Hello!

This week I closed the book on my analysis of the NOAA West Coast Ocean Acidification cruises for possible organic alkalinity inputs. Adhering to rather conservative uncertainties related to pH, total alkalinity, and DIC measurements and a 95% confidence interval, less than 10 stations per cruise year showed evidence of statistically significant organic alkalinity presence. These stations were also clustered geographically close to one another, and generally showed median organic alkalinity values less than 5 umol/kg seawater. These low median values as well as the station clusters indicate that what could be proposed as "organic alkalinity" input, is more likely something like instrument/analyst error. In any case, the "no result" result is actually important because I am now looking into doing the same analysis on MBARI's Biological Oceanography Group (BOG) dataset, which includes more estuarine and bay environments where organic alkalinity theoretically would be a more prevalent issue. I'm also learning some instrumentation in order to start processing my own samples from Elkhorn Slough and get even more data on which to do this organic alkalinity analysis.

Hello everyone!

It's been a busy week for us in the wet sea lab. We are trying to get the abalone experiment up and running by the beginning of next week so that means prepping all the conditioning tanks and tagging the animals. Yesterday we got a new shipment of abalone which contained different size classes (20-30mm, 30-40mm, etc.) that we are going to use for the experiment. Handling the juvenile abalone has been really interesting and I've learned that the bigger ones make it harder for us to tag them because they don't sit still! Today, we are going to finish tagging them up and help to prep other aspects of the experiment.

Hi all,

This past week, my work has slowed down on the FathomNet project, and I have started working on the Mesobot project. As for FathomNet, I've used my application to download frame grabs for the top 20 midwater and 20 benthic genera from VARS, which has turned up over 77 thousand images. I'm in the process of uploading those images to an MBARI FTP server.

The Mesobot project is an AUV project headed by the Wood's Hole Oceanographic Institution (WHOI) aiming to use a stereo camera setup to track mesopelagic processes. Right now, I'm working on writing code to calibrate the stereo setup through a computer vision library called OpenCV and testing it through a stereo camera setup that I've built.

Hi everyone!
This week I decided on a new method for estimating coral area that will be more accurate than what I was using before. Now I just need to learn to correct measurements for the field of view distortion caused by the camera angle. I spent a good part of this week re-measuring corals on the older tapes with the new method. Next week I'll start comparing footage of the same transects from different points in time, and hopefully output some numbers.
Monday was my first time visiting the Monterey Bay Aquarium, and I was surprised to find that the kelp forest exhibit was my favorite. (Those colors! And leopard sharks!) Speaking of color, talking to science illustrator Kirsten Carlson was interesting. I'm planning to learn more about collaborations between artists and scientists here at MBARI soon.

Hey everyone!

I finally got to start working at MBARI this week, and so far it has been really eventful and fun!
I now understand my project much better, and even started to implement some change-point algorithms into my test code. I have also read many papers in order to familiarize myself with empirical algorithms that are used here at MBARI in order to generate reference data for our collected data at depth, such as LIAR, LINR, and LIPHR.

Besides that, we got to go the Monterey Bay Aquarium on Monday, which was my first time there. I found it really amazing, and seeing a bit about how it looks above the tanks was also interesting.

Generally, I'm extremely excited to be here and work on this project, and my advisors are all really awesome and helpful in orienting me, although I started working late.

I hope everyone had an amazing week!

What's up everyone,

FRIDAY: Finished implementing Gene's feedback on the back-end SBD data processor. Set my computer to automatically kick off the script every 2 mins. Committed my functional code to BitBucket.
MONDAY: Went on a behind-the-scenes tour of the Monterey Bay Aquarium. Reviewed HTML and CSS fundamentals. Explored the use of Dash (by Plotly) as way to create the CPF MCS front end.
TUESDAY: Continued to experiment with Dash. Implemented a home page and navigation bar.
WEDNESDAY: I've finally got something that is starting to resemble a proper website. Added interactive time series plots to the dashboard page. Met with Gene and decided a web-based front-end created with Dash was viable.
THURSDAY: Attended my first bi-weekly lab meeting. Began work on the SBD Command Builder page. After much digging, I figured out how to send email attachments using Python:

smtpObj = smtplib.SMTP('mbarimail.mbari.org', 25)
smtpObj.sendmail(sender, recipients, msg.as_string())

I've been sick all week, and today is my first day back at MBARI since Monday! Sooooo... I don't have much to update you on other than I finally located my samples that I misplaced I'm meeting with my mentor today to discuss shifting my project since we didn't get any (maybe 1?) Chaetognaths on the last cruise. Stay tuned.

Hey everyone,

First week! This week has been dedicated to more exploration of humpback whale song, feature extraction using a python package called pyAudioAnalysis, and investigation into some automated detection methods for isolating humpback song units. We have found that common automated detection methods that have worked for simpler cetacean vocalizations are not effective for isolating humpback song units because of their high variability in time and frequency. Therefore, we are beginning to explore more complex approaches (general power law detector and blob detection).

Automating song unit isolation will likely be our biggest challenge, but once that is done the rest should (hopefully) fall into place. Fingers crossed.

