Service Description: Deep-sea corals are of particular conservation concern due to their slow growth rates and vulnerability to disturbance. Predictive modeling of deep-sea coral habitat can aid conservation planning, inform management of offshore activities affecting the seafloor, and guide exploration. Modeling can also lend insights into the environmental factors driving the distribution of deep-sea corals, helping to build our understanding of how these unique ecosystems function. This downloadable data package is an experimental product made available to the research community and interested individuals and organizations for research and evaluation purposes. The National Oceanic and Atmospheric Administration assumes no responsibility for the use of these data.

Service ItemId: b76b4edc5e4b4030a0491bc711e4f7a8

Has Versioned Data: false

Max Record Count: 2000

Supported query Formats: JSON

Supports applyEdits with GlobalIds: False

Supports Shared Templates: True

All Layers and Tables

Layers:

• Pennatulacea, Sessiliflorae suborder - Predicted Habitat Suitability (0)

Description: This dataset depicts the predicted likelihood of suitable habitat for deep-sea corals in the suborder Sessiliflorae (order Pennatulacea) in the U.S. Northeast Atlantic and Mid-Atlantic as coral habitat suitability likelihood classes derived from a categorical reclassification of the logistic output of a maximum entropy (MaxEnt) model derived from coral presence locations and environmental predictor variables at 370.65m resolution. The reclassification used calibrated thresholds based on different hypothetical scenarios of the false positive:false negative cost ratio, that is, the relative cost of false positive versus false negative errors. Thresholds were chosen to minimize error rates when tested on data left out of the model-fitting process (cross-validation datasets), after weighting false positive and false negative error rates by the indicatd cost ratio. The thresholds were then used to create coral habitat suitability likelihood classes depicting several categories of predicted likelihood of habitat suitability. Coral habitat suitability likelihood classes can be qualitatively described as follows: Low Likelihood (Category 0, FP:FN cost ratio 1:1 but less than 2:1), Medium Likelihood (Category 2, FP:FN cost ratio > 2:1 but less than 5:1), High Likelihood (Category 5, FP:FN cost ratio > 5:1 but less than 10:1), Very High Likelihood (Category 10, FP:FN cost ratio > 10:1). MaxEnt is a type of statistical model. The outputs depicted here represent statistical model predictions, and as such may vary greatly in their accuracy, and may or may not reflect actual conditions at any given location. This information is provided as-is for research and evaluation purposes.

Copyright Text: Brian Kinlan, Matthew Poti, Dan Dorfman, Chris Caldow (NOAA/NOS/NCCOS/CCMA Biogeography Branch); Amy Drohan, Dave Packer, Martha Nizinski (NOAA/NMFS Northeast Fisheries Science Center)

Spatial Reference: 26918 (26918)

Initial Extent:

XMin: 365209.476319999
YMin: 3938578.04849
XMax: 1379604.21548
YMax: 5022284.27671
Spatial Reference: 26918 (26918)

Full Extent:

XMin: 411318.3281
YMin: 3987837.4225
XMax: 1333495.3637
YMax: 4973024.9027
Spatial Reference: 26918 (26918)

Units: esriMeters

Child Resources:   Info   SharedTemplates

Supported Operations:   Query   ConvertFormat   Get Estimates   Create Replica