{"id":1076,"date":"2011-12-05T16:15:18","date_gmt":"2011-12-05T21:15:18","guid":{"rendered":"http:\/\/lukemiller.org\/?p=1076"},"modified":"2013-07-29T12:47:09","modified_gmt":"2013-07-29T16:47:09","slug":"loading-osus-vgpm-ocean-productivity-data-in-r","status":"publish","type":"post","link":"https:\/\/lukemiller.org\/index.php\/2011\/12\/loading-osus-vgpm-ocean-productivity-data-in-r\/","title":{"rendered":"Loading OSU’s VGPM ocean productivity data in R"},"content":{"rendered":"

Oregon State University makes a set of ocean productivity data derived from satellite data available for download and use by researchers. The Ocean Productivity website<\/a> explains the available data and how it was derived. I have put together a few R functions to open a subset of the available data files and plot the data.<\/p>\n

\"\"<\/a>
Average monthly net primary productivity along the west coast of North America during February 2003. Data derived from OSU’s Vertically Generalized Production Model.<\/figcaption><\/figure>\n

The data files used here are from the Vertically Generalized Production Model (VGPM) based on Behrenfeld and Falkowski 1997<\/a>. Most of the available files on the website are HDF5 format, which is currently difficult to deal with in the Windows version of R. The standard VGPM data derived from MODIS satellite data are also available in a simplified text file format (called .xyz files by OSU). These data files can be reached from this page.<\/a> The functions below work on the .xyz file type. The .xyz files are available in two resolutions, one with 1\/6-degree steps between grid cells (1080×2160 grid), and one with 1\/12-degree steps (2160×4320 grid). The files should first be downloaded to your local machine.<\/p>\n

There are 4 functions outlined below. The function vgpm.load() extracts a matrix of productivity values from a specified file. The function plot.ROI() will plot that matrix of data. The function vgpm.plot() will extract and plot a matrix of productivity values all in one operation, and return the data matrix. Finally, the function vgpm.raster() illustrates a method for plotting the productivity data as
\na raster image, which results in a smaller file size when saved as a pdf or when displayed on screen. On my machine, trying to plot data for the entire world as a vector graphic fails to display anything because of the large amount of memory necessary for such an operation, while displaying the same data in raster format works just fine. The raster display function is a bit of hack, as you’ll see if you check out the function code below.<\/p>\n

The functions below can be downloaded as a R script called OSU_VPGM_load_functions.R from my GitHub repository<\/a>. Run that script in a R session to gain access to the functions.<\/p>\n

\n
\n
# filename: OSU_VGPM_load_function.R<\/span>\r\n# This script contains 4 functions for loading data from OSU's VGPM .xyz data<\/span>\r\n# files and plotting the data. See below for further information.<\/span>\r\n# Author: Luke Miller Dec 2, 2011<\/span>\r\n###############################################################################<\/span>\r\n\r\n# This script is an attempt to extract data from Oregon State University's <\/span>\r\n# Ocean Productivity Standard Vertically Generalized Production Model (VGPM) <\/span>\r\n# files that are accessible here:<\/span>\r\n# http:\/\/www.science.oregonstate.edu\/ocean.productivity\/index.php<\/span>\r\n\r\n# This script deals specifically with files obtained from this page:<\/span>\r\n# http:\/\/orca.science.oregonstate.edu\/1080.by.2160.monthly.xyz.vgpm.m.chl.m.sst4.php<\/span>\r\n# or this page:<\/span>\r\n# http:\/\/orca.science.oregonstate.edu\/2160.by.4320.monthly.xyz.vgpm.m.chl.m.sst4.php<\/span>\r\n\r\n# This script is meant to work with .xyz files rather than .hdf files. The<\/span>\r\n# .xyz files simply store the data as a set of 3 columns of data with lat\/lon <\/span>\r\n# and the productivity value, in a simple text file format. <\/span>\r\n# The files will follow the naming format: vgpm.yyyyddd.all.xyz.gz<\/span>\r\n\r\n# Files are in an Equidistant Cylindrical projection, and the lat\/lon value of <\/span>\r\n# the center of each pixel is given. <\/span>\r\n# For 1080x2160 data (2332800 rows of data in the file), the grid spacing is at <\/span>\r\n# 1\/6 of a degree. There are 180 degrees of latitude starting at the north pole,<\/span>\r\n# and 360 degrees of longitude starting at -180 degrees (i.e. in the north <\/span>\r\n# Pacific ocean). For 2160x4320 data files (9331200 rows of data in the file), <\/span>\r\n# the grid spacing is a 1\/12\u00b0. All lat\/lon locations given in the file are for<\/span>\r\n# the center of a grid cell.<\/span>\r\n\r\n# See here: http:\/\/orca.science.oregonstate.edu\/faq01.php for more info.