| 1 |
+ |
|
| 2 |
|
/*=========================================== |
| 3 |
|
|
| 4 |
< |
The following 14 functions calculate the following spaceweather indices: |
| 4 |
> |
The following functions calculate these spaceweather indices from the vector magnetic field data: |
| 5 |
|
|
| 6 |
|
USFLUX Total unsigned flux in Maxwells |
| 7 |
|
MEANGAM Mean inclination angle, gamma, in degrees |
| 18 |
|
MEANPOT Mean photospheric excess magnetic energy density in ergs per cubic centimeter |
| 19 |
|
TOTPOT Total photospheric magnetic energy density in ergs per cubic centimeter |
| 20 |
|
MEANSHR Mean shear angle (measured using Btotal) in degrees |
| 21 |
+ |
CMASK The total number of pixels that contributed to the calculation of all the indices listed above |
| 22 |
+ |
|
| 23 |
+ |
And these spaceweather indices from the line-of-sight magnetic field data: |
| 24 |
+ |
USFLUXL Total unsigned flux in Maxwells |
| 25 |
+ |
MEANGBL Mean value of the line-of-sight field gradient, in Gauss/Mm |
| 26 |
+ |
CMASKL The total number of pixels that contributed to the calculation of USFLUXL and MEANGBL |
| 27 |
+ |
R_VALUE Karel Schrijver's R parameter |
| 28 |
|
|
| 29 |
|
The indices are calculated on the pixels in which the conf_disambig segment is greater than 70 and |
| 30 |
|
pixels in which the bitmap segment is greater than 30. These ranges are selected because the CCD |
| 94 |
|
|
| 95 |
|
for (i = 0; i < nx; i++) |
| 96 |
|
{ |
| 97 |
< |
for (j = 0; j < ny; j++) |
| 98 |
< |
{ |
| 97 |
> |
for (j = 0; j < ny; j++) |
| 98 |
> |
{ |
| 99 |
|
if ( mask[j * nx + i] < 70 || bitmask[j * nx + i] < 30 ) continue; |
| 100 |
|
if isnan(bz[j * nx + i]) continue; |
| 101 |
|
sum += (fabs(bz[j * nx + i])); |
| 102 |
|
err += bz_err[j * nx + i]*bz_err[j * nx + i]; |
| 103 |
|
count_mask++; |
| 104 |
< |
} |
| 104 |
> |
} |
| 105 |
|
} |
| 106 |
|
|
| 107 |
|
*mean_vf_ptr = sum*cdelt1*cdelt1*(rsun_ref/rsun_obs)*(rsun_ref/rsun_obs)*100.0*100.0; |
| 253 |
|
/*===========================================*/ |
| 254 |
|
/* Example function 5: Derivative of B_Total SQRT( (dBt/dx)^2 + (dBt/dy)^2 ) */ |
| 255 |
|
|
| 256 |
< |
int computeBtotalderivative(float *bt, int *dims, float *mean_derivative_btotal_ptr, int *mask, int *bitmask, float *derx_bt, float *dery_bt, float *bt_err, float *mean_derivative_btotal_err_ptr) |
| 256 |
> |
int computeBtotalderivative(float *bt, int *dims, float *mean_derivative_btotal_ptr, int *mask, int *bitmask, float *derx_bt, float *dery_bt, float *bt_err, float *mean_derivative_btotal_err_ptr, float *err_termAt, float *err_termBt) |
| 257 |
|
{ |
| 258 |
|
|
| 259 |
|
int nx = dims[0]; |
| 270 |
|
/* brute force method of calculating the derivative (no consideration for edges) */ |
| 271 |
|
for (i = 1; i <= nx-2; i++) |
| 272 |
|
{ |
| 273 |
< |
for (j = 0; j <= ny-1; j++) |
| 273 |
> |
for (j = 0; j <= ny-1; j++) |
| 274 |
|
{ |
| 275 |
< |
derx_bt[j * nx + i] = (bt[j * nx + i+1] - bt[j * nx + i-1])*0.5; |
| 275 |
> |
derx_bt[j * nx + i] = (bt[j * nx + i+1] - bt[j * nx + i-1])*0.5; |
| 276 |
> |
err_termAt[j * nx + i] = (((bt[j * nx + (i+1)]-bt[j * nx + (i-1)])*(bt[j * nx + (i+1)]-bt[j * nx + (i-1)])) * (bt_err[j * nx + (i+1)]*bt_err[j * nx + (i+1)] + bt_err[j * nx + (i-1)]*bt_err[j * nx + (i-1)])) ; |
| 277 |
|
} |
| 278 |
|
} |
| 279 |
|
|
| 280 |
|
/* brute force method of calculating the derivative (no consideration for edges) */ |
| 281 |
|
for (i = 0; i <= nx-1; i++) |
| 282 |
|
{ |
| 283 |
< |
for (j = 1; j <= ny-2; j++) |
| 283 |
> |
for (j = 1; j <= ny-2; j++) |
| 284 |
|
{ |
| 285 |
< |
dery_bt[j * nx + i] = (bt[(j+1) * nx + i] - bt[(j-1) * nx + i])*0.5; |
| 285 |
> |
dery_bt[j * nx + i] = (bt[(j+1) * nx + i] - bt[(j-1) * nx + i])*0.5; |
| 286 |
> |
err_termBt[j * nx + i] = (((bt[(j+1) * nx + i]-bt[(j-1) * nx + i])*(bt[(j+1) * nx + i]-bt[(j-1) * nx + i])) * (bt_err[(j+1) * nx + i]*bt_err[(j+1) * nx + i] + bt_err[(j-1) * nx + i]*bt_err[(j-1) * nx + i])) ; |
| 287 |
|
} |
| 288 |
|
} |
| 289 |
|
|
| 290 |
< |
|
| 291 |
< |
/* consider the edges */ |
| 290 |
> |
/* consider the edges for the arrays that contribute to the variable "sum" in the computation below. |
| 291 |
> |
ignore the edges for the error terms as those arrays have been initialized to zero. |
| 292 |
> |
this is okay because the error term will ultimately not include the edge pixels as they are selected out by the mask and bitmask arrays.*/ |
| 293 |
|
i=0; |
| 294 |
|
for (j = 0; j <= ny-1; j++) |
| 295 |
|
{ |
| 313 |
|
{ |
| 314 |
|
dery_bt[j * nx + i] = ( (3*bt[j * nx + i]) + (-4*bt[(j-1) * nx + i]) - (-bt[(j-2) * nx + i]) )*0.5; |
| 315 |
|
} |
| 316 |
< |
|
| 317 |
< |
|
| 316 |
> |
|
| 317 |
> |
// Calculate the sum only |
| 318 |
|
for (i = 1; i <= nx-2; i++) |
| 319 |
|
{ |
| 320 |
|
for (j = 1; j <= ny-2; j++) |
| 330 |
|
if isnan(derx_bt[j * nx + i]) continue; |
| 331 |
|
if isnan(dery_bt[j * nx + i]) continue; |
| 332 |
|
sum += sqrt( derx_bt[j * nx + i]*derx_bt[j * nx + i] + dery_bt[j * nx + i]*dery_bt[j * nx + i] ); /* Units of Gauss */ |
