Skip to content

R Functions For Accessing OOI Science Data Through The M2M System

License

Notifications You must be signed in to change notification settings

oceanobservatories/ooim2mr

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

49 Commits
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 

Repository files navigation

ooim2mr

License: CC0-1.0

The goal of ooim2mr is to provide users a programmatic method for accessing Ocean Observatories Initiative (OOI) scientific data via R. Please review the OOI Data Usage Policy if you plan to cite OOI data.

Installation

Install the development version of ooim2mr from this GitHub repository with:

#install.packages("devtools")
devtools::install_github("oceanobservatories/ooim2mr")

This package was built and tested in RStudio. It is recommended that you utilize this package through the RStudio integrated development environment.

Bugs, Issues, and Functionality Requests

General

For general bugs and issues, please use GitHub’s issue tracker. This package reads from a lookup table in the package data subfolder and has been curated to only give users science streams with variables that are L1 or greater data products. If you are unable to find the data that you want, the request stream and variables may have been inadvertently removed. If you believe this to the case, please submit this as an issue.

Issue: ncdf4 (Windows only)

There is a bug in the ncdf4 package (Version 1.17) that does not allow the reading of OpenDAP datasets. The work around to this is to first download the NetCDF file and open it locally. If you are running R on a MacOS or Linux machine, the ooi_download_data() function can generally be bypassed and the remote locations of the data can be input into the ooi_get_data() funciton. This brings data into the R environment without downloading any files.

Functionality Requests

For functionality requests, please submit those directly to the OOI Helpdesk with “ooim2mr” in the subject line: [email protected]

Examples

Example R scripts can be found on the OOI Data Explorations GitHub Repository.

Usage

Most functions in this package require an internet connection to submit requests or download data through the OOI API. The following native and CRAN packages are also needed when utilizing this package.

Required Package Install Command Use
httr install.packages(“httr”) For submitting requests to OOINet.
ncdf4 install.packages(“ncdf4”) For opening NetCDFs.
stringr require(stringr) For parsing the lookup table.
jsonlite require(jsonlite) For interpreting responses from OOINet.

ooi_site_info

# Example: Requests site information for the Oregon Inshore Surface Piercing Profiler (CE01ISSP).

user = "OOI-API-USERNAME"
token = "OOI-API-TOKEN"
info = ooi_site_info(site = 'CE01ISSP',user = user, token = token)

This function takes a supplied OOI site and submits a request using the provided credentials to obtain site metadata. The site must be supplied as an OOI eight (8) character site designator.

A list of OOI sites can be found here: https://oceanobservatories.org/site-list/

The return is a cascading list of lists that contains site metadata with respect to nodes, instruments, streams, etc.

You can create or find your OOINet credentials through the Log In tab here: https://ooinet.oceanobservatories.org/

ooi_check_availability

# Example: Requests deployment and annotation information for the Oregon Inshore Surface Piercing Profiler (CE01ISSP) CTD.

user = "OOI-API-USERNAME"
token = "OOI-API-TOKEN"
check = ooi_check_availability(site = 'CE01ISSP' ,node = 'PROFILER' ,instrument = 'CTD' ,user = user,token = token)

The supplied site, node, and instrument are used to submit a request for deployment and annotation information using the additionally given user credentials. For this function, a substring of the site, node, and instrument can be used.

You can create or find your OOINet credentials through the Log In tab here: https://ooinet.oceanobservatories.org/

ooi_create_url

# Example: Creates a request URL for the Oregon Inshore Surface Piercing Profiler (CE01ISSP) CTD for July 2019.

site = 'CE01ISSP'
node = 'PROFILER'
instrument = 'CTD'
method = 'recovered'
stream = ''
start_date = '2019-07-01'
start_time = '00:00:00'
stop_date = '2019-07-31'
stop_time = '23:59:59'
url = ooi_create_url(site = site ,
                     node = node ,
                     instrument = instrument, 
                     method = method,
                     stream = stream,
                     start_date = start_date,
                     start_time = start_time,
                     stop_date = stop_date,
                     stop_time = stop_time)

The supplied site, node, instrument, and method will be used to parse the lookup table for a request URL. If for some reason there exist multiple returned URLs, you can add a stream designator using the stream parameter.

By default, the site, node, instrument, method, and stream parameters are set to be empty strings. The function utilizes grep to parse the lookup table, so it could be used as a search function. For example, if you wanted to know what OOI Coastal Endurance Array assets telemeter dissolved oxygen data, the following would return a list of URLs which you could compare to the OOI site list.

site = 'CE'
instrument = 'DO'
method = 'telemetered'

url = ooi_create_url(site = site, instrument = instrument, method = method)

site

The site is generally an eight (8) character designator.

node

The node is generally a five (5) character designator. If inputting a node that is five (5) characters or less, the function assumes you are using the actual OOI node designator. If you use an input that is six (6) characters or more, the function assumes you are using a simplification of the node.

