Spatial autocorrelation analyses and species distribution modeling uncover biases in citizen science engagement patterns across Texas’ Blackland Prairies

Authors: Daniel LeVine*, University of Texas - Austin, Kelley Crews, University of Texas at Austin, Jennifer Miller, University of Texas at Austin
Topics: Spatial Analysis & Modeling, Biogeography, Human-Environment Geography
Keywords: volunteered geographic information (VGI), citizen science, species distribution modeling, spatial ecology, spatial autocorrelation, tallgrass prairie
Session Type: Paper
Day: 4/4/2019
Start / End Time: 8:00 AM / 9:40 AM
Room: Roosevelt 2, Marriott, Exhibition Level
Presentation File: No File Uploaded


The recent widespread adoption of citizen science represents a significant source of species occurrence data for spatial ecology and conservation biology. Many opportunities for the conservation of rare species and species assemblages have arisen from the plethora of data made available by sources such as iNaturalist, eBird, and MonarchWatch. However, biases towards large population centers and charismatic, easily-identifiable species present inconsistencies in the spatial distributions of citizen science data. To test both the accuracy and applicability of such datasets, this research identifies spatial patterns in citizen science observations across the Texas Blackland Prairies ecoregion, an area containing remnants of one of the rarest tallgrass prairie ecosystem in North America. The occurrences of citizen science data in public and private lands were discerned and mapped for three plant species representative of the Texas Blackland Prairies, 2 invasive grasses (Bothriochloa ischaemum and Sorghum halepense) and 1 native grass (Bothriochloa laguroides var. torreyana). The distribution of these species occurrences were assessed with regards to both large population centers and public/private land using bivariate local spatial autocorrelation analysis (LISA). To further assess the applicability of species occurrence datasets in conservation planning, blackland prairie grass species were modeled using a simple and commonly-employed species distribution modeling (SDM) approach, Maxent. The findings suggest opportunities to target spatial gaps and inconsistencies in the vast datasets offered by citizen science observations. This research provides valuable insight into the use of citizen science data for the conservation of vulnerable environments across public and private lands.

Abstract Information

This abstract is already part of a session. View the session here.

To access contact information login