| 333 |
< |
err += (((bt[(j+1) * nx + i]-bt[(j-1) * nx + i])*(bt[(j+1) * nx + i]-bt[(j-1) * nx + i])) * (bt_err[(j+1) * nx + i]*bt_err[(j+1) * nx + i] + bt_err[(j-1) * nx + i]*bt_err[(j-1) * nx + i])) / (16.0*( derx_bt[j * nx + i]*derx_bt[j * nx + i] + dery_bt[j * nx + i]*dery_bt[j * nx + i] ))+ |
| 334 |
< |
(((bt[j * nx + (i+1)]-bt[j * nx + (i-1)])*(bt[j * nx + (i+1)]-bt[j * nx + (i-1)])) * (bt_err[j * nx + (i+1)]*bt_err[j * nx + (i+1)] + bt_err[j * nx + (i-1)]*bt_err[j * nx + (i-1)])) / (16.0*( derx_bt[j * nx + i]*derx_bt[j * nx + i] + dery_bt[j * nx + i]*dery_bt[j * nx + i] )) ; |
| 333 |
> |
err += err_termBt[j * nx + i] / (16.0*( derx_bt[j * nx + i]*derx_bt[j * nx + i] + dery_bt[j * nx + i]*dery_bt[j * nx + i] ))+ |
| 334 |
> |
err_termAt[j * nx + i] / (16.0*( derx_bt[j * nx + i]*derx_bt[j * nx + i] + dery_bt[j * nx + i]*dery_bt[j * nx + i] )) ; |
| 335 |
|
count_mask++; |
| 336 |
|
} |
| 337 |
|
} |
| 348 |
|
/*===========================================*/ |
| 349 |
|
/* Example function 6: Derivative of Bh SQRT( (dBh/dx)^2 + (dBh/dy)^2 ) */ |
| 350 |
|
|
| 351 |
< |
int computeBhderivative(float *bh, float *bh_err, int *dims, float *mean_derivative_bh_ptr, float *mean_derivative_bh_err_ptr, int *mask, int *bitmask, float *derx_bh, float *dery_bh) |
| 351 |
> |
int computeBhderivative(float *bh, float *bh_err, int *dims, float *mean_derivative_bh_ptr, float *mean_derivative_bh_err_ptr, int *mask, int *bitmask, float *derx_bh, float *dery_bh, float *err_termAh, float *err_termBh) |
| 352 |
|
{ |
| 353 |
|
|
| 354 |
|
int nx = dims[0]; |
| 365 |
|
/* brute force method of calculating the derivative (no consideration for edges) */ |
| 366 |
|
for (i = 1; i <= nx-2; i++) |
| 367 |
|
{ |
| 368 |
< |
for (j = 0; j <= ny-1; j++) |
| 368 |
> |
for (j = 0; j <= ny-1; j++) |
| 369 |
|
{ |
| 370 |
< |
derx_bh[j * nx + i] = (bh[j * nx + i+1] - bh[j * nx + i-1])*0.5; |
| 370 |
> |
derx_bh[j * nx + i] = (bh[j * nx + i+1] - bh[j * nx + i-1])*0.5; |
| 371 |
> |
err_termAh[j * nx + i] = (((bh[j * nx + (i+1)]-bh[j * nx + (i-1)])*(bh[j * nx + (i+1)]-bh[j * nx + (i-1)])) * (bh_err[j * nx + (i+1)]*bh_err[j * nx + (i+1)] + bh_err[j * nx + (i-1)]*bh_err[j * nx + (i-1)])); |
| 372 |
|
} |
| 373 |
|
} |
| 374 |
|
|
| 375 |
|
/* brute force method of calculating the derivative (no consideration for edges) */ |
| 376 |
|
for (i = 0; i <= nx-1; i++) |
| 377 |
|
{ |
| 378 |
< |
for (j = 1; j <= ny-2; j++) |
| 379 |
< |
{ |
| 380 |
< |
dery_bh[j * nx + i] = (bh[(j+1) * nx + i] - bh[(j-1) * nx + i])*0.5; |
| 381 |
< |
} |
| 378 |
> |
for (j = 1; j <= ny-2; j++) |
| 379 |
> |
{ |
| 380 |
> |
dery_bh[j * nx + i] = (bh[(j+1) * nx + i] - bh[(j-1) * nx + i])*0.5; |
| 381 |
> |
err_termBh[j * nx + i] = (((bh[ (j+1) * nx + i]-bh[(j-1) * nx + i])*(bh[(j+1) * nx + i]-bh[(j-1) * nx + i])) * (bh_err[(j+1) * nx + i]*bh_err[(j+1) * nx + i] + bh_err[(j-1) * nx + i]*bh_err[(j-1) * nx + i])); |
| 382 |
> |
} |
| 383 |
|
} |
| 384 |
|
|
| 385 |
< |
|
| 386 |
< |
/* consider the edges */ |
| 385 |
> |
/* consider the edges for the arrays that contribute to the variable "sum" in the computation below. |
| 386 |
> |
ignore the edges for the error terms as those arrays have been initialized to zero. |
| 387 |
> |
this is okay because the error term will ultimately not include the edge pixels as they are selected out by the mask and bitmask arrays.*/ |
| 388 |
|
i=0; |
| 389 |
|
for (j = 0; j <= ny-1; j++) |
| 390 |
|
{ |
| 425 |
|
if isnan(derx_bh[j * nx + i]) continue; |
| 426 |
|
if isnan(dery_bh[j * nx + i]) continue; |
| 427 |
|
sum += sqrt( derx_bh[j * nx + i]*derx_bh[j * nx + i] + dery_bh[j * nx + i]*dery_bh[j * nx + i] ); /* Units of Gauss */ |
| 428 |
< |
err += (((bh[(j+1) * nx + i]-bh[(j-1) * nx + i])*(bh[(j+1) * nx + i]-bh[(j-1) * nx + i])) * (bh_err[(j+1) * nx + i]*bh_err[(j+1) * nx + i] + bh_err[(j-1) * nx + i]*bh_err[(j-1) * nx + i])) / (16.0*( derx_bh[j * nx + i]*derx_bh[j * nx + i] + dery_bh[j * nx + i]*dery_bh[j * nx + i] ))+ |
| 429 |
< |
(((bh[j * nx + (i+1)]-bh[j * nx + (i-1)])*(bh[j * nx + (i+1)]-bh[j * nx + (i-1)])) * (bh_err[j * nx + (i+1)]*bh_err[j * nx + (i+1)] + bh_err[j * nx + (i-1)]*bh_err[j * nx + (i-1)])) / (16.0*( derx_bh[j * nx + i]*derx_bh[j * nx + i] + dery_bh[j * nx + i]*dery_bh[j * nx + i] )) ; |
| 428 |
> |
err += err_termBh[j * nx + i] / (16.0*( derx_bh[j * nx + i]*derx_bh[j * nx + i] + dery_bh[j * nx + i]*dery_bh[j * nx + i] ))+ |
| 429 |
> |
err_termAh[j * nx + i] / (16.0*( derx_bh[j * nx + i]*derx_bh[j * nx + i] + dery_bh[j * nx + i]*dery_bh[j * nx + i] )) ; |
| 430 |
|
count_mask++; |
| 431 |
|
} |
| 432 |
|
} |
| 442 |
|
/*===========================================*/ |
| 443 |
|
/* Example function 7: Derivative of B_vertical SQRT( (dBz/dx)^2 + (dBz/dy)^2 ) */ |
| 444 |
|
|
| 445 |
< |
int computeBzderivative(float *bz, float *bz_err, int *dims, float *mean_derivative_bz_ptr, float *mean_derivative_bz_err_ptr, int *mask, int *bitmask, float *derx_bz, float *dery_bz) |
