Community Dynamics Toolbox

h1. Workshop Notes

are on etherpad: https://epad.nceas.ucsb.edu/p/commdyn-20140105

h1. Breakout session 1: Metrics brainstorming

* what are you currently using

* what would you like to use

* how widely is it used

* can it be applied to different biological community datasets (sampling approach)

* is it already coded {in R}

h2. Metrics

# *Diversity* (all of these are generally in R, mostly in vegan)

## Jaccard index

## Simpson's diversity 

## Shannons index

## Turnover - different ways to calculate

## Dominance 

## Evenness

## Richness

## Rank abundance shift

## Proportion of overall diversity

## Beta diversity

# *Community metrics/ordination*

## NMDS (vegan)

## PCA (vegan)

## Bray curtis (vegan)

## Variance tracking, quantify variability change

## Position in ordination-space

# *Spatial*

## patch scale 

## spatial autoregression

## Endemism

## Summary of species' positions within their ranges

## meta community statistics

# *Mechanistic models*

## MAR, needs driver matrix, problem auto-corelation, mostly fresh water or marine (Eli Holmes has state-space MAR in R implemented, not sure if it's on CRAN)   http://cran.r-project.org/web/packages/MARSS/index.html

## MANOVA (vegan? Also, permanova is in vegan)

## Ecosystem function (e.g. N deposition)

## interaction population models - inter specific competition (Ben Bolker's book and corresponding package)

## Economically/legally relevant metrics (e.g. Maximum sustainable yield)

# *Food webs*

## connectance

## network analysis

# *Traits/phylogentic*

## functional/phylogenetic diversity

## species aggregation (functional groups, trophic levels

## phylogenetic dispersion

## Native/exotic

## Phylogeographic history

# *Temporal indices*

## species turnover

## rate of return

## Variance ratio

## Mean-variance scaling

## Spectral analysis

## Regresssion windows (strucchange)

## time series models of abundance -- metric would be parameters of model

# *null models*

# *Comparative analysis of small noise vs large noise systems. What drives differences?*

h2. Issues:

# length of time series relative to lifespan of organisms

> WMI toolbox

# high frequency data needed

> sample too frequently then don't see signal, sample to far about miss all dynamics

# type of variable being measured

> abundance, biomass, production

# Rare species as background noise

h2. Coded in R

* Richness/diversity metrics: http://cran.r-project.org/web/packages/vegan/index.html

* Diversity metrics (alpha, beta, gamma): http://cran.r-project.org/web/packages/vegetarian/index.html

* Hubble metrics: http://cran.r-project.org/web/packages/untb/index.html

* Leading indicators, variance, autocorrelation, skew, heteroscedasticity: http://cran.at.r-project.org/web/packages/earlywarnings/index.html

not yet coded:

* state-space models and community level resilience

h1. Breakout Session 2: Identify research questions:

# Data set transformation to allow compute of many metrics

# Time series analysis of community level metrics (consider higher freq data too)(earlywarnings R package)

# New R code for capturing climate variance at seasonal and interannual scales and residuals

# Review of non-stationarity

## How do you figure out if 3 or more spp are changing together?

## Developing/adopting/adapting new techniques

h2. Report back from working groups

h3. Group 1 and 2

* Downloaded datasets, created common format

* Eric made script to calculate metrics over time

* lots of time to clean up and reformat the data

* Dataformat most useful: long format, some standardization is useful: NAN, date format, separate file with site conditions, spelling of words, outliers, how to deal with large missing data chunks, functional groups for species

* How to deal with data where individuals have not been keyed to species, but higher and varying taxonomic ranks

* more exploration is necessary

* specific site conditions were partly unknown (e.g. CAP landuse)

* Continue work on development of workflows for data cleanup

* Ecological analysis of temporal change data sets

h3. Group 3

* Eric's LTER project to develop the R model for analyzing more spatial variability, could collaborate with long term data

h3. Group 4

*Temporal and spatial variance*

* ecosystems strongly influenced by temporal and spatial variation bi-variate plot

* limitations: other data, e.g. biochemical conditions

* time turnover

* spatial heterogeneity

* cross site LTER datasets to see if there is some traction to the idea

* datasets in long form because they need to be manipulated a little for each new exploratory step

* Work on common data format for various communities

* Work on script to calculate variance in space and time metrics for each of those communities 

* One limitation was that many available data don't lend themselves to both the spatial and temporal analyses

*Variance partitioning*

* climate, intra annual seasonal, inter annual

* new metrics: variance components analysis

* explore different temporal scale

* which part of the annual 'weather' is important