<\/span>\r\n\r\n# To cite the data used here, cite:<\/span>\r\n#  Behrenfeld, MJ, PG Falkowski<\/span>\r\n#  Limnology and Oceanography<\/span>\r\n#  1997a, Volume 42: 1-20<\/span>\r\n#  Photosynthetic rates derived from satellite-based chlorophyll concentration<\/span>\r\n#<\/span>\r\n# as well as citing the website:<\/span>\r\n# \thttp:\/\/www.science.oregonstate.edu\/ocean.productivity\/index.php<\/span>\r\n\r\n# The plotting routines below make use of the fields package, which must be<\/span>\r\n# installed before running these functions. The fields package was produced by<\/span>\r\n#   Reinhard Furrer, Douglas Nychka and Stephen Sain (2011). fields:<\/span>\r\n#   Tools for spatial data. R package version 6.6.1.<\/span>\r\n#   http:\/\/CRAN.R-project.org\/package=fields<\/span>\r\n\r\n################################################################################<\/span>\r\n################################################################################<\/span>\r\n# The function vgpm.load() opens a vgpm .xyz data file and extracts the <\/span>\r\n# productivity data from the specified region of interest. <\/span>\r\n# The return value is a matrix of productivity values with associated grid cell<\/span>\r\n# latitudes and longitudes listed in the row and column names<\/span>\r\n# The options to vgpm.load() are as follows:<\/span>\r\n# file = file name (or substitute file.choose() to pick file interactively)<\/span>\r\n# w.lon = western longitude limit for region of interest (-180 to +180)<\/span>\r\n# e.lon = eastern longitude limit for region of interest (-180 to +180)<\/span>\r\n# n.lat = northern latitude limit for region of interest (+90 to -90)<\/span>\r\n# s.lat = southern latitude limit for region of interest (+90 to -90)<\/span>\r\n\r\nvgpm.load = function<\/span><\/a>(<\/span>file<\/span><\/a>,<\/span> w.lon,<\/span> e.lon,<\/span> n.lat,<\/span> s.lat)<\/span>{<\/span>\r\n\r\n# The value -9999 is used as a missing data marker, so all occurrences are <\/span>\r\n# changed to NA during the initial file read.<\/span>\r\n\tx = read.table<\/span><\/a>(<\/span>file<\/span><\/a>,<\/span> sep = ' '<\/span>,<\/span> skip = 1<\/span>,<\/span> na.strings = '-9999'<\/span>)<\/span>\r\n\r\n\tnames<\/span><\/a>(<\/span>x)<\/span> = c<\/span><\/a>(<\/span>'lon'<\/span>,<\/span>'lat'<\/span>,<\/span>'values'<\/span>)<\/span> #Rename input columns<\/span>\r\n\r\n\t#Establish which grid size is being used, 1080x2160 or 2160x4320<\/span>\r\n\tif<\/span> (<\/span>nrow<\/span><\/a>(<\/span>x)<\/span> ==<\/span> 2332800<\/span>)<\/span> {<\/span> f.size = '1080'<\/span>\r\n\t}<\/span> else<\/span> if<\/span> (<\/span>nrow<\/span><\/a>(<\/span>x)<\/span> ==<\/span> 9331200<\/span>)<\/span> {<\/span> f.size = '2160'<\/span>\r\n\t}<\/span> else<\/span> {<\/span>\r\n\t\twarning<\/span><\/a>(<\/span>'Unknown file type\\n<\/span>'<\/span>,<\/span> immediate. = TRUE<\/span>)<\/span>\r\n\t}<\/span>\r\n\r\n# The units for the column 'values' should be mg C \/ m^2 \/ Day<\/span>\r\n# The data are arranged so that the first value is at ~-180 longitude, <\/span>\r\n# +89.9166N latitude. The values then increase in longitude (encircling the <\/span>\r\n# globe at a fixed latitude) for 2160 rows total. After that, the data move to <\/span>\r\n# the next latitude value south, and the next 2160 values are for each longitude<\/span>\r\n# at the new latitude. This continues until you reach the southern-most latitude<\/span>\r\n# value. The process is the same for the higher-resolution 2160x4320 files.<\/span>\r\n\tif<\/span> (<\/span>f.size ==<\/span> '1080'<\/span>)<\/span> {<\/span>\r\n\t\tlons = x$<\/span>lon[<\/span>1<\/span>:<\/span>2160<\/span>]<\/span> #get set of longitude values<\/span>\r\n\t\tlats = x$<\/span>lat[<\/span>seq<\/span><\/a>(<\/span>1<\/span>,<\/span>2332800<\/span>,<\/span>by<\/span><\/a> = 2160<\/span>)<\/span>]<\/span> #get latitude values<\/span>\r\n\t\tvalues = matrix<\/span><\/a>(<\/span>x$<\/span>values,<\/span> nrow<\/span><\/a> = 1080<\/span>,<\/span> ncol<\/span><\/a> = 2160<\/span>,<\/span> byrow = TRUE<\/span>)<\/span>\r\n\t}<\/span> else<\/span> if<\/span> (<\/span>f.size ==<\/span> '2160'<\/span>)<\/span> {<\/span>\r\n\t\tlons = x$<\/span>lon[<\/span>1<\/span>:<\/span>4320<\/span>]<\/span> #get set of longitude values<\/span>\r\n\t\tlats = x$<\/span>lat[<\/span>seq<\/span><\/a>(<\/span>1<\/span>,<\/span>9331200<\/span>,<\/span>by<\/span><\/a> = 4320<\/span>)<\/span>]<\/span> #get latitude values<\/span>\r\n\t\tvalues = matrix<\/span><\/a>(<\/span>x$<\/span>values,<\/span> nrow<\/span><\/a> = 2160<\/span>,<\/span> ncol<\/span><\/a> = 4320<\/span>,<\/span> byrow = TRUE<\/span>)<\/span>\r\n\t}<\/span>\r\n\t#Insert the lat\/lon values as the 'values' matrix dimension names<\/span>\r\n\tdimnames<\/span><\/a>(<\/span>values)<\/span> = list<\/span><\/a>(<\/span>Latitude = lats,<\/span> Longitude = lons)<\/span>\r\n\r\n# Specify the boundaries of your lat\/lon of interest. Recall that<\/span>\r\n# longitude values run from -180E (international date line in the Pacific)<\/span>\r\n# to +180E, where Greenwich,England is at 0E. Latitude values range from<\/span>\r\n# +90N (north pole) to -90 (south pole). The first value for longitude must be<\/span>\r\n# the western-most edge of your region of interest, and the first value for the<\/span>\r\n# latitude must be the northern-most edge of the region of interest.