| 445 |
> |
int computeBzderivative(float *bz, float *bz_err, int *dims, float *mean_derivative_bz_ptr, float *mean_derivative_bz_err_ptr, int *mask, int *bitmask, float *derx_bz, float *dery_bz, float *err_termA, float *err_termB) |
| 446 |
|
{ |
| 447 |
|
|
| 448 |
|
int nx = dims[0]; |
| 450 |
|
int i = 0; |
| 451 |
|
int j = 0; |
| 452 |
|
int count_mask = 0; |
| 453 |
< |
double sum = 0.0; |
| 453 |
> |
double sum = 0.0; |
| 454 |
|
double err = 0.0; |
| 455 |
< |
*mean_derivative_bz_ptr = 0.0; |
| 455 |
> |
*mean_derivative_bz_ptr = 0.0; |
| 456 |
|
|
| 457 |
< |
if (nx <= 0 || ny <= 0) return 1; |
| 457 |
> |
if (nx <= 0 || ny <= 0) return 1; |
| 458 |
|
|
| 459 |
|
/* brute force method of calculating the derivative (no consideration for edges) */ |
| 460 |
|
for (i = 1; i <= nx-2; i++) |
| 461 |
|
{ |
| 462 |
< |
for (j = 0; j <= ny-1; j++) |
| 462 |
> |
for (j = 0; j <= ny-1; j++) |
| 463 |
|
{ |
| 464 |
< |
if isnan(bz[j * nx + i]) continue; |
| 465 |
< |
derx_bz[j * nx + i] = (bz[j * nx + i+1] - bz[j * nx + i-1])*0.5; |
| 464 |
> |
derx_bz[j * nx + i] = (bz[j * nx + i+1] - bz[j * nx + i-1])*0.5; |
| 465 |
> |
err_termA[j * nx + i] = (((bz[j * nx + (i+1)]-bz[j * nx + (i-1)])*(bz[j * nx + (i+1)]-bz[j * nx + (i-1)])) * (bz_err[j * nx + (i+1)]*bz_err[j * nx + (i+1)] + bz_err[j * nx + (i-1)]*bz_err[j * nx + (i-1)])); |
| 466 |
|
} |
| 467 |
|
} |
| 468 |
|
|
| 469 |
|
/* brute force method of calculating the derivative (no consideration for edges) */ |
| 470 |
|
for (i = 0; i <= nx-1; i++) |
| 471 |
|
{ |
| 472 |
< |
for (j = 1; j <= ny-2; j++) |
| 472 |
> |
for (j = 1; j <= ny-2; j++) |
| 473 |
|
{ |
| 474 |
< |
if isnan(bz[j * nx + i]) continue; |
| 475 |
< |
dery_bz[j * nx + i] = (bz[(j+1) * nx + i] - bz[(j-1) * nx + i])*0.5; |
| 474 |
> |
dery_bz[j * nx + i] = (bz[(j+1) * nx + i] - bz[(j-1) * nx + i])*0.5; |
| 475 |
> |
err_termB[j * nx + i] = (((bz[(j+1) * nx + i]-bz[(j-1) * nx + i])*(bz[(j+1) * nx + i]-bz[(j-1) * nx + i])) * (bz_err[(j+1) * nx + i]*bz_err[(j+1) * nx + i] + bz_err[(j-1) * nx + i]*bz_err[(j-1) * nx + i])); |
| 476 |
|
} |
| 477 |
|
} |
| 478 |
|
|
| 479 |
< |
|
| 480 |
< |
/* consider the edges */ |
| 479 |
> |
/* consider the edges for the arrays that contribute to the variable "sum" in the computation below. |
| 480 |
> |
ignore the edges for the error terms as those arrays have been initialized to zero. |
| 481 |
> |
this is okay because the error term will ultimately not include the edge pixels as they are selected out by the mask and bitmask arrays.*/ |
| 482 |
|
i=0; |
| 483 |
|
for (j = 0; j <= ny-1; j++) |
| 484 |
|
{ |
| 470 |
– |
if isnan(bz[j * nx + i]) continue; |
| 485 |
|
derx_bz[j * nx + i] = ( (-3*bz[j * nx + i]) + (4*bz[j * nx + (i+1)]) - (bz[j * nx + (i+2)]) )*0.5; |
| 486 |
|
} |
| 487 |
|
|
| 488 |
|
i=nx-1; |
| 489 |
|
for (j = 0; j <= ny-1; j++) |
| 490 |
|
{ |
| 477 |
– |
if isnan(bz[j * nx + i]) continue; |
| 491 |
|
derx_bz[j * nx + i] = ( (3*bz[j * nx + i]) + (-4*bz[j * nx + (i-1)]) - (-bz[j * nx + (i-2)]) )*0.5; |
| 492 |
|
} |
| 493 |
|
|
| 494 |
|
j=0; |
| 495 |
|
for (i = 0; i <= nx-1; i++) |
| 496 |
|
{ |
| 484 |
– |
if isnan(bz[j * nx + i]) continue; |
| 497 |
|
dery_bz[j * nx + i] = ( (-3*bz[j*nx + i]) + (4*bz[(j+1) * nx + i]) - (bz[(j+2) * nx + i]) )*0.5; |
| 498 |
|
} |
| 499 |
|
|
| 500 |
|
j=ny-1; |
| 501 |
|
for (i = 0; i <= nx-1; i++) |
| 502 |
|
{ |
| 491 |
– |
if isnan(bz[j * nx + i]) continue; |
| 503 |
|
dery_bz[j * nx + i] = ( (3*bz[j * nx + i]) + (-4*bz[(j-1) * nx + i]) - (-bz[(j-2) * nx + i]) )*0.5; |
| 504 |
|
} |
| 505 |
|
|
| 518 |
|
if isnan(bz_err[j * nx + i]) continue; |
| 519 |
|
if isnan(derx_bz[j * nx + i]) continue; |
| 520 |
|
if isnan(dery_bz[j * nx + i]) continue; |
| 521 |
< |
sum += sqrt( derx_bz[j * nx + i]*derx_bz[j * nx + i] + dery_bz[j * nx + i]*dery_bz[j * nx + i] ); /* Units of Gauss */ |
| 522 |
< |
err += (((bz[(j+1) * nx + i]-bz[(j-1) * nx + i])*(bz[(j+1) * nx + i]-bz[(j-1) * nx + i])) * (bz_err[(j+1) * nx + i]*bz_err[(j+1) * nx + i] + bz_err[(j-1) * nx + i]*bz_err[(j-1) * nx + i])) / |
| 523 |
< |
(16.0*( derx_bz[j * nx + i]*derx_bz[j * nx + i] + dery_bz[j * nx + i]*dery_bz[j * nx + i] )) + |
| 513 |
< |
(((bz[j * nx + (i+1)]-bz[j * nx + (i-1)])*(bz[j * nx + (i+1)]-bz[j * nx + (i-1)])) * (bz_err[j * nx + (i+1)]*bz_err[j * nx + (i+1)] + bz_err[j * nx + (i-1)]*bz_err[j * nx + (i-1)])) / |
| 514 |
< |
(16.0*( derx_bz[j * nx + i]*derx_bz[j * nx + i] + dery_bz[j * nx + i]*dery_bz[j * nx + i] )) ; |
| 521 |
> |
sum += sqrt( derx_bz[j * nx + i]*derx_bz[j * nx + i] + dery_bz[j * nx + i]*dery_bz[j * nx + i] ); /* Units of Gauss */ |
| 522 |
> |
err += err_termB[j * nx + i] / (16.0*( derx_bz[j * nx + i]*derx_bz[j * nx + i] + dery_bz[j * nx + i]*dery_bz[j * nx + i] )) + |
| 523 |
> |
err_termA[j * nx + i] / (16.0*( derx_bz[j * nx + i]*derx_bz[j * nx + i] + dery_bz[j * nx + i]*dery_bz[j * nx + i] )) ; |
| 524 |
|
count_mask++; |
| 525 |
|
} |
| 526 |
|
} |
| 527 |
|
|
| 528 |
< |
*mean_derivative_bz_ptr = (sum)/(count_mask); // would be divided by ((nx-2)*(ny-2)) if shape of count_mask = shape of magnetogram |
| 528 |
> |
*mean_derivative_bz_ptr = (sum)/(count_mask); // would be divided by ((nx-2)*(ny-2)) if shape of count_mask = shape of magnetogram |
| 529 |
|
*mean_derivative_bz_err_ptr = (sqrt(err))/(count_mask); // error in the quantity (sum)/(count_mask) |
| 530 |
|
//printf("MEANGBZ=%f\n",*mean_derivative_bz_ptr); |
| 531 |
|
//printf("MEANGBZ_err=%f\n",*mean_derivative_bz_err_ptr); |
| 566 |
|
|
| 567 |
|
|
| 568 |
|
// Comment out random number generator, which can only run on solar3 |
| 569 |
< |
//int computeJz(float *bx_err, float *by_err, float *bx, float *by, int *dims, float *jz, float *jz_err, float *jz_err_squared, |
| 569 |
> |
// int computeJz(float *bx_err, float *by_err, float *bx, float *by, int *dims, float *jz, float *jz_err, float *jz_err_squared, |
| 570 |
|
// int *mask, int *bitmask, float cdelt1, double rsun_ref, double rsun_obs,float *derx, float *dery, float *noisebx, |
| 571 |
|
// float *noiseby, float *noisebz) |
| 572 |
|
|
| 573 |
|
int computeJz(float *bx_err, float *by_err, float *bx, float *by, int *dims, float *jz, float *jz_err, float *jz_err_squared, |
| 574 |
< |
int *mask, int *bitmask, float cdelt1, double rsun_ref, double rsun_obs,float *derx, float *dery) |
| 574 |
> |
int *mask, int *bitmask, float cdelt1, double rsun_ref, double rsun_obs,float *derx, float *dery, float *err_term1, float *err_term2) |
| 575 |
|
|
| 576 |
|
|
| 577 |
|
{ |
| 588 |
|
|
| 589 |
|
for (i = 1; i <= nx-2; i++) |
| 590 |
|
{ |
| 591 |
< |
for (j = 0; j <= ny-1; j++) |
| 591 |
> |
for (j = 0; j <= ny-1; j++) |
| 592 |
|
{ |
| 593 |
< |
if isnan(by[j * nx + i]) continue; |
| 594 |
< |
derx[j * nx + i] = (by[j * nx + i+1] - by[j * nx + i-1])*0.5; |
| 593 |
> |
derx[j * nx + i] = (by[j * nx + i+1] - by[j * nx + i-1])*0.5; |
| 594 |
> |
err_term1[j * nx + i] = (by_err[j * nx + i+1])*(by_err[j * nx + i+1]) + (by_err[j * nx + i-1])*(by_err[j * nx + i-1]); |
| 595 |
|
} |
| 596 |
|
} |
| 597 |
|
|
| 598 |
|
for (i = 0; i <= nx-1; i++) |
| 599 |
|
{ |
| 600 |
< |
for (j = 1; j <= ny-2; j++) |
| 600 |
> |
for (j = 1; j <= ny-2; j++) |
| 601 |
|
{ |
| 602 |
< |
if isnan(bx[j * nx + i]) continue; |
| 603 |
< |
dery[j * nx + i] = (bx[(j+1) * nx + i] - bx[(j-1) * nx + i])*0.5; |
| 602 |
> |
dery[j * nx + i] = (bx[(j+1) * nx + i] - bx[(j-1) * nx + i])*0.5; |
| 603 |
> |
err_term2[j * nx + i] = (bx_err[(j+1) * nx + i])*(bx_err[(j+1) * nx + i]) + (bx_err[(j-1) * nx + i])*(bx_err[(j-1) * nx + i]); |
| 604 |
|
} |
| 605 |
|
} |
| 606 |
< |
|
| 607 |
< |
// consider the edges |
| 606 |
> |
|
| 607 |
> |
/* consider the edges for the arrays that contribute to the variable "sum" in the computation below. |
| 608 |
> |
ignore the edges for the error terms as those arrays have been initialized to zero. |
| 609 |
> |
this is okay because the error term will ultimately not include the edge pixels as they are selected out by the mask and bitmask arrays.*/ |
| 610 |
> |
|
| 611 |
|
i=0; |
| 612 |
|
for (j = 0; j <= ny-1; j++) |
| 613 |
|
{ |
| 614 |
< |
if isnan(by[j * nx + i]) continue; |
| 603 |
< |
derx[j * nx + i] = ( (-3*by[j * nx + i]) + (4*by[j * nx + (i+1)]) - (by[j * nx + (i+2)]) )*0.5; |
| 614 |
> |
derx[j * nx + i] = ( (-3*by[j * nx + i]) + (4*by[j * nx + (i+1)]) - (by[j * nx + (i+2)]) )*0.5; |
| 615 |
|
} |
| 616 |
|
|
| 617 |
|
i=nx-1; |
| 618 |
|
for (j = 0; j <= ny-1; j++) |
| 619 |
|
{ |
| 620 |
< |
if isnan(by[j * nx + i]) continue; |
| 610 |
< |
derx[j * nx + i] = ( (3*by[j * nx + i]) + (-4*by[j * nx + (i-1)]) - (-by[j * nx + (i-2)]) )*0.5; |
| 620 |
> |
derx[j * nx + i] = ( (3*by[j * nx + i]) + (-4*by[j * nx + (i-1)]) - (-by[j * nx + (i-2)]) )*0.5; |
| 621 |
|
} |
| 622 |
|
|
| 623 |
|
j=0; |
| 624 |
|
for (i = 0; i <= nx-1; i++) |
| 625 |
|
{ |
| 626 |
< |
if isnan(bx[j * nx + i]) continue; |
| 617 |
< |
dery[j * nx + i] = ( (-3*bx[j*nx + i]) + (4*bx[(j+1) * nx + i]) - (bx[(j+2) * nx + i]) )*0.5; |
| 626 |
> |
dery[j * nx + i] = ( (-3*bx[j*nx + i]) + (4*bx[(j+1) * nx + i]) - (bx[(j+2) * nx + i]) )*0.5; |
| 627 |
|
} |
| 628 |
|
|
| 629 |
|
j=ny-1; |
| 630 |
|
for (i = 0; i <= nx-1; i++) |
| 631 |
|
{ |
| 632 |
< |
if isnan(bx[j * nx + i]) continue; |
| 624 |
< |
dery[j * nx + i] = ( (3*bx[j * nx + i]) + (-4*bx[(j-1) * nx + i]) - (-bx[(j-2) * nx + i]) )*0.5; |
| 632 |
> |
dery[j * nx + i] = ( (3*bx[j * nx + i]) + (-4*bx[(j-1) * nx + i]) - (-bx[(j-2) * nx + i]) )*0.5; |
| 633 |
|
} |
| 634 |
+ |
|
| 635 |
|
|
| 636 |
< |
for (i = 1; i <= nx-2; i++) |
| 636 |
> |
for (i = 0; i <= nx-1; i++) |
| 637 |
|
{ |
| 638 |
< |
for (j = 1; j <= ny-2; j++) |
| 638 |
> |
for (j = 0; j <= ny-1; j++) |
| 639 |
|
{ |
| 640 |
< |
// calculate jz at all points |
| 632 |
< |
|
| 640 |
> |
// calculate jz at all points |
| 641 |
|
jz[j * nx + i] = (derx[j * nx + i]-dery[j * nx + i]); // jz is in units of Gauss/pix |
| 642 |
< |
jz_err[j * nx + i] = 0.5*sqrt( (bx_err[(j+1) * nx + i]*bx_err[(j+1) * nx + i]) + (bx_err[(j-1) * nx + i]*bx_err[(j-1) * nx + i]) + |
