Skip to content
Home
About Us
Resources
Profiles Metrics
Authors Directory
Institutions Directory
Top Authors
Top Institutions
Top Sponsors
AI Digest
Contact Us
Menu
Home
About Us
Resources
Profiles Metrics
Authors Directory
Institutions Directory
Top Authors
Top Institutions
Top Sponsors
AI Digest
Contact Us
Home
About Us
Resources
Profiles Metrics
Authors Directory
Institutions Directory
Top Authors
Top Institutions
Top Sponsors
AI Digest
Contact Us
Menu
Home
About Us
Resources
Profiles Metrics
Authors Directory
Institutions Directory
Top Authors
Top Institutions
Top Sponsors
AI Digest
Contact Us
Publication Details
AFRICAN RESEARCH NEXUS
SHINING A SPOTLIGHT ON AFRICAN RESEARCH
computer science
Untangling drivers of species distributions: Global sensitivity and uncertainty analyses of MaxEnt
Environmental Modelling and Software, Volume 51, Year 2014
Notification
URL copied to clipboard!
Description
Untangling drivers of systems and uncertainty for species distribution models (SDMs) is important to provide reliable predictions that are useful for conservation campaigns. This is particularly true for species whose habitat is threatened by climate change that enhances the uncertainty in future species distributions. Global sensitivity and uncertainty analyses (GSUA) is a robust method to globally investigate the uncertainty of SDMs and the importance of species distributions' drivers in space and time.Here we apply GSUA to MaxEnt that is one of the popular presence-only SDMs. We consider the Snowy Plover (Charadrius alexandrinus nivosus) (SP) in Florida that is a shorebird whose habitat is affected by sea level rise due to climate change. The importance of intrinsic and exogenous input factors to the uncertainty of the species distribution is evaluated for MaxEnt. GSUA is applied for three projections of the habitat (2006, 2060, and 2100) according to the A1B sea level rise scenario. The large land cover variation determines a moderate decrease in habitat suitability in 2060 and 2100 prospecting a low risk of decline for the SP. The regularization parameter for the environmental features, the uncertainty into the classification of salt-marsh, transitional marsh, and ocean beach, and the maximum number of iterations for the model training are in this order the most important input factors for the average habitat suitability. These results are related to the SP but, in general MaxEnt appears as a very non-linear model where uncertainty mostly derives from the interactions among input factors.The uncertainty of the output is a species-specific variable. Thus, GSUA need be performed for each case considering local exogenous input factors of the model. GSUA allows quantitative informed species-management decisions by providing scenarios with controlled uncertainty and confidence over factors' importance that can be used by resource managers. © 2013 Elsevier Ltd.
Authors & Co-Authors
Convertino, Matteo
United States, Minneapolis
University of Minnesota Twin Cities
United States, Gainesville
University of Florida
Muñoz-Carpena, Rafael
United States, Gainesville
University of Florida
Chu-Agor, Maria Librada
United States, St. Louis
Saint Louis University
Kiker, