Hi everyone,
This week was about MATLAB, Excel, and getting reading for the Western Flyer.  I ran a 5-Gaussian MATLAB function for pure water at 10 and 20 degrees Celsius with pressures varying from 0-1500psi.  After correcting for the baseline and choosing a high smoothing coefficient, it spits out a little box that has a bunch of useful information (amplitude, Raman shift, standard deviation, area and relative area) from the distribution. I diligently typed these into Excel and made some graphs from that and compared pressure to a bunch of different variables (Raman shift, speed of sound, relative peak area and hydrogen bond index). For the most part everything matched up pretty well to what we were expecting, but there was some funky stuff going on with the hydrogen bond index at 20 degrees C, but we'll worry about that later.  The past two days have been getting the laser ready in its housing, moving it on the Western Flyer, and calibrating it.  This afternoon I'm having lunch and taking a tour of the boat with my mentors, Alia, and Lizzie (she works for the Channel Islands National Marine Sanctuary).  We leave tomorrow at 9:30 and I'm pumped to see the ROV in action!

Hi everyone,
This week my mentor and I have been working together on writing a Matlab algorithm that'll automatically detect dial vertical migration patterns for day and night from the data set. So far, it seems to be working very well for the majority of the data. We had issues with the two datasets being on different depths which we adjusted by interpolation, but still need to adjust one of them for the time. Right now we're working on a function called peak finder that'll be used to find the min and max backscattering from surrounding data. besides that, I have also been working on REU timeline and Research planning guide assignments and having my mentor review them accordingly. Yesterday, I went out to sea with Steve Haddock and his team to deploy LAM and connect it on a MARS node with an ROV, which was pretty awesome!

Enjoy the rest of your week everyone!
Ash

Happy Thursday. I am happy to have my mentor back from travel, and we have been working on our respective machines to get set up with an integrated development environment (a place where we can write code and it will find errors and make it look nice) so that we can be on the same page. I have also been trying to work through some examples of data analysis, but (of course) I keep getting errors and having to troubleshoot. I got a crash course in using GitHub, and have successfully been able to edit a script and push the changes to the internet for all to see. I have found this process to be quite satisfying.

I'm still sort of in the process of getting situated and learning the basics. So far, it has been an interesting transition from being a scientist to being an engineer.

I spent most of the first half of the week obtaining around a total of 18,000 blue whale b calls. Hopefully this will be enough to train and test my CNN. The second half of the week I dedicated a lot to tuning the BLED to detect fin whale calls. It has been much more difficult to detect fins because their calls exist at the frequency where the hydrophone picks up a noise band (we don't know where that noise comes from). I found myself falling into a rabbit hole trying to use adaptive filters to exclude that background noise and tune the BLED (my mentor got me out eventually). I have been reading more literature for the for my project as well. The REU requires us to summarize a few papers and answer some questions about our project to help get us prepared for the paper and presentation. I hope by the end of today I can have at least 5,000 fin calls which should be enough to get me started with the machine learning for next week.

Morning!

This week has been an assortment of lab tasks for a couple different projects I'm working on. I've also continued to try and understand all the bioluminescence mechanisms, and what is happening when I run different assays. It feels like every time I learn something new I have to kind of reframe everything else in my brain so that it all makes sense again. I've run basic bioluminescence assays on our mystery mollusc and pannychia (sea cucumber with some WILD biolum patterns); testing for the presence of different photoproteins or luciferins (light producing molecules). We run different inhibition assays to try and identify the compounds in these organism's bioluminescence, because the assay that effectively inhibits light production will tell us something about the compounds involved. All of these reactions are done in the dark, so there is no extra light throwing our results off. I've gotten really good at pipetting and such without any light.Unfortunately, we don't have any positive results yet, so we're gonna try some new stuff.

The other project I'm sudo-working on is the sequencing/cloning of the vampire squid luciferin. I've never done extensive genetic work, so this is exciting and intimidating at the same time, but we'll see how that goes!

 Happy weekend!

Morning!

This week has been an assortment of lab tasks for a couple different projects I'm working on. I've also continued to try and understand all the bioluminescence mechanisms, and what is happening when I run different assays. It feels like every time I learn something new I have to kind of reframe everything else in my brain so that it all makes sense again. I've run basic bioluminescence assays on our mystery mollusc and pannychia (sea cucumber with some WILD biolum patterns); testing for the presence of different photoproteins or luciferins (light producing molecules). We run different inhibition assays to try and identify the compounds in these organism's bioluminescence, because the assay that effectively inhibits light production will tell us something about the compounds involved. All of these reactions are done in the dark, so there is no extra light throwing our results off. I've gotten really good at pipetting and such without any light.Unfortunately, we don't have any positive results yet, so we're gonna try some new stuff.

