This data package includes data from Gulf of Mexico Research Initiative Information and Data Cooperative (GRIIDC) Unique Dataset Identifiers (UDI) R6.x815.000:0023. It contains ZooScan zooplankton imagery data collected on board the R/V F.G. Walton Smith Cruise WS1017 during Natural Resource Damage Assessment (NRDA) Plankton Survey Walton Smith 3 (WS3) in the Gulf of Mexico from 2010-09-26 to 2010-10-01. 

This survey (WS3, chief scientist: Malinda Sutor) was part of a series of NRDA cruises conducted in 2010 and 2011 to evaluate the distribution and densities of ichthyoplankton and other zooplankton in Gulf of Mexico waters potentially affected by the Deepwater Horizon Oil Spill (DWHOS) and in surrounding areas.

Data provided in this data package are generated from the Natural Resource Damage Assessment Deepwater Horizon Oil Spill Plankton Processing Plan. Stations sampled are on the Southeast Area Monitoring and Assessment Program (SEAMAP) Gulf of Mexico grid.

There are both originators files and files that were converted by GRIIDC from the originally submitted format to an archival format. The converted files are located in the same directory as the original file. 

Converted File Types: *.csv- Comma-separated-value (CSV) files that GRIIDC converted from Microsoft Excel files. There is one CSV file for each Excel spreadsheet. For multi-spreadsheet files, the sheet name is appended to the file basename, separated by an underscore (_) and the worksheet name from the originator file.

Dataset directory naming structure and general contents: This data package contains one single directory within the main directory "WS3_LargeData/Data". 1) R6-x815-000-0023_ZooScan_Image_Data. This data directory is named based on the GRIIDC dataset UDIs [GoMRI RFPnumber.xGoMRI Project ID.GoMRI task ID:data set number, where the colons (:) and periods (.) are replaced with hyphens (-)] and data types. This directory is further organized into two separate subdirectories, 1) Data and 2) Documentation, where "Data" folders include the data files and the "Documentation" folders include the GRIIDC Standard ISO 19115-2 Metadata (XML file) for that particular dataset. Please note that some of the GRIIDC datasets include "GRIIDC cruise data documentation" in the GRIIDC database and those Excel files are not included in the package. However, all the related information and keywords from the files are entered in Send2NCEI (S2N), NCEI archiving tool while submitting these datasets. The general contents of the dataset is described below:

1. WS3_LargeData/Data/R6-x815-000-0023_ZooScan_Image_Data: This directory includes data from GRIIDC dataset R6.x815.000:0023, "ZooScan zooplankton image data, Walton Smith 3, WS3, NRDA plankton survey research cruise in the Gulf of Mexico, R/V F.G. Walton Smith WS1017, 2010-09-26 to 2010-10-01." The point of contact is Kelly L. Robinson and authors are Stacy Calhoun, Kelly Robinson, and Malinda Sutor. This dataset contains ZooScan image data of zooplankton with the accompanying Zooscan files so that others can re-analyze the images. 

Parameters included are: CruiseName [Survey vessel]; CruiseNo [Number of NRDA survey on above vessel]; StationID [Station name (B# stations are from the NOAA SEAMAP Grid)]; DayNightSample [Time of day associated with the sample. Day (D) is 1 hour after sunrise to 1 hour before sunset. Night (N) is 1 hr post sunset to 1 hr pre sunrise]; DeepShallowSample [General depths targeted by MOCNESS tow (not individual nets). Shallow (S) = 0-160 m; Deep (D) = 0-1500+ m]; DeploymentID [Unique identifier for each MOCNESS tow]; Field.Sample.ID [Unique identifier for each depth discrete net sample within the MOCNESS tow]; SampDateStart [Date sample was started (net entered the water/opened)]; SampDateEnd [Date sample was ended (net exited the water/closed)]; TowStartTime [Time of depth discrete net opening]; TowEndTime [Time of depth discrete net closing]; UpperDepth [Minimum depth sampled by net (m)]; LowerDepth [Maximum depth sampled by net (m)]; Volume [Volume of water filtered through the net (m3)]; StartLat [Start latitude of the tow]; StartLon [Start longitude of the tow]; EndLat [End latitude of the tow]; EndLon [End longitude of the tow]; Latitude [Latitude of the station (target coordinates, not actual sampling coordinates)]; Longitude [Longitude of the station (target coordinates, not actual sampling coordinates)]; NetNo [Net number (when applicable)]. 

Methods: Original zooplankton data were received as image data and Excel spreadsheets with density calculations for samples processed during the NRDA effort. Image files included in this dataset are stored as tif files (background scans, zooscan scans), and jpeg files (sorted and unsorted vignettes). Additional files required for Zooscan processing are also included. The image dataset includes neuston zooplankton samples in addition to the MOCNESS zooplankton samples. However, the dataset authors cannot confirm that all neuston samples taken during the cruise are represented in this dataset. Scans that were not separated into vignettes at the time of receipt were not processed by the dataset authors, but the files necessary for processing are included. Samples not scanned during the original NRDA effort were scanned by Robinson lab for this cruise (WS3) at University of Louisiana at Lafayette and were sorted manually in Ecotaxa and do not have "Learning Set" PID process files. Other PID process files produced by Zooscan are available in the PID process folder for each sample. Hence, vignettes sorted with PID are stored in taxa-specific folders in the PID_process/sorted_vignettes folder. 

