You may go for diversity index and consider the slope orientaions in case you with different slope orientaions mainly North ans South fasing,  this migh help you to get more accurate and logical results

I think you may would like to look into this paper  and try associated coverage-based (Hill's number transformed diversity measure) on your data to see if you can get more consistent results.

Anne Chao, Chun-Huo Chiu, and T. C. Hsieh 2012. Proposing a resolution to debates on diversity partitioning. Ecology 93:2037–2051. http://dx.doi.org/10.1890/11-1817.1

Dear IMADEDDIN

thanks for your comment, I think this conflict results may return to equation of PARTITIONING methods as well as dependence or independence that there is between diversity components (especially Beta diversity). but i am not sure.

warmth regards

Dear Xinghua

Thank you for valuable comment and your reference.

When i used the q-diversity based species richness, the results show a unique pattern that i don't see in any reference until. e.g. one of the results was that diversity of rare species mainly determined by diversity components in small scales (plot and site scales).

I think this is not true, because typically rare species introduce as the species that have very low frequency, therefore with increase scale level (e.g. plot to landscape) diversity components must regularly should increased.

I don't fund any reason for this results.

On the other hand using the additive partitioning methods our result were quite opposite,

Now I try to understand why different partitioning methods have different results.

Best regards

@ Reza - I think perhaps you are interpreting the results incorrectly. Beta diversity is always dependent on alpha, as it is calculated as the difference in alpha diversity at 2 adjacent levels in the hierarchy. However beta calculated by the additive method is much larger than beta calculated by the multiplicative method and should be interpreted differently. When you use additive partitioning, beta gives you the arithmetic difference between two different levels, and is logically a larger number than alpha diversity for the same level. However, when you use multiplicative partitioning, beta is a multiplication factor. i.e., a beta value of greater than 2 indicates that diversity more than doubles from the local to the regional level. 

I have used PARTITION in my own research. I have attached my paper (Golodets et al) for you. I used multiplicative partitioning with q-diversity values.

I have also attached an article by Veech and Crist 2010 which provides a good explanation of the different methods and when they should be applied, and the recommendation for using q-diversity values

Article Plant diversity partitioning in grazed Mediterranean grassla...

Dear Carly

Thank you for valuable comment.

I read your paper several time.It is a valuable article both from a theoretical and a practical point of view. it is very good reference about q-diversity and based your paper i decided to crate a study about impact of change in amount of q-value (from 0 to 5) on different diversity component by Species Richness and Shannon Diversity indices.

In this study for determine the diversity patterns, I used the multiplicative richness.

when i compare the multiplication results with additive results, it was complete conflict. I think its better that i read more references about diversity partitioning as well as basic references about difference between additive and multiplicative partitioning.

I recently finished my master's degree in Iran. In addition my study was the first using partitioning methods in Iran. This methods not known in Iran and can not be used. I've noticed that a lot of things that I need to learn in this regard.

Thank you very much for your valuable advice and valuable references.

With Best wishes

Reza

Dear Reza,

I suggest you contact Thomas Crist directly thomas.crist @ miamioh.edu

He developed the PARTITION software. He's very approachable and willing to help. I received very helpful advice from him wben I wrote my paper.

By the way I think I am the first Israeli researcher to use the partitioning technique.

Carly

Dear Carly

Thank you very much for your excellent offer, I contact with Prof. Crist  as soon as possible and use from his guidance.

Good Luck

Reza