mmellis / rSPACE

Power analysis for occupancy-based monitoring
7 stars 1 forks source link

Feedback and suggestions #30

Closed technoautotroph closed 6 years ago

technoautotroph commented 8 years ago

Your article says: "The current implementation focuses on occupancy-based monitoring of territorial carnivores. We anticipate providing additional options and templates for designs and analyses as these are developed...We welcome feedback on both current implementations and future suggestions."

In spirit of expressed goals of the paper I recommend adding an option to clump or aggregate populations as the next priority for rSPACE. A piece of software that might be useful in this regard is secr: http://www.otago.ac.nz/density/SECRinR.html

Combing the spatial heterogeneity capabilities of rSPACE with the option to change distribution of range centres in the population as in secr would be a huge added advantage. I note that secr treats density as a homogeneous Poisson point process and the code could be relatively easy to bring into rSPACE. I've been trying to read the code for rSPACE to see if I could do this myself, but I'm not code savvy enough and this would take me a long-time to figure out.

I have some concerns about your paper that have come to light after many attempts to simulate data that might be applicable to my study.

The package will be of use to ecologists...The program is designed to provide a flexible shell in which alternative approaches for a replicated population simulation, sampling design and analysis can be customized to match a given scenario...and illustrate its use with an example for wolverine in the Northern U.S. Rocky Mountains

and

Standard occupancy-based power analyses are non-spatial and thus assume uniform distribution of individuals across the landscape (e.g. Bailey et al. 2007; Guillera-Arroita, Ridout & Morgan 2010; Guillera-Arroita & Lahoz-Monfort 2012), which is unrealistic of natural populations.

and

Our approach is unique in that it (1) incorporates available spatial information on habitat and species biology,

The paper suggests that wolverines are just used as an example and implies that the program is designed to be more flexible for ecologists. However, quoted segments from your paper above run in contrast to information that you haves since provided in one of our previous discussions:

rSPACE was designed for working with territorial carnivores, so the distribution should be much more regular, rather than randomly distributed...Right now, we don't have the code to test models for non-territorial species.

The limitations of rSPACE should have been made explicit in the paper. While it has the advantage of incorporating spatial heterogeneity into the simulation, it is designed to simulate and provide power analysis for a territorial species and does not allow one to account for different kinds of distributional patterns that other taxa may exhibit (see Efford MG, Dawson DK, Jhala YV, Qureshi Q 2016. Density-dependent home range size revealed by spatially explicit capture-recapture. Ecography 39: 676–688).

The spatial distribution of a territorial species may be unrealistic for less territorial taxa or species exhibiting a clustered distribution. It would be beneficial to discuss sampling design and power analysis limitations in terms of seasonal variation in distribution from clustered (e.g., mating in breeding ponds, leks, or hibernacula) to more of a dispersed pattern (as species may spread out and move into upland areas) and how rSPACE may or may not be able to handle this.

More specifically, the package will be of use to mammalian ecologists working with territorial species like wolverines and avian ecologists may be interested in some of the features offered as well. The applicable biology needs to be more clearly laid out. Future implementations of the package (hopefully?) may be of more broad utility for ecologists working with taxa that may exhibit different kinds of spatial ecology, such as herpetofauna.

"We anticipate providing additional options and templates for designs and analyses as these are developed."

  • Yay!

I do not want you to be disheartened by my feedback here and I appreciate the goals of rSPACE - it is off to a wonderful beginning! Your feedback has been wonderful, prompt, and very much appreciated. I hope that my feedback/review is received in a positive light - science works best when we most brutally attack our hypotheses/positions and it is in this vein that my feedback is provided. My hope is that my critique can be used to improve on a next iteration of rSPACE.

Overall, expressed limitations of rSPACE is missing in the paper and may have been missed in the enthusiasm. which I can totally respect as this is an exciting project.

I have funding to work on the simulation work for my project and I am putting out more proposals. Please let me know if you would be interested in forming a collaboration to upgrade rSPACE to make it more compatible for analysis on herpetofauna.

"Through this effort, we hope to foster careful monitoring design and encourage multijurisdictional collaboration to enable large-scale monitoring efforts."

There are huge efforts of this type being undertaken by herpetologists (e.g., http://armi.usgs.gov/) where an rSPACE environment allowing for different kinds of distributions will be of great value.

Thank you!!!