Since 2016, the Port Orford project has collected Secchi disk measurements and GoPro video footage at each kayak sampling station. A Secchi disk is a simple tool with black and white quadrants that we lower into the water until it cannot be seen anymore. As we raise the disk out of the water, we count the marks on the line to calculate a measurement of water clarity (Figure 1). This long time-series of Secchi measurements is an excellent dataset, but what do these Secchi measurements actually reflect? Productivity in the water column, increased turbidity from river runoff, changes in zooplankton abundance? Additionally, what, if anything, does GoPro video color represent? My REU project aims to address these specific questions that the GEMM Lab needs answered. I will compare the Secchi disk measurements to the water color in GoPro video footage, collected at the same time and place, and satellite chlorophyll-a concentrations from MODIS<\/a>. The goal is to understand if there is a relationship between video color and visibility (Secchi disk data), or a relationship between video color and chlorophyll-a concentrations.<\/p>\n\n\n\n Figure 1. Secchi disk deployment (top) Secchi disk (bottom).<\/p>\n<\/div><\/div>\n\n\n\n I am using a programming language called Python to take screenshots of the GoPro footage at certain depths and extract color information. Originally, I extracted RGB values from each pixel and converted them to hex color codes. RGB stands for \u201cred, green, blue\u201d and represents the amount of each color present to achieve the color seen. Hex codes are unique codes for every color and contain six letters or numbers; the first two represent red, the second two represent green, and the final two represent green (Figure 2). However, to relate color to numeric data, I need to quantify the color values into a scale. Hex color codes do not have an obvious scale because they are so distinct and use both letters and numbers. On the other hand, RGB values have a numeric scale from 0\u2013255 for each of the three colors, so we ultimately decided to only use these.<\/p>\n\n\n\n Figure 2. Screenshot from MR17 GoPro video footage on August 23rd, 2021 and the hex color code extraction. The donut plot (left) shows the frequency of each hex code in the center GoPro image, and the table (right) lists the hex codes.<\/p>\n\n\n\n Figure 3. Screenshot from TC6 GoPro video footage on August 12th, 2021 (a) and its RGB color extraction histogram (b).<\/p>\n\n\n\n Every image has millions of pixels, and each pixel has an RGB value. My code separates the red, green, and blue values of each pixel and plots a histogram with the RGB color value on the x-axis and the number of pixels where that value is present on the y-axis (Figure 3). I am currently in the process of determining the best mode of summarizing the color values, whether that be the mean, maximum, or range of values. Once determined, the summarized values will be compared to Secchi disk values and satellite chlorophyll-a concentrations. I still have to iron out the code, but I am proud of what I have done so far and cannot wait for it to all come together!<\/p>\n\n\n\n Along with learning new methods of analysis, I am being challenged to learn new field techniques, such as self-rescue in a tandem kayak (Figure 4). I also have enjoyed performing the data collection that, until now, I have only been watching on my laptop. As this year\u2019s team collects data and reviews GoPro footage, which seems to be showing higher zooplankton abundance than in previous years, I get excited at the prospect of analyzing the data after the field season is complete.<\/p>\n\n\n\n Figure 4. Kayak safety training with the whale team and Marcus from South Coast Tours<\/a>.<\/p>\n\n\n\n At the beginning of the summer, I felt overwhelmed. Yet, I have come to realize that it is okay to not understand something as long as I put in the effort to learn and am not afraid to ask for repeated explanations. I have also learned what it is like to be part of a lab and that lab mates can be a great source of support and knowledge. The GEMM lab is collaborative and members enjoy helping each other brainstorm. I am very thankful that Clara Bird<\/a>, a GEMM lab PhD candidate, provided base code and additional guidance throughout my analysis. Additionally, I attended a lab meeting where many others provided helpful comments and suggestions that were crucial for my project.<\/p>\n\n\n\n My experience as a GEMM lab intern has allowed me to see my REU project through many phases. I have gained confidence in my R and Python programming skills, and confidence in my capabilities overall. Living and working at both the HMSC and Port Orford field stations has exposed me to a multitude of areas in marine science, from GEMM lab research on foraging behavior or acoustics, to other REU students\u2019 and mentors\u2019 research on seabird behavior or plankton ecology. Although there is still a month left of my internship, I have already affirmed my interest in marine biology, and hands-on exploration, and have a greater sense of what I may want to do in graduate school.<\/p>\n\n\n\n Did you enjoy this blog? Want to learn more about marine life, research and conservation? Subscribe to our blog and get weekly updates and more! Just add your name into the subscribe box below!<\/em><\/p>\n\n\n![](\"https:\/\/lh5.googleusercontent.com\/hRrtnNgce-4DaqCEDeWbi90yh_An3BHdfppKM5fbsaB8lzgW3MHgKOjATAmAN85ZqL7wO-H2tq8-08aUFUAMPTk6_5QLVze7-ZIFrcGWpli1XQy1PEEPNdv8NainplMKqsczk5XrhWA0\")
![\"\"](\"https:\/\/osu-wams-blogs-uploads.s3.amazonaws.com\/blogs.dir\/2115\/files\/2022\/08\/Screen-Shot-2022-08-01-at-4.31.49-PM-2.png\")
<\/a><\/figure><\/div>\n\n\n\n![](\"https:\/\/lh6.googleusercontent.com\/S9UBfgwL6bqE6EhV52-O9lyO_SAIurgbKMiVGLOePy8YWqele9h48JBfUxL_IdzTYjtd848BlgE2l-d1ES3EzwtedEZKNYJgaIUy_tRJ73ZfwsYzY3ZrfO6Lw2xr4xZ13DXzzxSa8D6xhDffDbvlgQ\")
![](\"https:\/\/lh3.googleusercontent.com\/9_P3JDsRYzMhkHdbQ82ldHjQXlVDXYLwCu_bStetpDtsCJX8amued6GGLjG6STashmvucSi-r9HyksquDC57t_VxIjuOY7u0KbGAGGJmeJ-1H4zYz5HzPiaug6EHNUGI3LzQ67sHtvJb\")
![](\"https:\/\/lh3.googleusercontent.com\/2lrLp_Nlno1AnyRQCWCR2a1rX_kvOJz3V7O8VGc2x-rir4_brCGLXC0dyQgVtJHwqlI2-4Ex1rG5m0YAD2h2gpHH2qARn_TnC8zeuI4w_VvKqSAgdZmRtVnoMH9jajivsaFi0yvkuEUTe-huhmmT4A\")