Hey everyone! This week started off great with kayaking in the slough. I got to buddy up with Karla and we got to see so many otters and seals. Tuesday I got to go to liquid robotics which was awesome! I was able to understand most of what they were saying even though I don't have an engineering background which was nice. Devin was great and gave really good advice on networking and life in general. Yesterday I continued to write up my results and create my plots. I am nearly finished so I have begun outlining my discussion. We had a lab meeting yesterday and laid out our final plans for open house and now I can't wait! See you all Saturday

Well after last week of boasting that my qPCR was going well, the Step One Plus machine broke the very same day - the first time its broken in the 9 years that our lab has had it... Disclaimer: not my fault, honest, the motor that operates the drawer burnt out, so the machine wouldn't open for love nor money. Despite this, my timeline was only shunted back a few days as we were able to get a loaner in on Monday, allowing me to carry on with my plates.

This week's PCR has been going smoothly - I have completed the assays for both the Steelhead trout and the Striped bass, and today I am finishing up my assays for the New Zealand Mud Snail. The latter two species we are looking at the quantity of their DNA as they are invasive species that in different ways outcompete the Steelhead, so we are hoping to see a relationship between the different amounts of eDNA for the species at different sites.

This week has been a really good one, I got a slow start because of kayaking and the tour of liquid robotics, but I really enjoyed both of those activities and really enjoyed talking to Devin about graduate school and how he navigated his career before and after MBARI. Apart from those small distractions, I think I am getting closer to the larvacean movement with the last few tail prototypes I have made, which is very exciting. I am hoping to finish my Matlab script today and finally do some flow measurements with the laser and some kinematics calculations with Matlab next week.

MONDAY: Began my career as a professional kayaker. Continued work on the CPF MCS GUI. Scrapped the tab idea because it's not scalable and wastes too much space. Moved to a dropdown menu instead.
TUESDAY: Toured Liquid Robotics! Wrote a function to reconstruct the complete pressure (and thus depth) vs. time history of a CPF.
WEDNESDAY: Updated my processor.py script to handle empty CTD measurements and duplicate SBDs. Accidentally messed up a GitHub commit and nearly lost the last three days of work when trying to undo it.
THURSDAY: Damage control... Hopefully wrapping up the website today...

It's been a fast week.

Monday after kayaking, I spent some more time working on the eddy covariance data analysis. I needed to make my mentors Matlab code run so I could see what he has already done, so I downloaded that and installed it and then started fixing all the things that didn't work on my machine.

Tuesday, I went on the Liquid Robotics tour and got a sweet hat. After that, I came back and worked on making the data extractor for my code. That has taken longer than I anticipated, largely because there is just so much going on in the original extractor and I need to reorganize the steps somewhat.

Wednesday, it was more of the same.

Enjoy the rest of your week!

I've been consolidating my figures over the past few days and searching for similar song structures within and between songs (data describing song unit types, duration etc) and by the end of the day I hope to have a nicely colored stacked bar graph of all the different unit types in all 10 songs I've labeled.

Yesterday, I met with Kim Fulton-Bennett, MBARI's Public Information Specialist, to discuss composing as I was hoping to compose a short piece, mimicking some of the whale song units on my violin as well as adding my own themes. I haven't had any composing experience so this should be a challenging but fun task!

Today, Tom and I met with Craig Dawe to learn about MARS for Open House this weekend. We also met with our mentor John to go over the sound system which we'll also be working at Open House. The sound system is very cool so be sure and stop by. There's even a humpback whale song!

This week has all been preparing for Open House on Saturday, where I can demonstrate my activities. I've made posters, created my phytoplankton sorting activity, and the 3D prints are in the post. We've also got some great toy plankton of various kinds (thanks George!) so come along for some fascinating phytoplankton facts. 

The director of IOOS is visiting the CeNCOOS office today, so I've also been helping prepare for that. 

See you all on Saturday!

