2008 Daily log




hey everyone this is my daily log for all the stuff I did this summer and to keep track of it all. Have fun browsing though it :)


June 2nd

Today was the first day on the job. I started learning the basics about IDL. Also I learned about the basics of my project for the summer. I am essentially taking different active regions of the sun. These regions are mainly areas where the sunspots are because of the fact that sunspots are the areas where the magnetic field breaks though the surface of the sun.  Dr. Longcope (my research advisor) gave me a task to help me learn IDL better. The task is to take a sunspot and using a couple different programs, find the perimeter of the different pieces of the sunspot. there are several things I need to focus on: find a way to focus on only that one region and ignore the others, and then find a way to make a table for the pixels in x and y coordiantes and finally use a method to sum up the perimeter.

June 3rd

To start off the day we went in and did a lot more with IDL to further lurn it. mainly we looked at loops and creating programs. After that I went back to work on the perimeter program. I started by looking around on the internet and also the help box to see if there was anything that looked remotely like it could help me. found a couple things but they were so complicated i couldnt understand each line so i just ingored those. I did find a way to convert pixels to x and y coordinates though.

June 4th

Today after another IDL session, i sat down and wrote down on paper the steps of what i did know and outline what i thought was needed and where. Next i started playing some more trying to get it to work. After many many many screw ups I think im starting to get into trouble. Going to have to talk to Dr. Longcope

June 5th


Today, after another IDL learning session, I met with Dr. Longcope for some major needed help. We sat down and with his help and pointing me in the right direction. I was able to figure out. The only thing is though I could only get it to work by typing everything in one line at a time. still have not gotten the full program to work.

June 6th

Today I presented my perimeter program and found out my program was almost complete, I just needed to add a couple of small things to it. I did not get a chance to add a loop to it so i could find out the perimeter of each of the regions. If i get time later this sumer I will try and add that. Heres a link to the program:

perimeter program


I started the new project today. i am going to take different magnetograms of active regions and make movies of the out of them so I can later on use them to gather data. To do this im starting with a program that Dr. Longcope gave me called ars_at_age. I am using this for the years between 2000 and 2007 and looking for regions that are crossing the 15 longitude line that are only one day old. After that I will be using a program called emerge_track, that Dr. Longcope also gave me, for the dates that look good. What I will consider good is regions that are not too big and there is only one sunspot there and not multiple ones. The emerge_track program will allow me to use the mouse to outline the region of interest. Once that is done i will use a program called ear_movie to compile all the images and form a movie of them. So far today i was able to get though august of 2000.

June 9th

I kept going with the movie project. I was able to get up through September of 2001

June 10th

I kept going with the movie project.  I was able to get though 2002.

June 11th

I kept going with the movie project. I was able to get through 2004.

June 12th

I kept going with the movie project. I was able to get through mid 2005.

June 13th

I finished the rest of the movies today. I met with Dr. Longcope to talk about next phaze of the project. He gave me a program called bp_emerge. I will use this program and apply it to all of the movies I have made. What this will do is go though the region that i outline and find the centroid of the both the positive and the negative region of the magnetogram. Then I will use another program called ebp_plot to see how the the positive and negative flux changes.  Also I will be using ebp_traj to see how the positive and negative regions move.

movie of sunspot number 9872


movie of sunspot number 10013


June 16th

I went though the movies and checked if there were any mistakes and fixed the ones that did. Next i started to do all the bp_emerges on all the files.

movie of sunspot 9872 with the center of the positive and negative regions marked

movie of sunspot 10013 with the center of the positive and negative regions marked

Next I did the ebp_plot

plot of sunspot 9872 of its flux and weighted average

plot of sunspot 10013 of its flux and weighted average

from those ebp_plots we see that from the top graph calulates the flux in the positive and negative regions as a function of time. From the bottom graph we see the weighted average. This means it finds the center of the flux and then finds distance.