<\/span>\r\n\tlonlim = c<\/span><\/a>(<\/span>w.lon,<\/span>e.lon)<\/span> # c(western edge, eastern edge)<\/span>\r\n\tlatlim = c<\/span><\/a>(<\/span>n.lat,<\/span>s.lat)<\/span>\t# c(northern edge, southern edge)<\/span>\r\n\r\n# Create vectors of lat\/lon indices<\/span>\r\n\tlonindx = 1<\/span>:<\/span>length<\/span><\/a>(<\/span>lons)<\/span> #make vector of longitude cell indices<\/span>\r\n\tlatindx = 1<\/span>:<\/span>length<\/span><\/a>(<\/span>lats)<\/span> #make vector of latitude cell indices<\/span>\r\n\r\n# Pull out 2 vectors that contain the indices of the lat\/lon coordinates<\/span>\r\n# of interest. We search for longitudes that are greater than the 1st value<\/span>\r\n# in lonlim, and longitudes that are less than the 2nd value in lonlim, and<\/span>\r\n# then grab the corresponding indices from lonindx to store in goodlons. The<\/span>\r\n# same is done for the latitudes<\/span>\r\n\tgoodlons = lonindx[<\/span>lons >=<\/span> lonlim[<\/span>1<\/span>]<\/span> &<\/span> lons <=<\/span> lonlim[<\/span>2<\/span>]<\/span>]<\/span>\r\n\tgoodlats = latindx[<\/span>lats >=<\/span> latlim[<\/span>2<\/span>]<\/span> &<\/span> lats <=<\/span> latlim[<\/span>1<\/span>]<\/span>]<\/span>\r\n\r\n# Extract a subset of the matrix 'values', call it the Region of Interest (ROI) <\/span>\r\n\tROI = values[<\/span>goodlats[<\/span>1<\/span>]<\/span>:<\/span>goodlats[<\/span>length<\/span><\/a>(<\/span>goodlats)<\/span>]<\/span>,<\/span>\r\n\t\t\tgoodlons[<\/span>1<\/span>]<\/span>:<\/span>goodlons[<\/span>length<\/span><\/a>(<\/span>goodlons)<\/span>]<\/span>]<\/span>\r\n# Add the latitudes and longitudes to the ROI matrix as dimension names<\/span>\r\n\tdimnames<\/span><\/a>(<\/span>ROI)<\/span> = list<\/span><\/a>(<\/span>Latitude = lats[<\/span>goodlats]<\/span>,<\/span> Longitude = lons[<\/span>goodlons]<\/span>)<\/span>\r\n\r\n\tROI # return the ROI matrix at the conclusion of the function<\/span>\r\n}<\/span> # end of vgpm.load function<\/span>\r\n\r\n################################################################################<\/span>\r\n################################################################################<\/span>\r\n# The plot.ROI function produces a plot of the productivity data stored in the<\/span>\r\n# output matrix from the vgpm.load() function above. The ROI matrix is <\/span>\r\n# expected to have latitudes in rows (from North to South) and longitudes in <\/span>\r\n# columns (from -180 to +180\u00b0). The associated latitudes and longitudes must<\/span>\r\n# be in the row\/column names of the matrix, as produced by vgpm.load().<\/span>\r\n\r\nplot.ROI = function<\/span><\/a>(<\/span>ROI,<\/span> log<\/span><\/a> = TRUE<\/span>,<\/span> color = tim.colors(<\/span>30<\/span>)<\/span>)<\/span>{<\/span>\r\n\r\n# For plotting with the image.plot() command, it is necessary to arrange ROI <\/span>\r\n# so that longitudes are in rows, and latitudes are in columns, with the <\/span>\r\n# latitudes order reversed so that they increase as you move across columns. <\/span>\r\n\tROI.plot = t<\/span><\/a>(<\/span>ROI)<\/span> #transpose the ROI matrix<\/span>\r\n\tROI.plot = ROI.plot[<\/span>,<\/span>rev<\/span><\/a>(<\/span>1<\/span>:<\/span>ncol<\/span><\/a>(<\/span>ROI.plot)<\/span>)<\/span>]<\/span> #reverse the latitudes<\/span>\r\n\tlons = as.numeric<\/span><\/a>(<\/span>rownames<\/span><\/a>(<\/span>ROI.plot)<\/span>)<\/span> #extract longitudes<\/span>\r\n\tlats = as.numeric<\/span><\/a>(<\/span>colnames<\/span><\/a>(<\/span>ROI.plot)<\/span>)<\/span> #extract latitudes<\/span>\r\n\r\n\trequire<\/span><\/a>(<\/span>fields<\/span><\/a>)<\/span>\r\n\tif<\/span> (<\/span>log<\/span><\/a>)<\/span>{<\/span>\r\n\t\timage.plot(<\/span>x=lons,<\/span> y=lats,<\/span> log10<\/span><\/a>(<\/span>ROI.plot)<\/span>,<\/span>\r\n\t\t\t\tcol<\/span><\/a> = color,<\/span>\r\n\t\t\t\tlegend.lab = expression<\/span><\/a>(<\/span>paste<\/span><\/a>(<\/span>log<\/span><\/a>[<\/span>10<\/span>]<\/span>,<\/span>\"(mg C \/ \"<\/span>,<\/span>m^<\/span>2<\/span>,<\/span>\"\/ day)\"<\/span>)<\/span>)<\/span>,<\/span>\r\n\t\t\t\tlegend.mar = 4.1<\/span>,<\/span>\r\n\t\t\t\txlab = \"Longitude\"<\/span>,<\/span>\r\n\t\t\t\tylab = \"Latitude\"<\/span>,<\/span>\r\n\t\t\t\tmain = 'Net Primary Production'<\/span>,<\/span>\r\n\t\t\t\tlas = 1<\/span>)<\/span>\r\n\t}<\/span> else<\/span> if<\/span> (<\/span>!