Benthic percent cover derived from image analysis for selected locations in the Pacific Ocean

2004 "American Samoa RAMP Cruise Report", http://www.pifsc.noaa.gov/library/pubs/cruise/Sette/CR0402-1.PSV.pdf.

2005 "Main Hawaiian Islands RAMP Cruise Report", http://www.pifsc.noaa.gov/library/pubs/cruise/Sette/CR0502-PSV.pdf. "Main Hawaiian Islands RAMP Cruise Report", http://www.pifsc.noaa.gov/library/pubs/cruise/Hiialakai/CRHI0505b-1.PSV.pdf. "Mariana Archipelago Cruise Report", http://www.pifsc.noaa.gov/library/pubs/cruise/Sette/CR0512-1.RES.pdf.

2006 "American Samoa RAMP Cruise Report", http://www.pifsc.noaa.gov/library/pubs/cruise/Hiialakai/CRHI0602-RS.pdf. "Main Hawaiian Islands RAMP Cruise Report", http://www.pifsc.noaa.gov/library/pubs/cruise/Hiialakai/CRHI0610-REB.pdf.

2007 "Mariana Archipelago RAMP Cruise Report", http://www.pifsc.noaa.gov/library/pubs/cruise/Hiialakai/CRHI0702-RS.pdf.

2008 "American Samoa RAMP Cruise Report", http://www.pifsc.noaa.gov/library/pubs/cruise/Hiialakai/CRHI0802-REB.pdf. "Main Hawaiian Islands RAMP Cruise Report", http://www.pifsc.noaa.gov/library/pubs/cruise/Sette/CR0810-BVA.pdf.

2009 "Mariana Archipelago RAMP Cruise Report", http://www.pifsc.noaa.gov/library/pubs/cruise/Hiialakai/CRHA0902-RES.pdf.

2010 "American Samoa RAMP Cruise Report", http://www.pifsc.noaa.gov/library/pubs/cruise/Hiialakai/CRHA1001II-KD.pdf. "Main Hawaiian Islands RAMP Cruise Report", http://www.pifsc.noaa.gov/library/pubs/cruise/Hiialakai/CRHA1008-KD.pdf.

2011 "Mariana Archipelago RAMP Cruise Report", http://www.pifsc.noaa.gov/library/pubs/cruise/Hiialakai/CRHA1101II-KD.pdf. "A Report Based on Underwater Visual Surveys in Guam and the Mariana Archipelago", http://www.pifsc.noaa.gov/library/pubs/tech/NOAA_Tech_Memo_PIFSC_33.pdf.

2013 "Small Boat Mission Report: Coral disease surveys in the vicinity of Kauai Island", http://intra.pifsc.gov/sites/default/files/SBR1309-BVA.pdf. "MHIRAMP Cruise Report", http://www.pifsc.noaa.gov/library/pubs/cruise/Hiialakai/CRHA1304I.pdf. "Small Boat Mission Report: 2013 Pacific Reef Assessment and Monitoring Program (continued), Oahu Operations", http://intra.pifsc.gov/sites/default/files/SBR1320-BVA.pdf. "Pacific Reef Assessment and Monitoring Program. Fish monitoring brief: main Hawaiian Islands 2013", http://www.pifsc.noaa.gov/library/pubs/DR-13-016.pdf.

2014 "Mariana Archipelago RAMP Cruise Report", Cruise report not yet published as of the publication date of this record. Contact CRED at nmfs.pic.credinfo@noaa.gov for a draft copy of the report. "Pacific Reef Assessment and Monitoring Program. Fish monitoring brief: southern Mariana Archipelago 2014.", http://www.pifsc.noaa.gov/library/pubs/DR-14-009.pdf. "Pacific Reef Assessment and Monitoring Program. Fish monitoring brief: northern Mariana Archipelago 2014.", http://www.pifsc.noaa.gov/library/pubs/DR-14-010.pdf.

"Coral Reef Ecosystem Division standard operating procedures: data collection for Rapid Ecological Assessment fish surveys", http://www.pifsc.noaa.gov/library/pubs/admin/PIFSC_Admin_Rep_11-08.pdf.