Also I used ebp_traj.pro. This revealed that the leading active region, rather it be positive or negative, moved away faster than the following region.

plot of sunspot 9872 and how the positive and negative regions move againist one another

plot of sunspot 10013 and how the positive and negative regions move againist one another


June 17th

Today Dr. Longcope gave me another assignment. He wants me to create another program that uses the same idea behind ebp_plot.  The idea behind this is that the origional plot that from all the origional plots we got some of the data is after it has already emerged so the change in the positive and negative regions flat out and some of them done start right off the bat. Some of them are also terrible plots in the first place so we want to just ingore those. So what we want to do is to create a program that will go through all the bp plots and allow us to keep the ones we want. Also it has to allow us to be able to click on the ones we want to keep and record the initial and final time. Then finally it has to list this information into a text file that will list in columns the name of the file, the initial time, and the final time.

So i started working on it. Was able to get a cursor type function up and running that will return the times. Also found that a function called file search will allow us to do a while loop to go though each of the different bp files. Still need to find a way to skip the unwanted graphs and also to get everything to print to a txt file.

June 18th

today i kept working on my new plot program that I was working on yesterday. I am right now having trouble getting the way to skip the unwanted graphs and the fit line. Meeting with Dr. Longcope, he sugguested that i use a cursor button that will just let me click left to keep and right to skip. Also he told me to look up the open and printf commands for new text. After some research and playing around I was able to get the program to work. He also told me to scrap the fit line and not worry about it because it is not relavant to what we are doing. heres a link to the program:

bp_loop.pro

This is the data we found. the first column tells us which program it is that we are using, the 2nd column tells us the initial time and the 3rd column tells us the final time.

bp_data.txt

June 19th

  Today Dr. Longcope gave me a new project to work on. What we are going to do is take the data we got and compiled into the file bp_data.txt, and use it to find different components. This includes the name of the spot, its latitude, the velocity of seperation of the positive and negative regions, the velocity of the positive region, the velocity of the negative region, the change in flux of the positive region, the time intercept of the positive region, the change in flux of the negative region, the time intercept of the negative region, the initial angle, and the final angle.

So far today i got the outline of the program set up. Also i got the program to read the file correctly and got the change in the positive flux and the positive time intercept up and running and used that to get the change in the negative flux and the positive time intercept up and running.

June 20th

Today i worked on the create_data program some more. I got the region number to print out right. also got the positive and negative regions velocities to work but i need to check the units to make sure it is correct. also realized there was a problem with the time intercepts and fixed that. After some tweaking i was able to get the seperation velocity to work and also the angles to work. Also to note for the angles i put the positive region at the orgion and then used the negative region to create a vector connecting the 2. I was able to get the full program to work. Still need to add a header to the columns and figure out how to convert everything over to a structure.

**Note**
I will put the create_data program and its column names on down farther on the log because there has been several changes to it since this entry.

June 23th

I started the day by fine tuning create_data.pro and making sure everything worked properly like i wanted too. Then I am now going to start taking each column and create plots comparing everything to each other and also coorelate each of them. Also I decided to create plots for the angle difference and the absolute angle difference to see if anything with that coorelates.

When I say coorelate I will be using the Spearman and Kendell coorelations. This will allow me to test the strength of the relationship between the 2 numbers. The smaller the number i get back the better. I started to do this by proving Joy's law that says the latitude will affect where the angle is. So i created a program for latitude and the final angle and am using it as a template to create everything else by just changing which column X and Y is set as. A good coorelation will be considered anything below 0.02 and the 2 variables must not contain anything that makes one a function of another, for example the difference between both initail and final angle vs the initial angle.

Heres a link to the latitude vs the final angle program I created to give an idea of what i did for each program.

lat_vs_anglef.pro

June 24th

couldnt do anything today because the servers were down.

June 25th

I started today by making more programs that will allow me to plot and to coorelate. I just took the lat_vs_anglef program and kept changing the X and Y. I was able to get though all of the programs up through the velocity of the negative region.

June 26th

Today i continued making more plot programs that compare the columns. I was able to get all the programs i been creating done for all the columns.

**Note will post results further down because of some changes that had to be made to create_data**

I met with Dr. Longcope today also. While going through the data he found a mistake with the create_data program where i only looked at the x coordinate and not the magnitude for the velocities of both the positive and negative regions and also both of the time intercepts. Also we decided to add a few more columns. since the time intercepts dont tell us much we decided to use that to find the change in time from the beginning and the end where the beginning number will be the max of the 2 time intercepts. another column we decided to at is the average of the absolute value of the positive and negative flux. also we created a column for the change in distance, and using the change in and the change in fluxes we made columns for the positive flux and the negative flux.

June 27th

Today i started by going through and finish fixing all the problems with the velcities of the positive and negative regions and also the time intercepts and rerunning those plots. Then i went through and added all of the new columns. After that i couldnt get much else done because my computer started to act up.

June 30th

After spending half the day trying to fix computer problems, I started to create all the new programs so i could run and compare all the new columns. Only got a few of them done.

July 1st

Finished running all of the programs for the new columns. After meeting with Dr. Longcope, we decided to add another column for dstar which is beginning displacement at the max of the 2 time intercepts for the fluxes. He also found a problem with the angles. I set the equation up in the wrong order so I need to go back and change that. He also wanted me to use the fact from the ebp_traj plots to find a rough estimate for the average velocity of seperation, the positive region, and the negative region. Also he wanted me to sit down and start thinking about why all of this is coming out the way it is.  Finished the day by fixing the create_data.pro

heres a link to the create_data program

create_data.pro

**note will put up a link to the data further down**

July 7th

Started the day by rerunning all the angles programs. Then i created programs for everything related to dstar. Looking back at my data i found the following plots coorilated based upon the standards i set earlier.
Program name
Link to graph
Spearman coorelation factor
lat_vs_anglef.pro
lat_vs_anglef graph
3.90216e-7
lat_vs_vpos.pro lat_vs_vpos graph
0.00626265
lat_vs_vneg.pro
lat_vs_vneg graph
0.0304054
lat_vs_anglei.pro
lat_vs_anglei graph
1.12862e-6
vsep_vs_vpos.pro
vsep_vs_vpos graph
0.000693894
vsep_vs_vneg.pro
vsep_vs_vneg graph
9.91860e-6
vsep_vs_absanglediff.pro
vsep_vs_absanglediff graph
0.0147858
vpos_vs_vneg.pro
vpos_vs_vneg graph
0.0124192
vpos_vs_anglei.pro
vpos_vs_anglei graph
0.00334096
vpos_vs_anglef.pro
vpos_vs_anglef graph
0.0102583
vneg_vs_anglei.pro
vneg_vs_anglei graph
0.161266
vneg_vs_anglef.pro
vneg_vs_anglef graph
0.0137599
vsep_vs_deltat.pro
vsep_vs_deltat graph
5.52269e-8
vpos_vs_deltat.pro
vpos_vs_deltat graph
0.0327515
vneg_vs_deltat.pro
vneg_vs_deltat graph
0.000607412

lat_vs_vneg is a little higher than .02 but it still looks like it works.

the only problems with the coorilations is that why didnt vpos_vs_deltat fall in those standards and also negi_vs_dstar so theres a chance those 2 may be a fluke and it would be wise to take them very lightly.

The next thing i did was outline on paper what i thought the program i needed to create to compute the velocities would be.

July 8th

Created the program to compute the velocities for the lead and follow regions and the seperation.

compute_velocity.pro

for average seperation velocity, it gave 179.69 m/s
for average lead velocity, it gave 140.21 m/s
for average follow velocity, it gave 43.0 m/s

This is same idea that we got from the ebp_traj program. If you look back at those graphs it looks like the lead moves away about 3 to 4 times faster than the following region. We get this from this program.

July 9th

Today I am to come up with an idea of what is going on. I did this by just looking at the northern hemisphere for a time. From looking at the different coorilations, the only conclusion I can come up with has to do with the fact the plasma moves faster at latitudes closer to the equator than the ones closer to the poles and also the fact that the sunspots are formed by magnetic fields. When the magnetic field comes out it can be in any random place but since the magnictic field is now weaker than the one below the surface, it wants to move with it so the sunspot moves so it will line up with that. Next the lead region will be moving faster because since it is further down the magnetic field line that goes to the equator and the plasma there is moving faster. The seperation is also affected by this because as the sunspots grow, both the positive and negative regions will have to get larger along with it.

I dont know if this is correct but based upon my knowlegde and what i have seen from the data, this is what im thinking is going on.

Well after meeting with Dr. Longcope, i found out my theory was a good one but not correct. It has more to do with the Coriolis force and the way the magnetic field line comes up. Also we discovered a problem with my dstar and also my compute velocity program. So therefore i need to fix the create_data.pro and the compute_velocity.pro. After looking though create_data.pro we also found the problem that was arising in via_data.txt so that will have to be fixed too.