The other project I'm sudo-working on is the sequencing/cloning of the vampire squid luciferin. I've never done extensive genetic work, so this is exciting and intimidating at the same time, but we'll see how that goes!

 Happy weekend!

Hello! I have officially changed up my thesis 3 times this week, but hey that's still progress... I decided to just focus on larvae as opposed to all the life stages of groundfish that have pelagic larvae. Looking into all of the life stages proved to be extremely difficult. I want to focus on Macruids for now, but will eventually end up looking at all of the species that are available within the CalCOFI datasets. Yesterday I spent most of the day downloading the 60 datasets available on the CalCOFI website, which took a long time because the datasets were so large. I will start doing data analysis today on the datasets, looking for any interesting patterns and comparing them to climate indices. Stay tuned! Hopefully I won't have to change my thesis again. I am excited to see what comes out of this project! Hope everyone is having a great week.

Hi Everyone!

This week started with heading out on the Rachel Carson with Chad Waluk & Kelly Benoit-Bird. The goal of the cruise was to capture simultaneous acoustic and camera data from the ROV. Luckily I got to see some of the images (along with Emma) while it was live! The points I got to see were at depths of 150 m and 220 m. A few squid happened to be captured on camera and the entirety of it was really interesting! We also saw a ton of whales that day! The rest of the week has been collecting data through VARS and working on a few assignments from the REU. We had to create a Research Planning guide (which included summarizing articles - luckily I had already read a few!) and a project timeline with our mentors. Last week I finished a draft of my introduction and I went over it with my mentor on Tuesday and we decided I'd get started on drafting up my methods and hopefully have an outline by Monday morning and begin analyzing whatever data I have so far to see if that's the direction I would like to go in. That's about it for this week!

Have an awesome weekend everyone! 

Hey everyone!

This week has been a pretty busy one! On Monday (with Victoria) and Tuesday (with Beverly) I went out with Kelly Benoit-Bird and Chad Waluk on the Carson. It was a really fun experience, especially Monday! It was very sunny (almost warm) and we saw A LOT of whales and some dolphins. Got to watch them use the ROV to do live imaging and it was pretty fun-- I saw a lot of squid and krill. On Tuesday, it was much colder and foggy, so we stayed inside most of the day. Unfortunately, I definitely get sea sick when I'm inside so not much work was done that day. I got some good sleep in though, and did have a chance to watch the screen from the common room, although going into the ROV room felt like a not so great idea. Yesterday I met up with Nancy and worked more on my different projects: Tom O'Reilly's autonomous tech, the blog, Benoit-Bird's web profile, and starting some brainstorming for my bigger project rewriting the ESP page for Jim Birch. Lots of stuff is almost done and then I'll be ready to really sink my teeth into the ESP project. Be sure to check out the blog too, which is going "live" Monday or Tuesday, hopefully!!

See you Monday,
Emma

Hi all!

This week has been pretty busy! On Sunday I came to MBARI with my mentor at 4:45am for the first of a series of 24 hour, high resolution sampling. We are sampling from the Chemostats at 5am, 9am, noon, 3pm, 7pm, 9pm, midnight, 3am, and 5am to look at the diel cycle of the algae. The other days I continued with daily sampling and working on finding the trends in our data. I've spent most afternoons reading papers and working on learning how to use R to analyze the data. On Monday we switched the light regime, so today we're doing another round of high resolution sampling to compare how the algae respond. Unfortunately, one of the four Chemostats had to be restarted and the cooling system stopped working for a few hours so those algae aren't very happy now. I'm also working on some photochemical analyses of the samples today. By this time next week "Phase 1" of the Chemostat experiment will hopefully be complete!
Hope everyone has a good weekend!

These are some of the things I've been working on this week!

Tensorflow

• Learned how to use a Tensorflow implementation of Faster R-CNN (Region Convolutional Neural Networks) on standardized datasets
• Made basic visualizer for model predictions
• Started learning how to use Tensorboard
• Realized that entire implementation is basically useless for our purposes and the codebase is basically useless. Scrapped the entire thing

Keras

• Found, started using, and modified a Keras implementation of Faster R-CNN to work with our data.
• Made more scripts to process and visualize data (such as processing interlaced video)
• Made a very small hand annotated dataset to test the Keras implementation
• Got first Faster R-CNN predictions

Random things

• Learned about basic web hosting with apache servers
• Annotated video data to improve our original image classifier

Nathan

Hi everyone,

So I'm finally starting my internship, super excited to be here. For the purposes of this confluence, I have to say I've had a total effective time of ~3 days at work, so this will be short. After doing some paperwork (talking to HR, getting my SSN, safety orientation), I started working on the OA2(located in Ano Nuevo) time-series in MATLAB, and these are the things I have accomplished:

• Clean data (reduce time period to when data is available for all variables, delete NaNs).
• EOF Principal Component Analysis for winds and plots to explore how winds drive other variables.
• Cross-covariance analysis between winds and all other variables (pH,pCO2,Temp,Salinity,O2,Fluor), as well as between some of the variables to investigate how they relate statistically.

