ASD Manual

[ Index | Overview |Core Cruises | Process Cruises | Projects | People | Cruise Dates | Equipment | Methods Manual | Data Tracking | Data Products ]

ASD
Boat Setup

Setup two gray plastic bins on the port side of the boat and secure with bungee cord.

The ASD box will sit on top of these bins.

To attach the RCR and 8° pistol grip:

Remove the black tip protectors from the fiber optic cables
Unscrew the female portion of the two piece cable clamp (gray) on the back of the two accessories and slide it over the appropriate fiber optic cable.
Cable 1 is long cable for 8º cable
Cable 2 is short cable for RCR

Insert the cable into the accessory and wiggle it until you are sure it is fully seated. For the 8º pistol grip, you should be able to see most of the tip of the cable when the 8º foreoptic is removed.
Slide the cable clamp into position and tighten gently until the cable is held firmly

Plug the computer into the extensions cords to power up the system (MAY NEED MORE DETAILS HERE)
Position the ASD box on the plastic bins and bungee cord down.
Open the box, and place the 98% plaque on the black surface on top

Data Acquisition

Turn on ASD first: toggle switch in right rear below the computer, then turn on the computer
Make directory in c:\dual\data\pb## (in windows or dos)
Type cd dual
Type dual 905 # (where # is the calibration number; Currently # = 9)
Before you take data, under the Auto heading, make sure that:

Dark Current before Spectrum is checked (MUST remain so for all spectra)

Optimize it before spectrum is NOT checked (for darks only)

Save spectrum after collection is checked

Under Spectrum Save heading make sure that:

The path name is c:\dual\data\pb##
The base name is pb##_0&. ## is the cruise number (ex: pb84) and & is the station number.
The starting spectrum is 001
The Number of spectra to collect is 001
The Interval between collections is 00:00:00

Under Control # 1 select Adjust Configuration and make sure

The foreoptic reads 8 degree
The integration time reads 544
Spectrum number and dark current number are 5

Under Control # 2 select Adjust Configuration and make sure

The foreoptic reads RCR (NOT AN OPTION IN THIS MENU : EITHER bare tip or 8 degree)
The integration time reads 544
Spectrum number and dark current number are

To take dark measurements, place each of the RCR and 8º pistol grips in their dark bags, close the bags, and set them inside the ASD box holding them firmly closed to prevent any stray light from entering the bags
Press the spacebar to acquire data. The cursor will become an hourglass when it is acquiring data and will return to an arrow when it is finished with the spectrum. Take 5 dark measurements.
Record these measurements on the ASD cast sheet.
Remove the RCR and 8º pistol grip from their dark bags.
Under Auto heading check Optimize IT before spectrum (if you forget to do this, then the spectrum will be processed to quickly)
Check for ship shadow, ship reflectance, fumes from the engine or anything else that will negatively contribute to the spectra and try to minimize it
All measurements must be taken 90º off of the sun and 45º from the horizontal! (the gun is mounted at 45 degree angle, so hold the pole parallel to the ocean surface)
To collect the 98% plaque spectra, hold the 8º pistol grip about 4 inches above the plaque (being careful not to shadow the plaque) with it pointed down at 45º.
To collect the water spectra, extend the 8º pistol grip over the water as far as possible with pointed down at 45º angle (watch for bubbles on the sea surface)
To collect sky spectra , run the 8º pistol grip over so it points 45º upwards
Alternate sky, water plaque and acquire as many spectra as time permits (5 is ideal)
After each spectra, note the type (water, plaque, sky) on the ASD cast sheet

NOTE: The file name being displayed is the data being acquired and saved. When the spectra has been saved, the NEXT filename will automatically appear in the lower left part of the screen. If you look on the screen and see the extension .005 then files .001 through .005 have been saved.

To quit, simply select Quit, turn off the computer, then turn off the ASD

Data Processing

**NOTE: All steps that occur in UNIX should be done on Sargasso**

Copy all files onto a floppy and then into a directory on a PC (this directory must contain the file asd.bat and the ASD utility Portspec which converts binary files to ascii)

After all the files are on the PC, open up a DOS prompt window, change directories to where ever the ASD binary files are, and type edit asd.bat.

Each line in this file should have the format:

copy pb##_01t.* b##a01t.* (notice the p is gone – this prevents files from being rewritten). There should 2 lines like this for each station (‘t’ and ‘r’). If a station is missing you must remove the lines which correspond to it so for example if you only did station one the asd.bat file should look like:

copy pb56_01t.* b56a01t.* (the a denotes ASCII)

copy pb56_01r.* b56a01r.*

for %%f in (b56a*.*) do portspec –a %%f p%%f

del b*

all you will ever need to change in the last two lines is the b56 to b## which corresponds to the appropriate cruise number. Save the changes and exit the edit window. At the prompt simply type asd. All you need to look for is the word ‘Successful’ flashing by on the screen!! When the program is done exit the DOS window.

On a Unix machine make a pb## directory in /home/data65/pb/ASD. Within this directory make a sub directory raw.
FTP the files over to UNIX – the binary files pb##_0*.* go in the /home/data65/pb/ASD/pb##/raw directory (***ftp as binary) and the ASCII files pb##a0*.* go in /home/data65/pb/ASD/pb## (***ftp as ASCII).
After you have ftp’ed the files, open a DOS window again, cd to the directory with the files, and type del pb*.* so you don’t clutter up your PC.
The rest of the processing steps happen in UNIX. Go to the /home/data65/pb/ASD/pb## directory and type sh. You should be where all the pb##a*.* files are and you will see a $ prompt rather than the one you normally see. At the prompt type extractasd and when it prompts you type pb##a (case sensitive). Including the ‘a’ is VERY important because otherwise the program will loop forever. As each file is extracted the new file name will be displayed on the screen:

pb##e0*.* (the e signifies extracted). These are the files that will be used from here on out.

