Minimum number of presence in MaxEnt

[Basma Sheta]

Is it possible to use MaxEnt if i have 6 presence data only for an animal?

and what is the minimum number of presence i can start with in MaxEnt?

[Martin Damus]

Hello,

What you want to do is use jackknifing or bootstrapping of your data points to see if your model is stable. If the loss of one of your points results in a totally different outcome, then the results are unreliable due to insufficient data points to create a stable model.

Also, you want to be sure that you've covered a good part of the ecological range of your species. Note that if this requirement is fulfilled then fulfilling the first one may be made more difficult -- hence more points are better. With 6 data points I would generally prefer to model a species with a very narrow 'climate niche' that is also geographically not widespread.

I am quite open to being told I'm wrong ... ; )

Cheers,

Martin Damus

---

From: Basma Sheta <egyptorn...@gmail.com>
To: max...@googlegroups.com
Sent: Monday, December 22, 2014 6:21 PM
Subject: Minimum number of presence in MaxEnt

Is it possible to use MaxEnt if i have 6 presence data only for an animal?

and what is the minimum number of presence i can start with in MaxEnt?

--
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 http://groups.google.com/group/maxent.
For more options, visit https://groups.google.com/d/optout.

[Basma Sheta]

Thanks Martin
What i have is the presence of new records for first time in our environment and my main goal is to predict other places for future survey to search for this species. That is why i have limited number of presence. According to what you said instead of using all Egypt map I can only use north Egypt map so make it for small area is this what you mean happy to correct for me too.
Thanks again

On Monday, December 29, 2014 6:56:53 PM UTC+2, Martin wrote:

Hello,

What you want to do is use jackknifing or bootstrapping of your data points to see if your model is stable. If the loss of one of your points results in a totally different outcome, then the results are unreliable due to insufficient data points to create a stable model.

Also, you want to be sure that you've covered a good part of the ecological range of your species. Note that if this requirement is fulfilled then fulfilling the first one may be made more difficult -- hence more points are better. With 6 data points I would generally prefer to model a species with a very narrow 'climate niche' that is also geographically not widespread.

I am quite open to being told I'm wrong ... ; )

Cheers,

Martin Damus

---

From: Basma Sheta <egyptorn...@gmail.com>
To: max...@googlegroups.com
Sent: Monday, December 22, 2014 6:21 PM
Subject: Minimum number of presence in MaxEnt

Is it possible to use MaxEnt if i have 6 presence data only for an animal?

and what is the minimum number of presence i can start with in MaxEnt?

--
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+unsubscribe@googlegroups.com.

[Alaa Eldeen]

Hi Basma,

Since Maxent is a machine learning tool, you can use it to model species with few observations and there is some publications supported this claim. bear in mind not to use cross-validation in this case.

Cheers

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 http://groups.google.com/group/maxent.
For more options, visit https://groups.google.com/d/optout.

--

Alaa Eldin Soultan,  B.Sc, MSc.
Wildlife Conservationist,
St Katherine Protectorate,
National parks of Egypt,
Nature Conservation Sector,
Egyptian Environmental Affairs Agency, EEAA
Tel-Fax: + 2069- 3470033
Mobil:+201225773922

[Husam El Alqamy]

Dear Dr. Basma

I have read two papers earlier that reports the same number (6 records ) as a viable number for estimating the SDM but you have to be aware of some limitations. How spread are your presence points in the target study are? If your records are clustered in a certain proportion of the study area you have to use a sampling bias file into your maxent modeling , if the points are fairly distributed over the study region then it is a better situation. Also it is much better if you include only environmental favriables that you are sure that they relate to the ecology of the species in hand otherwise you should try different possible combinations of variables and test each model for performance or best fit using some diagnostic such as AIC (Akaike Information Criterion) or TSS ( Total Skill Statistic) to select for the best performing model. Happy to help more if you have any further enquiries

 

From: max...@googlegroups.com [mailto:max...@googlegroups.com] On Behalf Of Basma Sheta
Sent: Tuesday, December 23, 2014 3:21 AM
To: max...@googlegroups.com
Subject: Minimum number of presence in MaxEnt

 

Is it possible to use MaxEnt if i have 6 presence data only for an animal?

and what is the minimum number of presence i can start with in MaxEnt?

--

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 http://groups.google.com/group/maxent.
For more options, visit https://groups.google.com/d/optout.