Summary of Analysis Cruise Oceanus 415-2 (EDDIES Tracer 1) 18 Jul 2005 - 5 Aug 2005 Rebecca Dell dell@post.harvard.edu All programs, except where otherwise noted, are saved in the directory Z:\oc415_2\adcp\Rebecca (1) Tracking the Center of the Eddy using ADCP data Method 1: Place eddy center at minima of ADCP speeds Method 2: Using program locate_center.m This program calculates a line perpendicular to each ADCP velocity, then selects and plots the segments of it that correspond to a reasonable rate of eddy rotation (currently assumes 5-7 days for a complete rotation, but this can be changed by altering the variables T_min and T_max, which denote the minimum and maximum rotational period in days). Finally, it asks you to click with your mouse where all the line segments intersect, and gives you the coordinates of the selected point. NOTES: >> locate_center.m can be used in conjunction with plot_os.m by commenting out the first 25 lines. >> You must always specify which ADCP 5-minute time bins you wish to include in the calculation by specifying the values of n1 and n2. These variables can be assigned either in plot_os.m or in locate_center.m. The variable 'limit' (defined in locate_center.m) has the beginning and end points of several of the earlier long and straight survey tracks. >> If you wish, you do not have to use the vertically averaged velocity. You can select some other type of velocity around line 15. (2) Plotting the Injection Pump Status Using program Z:\oc415_2\sled\pump_plot.m This program calls get_spd_on to find the points where the pump was working, then plots the track of the injection tow, with the shade of the track corresponding to the speed of the pump. Note that for injection #3, you need both the data files: Z:\oc415_2\sled\INJ_003_spd.mat Z:\oc415_2\sled\INJ_003_fl.mat but for injection #4, all the information you need is in the _fl file. Additionally, the pump speed for tow 3 is recorded in a variable named spd_2, while for tow 4 it is recorded in p2_spd. You must change this in BOTH pump_plot.m and in get_spd_on.m >> Output plots saved in: Z:\oc415_2\sled\INJ_003_pumpspeed.png Z:\oc415_2\sled\INJ_003_pumpspeed.fig Z:\oc415_2\sled\INJ_004_pumpspeed.png Z:\oc415_2\sled\INJ_004_pumpspeed.fig (3) CTD Survey Profiles This is just an excercise in making pretty pictures. The relevant matlab programs are: loadEW.m ; Plots the East-West CTD survey profile loadEWshallow.m ; Plots upper 200m of above loadNS.m ; Same for North-South survey loadNSshallow.m Note that you'll get a lot of Divide by Zero warnings--don't worry about them. The memory hogging lines are the nested loops in the first 20 lines, which you can comment out after you run the program once. The output plots are saved in: ctd_profiles\EWprofile_05-07-20.fig ctd_profiles\EWprofile_05-07-20.png ctd_profiles\EWprofile_shallow_05-07-20.fig ctd_profiles\EWprofile_shallow_05-07-20.png ctd_profiles\NSprofile_05-07-20.fig ctd_profiles\NSprofile_05-07-20.png ctd_profiles\NSprofile_shallow_05-07-20.fig ctd_profiles\NSprofile_shallow_05-07-20.png (4) SeaHorse Profiles Another exercise in pretty pictures, this time on the profiles produced by Blair Greenan's SeaHorse, so plotted in 3D. The relevant program is: Z:\oc415_2\SeaHorse\seahorse3D.m This draws on data in: Z:\oc415_2\SeaHorse\sh_ctd.mat Note that this program occasionally has trouble printing. You can solve this problem by switching graphics renderer with the command: >> set(gcf, 'renderer', 'zbuffer') The output plots are saved in: Z:\oc415_2\SeaHorse\profile_05-07-20.fig Z:\oc415_2\SeaHorse\profile_05-07-20.png (5) Calculating Rotational Velocity of the Eddy Using program azimuthalv.m This program calculates the azimuthal projection of ADCP velocities. Before running this program, you must run plot_os.m, which provides the velocity input vectors for azimuthalv.m. If you wish, you may alter plot_os to give you velocities at a specified depth or depth- averaged velocities, or anything else you prefer. Within azimuthalv.m, you must specify the beginning and end of the series you wish to calculate by setting the variables n1 & n2 (the same n1 & n2 from plot_os). You must also provide the location of the eddy center by setting the variables lonc & latc, which can be either a scalars or vectors of the same length as the ADCP velocity range. The program will plot the azimuthal velocity as a function of radial distance from the eddy center, and will do a linear best fit of the points. This is the same as assuming solid body rotation, and the slope of the best fit will give you the rotational velocity of the eddy. In the several calculations we performed, we found rotational periods in the range 5-10 days, with an overall average of 6.5 days. The output plots (though lacking the linear best fit) are in: Z:\oc415_2\adcp\Rebecca\azimuthalv.fig ; 19-21 Jul Data Z:\oc415_2\adcp\Rebecca\azimuthalv.png Z:\oc415_2\adcp\Rebecca\azimuthalv2.fig ; 29-31 Jul Data Z:\oc415_2\adcp\Rebecca\azimuthalv2.png (6) Plotting Injection and Sampling Tracks in Rotating Coordinates Using program plot_eddy.m This program will transform locations given in regular coordinates (latitude and longitude) into the eddy frame of reference, and then plot injection and sampling tracks on a polar-coordinate plane. The properly formatted injection and sampling track data are given in the file: Z:\oc415_2\adcp\Rebecca\tow_tracks.mat This program uses the medusa coordinates extracted by the routine load_med.m from the files Z:\oc415_2\sled\med#_pos.mat This program also uses the indices given by the file: Z:\oc415_2\adcp\Rebecca\med_ind.mat These provide a way to match the location of the individual Medusa bottle samples with the track of the entire sampling run. These indices were produced by the routine: Z:\oc415_2\adcp\Rebecca\match_day.m which matches medusa bottles with track locations based on the time. Near the end of plot_eddy (line ~155), you can set the variables named index_ms# to the medusa sample numbers that contained SF_6. A plot showing the output of this file is saved in Z:\oc415_2\adcp\Rebecca\plot_eddy.fig Z:\oc415_2\adcp\Rebecca\plot_eddy.png