This week has been busy! I got back from the Western Flyer on Sunday night. I spent most of the cruise in the control room annotating the dives in VARS. Highlights include: a pacific sleeper shark, octopuses, whale bones, a dolphin skeleton, some funky sponges, a newly-explored area, and of course beautiful deep sea corals! I've been studying video footage of Sur Ridge for weeks now, and getting to see it in real-time was amazing. The pilots also let me fly Doc Ricketts in the midwater and taught me to use the manipulator. Possibly even more exciting than learning to fly was when a long-held question about white polyps was answered by comparing what we were seeing on the dive with a photo of the same animal in the past. That moment really emphasized to me how special MBARI's capabilities are. Since getting back to dry land and taming the dock rock, I've cleaned up my data set and am now looking at general trends for species depth, habitat type, and related variables.

Hey guys! It's been a slow week, due to some tubes from a couple years ago (that contained the vectors and antibiotic that I need) being lost, and having to re-order some new primers and grow up some new cells with my vector in it. But I have my sequences all PCRed now, and will be confirming that they are the correct size, then ligating and transforming them today, tomorrow, and maybe a little on Saturday (shh). We're also busy packing up for our cruise that leaves on Tuesday, and that requires all hands on deck.

The only other update from this week is that I have been working with Steve on the design for DSBS, which is really exciting! We've been brainstorming which animal to use, the color palette, design style, etc. So look out for that this fall!

Hello all, this week has been exciting.

First, we got to go kayaking on Monday and see a lot of adorable critters, so that was fun. And Tuesday I toured Liquid Robotics in Silicon Valley, and even got an awesome hat (I have always wanted one that is relevant to what I am doing, so yay!). It was really cool to see what they are working on and how they have progressed and grown over time, and it seems like an awesome place to work!

In the remaining time, I have been continuing my MATLAB coding while continuously e-mailing my absent advisors about issues I have found, and how they would like me to proceed. I have made a lot of progress actually, and finished my script that automatically calculates and stores the best correction for every float for both pH and nitrate without breaking!

I also worked on changing my algorithms according to references and examples given by my advisors, and we are still working on correcting that. Since all these calculations are in a single function that is being called by all other scripts requiring it, I can easily change the calculations without breaking my completed script.

So once we find the calculations and algorithm that we are content with, it will be immediately applicable and runnable on all floats. But first we have to find that algorithm. We are making good progress, so I have hope that we will be able to create something that is usable and reliable.

Have a great day, everyone!

Guess what? I only have 2 more weeks here at MBARI! So with that crazy time crunch, this week I've spent a good amount of time working on project synthesis and organization, with some random assignments thrown in. As far as odd jobs go it's been pretty fun - yesterday I became the impromptu "figure artist" for a paper my mentor is submitting on Saturday. I got to spend the morning working in illustrator to fix a figure showing the community makeup and location of phytoplankton around upwelling eddies over a 3D model. (Might even get acknowledged in the paper!) More related to my project - I found a really interesting observation of a recent phenomenon called The Blob, in the BOG Cruises total alkalinity data. Basically, there was a section of measured TA data points that shared temp, salinity, density, and pressure characteristics with the rest of the data set, however when compared against predicted (via the algorithm I developed) values they were noticeably different. After a lot of trial and error I found out that these data points were all collected during a similar time period: 2013-2014, which happens to coincide with The Blob. The Blob is the fun name the science community gave to this massive 'blob' of warm water that persisted along the West Coast from 2013-2015 and is thought to be a contributor to the super long drought in California. It occurred because a large and, keyword, persistent high-pressure system formed over the Pacific and caused the common northwesterly winds that usually promote coastal upwelling to cease. The lack of upwelling in the Monterey Bay region during that time most likely caused the surfacing of the California Undercurrent to persist longer than its usual seasonal cycle, and thus introduce a different water source into the Bay that the BOG data caught! So long story short - I went down a rabbit hole of Blob info to come up with a reason to exclude ~60 data points from my algorithm test dataset. I would like to dedicate this post to Wikipedia & the Adobe/MatLab forums because they have definitely been my saving grace this week!

