California black rail - Species Distribution Model, DRECP

These data are statistical model outputs for Black rail (ssp. California black rail) (Laterallus jamaicensis coturniculus) species distribution, completed by Frank Davis’ Biogeography Lab at UC Santa Barbara. 
 
Based on examination of species observation data and consultation with biologists, CBI used the model's broad extent output masked to within 10 km of the Colorado River and the following USFS ecoregion subsections: 322Cd, 322Cc, 322Cb. The binary output was subsequently restricted to areas within 1km of the Salton Sea, a perennial stream/river, or artificial waterway based on the NHD flowlines dataset (canals, ditches, connectors, pipelines and other linear-based water features unlikely to provide suitable habitat were excluded from the proximity analysis) OR areas that were within 1km of the following vegetation type: 'Southwestern North American Riparian Flooded and Swamp Forest'. An additional area in the south western corner of the Salton Sea (along Hwy 86/78) was manually removed due to the low probability of it being suitable habitat.

Based on consultation with biologists, additional areas to the east of the agricultural fields in the Imperial Valley were determined not to be suitable habitat and were removed. These modifications were made by CBI on 6/25/2013.

The modifications listed above were performed post-modeling. Consequently the AUC generated for this model do not reflect the modified extent.

The UCSB Biogeography Lab used Maxent to generate predictions of habitat occupancy for ~70 species for the CA Energy Commission’s project “Cumulative Biological Impacts Framework for Solar Energy in the CA Desert”, 500-10-021.

Species distribution models were produced at 270 m resolution using a subset of 22 environmental variables. Models were evaluated with 10-fold cross validated AUC scores. Both continuous probability surfaces and binary layers are available for each species modeled. Binary layers depicting predicted suitable habitat were derived using the equal training sensitivity and specificity threshold. 
 
For Laterallus jamaicensis, Max Sensitivity + Specificity threshold = 0.0638; best AUC =0.9954; mean AUC = 0.9917. 

For more information on the environmental variables used, modeling process, and model diagnostics, please refer to the supporting document “Data Descriptions: UCSB DRECP Species Distribution Models, June 12, 2013" provided by Frank Davis.