Abstract:

This dataset reports In Situ Ichthyoplankton Imaging System (ISIIS) image segments, classifications, and oceanographic data collected on board the R/V Point Sur cruise PTS01 in the northern Gulf of Mexico on 2015-11-04. The chief scientists for this cruise are Dr. Frank Hernandez and Dr. Monty Graham. The data were collected using a towed in situ imaging just offshore of the Louisiana Bird Foot Delta. The ISIIS collected hydrographic data oceanographic parameters such as depth of the vehicle, temperature, salinity, dissolved oxygen, light levels (PAR), and chlorophyll-a fluorescence of the water, along with in situ images of the plankton. The dataset includes the R codes for the image classification. The data were collected during the day and night, and regions of interest (segments) were extracted and identified for comparison to continuous ADCP measurements collected in the same area. ADCP data is available under GRIIDC Unique Dataset Identifier (UDI) R4.x260.204:0004 (DOI: 10.7266/N7PG1Q49).

Suggested Citation:

Adam T. Greer, Christian Briseno-Avena, Sabrina M. Parra. 2019. In Situ Ichthyoplankton Imaging System (ISIIS) and oceanographic data collected on board the R/V Point Sur cruise PTS01 in the northern Gulf of Mexico on 2015-11-04. Distributed by: GRIIDC, Harte Research Institute, Texas A&M University–Corpus Christi. doi:10.7266/n7-yan1-f697

Publications:

Parra, S. M., Greer, A. T., Book, J. W., Deary, A. L., Soto, I. M., Culpepper, C., … Miles, T. N. (2019). Acoustic detection of zooplankton diel vertical migration behaviors on the northern Gulf of Mexico shelf. Limnology and Oceanography. doi:10.1002/lno.11171

Purpose:

The data were collected to describe the plankton community in comparison with acoustic backscatter diel variability obtained from upward-facing ADCPs.

Data Parameters and Units:

Physical oceanographic data (physical.csv): central (central time, CST, hh:mm:ss), Julian time (percentage of a day), UTC time (hh:mm:ss), Unix time, temperature (temp, degrees C), depth (meters), flor or fluor (fluorescence, volts), forward velocity (forvel, m/s), heading (degrees), dissolved oxygen concentration (o2conc, mg/L), dissolved oxygen voltage (o2volts, volts), Photosynthetically Available Radiation (PAR, µE/cm^2), PAR (parv, volts), pitch (degrees), roll (degrees), salinity (PSU), latitude (degrees), longitude (degrees), vertical velocity (vertvel, m/s), time (MATLAB time). Classification files (particleSum.csv: mean concentrations in 1m bins for the objects segmented from the ISIIS images): count id, rdepth (rounded depth of observations to the nearest 1m), totc (total count for the particles), mSi (mean size of the particle in that bin, pixels), volsamp (volume sampled within that bin, m^3), conc (concentration of particles within that bin). Classification files (particleStats.csv): Julian (Julian time that individual plankton was detected, percentage of a day, local time, CST), timestamp (YYYMMDDHHMMSS.SSS.tif:slice where slice is the repeated and is the frame number within an image stack), area (area of the segmented particle in pixels), perim (perimeter of the segmented particle in pixels), bx (x position of the top left corner of the bounding box), by (y position of the bounding box), width (width of bounding box in pixels), height (height of the bounding box in pixels), major (length of the major axis of fitted ellipse in pixels), minor (length of the minor axis of the fitted ellipse in pixels), angle (angle of the major axis for the particle in degrees (0 to 180) - 90 degrees is perfectly horizontal in real space), circ (circularity of the segmented particle; between 0 and 1, 1 is a perfect circle), slice (frame number in the image stack), AR (aspect ratio of the particle, major/minor axis length), round (roundness of the particle from 0 to 1), solidity (solidity of the particle from 0 to 1). Classification files (ID.csv and *IDsizes.csv): nearjul (nearest Julian time in the physical dataset used to merge data), Julian, ID (code for identification of the segment, performed manually), junk (section of the image segment), label (same as timestamp), slice (image number within the image stack), xloc (x location of the top left corner of the bounding box for the segmented particle, pixels), yloc = y location of the top left corner of the bounding box for the segmented particle, pixels), width, height, central time, UTC time, Unix time, temperature (temp, degrees C), depth (meters), fluor (fluorescence, volts), forward velocity (fvel, m/s), heading (degrees), dissolved oxygen concentration (mg/L), dissolved oxygen voltage (o2volts, volts), PAR (µE/cm^2), parvolts (PAR, volts), pitch (degrees), roll (degrees), salinity (psu), latitude (degrees), longitude (degrees), vertical velocity (vertvel, m/s), X1 (disregard this column), X (disregard this column), area (pixels), perim (perimeter, pixels), width (pixels), height (pixels), major (pixels), minor (pixels), angle (degrees), circ (0-1), slice, AR (aspect ratio of the particle), round (0-1), solidity (0-1), ESDmm (equivalent spherical diameter of the particle (converted to mm using the field of view of the camera - 120 mm). Classification file (1104DielConc.csv): rdepth (rounded depth, m), variable (taxonomic category from the ISIIS images), value (concentration of the taxonomic category, individuals/m^3), diel (categorical variable for the time of day when the measurement was taken). Folder "ClassifiedSegments1104all-bigcam": image segments classified to different taxonomic groups. Filename format: TaxonName_Plankton_YYYYHHMMSS.SSS.tif_slice_bx_by_width_height.tif. Folders "ParticleCounts1104daynight-bigcam" and "ParticleCounts-1104-smallcam": contain text files with information about the size of segmented particles. The columns correspond to the following variables: label, area (pixels), perim (perimeter, pixels), width (pixels), height (pixels), major (pixels), minor (pixels), angle (degrees), circ (0-1), slice, AR (aspect ratio of the particle), round (0-1), solidity (0-1). File “SegmentationIDs.xlxs”: depth range (m), a percentage of particles (chaeto = chaetognaths, copepods, cteno = ctenophores, diatoms, dolio=doliolids, marine snow, other, unknown, polychaetes, pteropods, shrimp, siphon=siphonophores, stomato=stomatop) For more details on these measurements, you can visit the website for ImageJ at https://imagej.nih.gov/ij/ Additional information about the files: particleSum are the mean concentrations in 1 m bins for the objects segmented from the ISIIS images. I made different subsets to look at the change in the aspect ratio of the particles (AR) and the size of the particles (pixels) and their impact on the vertical distribution. All of these files were generated by filtering the data in various ways, and the exact filters applied are detailed in the R script file, which includes all of the data analysis for the ISIIS images in this publication. For the other files, the names ARXless just refers to the data containing objects when an aspect ratio less than X. Aspect ratio is calculated by fitting an ellipse to each particle and dividing the long axis of the ellipse by the short one. 1104dayID.csv and 1104nightID.csv are the manual identifications of each object segmented, merged with the physical data corresponding to each object. Each row is 1 individual object or particle, and the row contains all of the relevant information associated with that particle. 1104DielConc.csv contains the same data as 1104nightID and 1104dayID but transformed so the data are presented as counts per taxa per 1 m vertical bin. This data frame was used in the plot of the vertical distributions of the taxa in Parra et al., (in Press). All of these files were generated with the R script included in the dataset (Analysis4ISIISdata.R).

