Use of other diversity indices

[FerreiraPSM]

Is it possible to insert another index of phylogenetic diversity (phylogenetic endemism) as a conservation objective instead of Faith's phylogenetic diversity?

[jeffreyhanson]

Not currently, but I could look into adding this functionality for you? Do you have a specific paper or blog post that outlines the exact objective function you need? AFAIK, phylogenetic endemism is a static/pre-computed quantity that pertains to each planning unit. If so, you could manually compute this metric for each planning unit and use phylogenetic endemism as a feature in your spatial prioritisation? What do you think?

[FerreiraPSM]

Dear Dr. Jeff, Thank you very much for your reply. I'm using the function provided by Greg Guerin < https://github.com/GregGuerin/biomap/blob/master/phylogenetic.endemism.R> that was published in this paper <https://www.sciencedirect.com/science/article/pii/S2352711015000126> (also this one <https://onlinelibrary.wiley.com/doi/full/10.1111/j.1365-294X.2009.04311.x>). Your suggestion makes complete sense. I was just wondering if we might add a function add_max_phyloEnd_objective similar with the current add_max_phylo_objective, so I can use the phylogenetic endemism to prioritize phylogenetic endemism instead of PD. By doing so, I could understand whether some regions have a unique phylogenetic history instead of more or less diversity. Thank you again for helping me. Best, Paulo Em qua, 7 de ago de 2019 às 10:41, Jeff Hanson <notifications@github.com> escreveu:

…

Not currently, but I could look into adding this functionality for you? Do you have a specific paper or blog post that outlines the exact objective function you need? AFAIK, phylogenetic endemism is a static/pre-computed quantity that pertains to each planning unit. If so, you could manually compute this metric for each planning unit and use phylogenetic endemism as a feature in your spatial prioritisation? What do you think? — You are receiving this because you authored the thread. Reply to this email directly, view it on GitHub <#113?email_source=notifications&email_token=AJ3OGNUEWLGY23B3JL3CV3DQDLGHJA5CNFSM4IJXLPC2YY3PNVWWK3TUL52HS4DFVREXG43VMVBW63LNMVXHJKTDN5WW2ZLOORPWSZGOD3YN3WY#issuecomment-519101915>, or mute the thread <https://github.com/notifications/unsubscribe-auth/AJ3OGNU5ASBQODXSTTJKKDDQDLGHJANCNFSM4IJXLPCQ> .

-- *Paulo Sérgio Monteiro Ferreira* Professor da Escola Estadual de Ensino Fundamental Professor Rangel - Ingá Doutorando em Ecologia e Recursos Naturais pela Universidade Federal do Ceará Mestre em Ecologia e Conservação pela Universidade Estadual da Paraíba Licenciado e Bacharel em Ciências Biológicas pela Universidade Estadual da Paraíba Currículo Lattes: http://lattes.cnpq.br/4024572354364629 Facebook: https://www.facebook.com/paulosergio.monteiro.39 Itatuba-PB Tel: (83) 987081172/CEP: 58378000

[jeffreyhanson]

Great - thank you very much for the links - that helps a lot! I think we could implement some phylogenetic endemism stuff but I want to make sure that I understand exactly what you want to achieve with the prioritisation. Please correct me if anything I've said isn't right. I can think of two approaches that phylogenetic endemism could be used to generate a prioritisation:

1. Calculate a phylogenetic endemism score for each planning unit separately. For a given budget, simply select the planning units that will give you the highest sum of the phylogenetic endemism scores. This could be achieved using the maximum utility objective function (https://prioritizr.net/reference/add_max_utility_objective.html) and a single feature layer denoting the phylogenetic endemism scores for each planning unit. The issue with this approach is that the planning units with the highest phylogenetic endemism scores might all contain exactly the same branches, and so simply selecting the planning units with the highest phylogenetic endemism scores would "double up" on the same branches (in a similar manner to prioritising areas using species richness might double up on the same species) --- this wouldn't follow the principle of complementarity.

2. For a given budget, find the set of planning units that will maximize the total phylogenetic endemism of the species that are adequately represented in the solution. In other words, find a solution that will provide an adequate amount of habitat for a set of species to maximize the overall phylogenetic endemism of the phylogenetic tree when considering only the species that have an adequate amount of habitat protected. This would be equivalent to maximizing phylogenetic diversity, but instead of using the length of each phylogenetic branch to parametrise its conservation value we use the length / spatial distribution size. If the first approach is a phylogenetic version of the maximum utility objective function, I suppose this approach would be akin to a phylogenetic version of the maximum features objective function (https://prioritizr.net/reference/add_max_features_objective.html).

Do either of these approaches match up with what you're trying to do? I would generally caution against the first approach though, because it doesn't follow the principle of complementarity. Please let me know if what I've written doesn't make sense, and I can try explaining it more clearly.

[FerreiraPSM]

Thank you very much for your availability and kindness in responding.

You're right. Your second suggestion fits exactly what I would like to do.

How would it be possible to modify the maximum features objective function to achieve this result?

Best regards
Paulo

[jeffreyhanson]

No worries - ok sounds awesome! The second approach would involve modifying the phylogenetic diversity objective function so that it uses phylogenetic branch length / phylogenetic branch spatial distribution size instead of just phylogenetic branch length to specify how "valuable" each phylogenetic branch is for conservation. This should be fairly straight forward (in theory).

How urgently do you need this? If you need it ASAP, you could implement this yourself by taking your phylogenetic tree and manually rescaling the branches according to their distribution size, and then just using the phylogenetic diversity objective function. Or if you don't mind waiting a week or so, I could implement this new functionality into prioritizr for you?

[FerreiraPSM]

Yes, I can wait a while!

It will be amazing if you can do that.

Thank you so much for your availability!

[jeffreyhanson]

No worries! I've just pushed a new version to GitHub with the add_max_phylo_end_objective function to maximize phylogenetic endemism (see https://prioritizr.net/reference/add_max_phylo_end_objective.html). You can download this with the code devtools::install_github("prioritizr/prioritizr"). Please feel free to reopen the issues if you have any issues or questions.