Accessing registry tables
The chronosphere provides access to thousands of data items. The available data can be queried with the client package, or can be inspected on the project’s (in-development) website. To access the available data in R, we need client package.
library("chronosphere")## Chronosphere - Evolving Earth System Variables
## Important: never fetch data as a superuser / with admin. privileges!
##
## Note that the package was split for efficient maintenance and development:
## - Plate tectonic calculations -> package 'rgplates'
## - Arrays of raster and vector spatials -> package 'via'Overview
The available data can be inspected with the datasets()
function. This function is connecting to the remote where the registry
files are and downloads an overview of the registered data items.
ds <- datasets()## Use datasets(src = <src>) to see available versions and resolutions.This function returns a a data.frame class object, which
includes an overview of the available data( default argumentation). The
function will also check for important changes to the remotes, which
might indicate that some action (e.g. update of the client) is required.
Every row represents a unique series (src-ser
combinations), and some information about their defaults (version,
resolution, class, etc):
str(ds)## 'data.frame': 54 obs. of 9 variables:
## $ topic : chr "Traits" "Traits" "Organism distribution" "Organism distribution" ...
## $ sourceName : chr "Ancient Reef Traits Database" "Ancient Reef Traits Database" "BioDeepTime" "BioDeepTime" ...
## $ src : chr "AncientReefTraits" "AncientReefTraits" "biodeeptime" "biodeeptime" ...
## $ seriesName : chr "Reference list" "Denormalized Trait Data" "Denormalized biogeographic observations" "Bchron ages for records from the Neotoma Paleoecology Database" ...
## $ ser : chr "refs" "traits" "denormalized" "neotoma-bchron" ...
## $ defaultSeries: logi FALSE TRUE TRUE FALSE FALSE TRUE ...
## $ defaultClass : chr "data.frame" "data.frame" "data.frame" "data.frame" ...
## $ infoURL : chr "https://chronosphere.info/data/AncientReefTraits/refs/" "https://chronosphere.info/data/AncientReefTraits/traits/" "https://chronosphere.info/data/biodeeptime/denormalized/" "https://chronosphere.info/data/biodeeptime/neotoma-bchron/" ...
## $ defaultVer : chr "1.04" "1.04" "1.0" "1.0" ...Typically the 1) most recent version of the series with the 2)
coarsest resolution and the 3) fastest expected loading time will be the
default, which means that some data can be already be downloaded with a
simple src - ser combination. Please check out
the basic overview
of the chronosphere’s data model to read more about what
src, ser etc mean.
Note that this was tutorial was build on 2024-11-15, therefore the exact result that you see might be different!
Source-specific registry files
Note that the object that we just created does not contain all
available items that come from a series. There can can be hundreds of
these. This information is tabulated for every source
(src), and needs to be accessed separately. This ensures
that the growning of the chronosphere’s data library will not cause
performance issues when it comes to finding data. To get these
source-specific registry files, all we have to do is provide the source
(src) argument. For instance, if we want to access the data
items that are related to the PALEOMAP
project (or, on the chronosphere), we
have to provide the src="paleomap" argument:
pm <- datasets(src="paleomap")This results in a different data.frame, which includes
all items of the paleomap source. You can either look into
the table with View() (or equivalent) funciton, or you can
list out the available src- ser -
ver - res combinations with this chunk of
code:
## src ser ver resolution
## 1 paleomap areas 7 NA
## 2 paleomap dem 20180801 1.0
## 3 paleomap dem 20180801 0.1
## 4 paleomap dem v24221 0.1
## 5 paleomap dem v24221 1.0
## 6 paleomap gmst scotese02a_v21321 1.0
## 7 paleomap model v3-GPlates NA
## 8 paleomap model v19o_r1c NA
## 9 paleomap paleoatlas v3 0.1
## 10 paleomap paleocoastlines 7 NA
## 11 paleomap rainfall scotese_02 1.0Note that this large data.frame object will also include
additional metadata, such as the long names of the series, the
associated references, the class, as well as the URL of the
datafile that is associated with the the item and the
code that will be used to instantiate the object.
Saving the registry files
By default the registry files here are saved to a temporary
directory, which will be destroyed when your R session quits. This means
that you will have to re-download the registry files whenever you
execute the datasets() function. This can be avoided, with
the datadir argument, which has to point to an existing
directory (where you have writing access). The registry files will be
saved to this place.
# create chronosphere directory in the user's home
dir <- "~/chronosphere"
dir.create(dir, showWarnings=FALSE)
pm <- datasets(src="paleomap", datadir=dir)The datadir argument will added as a package-wide
variable in the next update, so you would not have to define it
separately for either datasets() or fetch()
function calls.
Now that we know how to inspect the available data, it is time to see how we can actually access these.