Simplified Node Name OOI Node Designator
Surface Surface Buoys (SBC11, SBD11, SBD12, SBD17, SBS01)
Midwater Near Surface Instrument Frames (7m, 12m), Mooring Risers (RI000,RII01,RII11,RIM01, RIS01, RID16, RID27, RID26, RIC21, PC03A, PC01A, PC01B)
Seafloor Multi-function Nodes, Benthic Packages (MFD35, MFC31, MFD37, LJ01D, MJ01C, LJ01C, LV01C, PN01C, PN01D, LV01A, PN01B, LJ01A, PN01B, MJ01A,PN01A, LJ01B, PN03B, MJ03E, MJ03F, MJ03C, MJ03D, MJ03A)
Profiler Moorings With One Profiler (SP001, SP002, WFP01, SF03A, SF01A, DP01A, DP03A, DP01B, SF01B)
Upper_Profiler Moorings With Two Profilers (WFP02)
Lower_Profiler Moorings With Two Profilers (WFP03)

Glider nodes are the glider’s unique serial number. If you are looking for glider data, it is recommended that you use the Glider DAC to access OOI glider data.

Example: Using SEAFLOOR as the node is the same thing as MFD35.

instrument

The instrument is generally a twelve (12) character designator, with the third character being a dash (-). Substring replacement can be used when parsing the lookup table.

Example: Using 051 as the instrument is the same thing as 02-CTDMOH051.

method

The method is how you want to get the data. It can also be simplified using a substring when also considering the site, node, and instrument.

Example: Using host as the method is the same thing as recovered_host.

Method Description
recovered_host Data that the mooring controller collects from the instrument (think datalogger).
recovered_inst Data that the instrument records internally.
recovered_wfp Data that the McLane Moored Profiler controller collects from the instruments (think datalogger).
recovered_cspp Data that the CSPP controller collects from the instruments (think datalogger).
streamed Data streamed over the Cabled Array.
telemetered Data telemetered over cell or iridium. Often decimated in size.

stream

Some instruments are split into multiple data streams. Most of the time, just supplying the site, node, instrument, and method is enough to get a single request URL. If multiple URLs are returned, you can specify the stream to reduce the list to one.

start_date

This is the start date of your data request window in UTC. By default, it is set to ‘2010-01-01’, so if you do not specify a start date, the request will get all OOI data since the start of the project. start_date must follow the format of ‘YYYY-mm-dd’.

start_time

This is the start time of your data request window in UTC. By default, it is set to ‘00:00:00’. start_time must follow the format of ‘HH:MM:SS’.

stop_date

This is the stop date of your data request window in UTC. By default, it is set to ‘2040-12-31’, so if you do not specify a stop date, the request will get all OOI data up until the second that you submit the request. stop_date must follow the format of ‘YYYY-mm-dd’.

stop_time

This is the stop time of your data request window in UTC. By default, it is set to ‘23:59:59’. stop_time must follow the format of ‘HH:MM:SS’.

ooi_submit_request

#Generic Example: url is the url obtained via ooi_create_url, or as a string url built by hand.

user = "OOI-API-USERNAME"
token = "OOI-API-TOKEN"
url = 'https://ooinet.oceanobservatories.org.....'
response = ooi_submit_request(url = url,user = user, token = token)

This function takes the URL generated by ooi_create_url() and issues a request to OOINet for that data. The return is a response object that is used in a following ooi_get_location() function. If the request fails, a message with the failure reason is printed.

You can create or find your OOINet credentials through the Log In tab here: https://ooinet.oceanobservatories.org/

ooi_get_location

#Generic Example: response is the response object created via ooi_submit_request.

remote = ooi_get_location(resposne = response, drop_paired = TRUE)

This function queries the JSON response created by the ooi_submit_request() function until the data request is complete. Once the data request is complete, the OpenDAP urls (dodsC) of where the data is remotely located is returned. By default, the drop_paired flag is set to TRUE. This removes paired datasets from the return output, making it easier to concatenate multiple files or deployments. If the drop_paired flag is set to FALSE, the return includes all paired dataset locations.

ooi_download_data

#Generic Example: remote is the list of OpenDAP urls returned from the ooi_get_location function or as a hand-compiled list.

directory = 'C:\Users\Ian\Desktop\test'
local = ooi_download_data(remote, directory)

This function converts the dodsC OpenDAP URLs to fileServer URLs so that the data can be downloaded to the local machine. Specifying the directory will download the data to that directory. If left empty, the default action is to download data to the current R session working directory.

Files are renamed as site_node_instrument_method_deployment_time.nc

This function can be used if you want to look at data later while offline, but it was initially included to allow Windows users to access the data via local files through ncdf4. Version 1.17 of ncdf4 struggles with opening some OpenDAP urls on Windows machines. If you are a MacOS or Linux R user, you can likely bypass this function if you do not want to download data.

ooi_get_data

# Generic Example: local is the list of filepaths gained through ooi_download_data or a list of hand-crafted filepaths.
# Alternatively, you can use a list of OpenDAP URLs, like the ones you get from ooi_get_location, to bring data directly into the workspace.

data_variables = ooi_get_data(local, simplify_data = TRUE)
data = data.frame(data_variables[['data]])
variables = data.frame(data_variables[['variables_units]])

This function will bring data into the work space and attempt to merge it as a dataframe if the data is 1D. If the data is 2D, it remains a list of lists. The return for this function is a list of lists. List one contains the data. List two contains a dataframe of variables that were brought and the units associated with those variables.

By default, the simplify_data flag is set to TRUE. This removes confusing variables and leaves mostly first order or greater data products. Setting this flag to FALSE will bring in all variables available in the NetCDF.

About

R Functions For Accessing OOI Science Data Through The M2M System

Resources

License

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published