<\/span>log<\/span><\/a>)<\/span>{<\/span>\r\n\t\timage.plot(<\/span>x=lons,<\/span> y=lats,<\/span> ROI.plot,<\/span>\r\n\t\t\t\tcol<\/span><\/a> = color,<\/span>\r\n\t\t\t\tlegend.lab = expression<\/span><\/a>(<\/span>paste<\/span><\/a>(<\/span>\"mg C \/ \"<\/span>,<\/span>m^<\/span>2<\/span>,<\/span>\"\/ day\"<\/span>)<\/span>)<\/span>,<\/span>\r\n\t\t\t\tlegend.mar = 4.1<\/span>,<\/span>\r\n\t\t\t\txlab = \"Longitude\"<\/span>,<\/span>\r\n\t\t\t\tylab = \"Latitude\"<\/span>,<\/span>\r\n\t\t\t\tmain = 'Net Primary Production'<\/span>,<\/span>\r\n\t\t\t\tlas = 1<\/span>)<\/span>\r\n\t}<\/span>\r\n}<\/span> #end of plot.ROI function<\/span>\r\n\r\n################################################################################<\/span>\r\n################################################################################<\/span>\r\n# The function vgpm.plot() reads data from the desired region of interest and<\/span>\r\n# also optionally plots the data. <\/span>\r\n\r\n# The options supplied to vgpm.plot() are as follows:<\/span>\r\n# file = file name (or substitute file.choose() to pick file interactively)<\/span>\r\n# w.lon = western longitude limit for region of interest (-180 to +180)<\/span>\r\n# e.lon = eastern longitude limit for region of interest (-180 to +180)<\/span>\r\n# n.lat = northern latitude limit for region of interest (+90 to -90)<\/span>\r\n# s.lat = southern latitude limit for region of interest (+90 to -90)<\/span>\r\n# plot.data = , plots data from region of interest<\/span>\r\n# log = ,  log10 transform productivity data before plotting<\/span>\r\n# color = , specify color set to plot productivity data<\/span>\r\n# Function returns a matrix of productivity values for the specified region of<\/span>\r\n# interest with lat\/lon listed in the row and column names.<\/span>\r\nvgpm.plot = function<\/span><\/a>(<\/span>file<\/span><\/a>,<\/span> w.lon,<\/span> e.lon,<\/span> n.lat,<\/span> s.lat,<\/span> plot.data = TRUE<\/span>,<\/span>\r\n\t\tlog<\/span><\/a> = TRUE<\/span>,<\/span> color = tim.colors(<\/span>30<\/span>)<\/span>)<\/span>{<\/span>\r\n\r\n\t#Extract date from file title<\/span>\r\n\tfname = basename<\/span><\/a>(<\/span>file<\/span><\/a>)<\/span>\r\n\tdots = gregexpr<\/span><\/a>(<\/span>'\\\\<\/span>.'<\/span>,<\/span>fname)<\/span>\r\n\tyrday = substr<\/span><\/a>(<\/span>fname,<\/span>dots[<\/span>[<\/span>1<\/span>]<\/span>]<\/span>[<\/span>1<\/span>]<\/span>+<\/span>1<\/span>,<\/span>dots[<\/span>[<\/span>1<\/span>]<\/span>]<\/span>[<\/span>2<\/span>]<\/span>-<\/span>1<\/span>)<\/span>\r\n\tyr = substr<\/span><\/a>(<\/span>yrday,<\/span>1<\/span>,<\/span>4<\/span>)<\/span>\r\n\tdoy = substr<\/span><\/a>(<\/span>yrday,<\/span>5<\/span>,<\/span>7<\/span>)<\/span>\r\n\tday1 = as.Date<\/span><\/a>(<\/span>paste<\/span><\/a>(<\/span>yr,<\/span>doy,<\/span>sep = '-'<\/span>)<\/span>,<\/span> format<\/span><\/a> = '%Y-%j'<\/span>)<\/span>\r\n# The value -9999 is used as a missing data marker, so all occurrences are <\/span>\r\n# changed to NA during the initial file read.<\/span>\r\n\tx = read.table<\/span><\/a>(<\/span>file<\/span><\/a>,<\/span> sep = ' '<\/span>,<\/span> skip = 1<\/span>,<\/span> na.strings = '-9999'<\/span>)<\/span>\r\n\r\n\tnames<\/span><\/a>(<\/span>x)<\/span> = c<\/span><\/a>(<\/span>'lon'<\/span>,<\/span>'lat'<\/span>,<\/span>'values'<\/span>)<\/span> #Rename input columns<\/span>\r\n\r\n\t#Establish which grid size is being used, 1080x2160 or 2160x4320<\/span>\r\n\tif<\/span> (<\/span>nrow<\/span><\/a>(<\/span>x)<\/span> ==<\/span> 2332800<\/span>)<\/span> {<\/span> f.size = '1080'<\/span>\r\n\t}<\/span> else<\/span> if<\/span> (<\/span>nrow<\/span><\/a>(<\/span>x)<\/span> ==<\/span> 9331200<\/span>)<\/span> {<\/span> f.size = '2160'<\/span>\r\n\t}<\/span> else<\/span> {<\/span>\r\n\t\twarning<\/span><\/a>(<\/span>'Unknown file type\\n<\/span>'<\/span>,<\/span> immediate. = TRUE<\/span>)<\/span>\r\n\t}<\/span>\r\n\r\n# The units for the column 'values' should be mg C \/ m^2 \/ Day<\/span>\r\n# The data are arranged so that the first value is at ~-180 longitude, <\/span>\r\n# +89.9166N latitude. The values then increase in longitude (encircling the <\/span>\r\n# globe at a fixed latitude) for 2160 rows total. After that, the data move to <\/span>\r\n# the next latitude value south, and the next 2160 values are for each longitude<\/span>\r\n# at the new latitude. This continues until you reach the southern-most latitude<\/span>\r\n# value. The process is the same for the higher-resolution 2160x4320 files.<\/span>\r\n\tif<\/span> (<\/span>f.size ==<\/span> '1080'<\/span>)<\/span> {<\/span>\r\n\t\tlons = x$<\/span>lon[<\/span>1<\/span>:<\/span>2160<\/span>]<\/span> #get set of longitude values<\/span>\r\n\t\tlats = x$<\/span>lat[<\/span>seq<\/span><\/a>(<\/span>1<\/span>,<\/span>2332800<\/span>,<\/span>by<\/span><\/a> = 2160<\/span>)<\/span>]<\/span> #get latitude values<\/span>\r\n\t\tvalues = matrix<\/span><\/a>(<\/span>x$<\/span>values,<\/span> nrow<\/span><\/a> = 1080<\/span>,<\/span> ncol<\/span><\/a> = 2160<\/span>,<\/span> byrow = TRUE<\/span>)<\/span>\r\n\t}<\/span> else<\/span> if<\/span> (<\/span>f.size ==<\/span> '2160'<\/span>)<\/span> {<\/span>\r\n\t\tlons = x$<\/span>lon[<\/span>1<\/span>:<\/span>4320<\/span>]<\/span> #get set of longitude values<\/span>\r\n\t\tlats = x$<\/span>lat[<\/span>seq<\/span><\/a>(<\/span>1<\/span>,<\/span>9331200<\/span>,<\/span>by<\/span><\/a> = 4320<\/span>)<\/span>]<\/span> #get latitude values<\/span>\r\n\t\tvalues = matrix<\/span><\/a>(<\/span>x$<\/span>values,<\/span> nrow<\/span><\/a> = 2160<\/span>,<\/span> ncol<\/span><\/a> = 4320<\/span>,<\/span> byrow = TRUE<\/span>)<\/span>\r\n\t}<\/span>\r\n\t#Insert the lat\/lon values as the 'values' matrix dimension names<\/span>\r\n\tdimnames<\/span><\/a>(<\/span>values)<\/span> = list<\/span><\/a>(<\/span>Latitude = lats,<\/span> Longitude = lons)<\/span>\r\n\r\n# Specify the boundaries of your lat\/lon of interest. Recall that<\/span>\r\n# longitude values run from -180E (international date line in the Pacific)<\/span>\r\n# to +180E, where Greenwich,England is at 0E. Latitude values range from<\/span>\r\n# +90N (north pole) to -90 (south pole). The first value for longitude must be<\/span>\r\n# the western-most edge of your region of interest, and the first value for the<\/span>\r\n# latitude must be the northern-most edge of the region of interest.<\/span>\r\n\tlonlim = c<\/span><\/a>(<\/span>w.lon,<\/span>e.lon)<\/span> # c(western edge, eastern edge)<\/span>\r\n\tlatlim = c<\/span><\/a>(<\/span>n.lat,<\/span>s.lat)<\/span>\t# c(northern edge, southern edge)<\/span>\r\n\r\n# Create vectors of lat\/lon indices<\/span>\r\n\tlonindx = 1<\/span>:<\/span>length<\/span><\/a>(<\/span>lons)<\/span> #make vector of longitude cell indices<\/span>\r\n\tlatindx = 1<\/span>:<\/span>length<\/span><\/a>(<\/span>lats)<\/span> #make vector of latitude cell indices<\/span>\r\n\r\n# Pull out 2 vectors that contain the indices of the lat\/lon coordinates<\/span>\r\n# of interest. We search for longitudes that are greater than the 1st value<\/span>\r\n# in lonlim, and longitudes that are less than the 2nd value in lonlim, and<\/span>\r\n# then grab the corresponding indices from lonindx to store in goodlons. The<\/span>\r\n# same is done for the latitudes<\/span>\r\n\tgoodlons = lonindx[<\/span>lons >=<\/span> lonlim[<\/span>1<\/span>]<\/span> &<\/span> lons <=<\/span> lonlim[<\/span>2<\/span>]<\/span>]<\/span>\r\n\tgoodlats = latindx[<\/span>lats >=<\/span> latlim[<\/span>2<\/span>]<\/span> &<\/span> lats <=<\/span> latlim[<\/span>1<\/span>]<\/span>]<\/span>\r\n\r\n# Extract a subset of the matrix 'values', call it the Region of Interest (ROI) <\/span>\r\n\tROI = values[<\/span>goodlats[<\/span>1<\/span>]<\/span>:<\/span>goodlats[<\/span>length<\/span><\/a>(<\/span>goodlats)<\/span>]<\/span>,<\/span>\r\n\t\t\tgoodlons[<\/span>1<\/span>]<\/span>:<\/span>goodlons[<\/span>length<\/span><\/a>(<\/span>goodlons)<\/span>]<\/span>]<\/span>\r\n# Add the latitudes and longitudes to the ROI matrix as dimension names<\/span>\r\n\tdimnames<\/span><\/a>(<\/span>ROI)<\/span> = list<\/span><\/a>(<\/span>Latitude = lats[<\/span>goodlats]<\/span>,<\/span> Longitude = lons[<\/span>goodlons]<\/span>)<\/span>\r\n\t##########################<\/span>\r\n\t##########################<\/span>\r\n#Plotting section<\/span>\r\n# For plotting with the image.plot() command, it is necessary to arrange ROI <\/span>\r\n# so that longitudes are in rows, and latitudes are in columns, with the <\/span>\r\n# latitudes order reversed so that they increase as you move across columns. <\/span>\r\n\tif<\/span> (<\/span>plot.data)<\/span>{<\/span>\r\n\t\tROI.plot = t<\/span><\/a>(<\/span>ROI)<\/span> #transpose the ROI matrix<\/span>\r\n\t\tROI.plot = ROI.plot[<\/span>,<\/span>rev<\/span><\/a>(<\/span>1<\/span>:<\/span>ncol<\/span><\/a>(<\/span>ROI.plot)<\/span>)<\/span>]<\/span> #reverse the latitudes<\/span>\r\n\t\tlons = as.numeric<\/span><\/a>(<\/span>rownames<\/span><\/a>(<\/span>ROI.plot)<\/span>)<\/span> #extract longitudes<\/span>\r\n\t\tlats = as.numeric<\/span><\/a>(<\/span>colnames<\/span><\/a>(<\/span>ROI.plot)<\/span>)<\/span> #extract latitudes<\/span>\r\n\r\n\t\trequire<\/span><\/a>(<\/span>fields<\/span><\/a>)<\/span>\r\n\t\tif<\/span> (<\/span>log<\/span><\/a>)<\/span>{<\/span>\r\n\t\t\timage.plot(<\/span>x=lons,<\/span> y=lats,<\/span> log10<\/span><\/a>(<\/span>ROI.plot)<\/span>,<\/span>\r\n\t\t\t\t\tcol<\/span><\/a> = color,<\/span>\r\n\t\t\t\t\tlegend.lab = expression<\/span><\/a>(<\/span>paste<\/span><\/a>(<\/span>log<\/span><\/a>[<\/span>10<\/span>]<\/span>,<\/span>\r\n\t\t\t\t\t\t\t\t\t\"(mg C \/ \"<\/span>,<\/span>m^<\/span>2<\/span>,<\/span>\"\/ day)\"<\/span>)<\/span>)<\/span>,<\/span>\r\n\t\t\t\t\tlegend.mar = 4.1<\/span>,<\/span>\r\n\t\t\t\t\txlab = \"Longitude\"<\/span>,<\/span>\r\n\t\t\t\t\tylab = \"Latitude\"<\/span>,<\/span>\r\n\t\t\t\t\tmain = paste<\/span><\/a>(<\/span>'Net Primary Production'<\/span>,<\/span>\r\n\t\t\t\t\t\t\tstrftime<\/span><\/a>(<\/span>day1,<\/span>'%B %Y'<\/span>)<\/span>)<\/span>,<\/span>\r\n\t\t\t\t\tlas = 1<\/span>)<\/span>\r\n\t\t}<\/span> else<\/span> if<\/span> (<\/span>!