It is assumed that images have already undergone appropriate post-collection processing in the field to catalog and rename the images using standard naming conventions prior to the beginning of the benthic image analysis.

If images need to be corrected for photo analysis run series or calibration exercises, a copy of the original image is preserved. Generally, for stratified random coral and fish site survey photos, if the cameras were properly white-balanced in the field, no color-correction is required. TDS all photos analyzed have undergone color correction implementing Picasa. Corrections may include brightness/contrast adjustments, color adjustments, sharpening, etc. In general, the software's auto-adjustment capabilities are sufficient for the majority of images. Some additional manual adjustments may still be required. If color-correction doesn't improve the image, this image may be excluded from the analysis for that location or transect.

Images are quantitatively analyzed implementing the Coral Point Count with Excel extensions (CPCe; Kohler and Gill, 2006) software. Images are opened up in the CPCe program and the area for image analysis is set. Due to how the images are collected the area for image analysis may vary. Due to image distortion that occurs along the edges of the majority of images, all towed-diver survey imagery and REA survey imagery from 2014 and beyond are analyzed with a pixel buffer around the entire perimeter of the image; pixel offset (228,192) and (60,42) for TDS and REA respectively. Prior to that the entire image was used for stratified random fish and benthic REA site survey photos.

Ten stratified random points are calculated and projected on each image by the CPCe software. The benthic elements falling directly underneath each point are identified to two (TSD) or three (REA surveys) functional groups levels: Tier 1 (e.g., hard coral, soft coral, macroalgae, turf algae, etc.) and Tier 2 (e.g., Hard Coral = massive, branching, foliose, encrusting, etc.; Macroalgae = upright macroalgae, encrusting macroalgae, bluegreen macroalgae, and Halimeda, etc.), and Tier 3 (e.g., Hard Coral = Acropora, Montipora, Pocillopora, etc.; Macroalgae = Padina, Dictyota, Dictosphaeria, etc.). In cases where a coral colony exhibits more than one growth form, it is the morphology directly under the point that is classified, rather than the morphology of the colony overall. If a point falls precisely between two coral growth morphologies, or on the border between two benthic categories (e.g., coral-algae), the benthic category occupying the greatest area within symbol (circle wrapping the cross-hairs) is classified. Additionally, if the two benthic categories occupy equal space within the symbol, the benthos falling on the top left quadrant within the point symbol is classified. The category Shadow is used when the point falls on an area that is dark and the nature of the benthos cannot be assessed due to diminished light. The category Unclassified is used when the nature of the benthos cannot be determined due to image quality or unfamiliarity with the type of benthos. The category Wand is used when the point falls on a transect line/tape measure, or photo quadrat frame/spacing wand.

If any photo has more than 5 combined points classified as Shadow or Unclassified, or if a stratified random site photo has more than 1 point classified as Wand these data are not saved. Instead a new set of points to analyze is generated by using the Point Overlay menu and selecting Recalculate point coordinates. This step of regenerating points for analysis will be repeated up to three times before the image is removed from the production series. For tows with several photographs whose quality is insufficient for use, ~35% (40 images) of the total number of images to be analyzed is the minimum acceptable number that need to be analyzed to grant retention of those data in the production series.

Power analyses support the methodological approach outlined above, whereby statistically sound sub-jurisdictional (island) level, benthic cover estimates can be successfully derived from REA and TDS photos processed at: 300 points/survey (i.e., 10 points/photo x 30 photos) and 600-700 points/survey (i.e., 10 points/image x 60-70 photos), respectively. Percent cover at the island level can be determined from data derived from towed-diver survey images while percent cover is calculated at the site level for data derived from stratified random sampling images.

The complete Benthic Image Analysis Standard Operating Procedure (SOP) is available on the NOAA Wiki at https://www.st.nmfs.noaa.gov/confluence/display/CRED/Benthic+Image+Analysis. Non-NOAA users can contact CRED at nmfs.pic.credinfo@noaa.gov to request a copy of the SOP.