Development of species distribution models (SDMs) and application of them has been expanding very rapidly over the past few years. Often based on simple occurrence data like that provided by the Fishes of Texas project, they summarize and make these data sets useful in new ways and across large spatial extents. They have proven useful in diverse applications such as conservation planning, climate change studies, disease ecology, invasive species research, and community ecology.
As a first step toward many future landscape-scale geospatial analyses using Fishes of Texas data, we developed powerful predictive computer models of species' distributions using commonly accepted practices and modeling algorithms and provide them here so that others may use them in their own research and applications. Our models provide continuous coverages of probabilities of species occurrences across all cells of a fine-scale grid extending across all of Texas, thus effectively 'filling in the blanks' between the actual occurrences that we know to be distributed in non-random ways as a result of diverse historic factors such as collectors' interests, gears, landowner permission, etc.
We developed these models using only the most precisely located recent occurrence records in the Fishes of Texas database together with recent climate and physical environmental data. These models have now been thoroughly tested and demonstrated to be powerful predictors of actual occurrences under current conditions. They were constructed in such a way that the probability values in the models can be interpreted as indicators of suitability of habitat that are mostly independent of large scale land and water development influences such as diversions or dams.
Mapped modeled probabilities of species occurrences can be viewed and model outputs formatted for analysis may be downloaded via the model class table below.
When using models, please cite as suggested in the How to cite models section of the Fishes of Texas documentation.
Model classes available for Download : Fish
Model classes available for Download : Fish
| Class | available | spatial extent | resolution | model construction and interpretation reference | numbers of species available | Download Page |
|---|---|---|---|---|---|---|
| 01 | Yes | political boundary of Texas | 30 arc-seconds | Model Class 01 | 95 | Model Class 01 Downloads |
| 02 | No | The extent used for SDM construction included NHDplusV2 data regions 11 (Ark-Red-White), 12 (Texas), and 13 (Rio Grande). Model results from this extent was restricted to the political boundary of Texas for post-processing and symbolism. | 30 arc-seconds | Model Class 02 | 51 | Model Class 02 Downloads |
| 03 | No | Species Specific: within major river basin that intersect records | National Hydrography Dataset Plus Catchments | Model Class 03 | 0 | nla |
How to Cite Models:
Citation of models will be dependent on model class as model development will involve specific personnel contingent on application.
Citation for Using Class 01 Models
Labay, Ben J., Dean A. Hendrickson, and Adam E. Cohen. 2012. Fishes of Texas Project Class 01 Species Distributions Models (models). Published by Texas Natural History Collection, a division of Texas Natural Science Center, University of Texas at Austin. Accessed (insert date of data access).
There are many published applications using the maximum entropy algorithm and similar protocols as employed here (see SDM Background and Literature), but the most relavent publication to use for methodology citation would be:
Labay, B.J., A.E. Cohen, B. Sissel, D.A. Hendrickson, F.D. Martin, and S. Sarkar., 2011. Assessing historical fish community composition using surveys, historical collection data, and species distribution models. PLoS ONE 6, e25145.