If a plant species is distributed only in a very small area, would it be appropriate to model it in MaxEnt?
Добрый вечер
Вид, о котором я говорю, является национальным эндемичным растением площадью всего не более 2 кв км.
This is the result at Maxent. What you think. Is this true for an endemic species whose range does not exceed 2 sq. km?
Dear Bekhruz Khabibullaev
This issue is very critical, there are many plants whose geographical distribution does not exceed a square kilometer, which is normal when using coordinates in the Maxent program will give an incorrect result at all.
In this case, you must follow the following steps:
1) Remove the variables that give a zero contribution to the model
2) Adjust the program's settings so that 25% of the coordinates will be tested
3) Put 10 replicates in the replicates box
Then let the program give 10 different results
The program will send you a final product for the average of ten models and three ASCII files that represent the high, low and medium scenarios.
You will find that the average is the closest to the truth
I made a screenshot to help you understand what I mean (Attached)
Here is also 2 researches about using maxent in very small area for endemic plant species:
Article Ecological Niche Modeling As a Tool for Conservation Plannin...
Article Predicting the potential geographical distribution of Nepeta...
Also here is a manual for openmodeller which contain bioclim and garp algorithm:
Method Ecological Niche Modeling - openModeller User Manual: a simple start
All the best Bro.
Dear Karim Omar,
In the picture I posted: Random test percentage-25, Replicates- 10. And that picture is Averages.
Your suggestion to remove the variables that contribute zero to the model is very interesting, I will try it. I used soil parameters when making the model, how do you feel about that?
I will definitely read the articles you recommend.
Best wishes
Bekhruz
Dear Bekhruz Khabibullaev
Removing the variable that has a bad contribution will improve the situation and prevent overfitting ... In small areas that have no difference between their sites in the soil properities it is preferable not to use it .. Soil is usually useful in large areas to show the variance
I understand your point, should we use variables at maxent first, remove zero-variable variables and use them again at maxent?
- Badges
- Science topic
- Similar topics
- Biological Science
- Physiology
- Physiological Processes
- Stress