Eriogonum coloradense

January 24, 2025

Taxonomy

“The continued recognition of E. coloradense is probably dubious given what is already reduced here to synonymy under E. lonchophyllum” - Fl. North America

“It differs from E. lonchophyllum by its low, matted habit and capitate inflorescence.” - Eriogonum as a Rock Garden Plant

Habitat I

Habitat II

“Eriogonum coloradense is unusual in that it has an extremely broad ecological range. It has been documented on every soil texture, slope, and aspect. It has been found on sedimentary, granitic, and volcanic substrates, with Artemisia species (sagebrush) and Bouteloua gracilis (blue grama) and also with alpine cushion plants…”

- E. coloradense Technical Conservation Assessment

Occurrences

Modelled Suitable Habitat

Note raster pixels which have modelled suitability probabilities less than 0.45 have been ‘removed’ to save data so the map moves quickly online; the scattered points out to the East have had their suitability removed as well to save memory and keep the map interactive.

Computational Questions - Phase 1

  • How does the total amount of suitable habitat predicted vary among different spatial resolutions (90m, 30m, 10m)?
  • Can we predict plant abundance using gradient descent?
  • Can we delineate population boundaries using thresholding? How accurate are our estimates of a populations spatial extent?
  • Can we predict population occupancy of suitable habitat using naive landscape connectivity metrics?

Comp. Questions - Phase 1 - REU

  • How do predictions from existing records (herbaria and citizen science), compare to predictions from existing + supplemented records, and from supplemented and single iteration ground truthed?
  • Are pre-existing records biased? Do they represent the ‘lowest’ elevation occupied by each population? Does this affect model predictions?

Laboratory Questions - Phase 1

  • What does the species suitable habitat look like throughout the holocene at 10,000, 5,000, BC, and 0 AD? Can we use this to infer the age of colonization of the three major metapopulations?
  • Using spatial clustering techniques (DBSCAN, spatially weighted hierarchial) and landscape connectivity metrics can we predict the population genetic structure of the spatially-structured-population in the Elk Mountains?

Questions for respondents

“I love those who yearn for the impossible”


Goethe