Tracking Microbioturbation

While I did not spend this week tracking bioturbation at the lab, I have been talking about it with groups that come to tour the lab, since bioturbation is a really important service that nematodes provide in sediments. Previously I have shown some of the pictures from the pilot but I realized that I never showed the entire microbioturbation output photographs, so that's what I'm here to do this week. Thanks to Dr. Solan in the UK, I was able to automatically extract microbioturbation data from the pilot experiment photographs, rather than manually finding each luminophore and measuring its depth. This automated process involves stitching the photographs of each mesocosm together to create a photo of all four sides of the jars. Then I use ImageJ, which is a free image analysis software, to trace the luminophore layer on the top of the sediment. This step sets a baseline for where the luminophores started and from which all measurements will be taken. Finally, I use a filtering process to isolate the luminophores in the photograph by isolating their pigments--think accentuating their color by increasing or decreasing the lighting. After this process is done, the software creates an image which has isolated the luminophores and measures the distance each particle has traveled from its starting position, assuming a vertical transport. While vertical transport isn't always the case with bioturbation, since many animals dig at angles, this method does provide the most accurate measurements of how deep particles have moved in muds, even on the finest scales.

The image here is all the steps stitched together, starting with the initial image of the four sides of the mesocosm (with colors inverted in Photoshop to isolate the luminophores, shown as the red), then the filtration step which isolates luminophores as the blue, and then the generated photograph from the software with the luminophores as red dots on the black background. Additionally, this program creates an Excel spreadsheet that has all the luminophore as x,y coordinates in the final image. I can use these data to find the mean distance a luminophore has moved over the time step, the maximum distance a particle has traveled, and I can use known distance from the original image to convert these pixel measurements to centimeters, millimeters, etc. I purchased brand new luminophores to use for the full experiment and soon I will be testing these with the image analysis methods to make sure that they show up nicely. I'm hoping that these will not require the inverted color step in Photoshop, which will streamline this process even more.

Thanks for attending this week's corner of arts and crafts with Aaron. Stay tuned for next week's blog.

---