Corfrom ENMTools.pl - does this look right?

[Steve Research]

Hi,

I ran all of my layers through ENMTools using the Correlation function. I have a range of values in my output file 'AllASCv2_correlation.csv'

    (1,0.9999977298,0.9999993497,0.9929911605,0.9999962174,0.9999987349,0.9999915987,0.9999995263,0.9999993612,0.9999995022,0.9999995453,0.9991056607,0.9999669615,0.9999986717,0.9999976136,0.9997493077,0.9999975869,0.999997103,0.99975491,0.9904618922),
    (1,1,0.9999980987,0.9931743134,0.9999983484,0.9999941295,0.9999973878,0.9999958947,0.9999986687,0.9999987119,0.9999958049,0.9990491409,0.9999562608,0.9999971677,0.9999954431,0.9997213986,0.9999948096,0.9999937055,0.9997291389,0.9904258033),
    (1,1,1,0.9929535127,0.9999947787,0.9999987051,0.9999912489,0.9999989142,0.9999984073,0.9999985056,0.9999992044,0.9991006603,0.9999666209,0.9999997363,0.9999980023,0.9997465863,0.9999985706,0.9999981312,0.9997529677,0.9904545162),
    (1,1,1,1,0.9933089817,0.9928108825,0.9934368116,0.9929036305,0.9931089979,0.9931070175,0.9928792305,0.989469558,0.9924822678,0.9929134774,0.9929307399,0.9915099242,0.9928459648,0.9928012626,0.9915797009,0.9820301283),
    (1,1,1,1,1,0.9999908476,0.9999984595,0.9999936386,0.9999983978,0.9999984371,0.9999932337,0.9990490857,0.9999539814,0.9999931937,0.9999930325,0.999721861,0.9999904727,0.9999890286,0.9997290872,0.9904223283),
    (1,1,1,1,1,1,0.99998462,0.9999993579,0.999996642,0.9999967813,0.9999997468,0.9991438984,0.999973456,0.9999986575,0.999997826,0.9997679329,0.9999981722,0.9999981531,0.9997725948,0.9904796988),
    (1,1,1,1,1,1,1,0.9999880474,0.9999948625,0.9999948445,0.9999875502,0.9989984275,0.9999422689,0.9999896144,0.9999885449,0.9996951065,0.9999859279,0.9999839634,0.9997040462,0.9903911514),
    (1,1,1,1,1,1,1,1,0.9999981011,0.9999982747,0.999999747,0.9991111333,0.9999675414,0.9999983028,0.9999969742,0.9997515079,0.9999974404,0.9999971727,0.9997561972,0.9904717868),
    (1,1,1,1,1,1,1,1,1,0.999999882,0.9999979905,0.9990886592,0.9999635603,0.9999974474,0.9999968597,0.999741657,0.9999957167,0.9999947976,0.9997478383,0.990448224),
    (1,1,1,1,1,1,1,1,1,1,0.9999981371,0.9990880689,0.9999634952,0.9999975115,0.9999968699,0.9997413515,0.9999958439,0.9999950064,0.9997474611,0.9904495658),
    (1,1,1,1,1,1,1,1,1,1,1,0.9991293015,0.9999710426,0.9999988442,0.9999978649,0.9997606499,0.9999982222,0.9999981009,0.9997657388,0.9904734488),
    (1,1,1,1,1,1,1,1,1,1,1,1,0.9994054127,0.9991135497,0.9991600025,0.9997837333,0.9991313199,0.9991421243,0.9997771779,0.9899846624),
    (1,1,1,1,1,1,1,1,1,1,1,1,1,0.9999683458,0.9999783833,0.99989689,0.9999688335,0.9999704001,0.999899807,0.9904993506),
    (1,1,1,1,1,1,1,1,1,1,1,1,1,1,0.9999978186,0.9997516121,0.9999992523,0.999998773,0.9997583368,0.9904557115),
    (1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,0.9997820774,0.9999958322,0.9999956244,0.9997875414,0.9904636823),
    (1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,0.9997555956,0.999760759,0.9999968871,0.9904190682),
    (1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,0.9999997855,0.9997626415,0.9904606637),
    (1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,0.9997676535,0.9904645216),
    (1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,0.9903904734),
    (1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1)Enter code here...

Does this look like a good result and if so how do I interpret it?

Cheers,
Steve

[Samuel Veloz]

If I am reading this correctly you have 19 layers that are all almost perfectly correlated with each other, lowest correlation is about 0.98. That is really bad actually, there are rules of thumb about dropping 1 of a pair of layers with correlations >0.7 for example. I am guessing you may have made an error running the function. I am assuming these layers are your environmental predictors.

Sam

--
You received this message because you are subscribed to the Google Groups "Maxent" group.
To unsubscribe from this group and stop receiving emails from it, send an email to maxent+un...@googlegroups.com.
To post to this group, send email to max...@googlegroups.com.
Visit this group at https://groups.google.com/group/maxent.
For more options, visit https://groups.google.com/d/optout.

[Steve Research]

Sam,
I am running his tool with the standard 19 WorlClim layers for my area. I am getting good models with good TSS, Binomial P and AUC values.

I ran this again by using the 'Standardization' tool. Then ran it again with similar results.

I can't believe that all 19 layers are correlated so is this tool just giving me junk output? If so how am I using it wrongly? There don't to seem to be many options.
Cheers
Steve

[Dan Warren]

What value represents NODATA in your layers?  If it's not -9999, ENMTools may be including your NODATA values as data.

[Steve Research]

Hi Dan,

I checked (opened the *.asc files in notepad).

The header states "NODATA_value -9999 and this seems to match the actual data in the lines.

Cheers

Steve

Sent from my iPhone

--

[Dan Warren]

Rightyo.  And are you sure you haven't changed the nodata value in ENMTools?  It's in the ENMTools config panel.

[matea.vukelic31]

Hi everyone,

i have a problem with running correlation options in ENMTools. When I upload my asc rasters and click Go,an error occurs after a few minutes. It says: Error: Illegal division by zero at ...ENMTools 1.1.1,line 25704

Can someone explain how could i fix this.

Greeting from Croatia

[Arta]

Hi Matea

I had this problem and found that one layer caused it. when i ran the correlation without it, it works well! I don't know the reason?

Best;

"Arta'

[matea.vukelic31]

Hi,

thank you so much for your advice, i was able to fix it.

Much greetings

[Forough Goudarzi]

good news.

How you could fix it?

Best;

"Arta"

--
You received this message because you are subscribed to a topic in the Google Groups "Maxent" group.
To unsubscribe from this topic, visit https://groups.google.com/d/topic/maxent/qJKbrfcYUl0/unsubscribe.
To unsubscribe from this group and all its topics, send an email to maxent+un...@googlegroups.com.

[matea.vukelic31]

i just eliminated the layer that was causing that error (no idea why that particulary one) and run the model without it.