| 635 |
< |
(by_err[j * nx + (i+1)]*by_err[j * nx + (i+1)]) + (by_err[j * nx + (i-1)]*by_err[j * nx + (i-1)]) ) ; |
| 642 |
> |
jz_err[j * nx + i] = 0.5*sqrt( err_term1[j * nx + i] + err_term2[j * nx + i] ) ; |
| 643 |
|
jz_err_squared[j * nx + i]= (jz_err[j * nx + i]*jz_err[j * nx + i]); |
| 644 |
< |
count_mask++; |
| 638 |
< |
|
| 644 |
> |
count_mask++; |
| 645 |
|
} |
| 646 |
|
} |
| 647 |
|
return 0; |
| 648 |
|
} |
| 649 |
< |
|
| 649 |
> |
|
| 650 |
|
/*===========================================*/ |
| 651 |
|
|
| 652 |
|
/* Example function 9: Compute quantities on Jz array */ |
| 837 |
|
/* Example function 12: Sum of Absolute Value per polarity */ |
| 838 |
|
|
| 839 |
|
// The Sum of the Absolute Value per polarity is defined as the following: |
| 840 |
< |
// fabs(sum(jz gt 0)) + fabs(sum(jz lt 0)) and the units are in Amperes. |
| 840 |
> |
// fabs(sum(jz gt 0)) + fabs(sum(jz lt 0)) and the units are in Amperes per square arcsecond. |
| 841 |
|
// The units of jz are in G/pix. In this case, we would have the following: |
| 842 |
|
// Jz = (Gauss/pix)(1/CDELT1)(0.00010)(1/MUNAUGHT)(RSUN_REF/RSUN_OBS)(RSUN_REF/RSUN_OBS)(RSUN_OBS/RSUN_REF), |
| 843 |
|
// = (Gauss/pix)(1/CDELT1)(0.00010)(1/MUNAUGHT)(RSUN_REF/RSUN_OBS) |
| 853 |
|
int i=0; |
| 854 |
|
int j=0; |
| 855 |
|
int count_mask=0; |
| 856 |
< |
double sum1=0.0; |
| 856 |
> |
double sum1=0.0; |
| 857 |
|
double sum2=0.0; |
| 858 |
|
double err=0.0; |
| 859 |
< |
*totaljzptr=0.0; |
| 859 |
> |
*totaljzptr=0.0; |
| 860 |
|
|
| 861 |
< |
if (nx <= 0 || ny <= 0) return 1; |
| 861 |
> |
if (nx <= 0 || ny <= 0) return 1; |
| 862 |
|
|
| 863 |
< |
for (i = 0; i < nx; i++) |
| 863 |
> |
for (i = 0; i < nx; i++) |
| 864 |
|
{ |
| 865 |
< |
for (j = 0; j < ny; j++) |
| 865 |
> |
for (j = 0; j < ny; j++) |
| 866 |
|
{ |
| 867 |
|
if ( mask[j * nx + i] < 70 || bitmask[j * nx + i] < 30 ) continue; |
| 868 |
|
if isnan(bz[j * nx + i]) continue; |
| 874 |
|
} |
| 875 |
|
} |
| 876 |
|
|
| 877 |
< |
*totaljzptr = fabs(sum1) + fabs(sum2); /* Units are A */ |
| 877 |
> |
*totaljzptr = fabs(sum1) + fabs(sum2); /* Units are Amperes per arcsecond */ |
| 878 |
|
*totaljz_err_ptr = sqrt(err)*(1/cdelt1)*fabs((0.00010)*(1/MUNAUGHT)*(rsun_ref/rsun_obs)); |
| 879 |
|
//printf("SAVNCPP=%g\n",*totaljzptr); |
| 880 |
|
//printf("SAVNCPP_err=%g\n",*totaljz_err_ptr); |
| 881 |
|
|
| 882 |
< |
return 0; |
| 882 |
> |
return 0; |
| 883 |
|
} |
| 884 |
|
|
| 885 |
|
/*===========================================*/ |
| 905 |
|
int i = 0; |
| 906 |
|
int j = 0; |
| 907 |
|
int count_mask = 0; |
| 908 |
< |
double sum = 0.0; |
| 908 |
> |
double sum = 0.0; |
| 909 |
|
double sum1 = 0.0; |
| 910 |
|
double err = 0.0; |
| 911 |
|
*totpotptr = 0.0; |
| 912 |
< |
*meanpotptr = 0.0; |
| 912 |
> |
*meanpotptr = 0.0; |
| 913 |
|
|
| 914 |
< |
if (nx <= 0 || ny <= 0) return 1; |
| 914 |
> |
if (nx <= 0 || ny <= 0) return 1; |
| 915 |
|
|
| 916 |
< |
for (i = 0; i < nx; i++) |
| 916 |
> |
for (i = 0; i < nx; i++) |
| 917 |
|
{ |
| 918 |
< |
for (j = 0; j < ny; j++) |
| 918 |
> |
for (j = 0; j < ny; j++) |
| 919 |
|
{ |
| 920 |
|
if ( mask[j * nx + i] < 70 || bitmask[j * nx + i] < 30 ) continue; |
| 921 |
|
if isnan(bx[j * nx + i]) continue; |
| 942 |
|
//printf("TOTPOT=%g\n",*totpotptr); |
| 943 |
|
//printf("TOTPOT_err=%g\n",*totpot_err_ptr); |
| 944 |
|
|
| 945 |
< |
return 0; |
| 945 |
> |
return 0; |
| 946 |
|
} |
| 947 |
|
|
| 948 |
|
/*===========================================*/ |
| 973 |
|
double part2 = 0.0; |
| 974 |
|
double part3 = 0.0; |
| 975 |
|
*area_w_shear_gt_45ptr = 0.0; |
| 976 |
< |
*meanshear_angleptr = 0.0; |
| 976 |
> |
*meanshear_angleptr = 0.0; |
| 977 |
|
|
| 978 |
< |
if (nx <= 0 || ny <= 0) return 1; |
| 978 |
> |
if (nx <= 0 || ny <= 0) return 1; |
| 979 |
|
|
| 980 |
< |
for (i = 0; i < nx; i++) |
| 980 |
> |
for (i = 0; i < nx; i++) |
| 981 |
|
{ |
| 982 |
< |
for (j = 0; j < ny; j++) |
| 982 |
> |
for (j = 0; j < ny; j++) |
| 983 |
|
{ |
| 984 |
|
if ( mask[j * nx + i] < 70 || bitmask[j * nx + i] < 30 ) continue; |
| 985 |
|
if isnan(bpx[j * nx + i]) continue; |
| 1028 |
|
} |
| 1029 |
|
/* For mean 3D shear angle, area with shear greater than 45*/ |
| 1030 |
|
*meanshear_angleptr = (sumsum)/(count); /* Units are degrees */ |
| 1031 |
< |
*meanshear_angle_err_ptr = (sqrt(err)/count_mask)*(180./PI); |
| 1031 |
> |
|
| 1032 |
> |
// For the error in the mean 3D shear angle: |
| 1033 |
> |
// If count_mask is 0, then we run into a divide by zero error. In this case, set *meanshear_angle_err_ptr to NAN |
| 1034 |
> |
// If count_mask is greater than zero, then compute the error. |
| 1035 |
> |
if (count_mask == 0) |
| 1036 |
> |
*meanshear_angle_err_ptr = NAN; |
| 1037 |
> |
else |
| 1038 |
> |
*meanshear_angle_err_ptr = (sqrt(err)/count_mask)*(180./PI); |
| 1039 |
|
|
| 1040 |
|
/* The area here is a fractional area -- the % of the total area. This has no error associated with it. */ |
| 1041 |
|
*area_w_shear_gt_45ptr = (count_mask/(count))*(100.0); |
| 1042 |
|
|
| 1043 |
|
//printf("MEANSHR=%f\n",*meanshear_angleptr); |
| 1044 |
< |
//printf("MEANSHR_err=%f\n",*meanshear_angle_err_ptr); |
| 1044 |
> |
//printf("ERRMSHA=%f\n",*meanshear_angle_err_ptr); |
| 1045 |
|
//printf("SHRGT45=%f\n",*area_w_shear_gt_45ptr); |
| 1046 |
< |
|
| 1034 |
< |
return 0; |
| 1046 |
> |
return 0; |
| 1047 |
|
} |
| 1048 |
|
|
| 1049 |
|
/*===========================================*/ |
| 1050 |
+ |
/* Example function 15: R parameter as defined in Schrijver, 2007 */ |
| 1051 |
+ |
// |
| 1052 |
+ |
// Note that there is a restriction on the function fsample() |
| 1053 |
+ |
// If the following occurs: |
| 1054 |
+ |
// nx_out > floor((ny_in-1)/scale + 1) |
| 1055 |
+ |
// ny_out > floor((ny_in-1)/scale + 1), |
| 1056 |
+ |
// where n*_out are the dimensions of the output array and n*_in |
| 1057 |
+ |
// are the dimensions of the input array, fsample() will usually result |
| 1058 |
+ |
// in a segfault (though not always, depending on how the segfault was accessed.) |
| 1059 |
+ |
|
| 1060 |
|
int computeR(float *bz_err, float *los, int *dims, float *Rparam, float cdelt1, |
| 1061 |
|
float *rim, float *p1p0, float *p1n0, float *p1p, float *p1n, float *p1, |
| 1062 |
< |
float *pmap, int nx1, int ny1) |
| 1063 |
< |
{ |
| 1064 |
< |
|
| 1062 |
> |
float *pmap, int nx1, int ny1, |
| 1063 |
> |
int scale, float *p1pad, int nxp, int nyp, float *pmapn) |
| 1064 |
> |
|
| 1065 |
> |
{ |
| 1066 |
|
int nx = dims[0]; |
| 1067 |
|
int ny = dims[1]; |
| 1068 |
|
int i = 0; |
| 1069 |
|
int j = 0; |
| 1070 |
< |
int index; |
| 1070 |
> |
int index, index1; |
| 1071 |
|
double sum = 0.0; |
| 1072 |
|
double err = 0.0; |
| 1073 |
|
*Rparam = 0.0; |
| 1074 |
|
struct fresize_struct fresboxcar, fresgauss; |
| 1075 |
< |
int scale = round(2.0/cdelt1); |
| 1075 |
> |
struct fint_struct fints; |
| 1076 |
|
float sigma = 10.0/2.3548; |
| 1077 |
|
|
| 1078 |
+ |
// set up convolution kernels |
| 1079 |
|
init_fresize_boxcar(&fresboxcar,1,1); |
| 1056 |
– |
|
| 1057 |
– |
// set up convolution kernel |
| 1080 |
|
init_fresize_gaussian(&fresgauss,sigma,20,1); |
| 1081 |
< |
|
| 1082 |
< |
// make sure convolution kernel is smaller than or equal to array size |
| 1083 |
< |
if ( (nx < 41.) || (ny < 41.) ) return -1; |
| 1062 |
< |
|
| 1081 |
> |
|
| 1082 |
> |
// =============== [STEP 1] =============== |
| 1083 |
> |
// bin the line-of-sight magnetogram down by a factor of scale |
| 1084 |
|
fsample(los, rim, nx, ny, nx, nx1, ny1, nx1, scale, 0, 0, 0.0); |
| 1085 |
+ |
|
| 1086 |
+ |
// =============== [STEP 2] =============== |
| 1087 |
+ |
// identify positive and negative pixels greater than +/- 150 gauss |
| 1088 |
+ |
// and label those pixels with a 1.0 in arrays p1p0 and p1n0 |
| 1089 |
|
for (i = 0; i < nx1; i++) |
| 1090 |
|
{ |
| 1091 |
|
for (j = 0; j < ny1; j++) |
| 1101 |
|
p1n0[index]=0.0; |
| 1102 |
|
} |
| 1103 |
|
} |
| 1104 |
< |
|
| 1104 |
> |
|
| 1105 |
> |
// =============== [STEP 3] =============== |
| 1106 |
> |
// smooth each of the negative and positive pixel bitmaps |
| 1107 |
|
fresize(&fresboxcar, p1p0, p1p, nx1, ny1, nx1, nx1, ny1, nx1, 0, 0, 0.0); |
| 1108 |
|
fresize(&fresboxcar, p1n0, p1n, nx1, ny1, nx1, nx1, ny1, nx1, 0, 0, 0.0); |
| 1109 |
< |
|
| 1109 |
> |
|
| 1110 |
> |
// =============== [STEP 4] =============== |
| 1111 |
> |
// find the pixels for which p1p and p1n are both equal to 1. |
| 1112 |
> |
// this defines the polarity inversion line |
| 1113 |
|
for (i = 0; i < nx1; i++) |
| 1114 |
|
{ |
| 1115 |
|
for (j = 0; j < ny1; j++) |
| 1116 |
|
{ |
| 1117 |
|
index = j * nx1 + i; |
| 1118 |
< |
if (p1p[index] > 0 && p1n[index] > 0) |
| 1118 |
> |
if ((p1p[index] > 0.0) && (p1n[index] > 0.0)) |
| 1119 |
|
p1[index]=1.0; |
| 1120 |
|
else |
| 1121 |
|
p1[index]=0.0; |
| 1122 |
|
} |
| 1123 |
|
} |
| 1124 |
< |
|
| 1125 |
< |
fresize(&fresgauss, p1, pmap, nx1, ny1, nx1, nx1, ny1, nx1, 0, 0, 0.0); |
| 1126 |
< |
|
| 1124 |
> |
|
| 1125 |
> |
// pad p1 with zeroes so that the gaussian colvolution in step 5 |
| 1126 |
> |
// does not cut off data within hwidth of the edge |
| 1127 |
> |
|
| 1128 |
> |
// step i: zero p1pad |
| 1129 |
> |
for (i = 0; i < nxp; i++) |
| 1130 |
> |
{ |
| 1131 |
> |
for (j = 0; j < nyp; j++) |
| 1132 |
> |
{ |
| 1133 |
> |
index = j * nxp + i; |
| 1134 |
> |
p1pad[index]=0.0; |
| 1135 |
> |
} |
| 1136 |
> |
} |
| 1137 |
> |
|
| 1138 |
> |
// step ii: place p1 at the center of p1pad |
| 1139 |
> |
for (i = 0; i < nx1; i++) |
| 1140 |
> |
{ |
| 1141 |
> |
for (j = 0; j < ny1; j++) |
| 1142 |
> |
{ |
| 1143 |
> |
index = j * nx1 + i; |
| 1144 |
> |
index1 = (j+20) * nxp + (i+20); |
| 1145 |
> |
p1pad[index1]=p1[index]; |
| 1146 |
> |
} |
| 1147 |
> |
} |
| 1148 |
> |
|
| 1149 |
> |
// =============== [STEP 5] =============== |
| 1150 |
> |
// convolve the polarity inversion line map with a gaussian |
| 1151 |
> |
// to identify the region near the plarity inversion line |