Hey hey! This week I read a few more papers on different methods to analyze song units. With the data I manually labeled in Raven (12/07 song- the cleanest song ), Tom created a 2D plot using t-SNE and from that plot, we compared its clustering to my clustering of distinct units. From this data, it became evident that the plot grouped more than one type of unit into a cluster and it excluded some units, proving that feature selection is key. Also I added unit type (letters of the alphabet) and contour (distinct traceable shape- yes or no) to the selection tables. I listened to more songs and labeled more units to create more data to compare once we figure out a dependent way to automatically isolate units. Furthermore I read up on t-SNE, learning about the tuneable parameter called "perplexity", or how close the points are to one another. This morning I set up a GitHub account, a code hosting platform for version control and collaboration, and completed the Hello World tutorial.

hello hello!

This week has been a whirlwind. I was helping out a NatGeo film crew in the seawater lab so that they could film some of the specimens that were collected during the bluewater dives and cruises on Mon/Tues. I did some work looking through the vampy transcriptome during the day, and then worked until 1am both nights, so I'm a little sleepy!

Yesterday was our first Carson cruise of the summer, and it was a pretty good one! We got some nice samples for the film crew (a couple solmissus jellies), some assorted ctenos for the lab, and we found an Octopoteuthis squid that I may work on this summer to try and identify its bioluminescence enzymes.

I will finish working with the film crew tomorrow, and am excited to get back into the lab next week. Hope everyone's projects are going well! best of luck to you all(:

Hi everyone!

I started this week by finishing up my draft introduction of my paper and clearly wrote out my hypotheses. I got great feedback on my draft from my mentors so I have now moved into working on preparing my data for analysis. I'm preparing my data for analysis through R, which I have a very love/hate relationship with. Hopefully next week I will have my data in analyzable form! My mentor also requested that I present a poster of my work at open house which I'm pretty excited about!

This week has been a nice mix of work in the lab and meeting and networking with interesting people. I have focused my efforts this week on trying different drive mechanisms for my motors, and have tested a quick design I made that tranforms rotating motion from a dc motor to a sinusodial movement via a couple levers. Additionally I got some extra experience sautering and attaching connectors to wires to try out a positioning controller for the DC motor. Apart from that I had a great time on the boat on monday, and achieved one of my goals for the summer (see an ROV in action!) and I got to chat with an AUV engineer from Woods Hole who I was interested in working with in the past, the intern meeting with Kirsten Carlson was also a lot of fun and gave me a new perspective on some parts of my project. Hoping for an efficient week next week!

Hello everyone,

This week has been full of data. On Friday, we worked in a small project in C++ to read data from xmls files and store valuable annotations on vectors and maps. On Monday, my mentor was back from her vacation. YAY!! We looked over the small project and visualize this algorithm in big in order to achieve the tracking system. We prepared our introduction for the final paper and create an outline for the following weeks. The rest of the week, our mentors wanted to run some new commands in our machines, as well create new images in Docker with different updated files and --options. Today, we're implementing our small project to mbarivision in which we will adding extra command to obtain the xml files in ROV footage.

Hi everyone,

I've finally met both of my mentors this week so have really been able to get going on projects. I'm mainly working on a travelling/mobile exhibit about phytoplankton and harmful algal blooms. I ran a successful brainstorm with the team yesterday and I'm now putting that all in a brief before I start producing something. I've also been helping out with organization for a few meetings, listening into IOOS conversations about ocean acidification and data portals and reading a lot about the state of ocean education in the US.

And more importantly, I had an absolutely fantastic day on Monday seeing an ROV working, being out on the research ship and being able to see Atolla jellyfish, siphonophores and all my other favorite 'critters'.

Hi everyone,

This week, we counted ~1300 abalones. It was a tough work, but with working with Laurissa, we got to tag all of them before we got new abalones from the farm. Yesterday, we used bee tags (they stick really great on abalone shells) to tag around 500 abalones in total. The smallest abalones we tagged are smaller than 15mm and the bigger ones were around 50~60mm. Currently, we are preparing for the upwelling experiment that's going to start next week.

Hi everyone!

This week I have been exploring the LIDAR data and running it through the existing adjustment process using the MB-System programs. I had to compile MB-System first of course, and that was an adventure since it is composed of many parts from several different oranizations! I have also been familiarizing myself with the Point Cloud Library, and the different algorithms for point cloud regestration and matching. My sponsor is working on a way for me to get access to the LIDAR data at the right spot in the process so that I can start feature matching, so for now I am practicing with some example data.

Next week I should have the real data from the LIDAR deployments and be working on importing the data into a standard Point Cloud format and back again so that I can use the PCL on it.

Sorry for the late post, but I was on the Carson today and didn't get a chance to post there.

I had an up and down week, but I was productive and I had some great experiences. The project I am working on involves looking at time-based concentration gradients in relatively shallow waters, so I took some time on Friday to drive out to Point Pinos and look at the tide pools to get a better mental image of what goes on in this process. Monday and Tuesday I spent time working on working on learning Solidworks and some coding approaches to solving the equations that I need for my project. Wednesday, I spent more time looking for resources that could help with my project and found a whole class that someone taught that should be very helpful. Today, I was on the Carson, which was awesome. If you haven't gone yet, I highly recommend it.