 1st Qu.:118.63
# Median :136.69
# Mean :142.79
# 3rd Qu.:164.81
# Max. :229.56
# dimension(s):
# from to offset delta refsys point values x/y
# x 1 87 6478705 10 NZGD49 / New Zealand Map ... NA NULL [x]
# y 1 61 2668015 -10 NZGD49 / New Zealand Map ... NA NULL [y]
# volcano 1 3 NA NA NA NA v1, v2, v3
plot(v_b)
A multiband dataset with very small dimensions, with convenient polygons and points.
toy <- toy_multi()
# stars object with 3 dimensions and 1 attribute
# attribute(s):
# X
# Min. : 1.0
# 1st Qu.:115.5
# Median :234.0
# Mean :237.0
# 3rd Qu.:356.5
# Max. :500.0
# NA's :225
# dimension(s):
# from to offset delta refsys point values x/y
# x 1 10 0 1 NA NA NULL [x]
# y 1 10 10 -1 NA NA NULL [y]
# band 1 5 NA NA NA NA b1,...,b5
pts <- toy_points()
# Simple feature collection with 10 features and 2 fields
# geometry type: POINT
# dimension: XY
# bbox: xmin: 1.858451 ymin: 4.611483 xmax: 8.521713 ymax: 8.987802
# CRS: NA
# # A tibble: 10 x 3
# id band geometry
# <int> <chr> <POINT>
# 1 1 b3 (1.858451 6.031725)
# 2 2 b4 (1.981977 8.810914)
# 3 3 b1 (7.698303 5.102966)
# 4 4 b5 (4.586549 5.620073)
# 5 5 b2 (1.906566 6.416532)
# 6 6 b1 (8.521713 6.300849)
# 7 7 b3 (5.645839 8.987802)
# 8 8 b2 (4.17234 4.611483)
# 9 9 b3 (5.397651 6.953142)
# 10 10 b3 (4.491615 4.644204)
poly <- toy_polygon()
# Simple feature collection with 1 feature and 1 field
# geometry type: POLYGON
# dimension: XY
# bbox: xmin: 3 ymin: 2 xmax: 8 ymax: 8
# CRS: NA
# id geom
# 1 1 POLYGON ((5 8, 8 5, 5 2, 3 ...
plot(toy[,,,1], axes = TRUE, reset = FALSE)
plot(sf::st_geometry(pts), add = TRUE, pch = 19, col = "orange")
plot(poly, add = TRUE, border = "green", col = NA)