In 2002, there were 57 lines surveyed at 28 sites. Due to funding cuts, the number of surveys dropped to 8 in 2003 and 5 in 2004. Surveys typically consist of shallow (~3m) and deep (~10m) lines. This dataset contains surveys from 20 sites.
The CRAMP Assessment Protocol RAT is a highly abbreviated version of the CRAMP monitoring protocol, consisting of a single fish transect, a single benthic transect, a rugosity measurement, a sediment sample and various qualitative habitat observations. The assessment protocol is designed to produce quantitative spatial data that is consistent with and comparable to data taken at the permanent monitoring sites. The power of the RAT lies in large numbers of replicates taken over the spatial range of a given habitat. The assessment program expands our ability to describe habitats and spatial distributions of Hawaiian reef organisms in relation to various environmental factors. However, the assessment protocol requires a small fraction of the human effort and cost per site. This is critical due to the large number of sites needed to describe habitats along the entire coast of Hawaii. Considerable time saving is achieved because no permanent transect markers are needed and no permanent photo-quadrats are installed. Assessment data can be used with monitoring data for spatial comparisons, but the benthic assessment data does not have the statistical power needed to establish temporal change with the degree of precision involved in the monitoring effort unless an extremely large number of RATs are performed in a small area or repeatedly over time.
All of the CRAMP long-term monitoring sites have been established on hard bottom in coral habitats stratified at depths of 3 m and 10 m. In contrast, the RAT is used in all coral reef ecosystem habitats and at all depths being mapped.
The reef fish sampling method used in the RAT is identical to the monitoring method, but only one 25m x 5 m transect is measured. The benthic sampling effort is the same as used at the monitoring sites, but reduced to a single 10 m transect that only has sufficient power to describe habitat differences. The RAT requires the use of two divers to conduct the full survey (fish, benthic video recording, rugosity measurement, sediment and observations) in a single short dive. In contrast, establishing the monitoring sites took a team of 6 divers multiple dives to install and conduct the initial monitoring of the site. Data entry time for the assessment method is reduced to less than 2 person-hours per site for the assessment method compared to more than 20 person-hours for the monitoring sites. The monitoring site protocol must have sufficient statistical power to detect a less than 10% change in coral cover between samplings. The assessment protocol only requires sufficient statistical power to allow quantitative description of a given habitat. Generally from 3 to 7 RATs are needed to describe a given habitat.
1. Location of the assessment site is pre-determined using the habitat maps and other information to develop the experimental design. A stratified random sample is selected from within each habitat. Latitude and longitude are determined for each rapid assessment site and entered as way points into the GPS.
2. A field team consisting of 2 divers navigates to way point using GPS, marks the location with a lead weight and float and accurately establishes the location using GPS measurements. Divers descend together. Diver 1 carries one 25 m transect line. Diver 1 begins fish transect starting at the marked way point and moves along depth contour. The fish count method is identical to that described above for the monitoring method. Diver 2 carries digital video system and rugosity chain. As diver 1 lays out transect line, diver 2 video records the general environment through the full 360-degree panorama at the starting point. Diver 2 then begins to video one 10 m transect selected at random along the 25 m fish transect line. Diver 2 also runs rugosity on the the 10-m transects. Diver 1 completes the fish transect and assists Diver 2 in completion of the rugosity, sediment sampling and general observations. This produces a data set similar to the monitoring sites but with only one fish transects. Brown, E, E Cox, B Tissot, K Rodgers, and W Smith (1999). Evaluation of benthic sampling methods considered for the Coral Reef Assessment and Monitoring Program (CRAMP) in Hawaii. International Conference on Scientific Aspects of Coral Reef Assessment, Monitoring, and Restoration. April 14-16, Ft. Lauderdale, FL.
Green, R H and S R Smith (1997). Sample program design and environmental impact assessment on coral reef. Proc 8th International Coral Reef Symposium. 2: 1459-1464.
McCormick, Mark 1994. Comparison of field methods for measuring surface topography and their associations with a tropical reef fish assemblage. Marine Ecology Progress Series 112: 87-96.
1. Files in ../../data (multiple sites) Filename template:
where: SSS: site ID (see STATIONS above) yyyymmdd: survey date
Fields in these files: Site Name - usually NA (not available) Station - usually NA Frame No - usually NA Image Date - usually NA, get survey (image) date from filename ID Name - equivalent to TaxonName in PointCount99, this is the species recorded but for some organisms if not identifiable to the species or even genus level then just to taxanomic level ID Code - usually NA
The following are PhotoGrid parameters equivalent to PointCount parameters of the same name. Point - Point number on the frame X - X coordinate on the image for each point Y - Y coordinate on the image for each point Intensity - value for the point Red - RGB value on the image Green - RGB value on the image Blue - RGB value on the image
Notes from Kuulei Rodgers concerning these paramters: "Point X and Y are the coordinates for each of the 50 points that are generated on an image. This way if you want to go back and check if it is correct or what someone called some organism it will regenerate the frame with the random points that were originally used. If for example you see Pavona maldivensis and want to see if that is correct because you don't think it is at that site and may have been interpreted, you can go back and look at point number 7 to see what is under it. The program will use the coordinates to reconstruct the original random points on that frame. Red, Green and Blue are just the exact colors as the person who first did the analysis saw it. Since you can adjust the color balance and the contrast, the program saves the adjustments so it can be revisited if need be later."
Filename - this is a critical parameter. It is the name of the image file. Convention is yyIISSSDDmTTFFF, yy : last two digits of year II : island SSS : site (see STATIONS above) DDm : depth in meters TT : transect number FFF : frame number
The remaining parameters can be ignored and are usually NA: Total Points,ID Date, Site Despite Code, Time Code, Institution, User Name, Habitat, WQS, Length, Depth
Taxa codes in PhotoGrid output are provided in MS Excel97 file: CRAMP99codesum.xls
The codes sheet was dumped into a CSV format in text file: taxacodes.csv
Potential frequently asked question: -The CRAMP website lists available benthic data parameters: coral and substrate cover. Can these be derived from the PhotoGrid data given? How?
Reply from Ku'ulei Rodgers: We use ACCESS, a relational database that calculates these for us but it can be done in EXCEL as well by sorting alphabetically and deriving a percent of the total for each substrate type. For example if you have 10 points that are Porites compressa and there are 20 frames with 50 points on each, this would be 10 out of 1000 points for the whole transect so 1% cover. This is then done for each substrate type. Then all the coral species percentages are added together for a total coral cover number.
2. Files in ../../data/waikiki2002
Filename template: SSSTTDDmyyyymmdd.csv
SSS : site (see #STATIONS above) yyyymmdd : survey date TT : transect number DDm : depth in meters
Fields in these files: same as above