For next meeting he also wants me to check out the total flux and see what I get and also try and calculate a change in distance for the region and from that try and come up with a rough size.


July 10th


Started off the day by rerunning create_data.pro now that its fixed and got the new via_data.txt. Here it is:

via_data.txt


The columns in this program go as follow:

the spot number
the latitude
the seperation velocity
the positive region velocity
the negative region velocity
the change in positive flux
the time intercept of the positive region
the change in negative flux
the time intercept of the negative region
the initail angle
the final angle
change in time
the average change in flux
the positive flux
the negative flux
dstar


Next I started to fix the compute_velocity program. The problem is that with the lead and the follow i just put the higher velocity in the lead position and the lower velocity in the follow position. The better way to do it though is to look at either the initial or final angle and coorelate it with that. that is because how i got the angle set up it will always tell which one is which by who is over pi/2 or not.

**I put the compute_velocity program up above where the pervious link was.

For my new velocities i got:

for average seperation velocity, it gave 179.69 m/s
for average lead velocity, it gave 126.97 m/s
for average follow velocity, it gave 54.66 m/s


When looking back at the couple different pics that we got from the ebp_traj program, we can still see these numbers are still good. From the two that I put up here earlier in the log, look a little bit off but looking at the rest I didnt put up these numbers look a little off but have to remember theres 54 more of them that we dont see because of the fact that i dont have time to put them all up here.

Next thing I did today was I realized there was a small problem with my dstar calculations. So i went though and fixed that and reran the create_data program to get the new dstar calculations. Then i went though and reran the correlations and found out a lot of them actually did coorelate with dstar:

Program Name
Link to graph
Spearman coorelation factor
vsep_vs _dstar
vsep_vs_dstar graph
0.000692404
vpos_vs_dstar
vpos_vs_dstar graph
0.0175651
dip_vs_dstar
dip_vs_dstar graph
0.00121176
din_vs_dstar
din_vs_dstar graph
0.00128503
deltat_vs_dstar
deltat_vs_dstar graph
0.000960694
avgdi_vs_dstar
avgdi_vs_dstar graph
0.00101390
posi_vs_dstar
posi_vs_dstar graph
3.35068e-7
negi_vs_dstar
negi_vs_dstar graph
1.93877e-9


There is a couple problems i have with this list though. With vpos_vs_dstar, vneg didnt coorelate with dstar so is it just a fluke or something else? Also deltat and dstar have the same variable tstar so of course its going to coorelate so I have to look at that with a grain of salt. Also because of that reason right there I have to also do the same with posi and negi because they were both made from deltat.

The next thing I did today was computed the total flux. I did this by using the equation:

total flux= avgdi* deltat

After i did this i took the median and got an average of 2.17036e13 V*s

Next did the calculation:

deltad=vsep*deltat

This gave me a average of 2.10248e7 meters

I did find that deltad and total flux does coorelate but that could easily be because they both have deltat as part of their calculation.


July 11th

The next calculation I did today is to find the total distance between the positive and negative regions. This is done with the equation:

d=dstar+deltad

For the median d, I found it to be 3.51827e7

When I ploted distance against the total flux i found that in general as the distance increases between the positive and negative regions, the greater the flux. I also got a correlation of 5.02634e-8
Heres a link to the plot:

totalflux_vs_distance.jpg

Now Dr. Longcope gave me a new project to begin. I am going to go back though all the different directories and find out how many different poles there are in the positive and negative region. I will then use this to do other calculations. One of the big things with this new program is the fact that I will have to change bthr and h a lot.

bthr is a value that tells the program to include only the values of the flux over a certain amount. this will help eliminate the noise which is just a bunch of junk.
h is a smearing value that smears the pixels a little bit. this also helps to get rid of some of the noise.

July 12th

Today i worked on creating a program that will only find the number of poles for each individual member of the directory. Its called pole_num1.pro. I was able to get it up and running and did several different test just to see what it will do. This program will help me later on when i create a program that will go through all the directories and let me see what is going on at each point.

(insert a couple pictures of what i am talking about here)

I still need to take this program and incorperate it into a bigger program that will scan all the files in a directories at once.