I think this is what my project is going to be, but my mentor has been away this week so I have to confirm with him later today.

Enjoy your long weekend,
Diego

Week 3

After measuring the max current of the SeaStar at 3 A and min current at 500 µA, we are moving on from the SeaStar II model. Ideally, the goal is 10 µA. As a side project and mini-learning experience, I have been tinkering with SeeStar's sleep protocol to try to enable/disable when SeaStar is idle. 

Now instead of the Pro Mini as the Arduino, we are using the Feather M0 to try to gain more peripheral availability through pin muxing. A  good factor of applying all the equipment integrated is reading the datasheets first. I have been reading the DS3234 datasheet for the sleep protocol and Feather M0 specs. and assembly for the continuation of the project.

I also learned more soldering applications. I can now apply hooked and straight attachments on header pins. I also learned how to solder resistors, small chips (op-amps?) and even smaller capacitors onto a circuit board!

Lastly, I rode on the Rachel Carson and spent most of my time sleeping to avoid nausea. I think ROVs are the coolest, but my experience has taught me that the remote operation should be done even more remotely. Some day, I would want to be able to control an ROV without being on a ship! I think this shows that there is room for innovation towards communication across the ocean.

Hi All,

Coming to the end of my second week on this project. Last week I said I felt good about the prospect of finishing the Benchtop Depth Simulator by the end of this week. That turned out to be an accurate prediction, I have a full working system that accurately feeds the CPF pressure based on the fluid displaced by the motor and the speed it is trying to achieve. My mentor then tasked me with changing the simulator to have a pressure response calculated from a dynamic model of the system to estimate the change in force based only on fluid displaced by the buoyancy engine. This is seen as an effective way to test the accuracy of our dynamic model. Initially, results from the model were very far off from where the real world data said they should be. So I set up a parameter estimation problem in Simulink to see what we could adjust to make our model results better approximate the real world values. Here I hit a roadblock, the Simulink tool I wanted to use was not licensed by MBARI. I spoke with my mentor about it and he may just buy a license for it. In the meantime I just manually tweaked the model to see how we could change the results. After some analysis, I found a way to introduce a corrective term that so far seems to vastly improve the model.

So the Benchtop Simulator is now essentially done, there are some finishing touches but for now I am moving on. The next project I will be working on is to try and use an acceleration sensor to detect underwater currents hitting the float. Lots to be done here with signal conditioning and hardware interfacing that I'm certain will be a great challenge.

Looking forward to our MBA tour on Monday

-Pierre

Aloha everyone, sorry this post is late with all the excitement going on the Western Flyer I totally forgot!  This week was great.  On Monday I had a meeting with Mary Miller at the San Francisco Exploratorium where I viewed the exhibits that related to plankton and ocean habitats.  It was interesting to speak with her about how to design educational resources that are informal and for an older age group than what I am use to.  On Tuesday I was able to attend the MBARI EARTH spin off workshop about climate change.  I really liked this workshop and the speakers I thought did a great job conveying why it is important to study climate change from a Pacific Islander point of view.  On Wednesday, my mentor and I spoke again about what my project might be and I sent out more emails to get information and help with the HABs resources that are already on the CeNCOOS website.  Of course today I am on the Western Flyer and will be shooting video to do a mini doc about ground truthing and why it is important in science.  Hope you all have a great week/weekend and see you when we return on Tuesday!

Mahalo

Alia

A week of milestones

So in this week of milestones, I forgot my weekly update...

In the world of research, my project is progressing pretty well. I've run all of my PCR and next week will start to clean samples for sequencing. We're also going out on the Carson on Tuesday and are hoping to collect some more pteropods- I'm staying away from Dramamine naturals this time

This week, I spent most of my time at the Stevenson School in Pebble Beach. I sat in on the EARTH workshop and learned a lot! This workshop was focused on culturally appropriate lesson plans about climate change for students in the Pacific Islands. There were speakers from American Samoa, the Marshall Islands, and a scientist from Stanford University. Most of the attendees were affiliated with World Teach- a program where people volunteer to teach in remote areas, like the Pacific Islands.

Now for some non-MBARI related things that were exciting this week- I celebrated my 5 year wedding anniversary and closed on my new house! (Both on Thursday-hence the forgetting my update).