Mapping!

This week has been a refreshing change of pace. I have enjoyed organizing the float data and thinking of creative ways to display it. I started this week by downloading some mapping software for matlab. It took me a couple hours to learn the ins and outs of the program. I am able to make some nice graphs using this software. These graphs can display the trajectory and current location of the floats. The software has a number of different style maps and a built in coastline generator. Here are a few Lambert maps I made displaying information on float trajectory.

I think information like this for each float would look nice on the website. This would be included along with graphs, tables, and raw data for people to download. I also mapped all the floats on a stereographic map.

 

The way I am envisioning it, this will be on the float homepage and each float (red dot) will be click-able, bringing the user to information about that float. I made the dots red on purpose, I will explain this later. I have started learning html in order to make the websites. I downloaded Microsoft Frontpage express, but have not used it much at all. I’m finding html not that difficult and it’s probably more beneficial to learn code rather than use Frontpage. So far it is going well. The commands are pretty strait forward, and as long I have a list of commands I can do pretty much whatever I need to.

After reviewing the University of Washington float page I got some ideas. It would be best is to have the website auto update. The University of Washington page updates itself with no human intervention. I believe I can make it so our float page requires no manual updating as well.

The first problem I ran into is much of html code is repetitive. Putting large amounts of data into a table using html requires you to type LOTS of code. So I used matlab to write code for me. This is the easiest way to make auto updating code. I can insert variables into a repetitive string, and then export it into a .htm file. This is good, because a matlab script can be used to write text to the .htm file that the site reads from, thus updating the information.

I spent a lot of time this week setting up the stereographic map above to update automatically. The way I have the map set up now, each red dot is a click-able region that opens a web page containing float information. However, when the float changes location, the map must update. Having the map itself update was quite simple, but in the html text the click-able region must be identified using pixels. So when the float moves location, the click-able regain must move as well. This meant correlating pixels with latitude and longitude. The way I solved this was to generate the map in matlab using the mapping software I downloaded, then exporting it out of matlab. Then I read it back into matlab as a grayscale image, and then used the intensity values to find the red dots. Red dots made this easy, because they are the only red pixels on the map. Finally, I wrote a string in matlab containing the html text and inserted the pixel locations into the string. Then I exported the string into an .htm file.

I don’t know how I am going to work everything out yet, but I got a pretty good start this week. I really like the idea of the website self updating. This may not be totally necessary, but I want to set it up this way. I believe the float information is sent to the ftp site that Emmanuel sent to me. I am banking on this, because my plan is to pull the data off that site (or any ftp site), then run some scripts and spit out html. However, I haven’t seen any updated information on the site since it was sent to me. Looking at past data, it appears that data sent almost daily. Also the more recent files from the float in the Atlantic contain no data.

Last but not least, I started writing up a report on the Didemnum spectra we took down at DMC in June. I didn’t spend too much time on this, but I got a start on it. One thing that would be good to include is what kind of dive light was used for the illuminated spectra. I’m guesses it was a halogen light but, I will have to double check. That about sums it up…