Differences in HYSPLIT Results When Using MATLAB

[ppandaa33]

I am using MATLAB to run hyts_std for HYSPLIT. However, when I input the same data, I get a slight difference of about 0.001 in the latitude, longitude, and altitude values compared to the actual HYSPLIT results. For example, here is a part of the control file and the tdump result file generated using HYSPLIT:

21 10 02 08
1
35.830470 128.631560 500.0
-72
0
10000
5
./
oct1618.BIN
G:/hysplit_files/
gdas1.oct21.w1
G:/hysplit_files/
gdas1.oct21.w2
G:/hysplit_files/
gdas1.sep21.w4
G:/hysplit_files/
gdas1.sep21.w5
./
tdump


5 1
GDAS 21 9 22 0 0
GDAS 21 9 29 0 0
GDAS 21 10 1 0 0
GDAS 21 10 8 0 0
NGM 95 10 16 0 0
1 BACKWARD OMEGA
21 10 2 8 35.830 128.632 500.0
1 PRESSURE
1 1 21 10 2 8 0 2 0.0 35.830 128.632 500.0 929.5
1 1 21 10 2 7 0 1 -1.0 35.832 128.537 525.6 924.2
1 1 21 10 2 6 0 0 -2.0 35.865 128.424 589.2 913.2
1 1 21 10 2 5 0 1 -3.0 35.923 128.280 682.6 897.6
1 1 21 10 2 4 0 2 -4.0 36.000 128.084 808.8 875.9
1 1 21 10 2 3 0 3 -5.0 36.099 127.835 999.2 864.4
1 1 21 10 2 2 0 2 -6.0 36.213 127.553 1203.1 857.7
1 1 21 10 2 1 0 1 -7.0 36.334 127.226 1357.4 856.2
1 1 21 10 2 0 0 0 -8.0 36.436 126.856 1448.8 855.3
1 1 21 10 1 23 0 1 -9.0 36.535 126.462 1489.8 850.6
1 1 21 10 1 22 0 2 -10.0 36.617 126.056 1444.8 854.7
1 1 21 10 1 21 0 3 -11.0 36.663 125.640 1318.1 867.0

Now, here are the control and result files produced when running with MATLAB:

21 10 02 08
1
35.830470 128.631560 500.0
-72
0
10000
5
G:/hysplit_files/
gdas1.oct21.w1
G:/hysplit_files/
gdas1.sep21.w5
G:/hysplit_files/
gdas1.sep21.w5
G:/hysplit_files/
gdas1.oct21.w1
G:/hysplit_files/
gdas1.oct21.w1
C:\Users\chanh\Desktop\Matlab\2024\Location_35.8305_128.6316\2021_10\
Backward_dump_21100208

5 1
GDAS 21 9 29 0 0
GDAS 21 9 29 0 0
GDAS 21 10 1 0 0
GDAS 21 10 1 0 0
GDAS 21 10 1 0 0
1 BACKWARD OMEGA
21 10 2 8 35.830 128.632 500.0
1 PRESSURE
1 1 21 10 2 8 0 2 0.0 35.830 128.632 500.0 929.5
1 1 21 10 2 7 0 1 -1.0 35.832 128.538 525.8 924.2
1 1 21 10 2 6 0 0 -2.0 35.865 128.425 589.4 913.2
1 1 21 10 2 5 0 1 -3.0 35.923 128.281 683.3 897.5
1 1 21 10 2 4 0 2 -4.0 36.001 128.084 811.9 875.6
1 1 21 10 2 3 0 3 -5.0 36.099 127.838 998.7 864.4
1 1 21 10 2 2 0 2 -6.0 36.218 127.551 1198.4 858.3
1 1 21 10 2 1 0 1 -7.0 36.338 127.221 1352.1 857.0
1 1 21 10 2 0 0 0 -8.0 36.445 126.848 1448.2 855.4
1 1 21 10 1 23 0 1 -9.0 36.544 126.451 1486.2 851.0
1 1 21 10 1 22 0 2 -10.0 36.625 126.041 1432.8 855.9
1 1 21 10 1 21 0 3 -11.0 36.676 125.625 1314.2 867.4

As you can see, there is a slight difference of about 0.001 which is not much, but it's still there. Is it normal to have such small discrepancies when running with hyts_std compared to using the HYSPLIT application directly? If not, any suggestions on how to resolve this would be greatly appreciated.Could the issue be related to the number of decimal places? Currently, I am inputting latitude and longitude without limiting the decimal places. Could this be causing the problem?

[sonny.zinn]

If HYSPLIT trajectory simulations are done with the same SETUP.CFG, CONTROL, and meteorological data files, trajectory dump files should be identical. If simulations are done with different computers, a small difference is expected. This is because floating-point computations may be handled differently.

Your CONTROL files are not identical. Do you see the difference even when the CONTROL files are identical?