| 1152 |
> |
// the resultant array is called pmap |
| 1153 |
> |
fresize(&fresgauss, p1pad, pmap, nxp, nyp, nxp, nxp, nyp, nxp, 0, 0, 0.0); |
| 1154 |
> |
|
| 1155 |
> |
|
| 1156 |
> |
// select out the nx1 x ny1 non-padded array within pmap |
| 1157 |
> |
for (i = 0; i < nx1; i++) |
| 1158 |
> |
{ |
| 1159 |
> |
for (j = 0; j < ny1; j++) |
| 1160 |
> |
{ |
| 1161 |
> |
index = j * nx1 + i; |
| 1162 |
> |
index1 = (j+20) * nxp + (i+20); |
| 1163 |
> |
pmapn[index]=pmap[index1]; |
| 1164 |
> |
} |
| 1165 |
> |
} |
| 1166 |
> |
|
| 1167 |
> |
// =============== [STEP 6] =============== |
| 1168 |
> |
// the R parameter is calculated |
| 1169 |
|
for (i = 0; i < nx1; i++) |
| 1170 |
|
{ |
| 1171 |
|
for (j = 0; j < ny1; j++) |
| 1172 |
|
{ |
| 1173 |
|
index = j * nx1 + i; |
| 1174 |
< |
sum += pmap[index]*abs(rim[index]); |
| 1174 |
> |
if isnan(pmapn[index]) continue; |
| 1175 |
> |
if isnan(rim[index]) continue; |
| 1176 |
> |
sum += pmapn[index]*abs(rim[index]); |
| 1177 |
|
} |
| 1178 |
|
} |
| 1179 |
|
|
| 1181 |
|
*Rparam = 0.0; |
| 1182 |
|
else |
| 1183 |
|
*Rparam = log10(sum); |
| 1184 |
< |
|
| 1184 |
> |
|
| 1185 |
> |
//printf("R_VALUE=%f\n",*Rparam); |
| 1186 |
> |
|
| 1187 |
|
free_fresize(&fresboxcar); |
| 1188 |
|
free_fresize(&fresgauss); |
| 1189 |
+ |
|
| 1190 |
+ |
return 0; |
| 1191 |
+ |
|
| 1192 |
+ |
} |
| 1193 |
+ |
|
| 1194 |
+ |
/*===========================================*/ |
| 1195 |
+ |
/* Example function 16: Lorentz force as defined in Fisher, 2012 */ |
| 1196 |
+ |
// |
| 1197 |
+ |
// This calculation is adapted from Xudong's code |
| 1198 |
+ |
// at /proj/cgem/lorentz/apps/lorentz.c |
| 1199 |
+ |
|
| 1200 |
+ |
int computeLorentz(float *bx, float *by, float *bz, float *fx, float *fy, float *fz, int *dims, |
| 1201 |
+ |
float *totfx_ptr, float *totfy_ptr, float *totfz_ptr, float *totbsq_ptr, |
| 1202 |
+ |
float *epsx_ptr, float *epsy_ptr, float *epsz_ptr, int *mask, int *bitmask, |
| 1203 |
+ |
float cdelt1, double rsun_ref, double rsun_obs) |
| 1204 |
+ |
|
| 1205 |
+ |
{ |
| 1206 |
+ |
|
| 1207 |
+ |
int nx = dims[0]; |
| 1208 |
+ |
int ny = dims[1]; |
| 1209 |
+ |
int nxny = nx*ny; |
| 1210 |
+ |
int j = 0; |
| 1211 |
+ |
int index; |
| 1212 |
+ |
double totfx = 0, totfy = 0, totfz = 0; |
| 1213 |
+ |
double bsq = 0, totbsq = 0; |
| 1214 |
+ |
double epsx = 0, epsy = 0, epsz = 0; |
| 1215 |
+ |
double area = cdelt1*cdelt1*(rsun_ref/rsun_obs)*(rsun_ref/rsun_obs)*100.0*100.0; |
| 1216 |
+ |
double k_h = -1.0 * area / (4. * PI) / 1.0e20; |
| 1217 |
+ |
double k_z = area / (8. * PI) / 1.0e20; |
| 1218 |
+ |
|
| 1219 |
+ |
if (nx <= 0 || ny <= 0) return 1; |
| 1220 |
+ |
|
| 1221 |
+ |
for (int i = 0; i < nxny; i++) |
| 1222 |
+ |
{ |
| 1223 |
+ |
if ( mask[i] < 70 || bitmask[i] < 30 ) continue; |
| 1224 |
+ |
if isnan(bx[i]) continue; |
| 1225 |
+ |
if isnan(by[i]) continue; |
| 1226 |
+ |
if isnan(bz[i]) continue; |
| 1227 |
+ |
fx[i] = bx[i] * bz[i] * k_h; |
| 1228 |
+ |
fy[i] = by[i] * bz[i] * k_h; |
| 1229 |
+ |
fz[i] = (bx[i] * bx[i] + by[i] * by[i] - bz[i] * bz[i]) * k_z; |
| 1230 |
+ |
bsq = bx[i] * bx[i] + by[i] * by[i] + bz[i] * bz[i]; |
| 1231 |
+ |
totfx += fx[i]; totfy += fy[i]; totfz += fz[i]; |
| 1232 |
+ |
totbsq += bsq; |
| 1233 |
+ |
} |
| 1234 |
+ |
|
| 1235 |
+ |
*totfx_ptr = totfx; |
| 1236 |
+ |
*totfy_ptr = totfy; |
| 1237 |
+ |
*totfz_ptr = totfz; |
| 1238 |
+ |
*totbsq_ptr = totbsq; |
| 1239 |
+ |
*epsx_ptr = (totfx / k_h) / totbsq; |
| 1240 |
+ |
*epsy_ptr = (totfy / k_h) / totbsq; |
| 1241 |
+ |
*epsz_ptr = (totfz / k_z) / totbsq; |
| 1242 |
+ |
|
| 1243 |
+ |
//printf("TOTBSQ=%f\n",*totbsq_ptr); |
| 1244 |
+ |
|
| 1245 |
+ |
return 0; |
| 1246 |
+ |
|
| 1247 |
+ |
} |
| 1248 |
+ |
|
| 1249 |
+ |
/*===========================================*/ |
| 1250 |
+ |
|
| 1251 |
+ |
/* Example function 17: Compute total unsigned flux in units of G/cm^2 on the LOS field */ |
| 1252 |
+ |
|
| 1253 |
+ |
// To compute the unsigned flux, we simply calculate |
| 1254 |
+ |
// flux = surface integral [(vector LOS) dot (normal vector)], |
| 1255 |
+ |
// = surface integral [(magnitude LOS)*(magnitude normal)*(cos theta)]. |
| 1256 |
+ |
// However, since the field is radial, we will assume cos theta = 1. |
| 1257 |
+ |
// Therefore the pixels only need to be corrected for the projection. |
| 1258 |
+ |
|
| 1259 |
+ |
// To convert G to G*cm^2, simply multiply by the number of square centimeters per pixel. |
| 1260 |
+ |
// As an order of magnitude estimate, we can assign 0.5 to CDELT1 and 722500m/arcsec to (RSUN_REF/RSUN_OBS). |
| 1261 |
+ |
// (Gauss/pix^2)(CDELT1)^2(RSUN_REF/RSUN_OBS)^2(100.cm/m)^2 |
| 1262 |
+ |
// =Gauss*cm^2 |
| 1263 |
+ |
|
| 1264 |
+ |
int computeAbsFlux_los(float *los, int *dims, float *absFlux_los, |
| 1265 |
+ |
float *mean_vf_los_ptr, float *count_mask_los_ptr, |
| 1266 |
+ |
int *bitmask, float cdelt1, double rsun_ref, double rsun_obs) |
| 1267 |
+ |
|
| 1268 |
+ |
{ |
| 1269 |
|
|
| 1270 |
+ |
