| Nellie: Ecological Null Models for the PC |
|
Please note: This program was compiled for windows98 and it does not work very well under current versions of windows.
Tony Swenson and Evan Weiher
We are currently developing this program to work within the R environment.
Dept. of Biology
University of Wisconsin - Eau Claire
Ecological null models are powerful tools for testing non-standard hypotheses about patterns in ecological data. Here, we provide a free downloadable null model that has a battery of community-level tests.
When designing this program, we tried to minimize similarity and overlap with the other
null models that are currently available. Please also consider using:
EcoSim by Nick Gotelli and Gary
Entsminger
Nestedness Temperature
Calculator by Wirt Atmar and Bruce Patterson
Download Nellie
(version 1) here.
Please note that this is a first edition and it should be considered a beta release. We have tested the program for accuracy and stability, but you might find problems. Some of the procedures are extremely memory intensive, if the program keeps turning itself off (a euphemism for crashing), try rebooting your computer and running Nellie without other programs running simultaneously.
Features:
| Tests | questions addressed | additional output |
| Checkerboard Score (species co-occurrence) |
Do some species combinations occur less often
than expected by chance? |
ordered C scores for each species pair |
| Nestedness (Wright and Reeves method) | Are rare species more likely to be found in the most diverse observations? | |
| Variance in Species Richness | Is the variance in species richness less than expected by chance? | |
| Variance in Evenness Evar and EH -- both are independent of richness |
Is the variance in evenness less than expected by chance? | evenness for each observation & summary statistics |
| Variance in Biomass | Is the variance in biomass (or abundance) less than expected by chance? | |
| Guild Proportionality - Richness | Is the variance in the proportion of species from each of several guilds less than expected by chance? Is the proportion of species from each of several guilds greater or less than what was expected by chance? | guild proportions for each observation & summary statistics |
| Guild Proportionality - Biomass | Is the variance in the proportion of abundance from each of several guilds less than expected by chance? Is the proportion of abundance from each of several guilds greater or less than what was expected by chance? | guild proportions for each observation & summary statistics |
| Trait Maxima & Minima | Is the variance in trait maxima & minima less than expected by chance? Are the average trait maxima & minima less than or greater than expected by chance? | trait maxima & minima for each observation & summary statistics |
| Mean Nearest-Neighbor Distance in Trait Space | Is the among observation average of the community mean NND greater than or less than expected by chance (are coexisting species more or less different than expected by chance)? Is the among observation variance in mean NND less than expected by chance? | mean NND for each observation & summary statistics |
| Minimum Spanning Tree Distance in Trait Space | Is the average MST (morphological diversity) greater than or less than expected by chance (do coexisting species occupy a larger or smaller volume of morphological space than expected by chance)? Is the variance in MST less than expected by chance? | MST for each observation & summary statistics |
Randomization (Shuffling) Procedures:
| row and column constrained: | checkerboard swap method** | ** recommended default |
| knight's tour method | ||
| swenson's method | ||
| row constrained | ||
| column constrained | ||
| complete random |
Data input:
tab delimited text files
200 species by 200 observations
System Requirements:
Windows 95 or NT or better.
This was work was supported by a Faculty-Student collaboration grant from the University of Wisconsin - Eau Claire Office of University Research.
 |
 |