<\/span>log<\/span><\/a>)<\/span>{<\/span>\r\n\t\t\timage.plot(<\/span>x=lons,<\/span> y=lats,<\/span> ROI.plot,<\/span>\r\n\t\t\t\t\tcol<\/span><\/a> = color,<\/span>\r\n\t\t\t\t\tlegend.lab = expression<\/span><\/a>(<\/span>paste<\/span><\/a>(<\/span>\"mg C \/ \"<\/span>,<\/span>m^<\/span>2<\/span>,<\/span>\"\/ day\"<\/span>)<\/span>)<\/span>,<\/span>\r\n\t\t\t\t\tlegend.mar = 4.1<\/span>,<\/span>\r\n\t\t\t\t\txlab = \"Longitude\"<\/span>,<\/span>\r\n\t\t\t\t\tylab = \"Latitude\"<\/span>,<\/span>\r\n\t\t\t\t\tmain = paste<\/span><\/a>(<\/span>'Net Primary Production'<\/span>,<\/span> \r\n\t\t\t\t\t\t\tstrftime<\/span><\/a>(<\/span>day1,<\/span>'%B %Y'<\/span>)<\/span>)<\/span>,<\/span>\r\n\t\t\t\t\tlas = 1<\/span>)<\/span>\r\n\t\t}<\/span>\r\n\t}<\/span>\r\n\r\n\tROI # return the ROI matrix at the conclusion of the function<\/span>\r\n}<\/span> # end of vgpm.load function<\/span>\r\n\r\n################################################################################<\/span>\r\n################################################################################<\/span>\r\n# If you want to plot a very large area, such as the whole globe, plotting the <\/span>\r\n# data using the regular image() or image.plot() functions produces very large<\/span>\r\n# file sizes due to the large number of vectors used to make up the figure. <\/span>\r\n# Often the figure may not even display correctly on the screen due to the large <\/span>\r\n# file size. For small areas this won't be as much of a problem.<\/span>\r\n# It is more space-efficient to plot large data sets as a raster image with the <\/span>\r\n# useRaster = TRUE option of the image plotting tools. <\/span>\r\n# To make use of the useRaster option for plotting, the lat\/lon grid must be on<\/span>\r\n# a completely regular grid, but due to the limitations of precision by the <\/span>\r\n# computer, the 1\/6\u00b0 (or 1\/12\u00b0) steps stored as characters in the input file <\/span>\r\n# aren't converted to exact 1\/6\u00b0 steps for latitude and longitude.<\/span>\r\n# We can make a set of approximate lat\/lon values for plotting. For the 1\/6\u00b0 <\/span>\r\n# the maximal error for the latitudes calculated here (~1\/30000\u00b0) versus those <\/span>\r\n# reported in the original input file is a few meters of latitude <\/span>\r\n# near the south pole. <\/span>\r\n# The error is smallest in the northwest corner of the plot and largest in the <\/span>\r\n# southeast corner. Remember this only applies to the displayed figure, not the<\/span>\r\n# data and lat\/lon values extracted above.<\/span>\r\n\r\n# The options supplied to vgpm.raster() are as follows:<\/span>\r\n# file = file name (or substitute file.choose() to pick file interactively)<\/span>\r\n# w.lon = western longitude limit for region of interest (-180 to +180)<\/span>\r\n# e.lon = eastern longitude limit for region of interest (-180 to +180)<\/span>\r\n# n.lat = northern latitude limit for region of interest (+90 to -90)<\/span>\r\n# s.lat = southern latitude limit for region of interest (+90 to -90)<\/span>\r\n# log = TRUE - log10 transform productivity data before plotting<\/span>\r\n# color = specify color set to plot productivity data<\/span>\r\n# Function returns a matrix of productivity values for the specified region of<\/span>\r\n# interest with lat\/lon listed in the row and column names.<\/span>\r\nvgpm.raster = function<\/span><\/a>(<\/span>file<\/span><\/a>,<\/span> w.lon,<\/span> e.lon,<\/span> n.lat,<\/span> s.lat,<\/span> log<\/span><\/a> = TRUE<\/span>,<\/span> \r\n\t\tcolor = tim.colors(<\/span>30<\/span>)<\/span>)<\/span>{<\/span>\r\n\r\n\t#Extract date from file title<\/span>\r\n\tfname = basename<\/span><\/a>(<\/span>file<\/span><\/a>)<\/span>\r\n\tdots = gregexpr<\/span><\/a>(<\/span>'\\\\<\/span>.'<\/span>,<\/span>fname)<\/span> #find locations of . in file name<\/span>\r\n\tyrday = substr<\/span><\/a>(<\/span>fname,<\/span>dots[<\/span>[<\/span>1<\/span>]<\/span>]<\/span>[<\/span>1<\/span>]<\/span>+<\/span>1<\/span>,<\/span>dots[<\/span>[<\/span>1<\/span>]<\/span>]<\/span>[<\/span>2<\/span>]<\/span>-<\/span>1<\/span>)<\/span> #extract yearday combo<\/span>\r\n\tyr = substr<\/span><\/a>(<\/span>yrday,<\/span>1<\/span>,<\/span>4<\/span>)<\/span> #extract year<\/span>\r\n\tdoy = substr<\/span><\/a>(<\/span>yrday,<\/span>5<\/span>,<\/span>7<\/span>)<\/span> #extract day of year<\/span>\r\n\tday1 = as.Date<\/span><\/a>(<\/span>paste<\/span><\/a>(<\/span>yr,<\/span>doy,<\/span>sep = '-'<\/span>)<\/span>,<\/span> format<\/span><\/a> = '%Y-%j'<\/span>)<\/span> #convert to Date<\/span>\r\n\r\n\t#Read data from input file<\/span>\r\n\tx = read.table<\/span><\/a>(<\/span>file<\/span><\/a>,<\/span> sep = ' '<\/span>,<\/span> skip = 1<\/span>,<\/span> na.strings = '-9999'<\/span>)<\/span>\r\n\r\n\tnames<\/span><\/a>(<\/span>x)<\/span> = c<\/span><\/a>(<\/span>'lon'<\/span>,<\/span>'lat'<\/span>,<\/span>'values'<\/span>)<\/span> #Rename input columns<\/span>\r\n\r\n\tif<\/span> (<\/span>nrow<\/span><\/a>(<\/span>x)<\/span> ==<\/span> 2332800<\/span>)<\/span> {<\/span> f.size = '1080'<\/span>\r\n\t}<\/span> else<\/span> if<\/span> (<\/span>nrow<\/span><\/a>(<\/span>x)<\/span> ==<\/span> 