This is my last week in Aubert, so this week I tried to get everything in order before I move out of my office. The start of this week I finished up work on the spectral index site, which came out pretty well. There is also room to add data if need be. Earlier this week I also took a look at the MVCO stress data that was sent to me. I matched it up with dive data and plotted it against time,

The results are similar to what I saw before, unfortunately. I haven’t had a chance to add this into the paper yet, because I will need to learn a little more about the topic before I can explain it in the paper. How was it calculated? Is it stress from waves or currents? Anyway, I can always add that at a later date. I took your offer up on borrowing the computer since I was not able to get matlab onto my computer. Student employees cannot download matlab from the IT site. Also I can now access the server with this computer so I will need it in order to run my website program.

I met with Lisa Taylor on Wednesday and discussed the websites I wanted add. It went well and she gave me the templates for the MISC lab websites. I had to do a bit of work to fit things into the template nicely. However, I was only able to get the HTML template to work. Lisa also sent me a .php template, but I cannot get it to work. I assuming she would rather have me use the .php template so we are currently working on getting that up and running. The link to the spectral index site is,

http://misclab.umeoce.maine.edu/SpecIndex/SpecIndexPage2.htm


On Thursday I met with Steve and he set me up with access to the server where I have uploaded my websites. Once I had access to the server I got working on updating my float program. The directories had to be changed to the location of the folders on the server. I worked hard today and yesterday on making the directories as variables in the script. This will give Lisa a bit more freedom in moving and renaming files. It should be rather simple to update the script if the files are renamed or moved. I realize this program may be a bit hard to work with so I have made every effort to make it as user friendly as possible. I do still need to change it to the .php template when I figure out why that template is not working. It is done for the most part, just a couple tweaks here and there and updating the site will take only a couple key strokes!

Today I uploaded all the folders, files, and m files to the server and ran the program a couple times. I believe I have most of the bugs out. I will run myself for the next week or so, to make sure it functioning right. I reorganized the images to fit into the template, thumbnails I figured were the easiest way, but it doesn’t necessarily have to be thumbnails. I am open to suggestions if anyone has a better idea. I threw together I quick float homepage which I will improve if Lisa decides to keep it. The link to the homepage and my other pages can be found here,


http://misclab.umeoce.maine.edu/FloatUpdates/FloatHome.htm


A finer detail about the program is that in order to run the program, you must network with the server. The server must be mounted and be under a directory on the computer you are running the program from. The default directory is ‘Z’, but just to be safe the script prompts you to enter the directory. This is in case for some reason you cannot mount the server under ‘Z’. The program takes a couple minutes, but it will do everything I’ve promised.

Well, it’s been a good summer. I am interested in working with both of you again in the spring or next summer. I am glad I was able to be a part of the MISC lab, even for such a brief period. I will certainly be in touch. I am going to need help getting my presentation ready in the next couple weeks, so I’m sure we will be seeing each other again soon. I will return the computer as soon as I get the websites squared away.

Adiós

On Monday of this week, I got equipment for the DMC trip ready and packed. I also started to set up the website that will contain all the spectral data we have collected. I have already added most of the new data we collected on Tuesday to the site. I have pictures and names corresponding to the each spectrum. The names are specific as possible, some organisms I have tagged with genus and species, others just genus. Then I have just substrates (e.g. sand, shell, mud). All data on the site is referenced to the 99% reflectance standard. I applied the equitation I derived last week to all the Satalantic radiometer data. It’s looking pretty good so far, come check it out next week sometime. Hopefully next week I can work with Lisa Taylor to get both the float site and the reflectance data site up and running.


Ok, now to the data we collected on Tuesday. The Satalantic data matched up excellent with Mike’s notes! No trouble there. The DiveSpec data does not match up quite as well, the headings and Mike’s notes differ a couple times, but I have being using the headings entered by Emmanuel as the final decider in these situations. With the DiveSpec, six Didemnum spectra were taken,

I Hope these look familiar to you by now, it’s quite obvious when you are looking at Didemnum. Only one seems slightly different than the rest. Emmanuel told me that “D2” was not a prime specimen of Didemnum, however the outlier we see here is not “D2”, just so you know. Wayne and I also did some work in the touch tank with the DiveSpec, these were our results.

The “unknown tunicate” was just that. We took two pictures of it, but they came out blurry. Besides the unknown tunicate everything is significantly different from Didemnum. There is a lot of DiveSpec data, so I am not going to put all the graphs up. I will work on condensing and adding it all to the website next week. With the Satlantic radiometer we collected two Didemnum spectra.

The dashed lines are the original spectra; the solid line is after it had been referenced to the 99% reflectance standard. These look similar to our light data we collected last time we were at DMC. Here is another comparative plot but using the some of the Satlantic data. The numbers next to the name in legend are the 550/500 ratio.

 After I organized all the Didemnum spectra I calculated the 550/500 ratio again.


Dive                       Mean               Stand Dev.

1(DiveSpec)          2.5822              0.6986

2(HyperOCR)       2.2518              0.2441

3(DiveSpec)         1.7056               0.1788

4(HyperOCR)      1.5939               0.1774

Total                     2.1514              0.5034



Dive 1 also includes the Didemnum spectra Wayne and I took in the lab. The first time we went, both the means are above two and the second time we went, both are below two. It is possible that we see a difference between the two trips to DMC because conditions were different. Also perhaps we sampled old growth the first time and new growth on the second trip. Regardless, dive 1 is definitely an outlier. The mean is very high and so is the standard deviation. We may have to reevaluate that data, because some spectra have very low reflectance, perhaps the DiveSpec was not functioning correctly.

We also have a significant amount of fluorescence data. Emmanuel took four fluorescence readings of Didemnum.

There all seem to have a small bump around 550, it is rather small however. I suppose this is good news. Wayne and I also collected fluorescence data in the touch tanks. Here are four the six fluorescence readings we took.

Note that the scale is different for each plot. Just like Didemnum, the sea star shows a similar rise at 550, which is unfortunate. As you can see both macroalgae show a large spike around 680.

I have most of the data we collected organized and named. As you know, next week is my last week of work. I will compile all the data and some of the m files I wrote and pass them on to Wayne. I will try to organize everything so he will know what is what. Hopefully i can wrap everything up next week, I will try my best to get everything done, websites finished and launch, stress data added to my current MVCO data, and all spec data organized and in Wayne's hands. See you next week.

Ok, quite a bit of stuff to share with you this week. Most if it is the data I collected in the lab this week. First off, the paper on the data we collected at DMC is done, but you already knew that. It is going to be exciting after we collect data next week to see if the new data reinforces our 550/500 ratio. So, that brings me to my first topic which is our new light source.


In the lab this week I measured the output of our new light with the radiometer. I placed both the radiometer and the light source in the sink and filled it with water. I propped the spectralon plate up against the far side of the sink. The setup was similar to when Wayne and I and looked at LED lights in the lab, except this time the light and radiometer were submerged. First I logged data for only a minute or so to see what the spectra looked like. I also pointed the LEDs of the DiveSpec and the radiometer at the spectralon plate. However this was not done underwater. The graph below displays spectra from the AquaSun, the DiveSpec, and from Wayne’s LED flashlight.

Interestingly enough, they all appear to be quite different. Next I let the light run for ten minutes in the sink while logging data. This is what the spectra looked like. The numbers on the color bar are in seconds.

Over the ten minute period the brightness of the light had dropped! I assume the brightness of the light will drop to a certain level then remain constant. I looked over the box after I was done and discovered that this is mentioned on the back of the box. On the back of the box there is a small graph of lumens vs. time. It shows that it starts high and drops slightly for the first 10-15 minutes, then levels out. Hopefully this won’t be too big an issue. I figure as long as the shape of the spectra remains relatively constant, we are going to be OK.
In the lab I also measured the blue LEDs of the DiveSpec used during the fluorescence function with the radiometer. I measured them the same way i measured the white LED lights. Just like with the white LED lights this measurement was not done underwater. Below is the spectrum of the blue LEDs.

It appears that Didemnum may fluoresce! The spike at 550 was not produced by the DiveSpec LEDs. This is what I expected, but I wanted to double check.

Last but not least, I measured our spectralon plate with the DiveSpec. After calibrating the DiveSpec with the 99% reflectance standard, I measured each corner and the center of the plate on each side. Using this data I was able to reshape the radiometer data to how it would look, had it been referenced with the 99% standard. I did this by taking the median of the 10 DiveSpec measurements I took and fitting a linear equation to that measurement.

After that I took the existing data and put it into the equation Reflectance = Lw(sample)*(1.07 – 0.00022713*lambda). This shifted the spectra to what they would have looked like had they been referenced to the 99% reflectance standard. The two graphs show how the spectra changed after applying the above equation. The dashed lines are the original spectra.

These new spectra brought the total standard deviation of our 550/500 ratio down by small amount.

Also this week, I added information on float 6810 to my website. Unfortunately the files sent by 6810 contain more NaNs than actual data, but it did report its location. So I added a map to show its trajectory while it was deployed. The website should be up soon. I will need help from Lisa to add the umaine header and navigation bars, but once that’s done it should be good to go.

Anyway, there is game of disc with my name on it, see you next week!