Likewise, vignettes sorted in Ecotaxa are stored in taxa-specific folders in the Ecotaxa_sorted_vignettes folder. *Present only in samples sorted with Ecotaxa software.

Data are organized as per tow in a folder labeled with NRDA sample ID and each tow includes both source and the processing data. The naming convention for NRDA sample ID is: [cruise ID]-[StationID (B# stations are from the NOAA SEAMAP Grid)Day/Nighttime of the day during sampling where D (day) is 1 hour after sunrise to 1 hour before sunset, N (night) is 1-hour post-sunset to 1-hour pre-sunrise]-[MOC (for MOCNESS)]-for each depth discrete net sample within the MOCNESS tow. 

Each sample scanned using the Zooscan instrument have its own sample specific electronic folder within containing all of the data associated with that particular sample. Data are further organized into 6 subfolders:  

1. PID_process - Contains all of the information pertaining to extracted particles that will be needed for use in Plankton Identifier, the automated image identification software.

2. Zooscan_back - Contains original 16bit tif background scan images, as well as processed 8bit background scan images. Also includes a text log file of the settings used for taking these background scans.

3. Zooscan_config - Contains the configuration text files used for the project.

4. Zooscan_meta - Contains “metadata” text file for each individual scan. File contains information associated with the sample (gear type, split, mesh size, sample site, etc.)  

5. Zooscan_results - Contains validated pid files from Plankton Identifier software. These files contain the identifications made for each particle in the scans.

6. Zooscan_scan - Contains the following:
•  “_raw” folder --> original 16bit uncompressed tif scans, their associated “metadata” text file, and the log file containing the settings used for that particular scan.
•  “_work” folder --> subfolders for every scan taken of the sample. Each folder contains files needed for Plankton Identifier software. Also includes .tsv files for Ecotaxa sorted samples.
•  processed 8bit tif scan images.

File format included are: csv, dat, ext, gif, html, ini, jpg, mat, pid, tdm, tif, tsv, txt, xlsx

Note_1: Please note that the original dataset in GRIIDC includes 84 empty folders that are byproduct of classification workflow, and the folders for detections are created even if no detections for that species is found. However, these folders were removed and are not included in this data package submitted to NCEI, and the list of those removed empty folders is provided in the text file [R6-x815-000-0023-emptyfolders-deleted.txt].

Note_2: There are a number of close to duplicate names for sample directories, example: WS3-B175D-MOC5-587-399 and WS3_B175D_MOC5_587_399. From the phone call conversation with dataset author (Stacy Calhoun), GRIIDC found that the names with "underscores" should be the ones processed by the Robinson lab; those also contain a subdirectory called "Ecotaxa_Sorted_Vignettes". Those are manually sorted images scanned from fixed sample jars provided by Melinda Sutor. The subdirectories "_raw" and "_work" are the raw scans and scans with background removed, respectively. Dataset author did not feel there was a significance to any samples without depth ranges (_587_399 in the above example); they simply may not have had that information provided to them on the sample jars.

Note_3:  The metadata file originally submitted with the data listed the metadata for every sample rather than every station; it was also incomplete.  We have retrieved the station metadata from companion datasets (GRIDDC UDI R6.x815.000:0014 and R6.x815.000:0009) to create the file "WS3_MOC_summary.csv."

Note_4:  The related datasets appear to use "pressure" and "depth" interchangeably.  

The submitter provided keywords: Zooplankton, ZooSCAN imagery, MOCNESS net tows, Neuston, Plankton survey, Deepwater Horizon Plankton Assessment Archive (DWHPAA), Southeast Area Monitoring and Assessment Program (SEAMAP), Natural Resource Damage Assessment (NRDA). 

"Documentation" folder in all three data directories includes the GRIIDC Standard ISO 19115-2 Metadata (XML file) for that particular dataset UDI. The naming convention for GRIIDC Standard ISO 19115-2 Metadata is: GoMRI RFP number-xGoMRI Project ID-GoMRI task ID-dataset number-metadata.xml

MOCNESS profile and other zooplankton, ichthyoplankton and decapod specimen identification data collected during NRDA Plankton Survey Walton Smith 3 (WS3), R/V F.G. Walton Smith Cruise WS1017 are archived at NCEI under NCEI Accession Number 0247549.

This ReadMe file is created on 2022-03-11 by Bipana Sigdel from GRIIDC and updated on 2022-06-29 by Deborah LeBel of GRIIDC lsto submit data package to NCEI via Send2NCEI for long term archival.