[ncl-talk] Calculating the area below the curve

Tue Dec 10 22:32:06 MST 2019

I think the numbers derived by Rick are "reasonable".

I am no sure what should be expected when integrating the Seawater values.
There is a LARGE gap between diameters 187.7 and 710.5. The seawater values
are all missing. To integrate across the seawater values would mean drawing
a straight line between 1244.69 and 384.31

Diam   Sea       Fresh
[SNIP]
168.5   1241.65  819.803
174.7   1244.69  813.677
181.1   *9999 *    862.247
187.7   *9999 *    828.698
[SNIP]
710.5   *9999*     135.773
736.5   *9999 *    107.602
835     384.306   80.561
898     356.98    63.243
[SNIP]

Ricks results:
Fresh:  363163.4
Sea:    1067454

%> ncl TEST_trap.ncl                    <== Basically, the trapezoidal rule
[lots of print statements]
(0) trap_fresh=     362866.156
(0) trap_sea  =    1125439.000

%> ncl TEST_simpne.ncl               <== Implements simpsons method  for
unevenly sampled data in NCL
(0) simp_fresh=     363163.344
(0) simp_sea  =    1067453.250

Again, I am not sure how one would interpret the Seawater result due to the
gap.

Good Luck

On Mon, Dec 9, 2019 at 8:54 PM Rick Brownrigg via ncl-talk <
ncl-talk at ucar.edu> wrote:

> Hi Anahita,
>
> In stepping through the code to where this error message originates, I
> believe it to be due to a bug that was introduced several years ago. Note
> that the examples given on the documentation for simpne() (
> http://ncl.ucar.edu/Document/Functions/Built-in/simpne.shtml) also fail
> with the same message (this function must not get used much).  The test for
> "3 non-missings" occurs *after* the call to the underlying fortran routine
> that actually calculates the area. With a debugger, I was able to side-step
> the test, and for your particular dataset, I get:
>
> print(simpne_sea)
> Variable: simpne_sea
> Type: float
> Total Size: 4 bytes
>             1 values
> Number of Dimensions: 1
> Dimensions and sizes:   [1]
> Coordinates:
> (0)     1067454
>
> ncl 11> print(simpne_fresh)
> Variable: simpne_fresh
> Type: float
> Total Size: 4 bytes
>             1 values
> Number of Dimensions: 1
> Dimensions and sizes:   [1]
> Coordinates:
> (0)     363163.4
>
> Do these values make sense? If I similarly do the "side step" for the
> examples on the docs page, I get the results shown there, so I'm fairly
> confident in the analysis that its the test for non-missings, not the area
> computations, that is erroneous.
>
> Unfortunately, this bug occurs in compiled code. I am no longer in a
> position to i) officially address the bug, and ii) offer you a temporary
> version of NCL until the big-fix appears in the next release. The best I
> can do is file an issue on the github site with as much detail as I know
> now, and maybe a fix will appear in a future release.
>
> I wish I had a better answer.
>
> Rick
>
> On Mon, Dec 9, 2019 at 3:31 PM Anahita Amiri Farahani via ncl-talk <
> ncl-talk at ucar.edu> wrote:
>
>> Hello All,
>>
>> I tried to calculate the area below the curve but I got the error, I used
>> simpne finction. Here is my code and my data:
>>
>> ana = asciiread("test.csv",(/146,3/),"float")
>> ana at _FillValue = 9999
>>
>> xi= ana(:,0) ; Diameter
>> yi= ana(:,1) ;synthetic sea spray
>> zi= ana(:,2) ; Fresh water
>> simpne_sea = simpne(xi,yi)
>> simpne_fresh = simpne(xi,zi)
>>
>> The error that I got is:
>>
>> fatal:simpne: Must have three or more non-missing values.
>>
>> Diameter    Seawater   Freshwater
>> 20.2 379.07 62.8549
>> 20.9 363.27 91.3129
>> 21.7 455.041 97.3195
>> 22.5 451.004 106.884
>> 23.3 490.321 135.746
>> 24.1 571.821 136.587
>> 25 619.093 135.276
>> 25.9 616.577 153.029
>> 26.9 691.639 164.088
>> 27.9 664.933 157.471
>> 28.9 717.085 196.833
>> 30 769.125 203.958
>> 31.1 784.47 192.036
>> 32.2 871.042 242.472
>> 33.4 881.308 224.274
>> 34.6 871.322 236.221
>> 35.9 902.276 283.161
>> 37.2 1005.14 277.881
>> 38.5 997.639 302.734
>> 40 1007.47 320.725
>> 41.4 1060.78 318.618
>> 42.9 1046.8 328.856
>> 44.5 1095.6 336.364
>> 46.1 1164.61 381.286
>> 47.8 1192.6 393.798
>> 49.6 1226.13 410.977
>> 51.4 1184.22 416.643
>> 53.3 1253.15 419.43
>> 55.2 1297.6 476.518
>> 57.3 1355.38 484.361
>> 59.4 1336.36 496.661
>> 61.5 1409.51 492.204
>> 63.8 1399.77 534.932
>> 66.1 1441.49 551.7
>> 68.5 1537.13 559.25
>> 71 1470.6 557.561
>> 73.7 1454.18 530.684
>> 76.4 1546.17 574.684
>> 79.1 1526.6 572.641
>> 82 1545.55 550.367
>> 85.1 1556.94 584.655
>> 88.2 1527.92 580.766
>> 91.4 1544.07 585.633
>> 94.7 1621.34 595.263
>> 98.2 1628.2 603.119
>> 101.8 1703.88 639.341
>> 105.5 1686.83 610.778
>> 109.4 1679.66 658.014
>> 113.4 1595.34 660.075
>> 117.6 1590.48 677.112
>> 121.9 1554.42 680.088
>> 126.3 1690.79 713.833
>> 131 1714.44 770.754
>> 135.8 1661.12 798.023
>> 140.7 1574.7 804.933
>> 145.9 1463.69 828.813
>> 151.2 1359.23 798.361
>> 156.8 1319.38 812.687
>> 162.5 1231.97 784.559
>> 168.5 1241.65 819.803
>> 174.7 1244.69 813.677
>> 181.1 9999 862.247
>> 187.7 9999 828.698
>> 194.6 9999 808.084
>> 201.7 9999 778.683
>> 209.1 9999 768.083
>> 216.7 9999 810.226
>> 224.7 9999 786.788
>> 232.9 9999 812.941
>> 241.4 9999 758.239
>> 250.3 9999 744.37
>> 259.5 9999 704.377
>> 269 9999 637.229
>> 278.8 9999 648.659
>> 289 9999 648.906
>> 299.6 9999 636.226
>> 310.6 9999 606.987
>> 322 9999 616.206
>> 333.8 9999 611.99
>> 346 9999 582.636
>> 358.7 9999 518.618
>> 371.8 9999 510.237
>> 385.4 9999 464.282
>> 399.5 9999 443.98
>> 414.2 9999 407.32
>> 429.4 9999 461.604
>> 445.1 9999 440.411
>> 461.4 9999 397.705
>> 478.3 9999 382
>> 495.8 9999 318.854
>> 514 9999 323.109
>> 532.8 9999 292.444
>> 552.3 9999 263.288
>> 572.5 9999 234.776
>> 593.5 9999 218.003
>> 615.3 9999 208.224
>> 637.8 9999 178.02
>> 661.2 9999 167.827
>> 685.4 9999 141.841
>> 710.5 9999 135.773
>> 736.5 9999 107.602
>> 835 384.306 80.5616
>> 898 356.98 63.2435
>> 965 328.19 49.151
>> 1037 301.18 36.6713
>> 1114 274.241 26.4571
>> 1197 242.377 20.198
>> 1286 215.253 14.8797
>> 1382 193.643 11.8462
>> 1486 174.463 8.71663
>> 1596 159.423 8.04464
>> 1715 150.035 5.81748
>> 1843 137.426 4.74231
>> 1981 121.248 4.20472
>> 2129 107.02 3.18714
>> 2288 95.7569 2.84154
>> 2458 82.8993 2.03516
>> 2642 69.2672 1.38237
>> 2839 57.6725 0.748785
>> 3051 48.669 0.383992
>> 3278 40.8249 0.211196
>> 3523 32.6918 0.211196
>> 3786 23.9757 0.0575989
>> 4068 16.7131 0.0191996
>> 4371 11.5676 0.0383992
>> 4698 7.72549 0
>> 5048 4.75836 0
>> 5425 2.59956 0
>> 5829 1.29723 0
>> 6264 0.619433 0
>> 6732 0.263469 0
>> 7234 0.100302 0
>> 7774 0.0339347 0
>> 8354 0.010811 0
>> 8977 0.0043044 0
>> 9647 0.00070072 0
>> 10370 0.0001001 0
>> 11140 0.0001001 0
>> 11970 0.0001001 0
>> 12860 0.00020021 0
>> 13820 0 0
>> 14860 0 0
>> 15960 0.00020021 0
>> 17150 0 0
>> 18430 0 0
>> 19810 0 0
>> _______________________________________________
>> ncl-talk mailing list
>> ncl-talk at ucar.edu
>> List instructions, subscriber options, unsubscribe:
>> http://mailman.ucar.edu/mailman/listinfo/ncl-talk
>
> _______________________________________________
> ncl-talk mailing list
> ncl-talk at ucar.edu
> List instructions, subscriber options, unsubscribe:
> http://mailman.ucar.edu/mailman/listinfo/ncl-talk
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mailman.ucar.edu/pipermail/ncl-talk/attachments/20191210/4553758d/attachment-0001.html>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: TEST_trap.ncl
Type: application/octet-stream
Size: 1511 bytes
Desc: not available
URL: <http://mailman.ucar.edu/pipermail/ncl-talk/attachments/20191210/4553758d/attachment-0002.obj>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: TEST_simpne.ncl
Type: application/octet-stream
Size: 1877 bytes
Desc: not available
URL: <http://mailman.ucar.edu/pipermail/ncl-talk/attachments/20191210/4553758d/attachment-0003.obj>