This week I finally got to run my topic model and visualize the results! I have built a workflow that runs the topic model with configurable parameters and produces a stacked bar plot of topic distributions over documents, and I am comparing the bar plot to a spectrogram of a humpback whale song. The desired result is to see topic probabilities react appropriately to changes in the spectrogram (by "appropriately" I mean certain topics being in high probability in documents containing song units, and others in high probability for documents containing noise). From here on out I will be tuning parameters and cleaning up the upstream workflow to iterate quickly over different parameterizations of the model, different features, and other configurables in the pipeline.

Happy Thursday everyone
It's been a good week working on my chaetognath time sequence. My mentor helped me plot my data in R (the data didn't look great, so I'm running another experiment). I messed it up yesterday and accidentally used C.macrocephala instead of coelenterazine (((( I was really upset. I basically lost a whole day from that. ANYWAY. Today is a new day, and I'm completing the experiment anyway. I've helped pack a little bit for our upcoming WF cruise Monday, and I'm just making sure I have everything I'm going to need. Sorry this is poorly written. I'm currently 3 minutes late, and I'm trying to hurry!!

Hello! (late as usual)

Monday started off with a great tour of the slough (thanks again for setting that up for us!) follows by some more tests of the prototype design I had. Tuesday I got to use some of the saws on the shop floor to create some more prototypes that use a different design. Wednesday I spent actually making a couple more prototypes that implement a two spring design while also figuring out which kinds of springs I wanted to order from McMaster-Carr. Today I finished up the drilling of the current prototype pieces and am waiting for the springs to come in. Simultaneously brainstorming over what kind of connections for our wires we want to have, in order for our jellyfish tag design to be reusable and easy to use! Next week, I have scheduled to learn how to use the mill machine in the community shop room in order to get more precise cuts for the final prototype / early iTAG.

Hey everyone,

This past week as been as much fun as it has been fast. Monday's kayak trip to the slough was a great start to the week, and Tuesday's trip to Liquid Robotics was fascinating.

My work for the past week has consisted of writing code to create visualizations of the FathomNet database. I'm making a web page where one can see the amount of data available for each class available, as well as some sample images and a taxonomic classification. My work with the MesoBot project has slown down a bit as I have been working on this first.

Hello Folks,

Apart from doing daily abalone husbandry routine, I have started to make my Prototype for my respirometry chambers for my main project. The respirometry chambers mimics the one at Hopkins marine Lab. We have the materials for the chambers and we are currently running how gas tight it is.

This week I started out by finishing the "fitness" calculations for the alignment algorithms. The fitness is the sum of the squared distance between points in the two map portions getting aligned, giving me a quantitative way of comparing different alignment methods.

I then refactored everything to use command line arguments for the different algorithm settings, and made it capable of processing entire missions at once. Running the algorithm on all 932 overlapping map points took about 6 hours, and in the end I had a CSV file with the results of each run. I used that data to try and find any correlation between good alignments and map attributes (such as overlap percentage, map size, etc). Hopefully I con use this information to make better default settings, and inform how to better steer the algorithm.

 6 hours is a long time, so I will also look into multi-threading the program for better performance next week.

Hi everyone,

On Monday, I met with my mentors to discuss about this week's plan and worked on setting up my Docker files with Danelle to have it ready to run my code. On Tuesday, I planned with Mayra on how to create our Voting Classifier class by sketching out what we needed from it and managed to display the results of each frame from the video footage. On Wednesday, I met with Danelle Cline to discuss our plan on the Voting Classifier class and made adjustments to other areas of the code together. On Thursday, I continued our work from the day before and went to a meeting with other workers about "Class Activation Map". From there, I spoke with Danelle and went over our plans for the project.

Last week, I wasn't gone for a couple days, but Monday I just catch up with some updates with our project. I had two meetings with mentors that day. We discussed about some docker commands to run and see the results in the classifier classes. We had trouble displaying the results due to missing command in the internal files. Tuesday, I started creating the algorithm for a class which is going to be on charge to vote and classify the appearance of one event between all the frames produced.

Last week, I wasn't gone for a couple days, but Monday I just catch up with some updates with our project. I had two meetings with mentors that day. We discussed about some docker commands to run and see the results in the classifier classes. We had trouble displaying the results due to missing command in the internal files. Tuesday, I started creating the algorithm for a class which is going to be on charge to vote and classify the appearance of one event between all the frames produced.