Run one r and one t file and check the header

If # signs extent into data field you must edit the script extractasd

Edit: sed ‘1,20... Change "20" to approprate number

Exit bash shell

The next set of steps will need to happen twice for each station – they are generally easiest if you have 2 UNIX windows open – one with matlab running and one in the /home/data65/pb/ASD/pb## directory

In the /home/data65/pb/ASD/pb## directory type:

sfload –a pb##e01t (case sensitive) (you will do this next for the ‘r’ files) – it is always best to start with the ‘t’ files for each station because they are easier to identify scan types

In matlab cd to the above directory (where the pb##e*.* files are)
If you are working on the ‘t’ files and just sfloaded them type firstfov and if you are working on the ‘r’ files type firstrcr – the files matlab is working on will be displayed to the screen

Check to make sure that the number of files loaded into matlab matches the cast sheet.

You will be prompted for ‘bad scan files’ – these are ones that on the cast sheet say no data or something to that sort: while taking the files you put your hand over the sensor, a whale swam under it, etc. The one thing to remember is that this program assumes that the first file you took is numbered .001. A common mistake that is made is that people often start with .000. You MUST know the starting number to do this step (just ls all the pb##e01t*.* files to see). Also the program assumes that the numbers increase systematically, it does not know what to do if you for example deleted file .005 already. If you started with .001 just enter in the bad files to be removed starting with the highest number first (ex. if .005 is bad type 5). If you started with .000 and .005 is bad then you will need to type 6. If you skipped files or deleted some, you will need to figure out which files correspond to which numbers – essentially it is one big matrix of data and you are telling it which column to delete. After you delete a file the program will display the remaining file numbers to the screen. (the first number displayed is a place holder 0, which corresponds to wavelengths BUT, if there are 2 zeros then you know that your first file was .000), otherwise if there is only one you are OK. Double check and make sure that you removed the scan you intended to. If you screw-up ctrl^c, clear, and start again, otherwise just press return when you are done.
The next section prompts you to identify the file types based on the information on the cast sheet and the integration times. Pay attention here –the cast sheets are always screwy so look at the integration times too.

Water is always up in the 1000’s to 2000’s,
Sky is usually around 136 or 272
Plaque is around 34 or so.
If there is a change from the cast sheet write it down because when you process the ‘r’ files the integration times are all the same. You get a sense of the values as you go along although they vary cruise to cruise. Also, the first 5 scans are darks – you will know that the files are no longer darks when the integration time finally changes.

The next section asks you about calibration files and until the instrument returns to ASD again, simply press return to all the ?s. (we’ll talk when it actually goes back). After this, the program will do its thing and spit out a *.dat filename and clear the memory.
In the /home/data65/pb/ASD/pb## directory type sfload –a pb##e01r this time and repeat the process but with firstrcr. Repeat this process for all stations of the cruise.

Part 2 – go to the /home/data65/pb/ASD/pb## directory and do a ls *.dat – make sure that there are two files for each station (an ‘r’ file and a ‘t’ file)

In matlab, you will run this program ONCE for each station of the cruise – simply type second.

It will prompt you for the radiance file (pb##_01t) and the irradiance files (pb##_01r) – enter these WITHOUT the *.dat extension.
Next the program will double check that you have the correct line indicating file types (should be all 1,2,4, and 5’s). (you shouldn’t have to worry about this changing until the calibration changes) Press return to go on.
This step is very subjective and a little tricky – here you are asked to pick out ‘outlier’ spectra. All the spectra, with means and +/- standard deviations, are plotted. I get rid of ones that are ‘obvious’, outside the 2 standard deviation limits, and also ones that look…weird. Sometimes because of foam or exhaust from the boat certain sections of the spectra will be smeared out – this mainly occurs in the 450-550 range but you REALLY have to watch for it. What appears minor now will produce a huge dip in the Rrs spectra in this range (you’ll know it when you see it). If this happens re-run second and find the offending spectra and get rid of it!! Press return when there are no more outliers to remove.
Next, the program will display the various determinations of Rrs. The Rrs spectra will have a ton of variability in them because of differing amounts of sky but we deal with that later so don’t worry about it – just press return.
The program will ask you to verify that 750 nm is the beginning of the range that we normalize over (just press return – it shouldn’t change until ASD sees the instrument again).
Repeat this process for every station!

After all the stations have been processed go the /home/data65/pb/ASD/mfiles/pb## window and type ls *.mat to verify that there is an anpb##_0#.mat files for each station!

In matlab type close to get rid of any figures
In the UNIX window type sfload –a anpb##

Go to /home/data65/pb/ASD and xedit rsrcomb.m – this files simply brings all the data together as one big file but it is set to deal with 7 files. If there are less than 7 stations you will need to comment out the lines for plotting and writing text for the non-existent station(s) (Ed and 98% parts). Save the changes and exit.
In matlab type rsrcomb – this will bring in all the data, plot it, and save it as pb##rsr.mat.
Clear matlab and load in the pb##rsr.mat file:

Make a st variable (cruise.station) which corresponds to the anpb##_0#.mat files – station one is always loaded in first so for example pb56 would be

st=[56.1 56.2 56.3 56.4 56.5];

(there were no stations 6 or 7)

NEXT type:

arsr98wst=[st,

arsr98(4:229,:)];

it is VERY important that you do it exactly like this – I have many programs which look specifically for these variables ( the final dimensions of arsr98wst should be 227 x the number of stations)

Re-save the pb##rsr.mat file

Exit matlab.

If the pb##rsr.mat files is not ALL lowercase letters move it so it is and copy it to /home/data65/pb/ASD/rsrs and you are done J