9331200<\/span>)<\/span> {<\/span> f.size = '2160'<\/span>\r\n\t}<\/span> else<\/span> {<\/span>\r\n\t\twarning<\/span><\/a>(<\/span>'Unknown file type\\n<\/span>'<\/span>,<\/span> immediate. = TRUE<\/span>)<\/span>\r\n\t}<\/span>\r\n\r\n\tif<\/span> (<\/span>f.size ==<\/span> '1080'<\/span>)<\/span> {<\/span>\r\n\t\tlons = x$<\/span>lon[<\/span>1<\/span>:<\/span>2160<\/span>]<\/span> #get set of longitude values<\/span>\r\n\t\tlats = x$<\/span>lat[<\/span>seq<\/span><\/a>(<\/span>1<\/span>,<\/span>2332800<\/span>,<\/span>by<\/span><\/a> = 2160<\/span>)<\/span>]<\/span> #get latitude values<\/span>\r\n\t\tvalues = matrix<\/span><\/a>(<\/span>x$<\/span>values,<\/span> nrow<\/span><\/a> = 1080<\/span>,<\/span> ncol<\/span><\/a> = 2160<\/span>,<\/span> byrow = TRUE<\/span>)<\/span>\r\n\t}<\/span> else<\/span> if<\/span> (<\/span>f.size ==<\/span> '2160'<\/span>)<\/span> {<\/span>\r\n\t\tlons = x$<\/span>lon[<\/span>1<\/span>:<\/span>4320<\/span>]<\/span> #get set of longitude values<\/span>\r\n\t\tlats = x$<\/span>lat[<\/span>seq<\/span><\/a>(<\/span>1<\/span>,<\/span>9331200<\/span>,<\/span>by<\/span><\/a> = 4320<\/span>)<\/span>]<\/span> #get latitude values<\/span>\r\n\t\tvalues = matrix<\/span><\/a>(<\/span>x$<\/span>values,<\/span> nrow<\/span><\/a> = 2160<\/span>,<\/span> ncol<\/span><\/a> = 4320<\/span>,<\/span> byrow = TRUE<\/span>)<\/span>\r\n\t}<\/span>\r\n#Insert the lat\/lon values as the 'values' matrix dimension names<\/span>\r\n\tdimnames<\/span><\/a>(<\/span>values)<\/span> = list<\/span><\/a>(<\/span>Latitude = lats,<\/span> Longitude = lons)<\/span>\r\n\r\n# Specify the boundaries of your lat\/lon of interest. Recall that<\/span>\r\n# longitude values run from -180E (international date line in the Pacific)<\/span>\r\n# to +180E, where Greenwich,England is at 0E. Latitude values range from<\/span>\r\n# +90N (north pole) to -90 (south pole). The first value for longitude must be<\/span>\r\n# the western-most edge of your region of interest, and the first value for the<\/span>\r\n# latitude must be the northern-most edge of the region of interest.<\/span>\r\n\tlonlim = c<\/span><\/a>(<\/span>w.lon,<\/span>e.lon)<\/span> # c(western edge, eastern edge)<\/span>\r\n\tlatlim = c<\/span><\/a>(<\/span>n.lat,<\/span>s.lat)<\/span>\t# c(northern edge, southern edge)<\/span>\r\n\r\n# Create vectors of lat\/lon indices<\/span>\r\n\tlonindx = 1<\/span>:<\/span>length<\/span><\/a>(<\/span>lons)<\/span> #make vector of longitude cell indices<\/span>\r\n\tlatindx = 1<\/span>:<\/span>length<\/span><\/a>(<\/span>lats)<\/span> #make vector of latitude cell indices<\/span>\r\n\r\n# Pull out 2 vectors that contain the indices of the lat\/lon coordinates<\/span>\r\n# of interest. We search for longitudes that are greater than the 1st value<\/span>\r\n# in lonlim, and longitudes that are less than the 2nd value in lonlim, and<\/span>\r\n# then grab the corresponding indices from lonindx to store in goodlons. The<\/span>\r\n# same is done for the latitudes<\/span>\r\n\tgoodlons = lonindx[<\/span>lons >=<\/span> lonlim[<\/span>1<\/span>]<\/span> &<\/span> lons <=<\/span> lonlim[<\/span>2<\/span>]<\/span>]<\/span>\r\n\tgoodlats = latindx[<\/span>lats >=<\/span> latlim[<\/span>2<\/span>]<\/span> &<\/span> lats <=<\/span> latlim[<\/span>1<\/span>]<\/span>]<\/span>\r\n\r\n# Extract a subset of the matrix 'values', call it the Region of Interest (ROI) <\/span>\r\n\tROI = values[<\/span>goodlats[<\/span>1<\/span>]<\/span>:<\/span>goodlats[<\/span>length<\/span><\/a>(<\/span>goodlats)<\/span>]<\/span>,<\/span>\r\n\t\t\tgoodlons[<\/span>1<\/span>]<\/span>:<\/span>goodlons[<\/span>length<\/span><\/a>(<\/span>goodlons)<\/span>]<\/span>]<\/span>\r\n# Add the latitudes and longitudes to the ROI matrix as dimension names<\/span>\r\n\tdimnames<\/span><\/a>(<\/span>ROI)<\/span> = list<\/span><\/a>(<\/span>Latitude = lats[<\/span>goodlats]<\/span>,<\/span> Longitude = lons[<\/span>goodlons]<\/span>)<\/span>\r\n\tn.lats = as.numeric<\/span><\/a>(<\/span>rownames<\/span><\/a>(<\/span>ROI)<\/span>)<\/span>\r\n\tn.lons = as.numeric<\/span><\/a>(<\/span>colnames<\/span><\/a>(<\/span>ROI)<\/span>)<\/span>\r\n\r\n\t# Generate a new set of lats and longs on a regular grid spacing for plot.<\/span>\r\n\tif<\/span> (<\/span>f.size ==<\/span> '1080'<\/span>)<\/span> {<\/span>\r\n\t\tlats2 = seq<\/span><\/a>(<\/span>n.lats[<\/span>1<\/span>]<\/span>,<\/span>(<\/span>n.lats[<\/span>length<\/span><\/a>(<\/span>n.lats)<\/span>]<\/span>-<\/span>0.1666667<\/span>)<\/span>,<\/span>by<\/span><\/a>=-<\/span>0.1666667<\/span>)<\/span>\r\n\t\tlons2 = seq<\/span><\/a>(<\/span>n.lons[<\/span>1<\/span>]<\/span>,<\/span>(<\/span>n.lons[<\/span>length<\/span><\/a>(<\/span>n.lons)<\/span>]<\/span>+<\/span>0.1666667<\/span>)<\/span>,<\/span>by<\/span><\/a>=0.1666667<\/span>)<\/span>\r\n\t}<\/span> else<\/span> if<\/span> (<\/span>f.size ==<\/span> '2160'<\/span>)<\/span> {<\/span>\r\n\t\tlats2 = seq<\/span><\/a>(<\/span>n.lats[<\/span>1<\/span>]<\/span>,<\/span>(<\/span>n.lats[<\/span>length<\/span><\/a>(<\/span>n.lats)<\/span>]<\/span>-<\/span>0.0833333<\/span>)<\/span>,<\/span>by<\/span><\/a>=-<\/span>0.0833333<\/span>)<\/span>\r\n\t\tlons2 = seq<\/span><\/a>(<\/span>n.lons[<\/span>1<\/span>]<\/span>,<\/span>(<\/span>n.lons[<\/span>length<\/span><\/a>(<\/span>n.lons)<\/span>]<\/span>+<\/span>0.0833333<\/span>)<\/span>,<\/span>by<\/span><\/a>=0.0833333<\/span>)<\/span>\r\n\t}<\/span>\r\n\tif<\/span>(<\/span>length<\/span><\/a>(<\/span>lats2)<\/span> ><\/span> length<\/span><\/a>(<\/span>n.lats)<\/span>)<\/span> lats2 = lats2[<\/span>1<\/span>:<\/span>length<\/span><\/a>(<\/span>n.lats)<\/span>]<\/span>\r\n\tif<\/span>(<\/span>length<\/span><\/a>(<\/span>lons2)<\/span> ><\/span> length<\/span><\/a>(<\/span>n.lons)<\/span>)<\/span> lons2 = lons2[<\/span>1<\/span>:<\/span>length<\/span><\/a>(<\/span>n.lons)<\/span>]<\/span>\r\n\tROI.plot = t<\/span><\/a>(<\/span>ROI)<\/span> # swap longs and lats in 'ROI', so lats are in columns<\/span>\r\n\tROI.plot = ROI.plot[<\/span>,<\/span>rev<\/span><\/a>(<\/span>1<\/span>:<\/span>length<\/span><\/a>(<\/span>lats2)<\/span>)<\/span>]<\/span> # reverse latitudes so that <\/span>\r\n\t # southern lats are listed first<\/span>\r\n\tif<\/span> (<\/span>log<\/span><\/a>)<\/span> {<\/span>\r\n\t\timage.plot(<\/span>lons2,<\/span> rev<\/span><\/a>(<\/span>lats2)<\/span>,<\/span> log10<\/span><\/a>(<\/span>ROI.plot)<\/span>,<\/span> useRaster = TRUE<\/span>,<\/span> \r\n\t\t\t\tcol<\/span><\/a> = color,<\/span>\r\n\t\t\t\txlab = 'Longitude'<\/span>,<\/span> ylab = 'Latitude'<\/span>,<\/span> \r\n\t\t\t\tmain = paste<\/span><\/a>(<\/span>'Net Primary Production'<\/span>,<\/span> strftime<\/span><\/a>(<\/span>day1,<\/span>'%B %Y'<\/span>)<\/span>)<\/span>,<\/span> \r\n\t\t\t\tlegend.lab = expression<\/span><\/a>(<\/span>paste<\/span><\/a>(<\/span>log<\/span><\/a>[<\/span>10<\/span>]<\/span>,<\/span>'(mg C \/'<\/span>,<\/span> m^<\/span>2<\/span>,<\/span>'\/ day)'<\/span>)<\/span>)<\/span>,<\/span>\r\n\t\t\t\tlegend.mar = 4.1<\/span>)<\/span>\r\n\t}<\/span> else<\/span> if<\/span> (<\/span>!<\/span>log<\/span><\/a>)<\/span>{<\/span>\r\n\t\timage.plot(<\/span>lons2,<\/span> rev<\/span><\/a>(<\/span>lats2)<\/span>,<\/span> ROI.plot,<\/span> useRaster = TRUE<\/span>,<\/span> \r\n\t\t\t\tcol<\/span><\/a> = color,<\/span>\r\n\t\t\t\txlab = 'Longitude'<\/span>,<\/span> ylab = 'Latitude'<\/span>,<\/span> \r\n\t\t\t\tmain = paste<\/span><\/a>(<\/span>'Net Primary Production'<\/span>,<\/span> strftime<\/span><\/a>(<\/span>day1,<\/span>'%B %Y'<\/span>)<\/span>)<\/span>,<\/span> \r\n\t\t\t\tlegend.lab = expression<\/span><\/a>(<\/span>paste<\/span><\/a>(<\/span>'mg C \/'<\/span>,<\/span> m^<\/span>2<\/span>,<\/span>'\/ day'<\/span>)<\/span>)<\/span>,<\/span>\r\n\t\t\t\tlegend.mar = 4.3<\/span>)<\/span>\r\n\t}<\/span>\r\n\tROI # return region of interest data to workspace<\/span>\r\n}<\/span>  # end of vgpm.raster() function<\/span><\/pre>\n<\/div>\n<\/div>\n
Created by Pretty R at inside-R.org<\/a><\/p>\n","protected":false},"excerpt":{"rendered":"
Oregon State University makes a set of ocean productivity data derived from satellite data available for download and use by researchers. The Ocean Productivity website explains the available data and how it was derived. I have put together a few R functions to open a subset of the available data files and plot the data. […]<\/p>\n","protected":false},"author":2,"featured_media":0,"comment_status":"closed","ping_status":"closed","sticky":false,"template":"","format":"standard","meta":{"footnotes":""},"categories":[218],"tags":[107,108,109,113,111,110,112,106],"class_list":["post-1076","post","type-post","status-publish","format-standard","hentry","category-r-project","tag-image-plot","tag-latitude","tag-longitude","tag-modis","tag-net-primary-productivity","tag-osu","tag-satellite-data","tag-vgpm"],"_links":{"self":[{"href":"https:\/\/lukemiller.org\/index.php\/wp-json\/wp\/v2\/posts\/1076","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/lukemiller.org\/index.php\/wp-json\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/lukemiller.org\/index.php\/wp-json\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/lukemiller.org\/index.php\/wp-json\/wp\/v2\/users\/2"}],"replies":[{"embeddable":true,"href":"https:\/\/lukemiller.org\/index.php\/wp-json\/wp\/v2\/comments?post=1076"}],"version-history":[{"count":10,"href":"https:\/\/lukemiller.org\/index.php\/wp-json\/wp\/v2\/posts\/1076\/revisions"}],"predecessor-version":[{"id":1085,"href":"https:\/\/lukemiller.org\/index.php\/wp-json\/wp\/v2\/posts\/1076\/revisions\/1085"}],"wp:attachment":[{"href":"https:\/\/lukemiller.org\/index.php\/wp-json\/wp\/v2\/media?parent=1076"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/lukemiller.org\/index.php\/wp-json\/wp\/v2\/categories?post=1076"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/lukemiller.org\/index.php\/wp-json\/wp\/v2\/tags?post=1076"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}