July 14th

Worked on getting the file scaning program up and running. I named it pole_num.pro. Once I got it up and running, I did a few tests runs. What i found was several glitches that needed to be fixed and im not sure what is wrong with it. The errors is coming from bthr and h because when i change them, I can usually get the program to run eventually. I need to create a command to get around those. So until I can get it fixed i will just ingore the directories that have a problem. Just to see what happens I set bthr=80 and h=2 and start running though directories to see how the number changes. Spent the rest of the day doing this just to see how the positive and negative changes. For most of them the number increased from the beginning to the end. a couple of them looked like they may have decreased a little bit.

July 15th

Today I went though and did more of those plots and got the same results.

I then went back through some of the directories with the pole_num1 program to see why different things failed like it did. What I found is that some of the pictures failed because the bthr would be so high that it only detected a postive region but not a negative region. So what is happening is that is not dectecting anything so it just fails

After talking with Dr. Longcope, we found the problem with the individual programs that were causing it to fail. One of the functions he gave me called phi_entropy.pro used the total for either the postive or negative region for both of them instead of doing individual ones. So I went in and changed that to get rid of that problem

heres a link to the new program I call sphi_entropy.pro

sphi_entropy.pro

Also I put in an if statement into the pole_num.pro to tell it that if it did not dectect a positve or negative region to skip it and go to the next one. This fixed all the problems i was having. Heres a link to both pole_num1.pro and pole_num.pro.

pole_num1.pro

pole_num.pro

like i said before both of these files are set up to only run with a bthr=80,  h=2

I also created another program called finish_pole.pro. What this did is went through and determined the number of times it decreased instead of increased. It also ploted the poles and found the linear slope.

finish_poles.pro

(add a couple links of pictures of the different changes in pole_num)

July 16th

Today I went through and redid the calculations I did before this time finding the slope and the humps. Eventually i got tired of doing this by hand so i went in to create a new program called findhumps1.pro
This program just does it all the directories at once and puts all the data into one big file for me to look at later. Heres a link to it:

findhumps1.pro

heres a link to the data i got for this:

hump_data.txt

I found that the median of the positive region decreases is 20 and its the same for the negative region. When I divided the positive and negative decreases columns by the number of files in that directory, I found that I ended up with a median of about .40 for both. So does that mean the average sunspot growth for number of poles in a sunspot is 60 percent.

After meeting with Dr. Longcope, I found out i should look at lead and follow again instead of pos and neg so I would have to use the position of the x corredinate and the flux to do this by finding essientially the center of flux in the x coordinate. So i had to go back into find humps and start playing with it to fix it to the new point of what i wanted before.

July 17th

Started the day off by changing findhumps program into lead and follow and got it to work properly.

found for median of total of humps:
lead=19.5
follow=20

lead/number of members=0.398
follow/number of members=0.394

Next came up with a way to find midpoint and then divide each set in half. created two seperate programs for this. called fh_findhumps and sh_findhumps. found a problem with the center fo flux in the x coordinate calculations. Where i was calculating it in the program was the problem. changed it to a different spot.

July 18th

Went in and fixed all the rest of the problems i found in all 3 programs. then went and reran it. After meeting with Dr. Longcope, found out i was calculating the wrong thing. Need to find the median number of poles in the first half and the median number of poles in the second half. I also decide to fix it so we have the median number over all.

heres the final programs that I got for all 3 of them:

findhumps.pro

fh_findhumps.pro

sh_findhumps.pro

for bthr=80, h=2, I ended up getting in fh_findhumps:
lead=1.613757
follow=1.71894

for sh_findhumps:
lead=2.62973
medfollow=2.47977

From looking at those numbers it looks like the number of poles in the leading region grows faster than the one in the following.

next i started to change bthr and h to see what would happen. I decided to run them all from bthr=50 to 100 at increments of 5 and h to be either 0, 1, 2, 3.

spent rest of the day working on this

July 21th

kept working on running the 3 programs for different bthr and h


July 22th

kept working on running the 3 programs for different bthr and h

July 23th

kept working on running the 3 programs for different bthr and h.

finshed it today. heres the results I got.


