ha0ye / portalds

Dynamic Stability functions and Portal analysis

portalDS

Overview

The goal of portalDS is to apply “dynamic stability” (Ushio et al. 2018) to the long-term rodent time series at Portal, to test whether this indicator of whole-ecosystem change predicts or lines up with abrupt community shifts.

Project Organization

This project is set up as an R package that implements functions for computing “dynamic stability”, mostly following Ben Marwick’s rrtools setup:

• The maizuru-dynamic-stability vignette reproduces the Maizuru Bay analysis from Ushio et al. 2018 as an example.
• The analysis folder contains R scripts and objects for the Portal dynamic stability analysis.
• The data, output, and figures folders contain data, results, and figures for both the Portal and the Maizuru Bay analysis.

Installation

You can install the current version of portalDS from GitHub with:

# install.packages("remotes")
remotes::install_github("ha0ye/portalDS")

portalds's Issues

Roadmap

Short-term goals:

• develop Drake plan version of workflow
  • figure out parallelization within targets
• update Maizuru Rmarkdown
• round off tests
• implement simulations

Medium-term goals:

• develop and check alternative metrics
• compare against DLM approach for estimating dynamic Jacobians
• outline paper
• draft preprint
• split package components for archiving

Long-term goals:

• paper submission
• methods writeup

some questions about the function "compute_smap_coeffs"

Hi, Mr Ye. I am running the code you provided me. It just fits my research. And thank you again. It helped me a lot. But I had some problems and I wish you could help me if you have time.
1．I running function “compute_smap_coeffs” successfully. But it have “NA” value the first few rows and I don‘t know the reason. If you had seen this before
2. The results I got make me confused. This picture is my result. “SP1”,“SP2”,”SP3”…. represent different species in diatom community.
And I guess:
SP1.SP1 means the intraspecific density restriction of SP1->SP2
SP1.SP2 means the interaction strength of SP1->SP2.
But I don’t know my understanding right or wrong so I want to check with you

Besides that, it was difficult for me to guess what SP1.const meant, if you could tell me about this I would be very grateful.

Add more ways to specify parameter sets for the analysis

Since there are multiple steps to the full dynamic stability analysis, being able to load and save the configuration of parameters would be handy for reproducibility reasons. (Also, to make sure that the description in the paper matches the results presented.)

• create a data structure for the params
• include the params data structure with saved results
• use the data structure for the Portal and Maizuru example runs
• be able to read in the params from an external file (yaml or some other format? @juniperlsimonis might have some suggestions ~ portalcasting)

there is an error when running the compute_ccm function

Hello, I found there is a bug when running the compute_ccm function. I have checked the codes and found that if we changed the following codes in the compute_ccm function, the codes could be run successfully:

The origin codes in the compute_ccm function:

   lib_ts <- simplex_results[[from_idx,"data"]]$abundance
   pred_ts <- simplex_results[[to_idx,"data"]]$abundance 

And i replaced them using the following codes:

    lib_ts <- simplex_results$data[[from_idx]]$abundance
    pred_ts <- simplex_results$data[[to_idx]]$abundance 

I hope it will be helpful for your package.