int nx = dims[0]; |
| 1271 |
+ |
int ny = dims[1]; |
| 1272 |
+ |
int i = 0; |
| 1273 |
+ |
int j = 0; |
| 1274 |
+ |
int count_mask_los = 0; |
| 1275 |
+ |
double sum = 0.0; |
| 1276 |
+ |
*absFlux_los = 0.0; |
| 1277 |
+ |
*mean_vf_los_ptr = 0.0; |
| 1278 |
+ |
|
| 1279 |
+ |
|
| 1280 |
+ |
if (nx <= 0 || ny <= 0) return 1; |
| 1281 |
+ |
|
| 1282 |
+ |
for (i = 0; i < nx; i++) |
| 1283 |
+ |
{ |
| 1284 |
+ |
for (j = 0; j < ny; j++) |
| 1285 |
+ |
{ |
| 1286 |
+ |
if ( bitmask[j * nx + i] < 30 ) continue; |
| 1287 |
+ |
if isnan(los[j * nx + i]) continue; |
| 1288 |
+ |
sum += (fabs(los[j * nx + i])); |
| 1289 |
+ |
count_mask_los++; |
| 1290 |
+ |
} |
| 1291 |
+ |
} |
| 1292 |
+ |
|
| 1293 |
+ |
*mean_vf_los_ptr = sum*cdelt1*cdelt1*(rsun_ref/rsun_obs)*(rsun_ref/rsun_obs)*100.0*100.0; |
| 1294 |
+ |
*count_mask_los_ptr = count_mask_los; |
| 1295 |
+ |
|
| 1296 |
+ |
printf("USFLUXL=%f\n",*mean_vf_los_ptr); |
| 1297 |
+ |
printf("CMASKL=%f\n",*count_mask_los_ptr); |
| 1298 |
+ |
|
| 1299 |
|
return 0; |
| 1300 |
|
} |
| 1301 |
|
|
| 1302 |
+ |
/*===========================================*/ |
| 1303 |
+ |
/* Example function 18: Derivative of B_LOS (approximately B_vertical) = SQRT( ( dLOS/dx )^2 + ( dLOS/dy )^2 ) */ |
| 1304 |
+ |
|
| 1305 |
+ |
int computeLOSderivative(float *los, int *dims, float *mean_derivative_los_ptr, int *bitmask, float *derx_los, float *dery_los) |
| 1306 |
+ |
{ |
| 1307 |
+ |
|
| 1308 |
+ |
int nx = dims[0]; |
| 1309 |
+ |
int ny = dims[1]; |
| 1310 |
+ |
int i = 0; |
| 1311 |
+ |
int j = 0; |
| 1312 |
+ |
int count_mask = 0; |
| 1313 |
+ |
double sum = 0.0; |
| 1314 |
+ |
*mean_derivative_los_ptr = 0.0; |
| 1315 |
+ |
|
| 1316 |
+ |
if (nx <= 0 || ny <= 0) return 1; |
| 1317 |
+ |
|
| 1318 |
+ |
/* brute force method of calculating the derivative (no consideration for edges) */ |
| 1319 |
+ |
for (i = 1; i <= nx-2; i++) |
| 1320 |
+ |
{ |
| 1321 |
+ |
for (j = 0; j <= ny-1; j++) |
| 1322 |
+ |
{ |
| 1323 |
+ |
derx_los[j * nx + i] = (los[j * nx + i+1] - los[j * nx + i-1])*0.5; |
| 1324 |
+ |
} |
| 1325 |
+ |
} |
| 1326 |
+ |
|
| 1327 |
+ |
/* brute force method of calculating the derivative (no consideration for edges) */ |
| 1328 |
+ |
for (i = 0; i <= nx-1; i++) |
| 1329 |
+ |
{ |
| 1330 |
+ |
for (j = 1; j <= ny-2; j++) |
| 1331 |
+ |
{ |
| 1332 |
+ |
dery_los[j * nx + i] = (los[(j+1) * nx + i] - los[(j-1) * nx + i])*0.5; |
| 1333 |
+ |
} |
| 1334 |
+ |
} |
| 1335 |
+ |
|
| 1336 |
+ |
/* consider the edges for the arrays that contribute to the variable "sum" in the computation below. |
| 1337 |
+ |
ignore the edges for the error terms as those arrays have been initialized to zero. |
| 1338 |
+ |
this is okay because the error term will ultimately not include the edge pixels as they are selected out by the mask and bitmask arrays.*/ |
| 1339 |
+ |
i=0; |
| 1340 |
+ |
for (j = 0; j <= ny-1; j++) |
| 1341 |
+ |
{ |
| 1342 |
+ |
derx_los[j * nx + i] = ( (-3*los[j * nx + i]) + (4*los[j * nx + (i+1)]) - (los[j * nx + (i+2)]) )*0.5; |
| 1343 |
+ |
} |
| 1344 |
+ |
|
| 1345 |
+ |
i=nx-1; |
| 1346 |
+ |
for (j = 0; j <= ny-1; j++) |
| 1347 |
+ |
{ |
| 1348 |
+ |
derx_los[j * nx + i] = ( (3*los[j * nx + i]) + (-4*los[j * nx + (i-1)]) - (-los[j * nx + (i-2)]) )*0.5; |
| 1349 |
+ |
} |
| 1350 |
+ |
|
| 1351 |
+ |
j=0; |
| 1352 |
+ |
for (i = 0; i <= nx-1; i++) |
| 1353 |
+ |
{ |
| 1354 |
+ |
dery_los[j * nx + i] = ( (-3*los[j*nx + i]) + (4*los[(j+1) * nx + i]) - (los[(j+2) * nx + i]) )*0.5; |
| 1355 |
+ |
} |
| 1356 |
+ |
|
| 1357 |
+ |
j=ny-1; |
| 1358 |
+ |
for (i = 0; i <= nx-1; i++) |
| 1359 |
+ |
{ |
| 1360 |
+ |
dery_los[j * nx + i] = ( (3*los[j * nx + i]) + (-4*los[(j-1) * nx + i]) - (-los[(j-2) * nx + i]) )*0.5; |
| 1361 |
+ |
} |
| 1362 |
+ |
|
| 1363 |
+ |
|
| 1364 |
+ |
for (i = 0; i <= nx-1; i++) |
| 1365 |
+ |
{ |
| 1366 |
+ |
for (j = 0; j <= ny-1; j++) |
| 1367 |
+ |
{ |
| 1368 |
+ |
if ( bitmask[j * nx + i] < 30 ) continue; |
| 1369 |
+ |
if ( (derx_los[j * nx + i] + dery_los[j * nx + i]) == 0) continue; |
| 1370 |
+ |
if isnan(los[j * nx + i]) continue; |
| 1371 |
+ |
if isnan(los[(j+1) * nx + i]) continue; |
| 1372 |
+ |
if isnan(los[(j-1) * nx + i]) continue; |
| 1373 |
+ |
if isnan(los[j * nx + i-1]) continue; |
| 1374 |
+ |
if isnan(los[j * nx + i+1]) continue; |
| 1375 |
+ |
if isnan(derx_los[j * nx + i]) continue; |
| 1376 |
+ |
if isnan(dery_los[j * nx + i]) continue; |
| 1377 |
+ |
sum += sqrt( derx_los[j * nx + i]*derx_los[j * nx + i] + dery_los[j * nx + i]*dery_los[j * nx + i] ); /* Units of Gauss */ |
| 1378 |
+ |
count_mask++; |
| 1379 |
+ |
} |
| 1380 |
+ |
} |
| 1381 |
+ |
|
| 1382 |
+ |
*mean_derivative_los_ptr = (sum)/(count_mask); // would be divided by ((nx-2)*(ny-2)) if shape of count_mask = shape of magnetogram |
| 1383 |
+ |
|
| 1384 |
+ |
printf("MEANGBL=%f\n",*mean_derivative_los_ptr); |
| 1385 |
+ |
|
| 1386 |
+ |
return 0; |
| 1387 |
+ |
} |
| 1388 |
|
|
| 1389 |
|
/*==================KEIJI'S CODE =========================*/ |
| 1390 |
|
|