operators
findhumps
fh_findhumps
sh_findhumps
bthr
h
lead
follow
lead
follow
lead
follow
50
0
5.13311
6.05836
3.48219
4.23874
7.24149
7.62029
50
1
3.26716
3.55522
2.31227
2.66881
4.65359
4.47667
50
2
2.25048
2.20922
1.74768
1.83710
2.84558
2.65175
50
3
1.77081
1.63486
1.28253
1.28182
2.08976
1.85093
55
0
4.90973
5.88764
3.25669
4.09585
7.05760
7.45241
55
1
3.24135
3.48980
2.27750
2.63571
4.61540
4.42270
55
2
2.21638
2.15461
1.72439
1.78605
2.82073
2.60713
55
3
1.77580
1.66368
1.32494
1.28843
2.12765
1.90433
60
0
4.66585
5.81543
3.06197
3.99465
6.77649
7.31347
60
1
3.15586
3.42509
2.25143
2.60561
4.50114
4.36129
60
2
2.21830
2.08661
1.68748
1.76267
2.80596
2.60576
60
3
1.77706
1.63307
1.39710
1.28973
2.12071
1.87558
65
0
4.54080
5.53426
3.00329
3.85317
6.53681
7.15372
65
1
3.09521
3.35978
2.13794
2.51785
4.40566
4.29025
65
2
2.19511
2.05573
1.68386
1.75580
2.74274
2.59115
65
3
1.75775
1.62449
1.28848
1.28935
2.06795
1.88350
70
0
4.29292
5.46811
2.95357
3.79788
6.44433
6.79449
70
1
2.95629
3.34659
2.10911
2.48461
4.26416
4.18450
70
2
2.16426
2.05253
1.67694
1.73546
2.72354
2.55713
70
3
1.75432
1.59605
1.34326
1.28889
2.06992
1.87221
75
0
4.24009
5.31635
2.86056
3.70081
6.07342
6.62959
75
1
2.95517
3.23892
2.01771
2.44377
4.17191
4.16541
75
2
2.13000
2.00119
1.66479
1.72736
2.62794
2.50551
75
3
1.75234
1.59175
1.30368
1.28335
2.01508
1.85077
80
0
4.07679
5.19161
2.81869
3.54674
6.01628
6.50718
80
1
2.88140
3.20246
2.00896
2.43043
4.09491
4.08680
80
2
2.08769
1.97964
1.61357
1.71894
2.62973
2.47977
80
3
1.75066
1.57895
1.29246
1.26457
2.05192
1.83476
85
0
3.94599
5.05729
2.75462
3.33510
5.72161
6.43959
85
1
2.83606
3.08406
1.97532
2.34648
4.00904
4.06979
85
2
2.03717
1.97689
1.57752
1.69907
2.60870
2.40805
85
3
1.75929
1.57748
1.27314
1.27187
2.01618
1.84699
90
0
3.94374
4.98632
2.70720
3.32757
5.63879
6.16561
90
1
2.78992
3.05897
1.96742
2.29914
3.95180
4.02035
90
2
1.99339
1.97767
1.53257
1.66498
2.59083
2.39011
90
3
1.74957
1.58900
1.26061
1.22524
2.00915
1.87327
95
0
3.83541
4.72919
2.50276
3.05033
5.58477
6.14842
95
1
2.70041
3.01759
1.95873
2.29113
3.81233
3.89876
95
2
1.97513
1.97843
1.56062
1.66517
2.58202
2.34793
95
3
1.72957
1.58247
1.25317
1.19473
2.00861
1.86873
100
0
3.66102
4.69464
2.48948
3.00057
5.40425
6.03023
100
1
2.65416
2.98497
1.94205
2.23018
3.77315
3.82016
100
2
1.96344
1.97644
1.55709
1.65479
2.57447
2.30762
100
3
1.71683
1.59257
1.24071
1.18266
2.01822
1.87038

finally i created a few functions to fill compute this table so i could compare it. added keywords to all 3 so i could change bthr and h easier

fffindhumps.pro (tells total median)

ffh_findhumps.pro (tells first half median)

fsh_findhumps.pro (tells second half median)

then used a program called create_med_data.pro. set up a for loop to do all of these and put it into one big file.

create_med_data.pro

final outcome:

med_data.txt



So after looking over the med data i went back and did correlations for everything i did before from via_data.txt and did it with the first half for each lead and follow, the second half, and then the difference between the two. what i found is that nothing correlated. This means that the only thing that could possibly be effecting the number of poles is just the passage of time.



Thats it for my daily log. hope u enjoyed