Methods:

The In Situ Ichthyoplankton Imaging System (ISIIS) was towed behind the RV Point Sur on November 4, 2015, as part of the interdisciplinary research consortium CONCORDE (see Greer et al. 2018 for details on the sampling design and instruments used). The ISIIS uses shadowgraph illumination and a line scan camera to image a continuous strip of ocean that is parsed into individual frames that are approximately 13 cm by 13 cm for the large camera and 4.3 cm by 4.3 cm for the small camera (50 cm depth of field and 8.9 cm depth of field, respectively). These images are accompanied by several oceanographic sensors mounted to the imaging system (CTD, dissolved oxygen, PAR, and chlorophyll-a) that collect data at 6 Hz, slightly slower than the speed of image acquisition (~17 Hz for the large camera, ~49 Hz for the small camera). The two datasets are merged using the nearest common time stamp. Counts of different organisms are binned by depth and volume sampled to produce concentrations or abundances. ISIIS images were segmented using ImageJ, which automatically extracts particle sizes and shape features. A subset of these segments was identified manually using ImageJ shortcuts for each taxon. Smaller size classes were subsampled randomly (see Analysis4ISIISdata.R and SegmentationIDs.xlsx), where 1000 were randomly chosen and identified from 10 m depth bins. The goal of these methods was to compare the abundances of different groups to bottom mounted ADCPs on November 4, 2015 (ADCP data: DOI: 10.7266/N7PG1Q49).

Instruments:

In Situ Ichthyoplankton Imaging System (ISIIS)

Provenance and Historical References:

Greer, A.T., A.M. Shiller, E.E. Hofmann, J.D. Wiggert, S.J. Warner, S.M. Parra, C. Pan, J.W. Book, D. Joung, S. Dykstra, J.W. Krause, B. Dzwonkowski, I.M. Soto, M.K. Cambazoglu, A.L. Deary, C. Briseño-Avena, A.D. Boyette, J.A. Kastler, V. Sanial, L. Hode, U. Nwankwo, L.M. Chiaverano, S.J. O’Brien, P.J. Fitzpatrick, Y.H. Lau, M.S. Dinniman, K.M. Martin, P. Ho, A.K. Mojzis, S.D. Howden, F.J. Hernandez, I. Church, T.N. Miles, S. Sponaugle, J.N. Moum, R.A. Arnone, R.K. Cowen, G.A. Jacobs, O. Schofield, and W.M. Graham. (2018). Functioning of Coastal River-Dominated Ecosystems and Implications for Oil Spill Response: From Observations to Mechanisms and Models. Oceanography, 31(3): 90-103. doi:10.5670/oceanog.2018.302 Parra, S.M., A.T. Greer, A.L. Deary, J.W. Book, C. Briseño-Avena, A.D. Boyette, T.N. Miles, F.J. Hernandez. Acoustic detection of zooplankton diel vertical migration behaviors on the northern Gulf of Mexico shelf (in press) Limnology & Oceanography.

Individual Files: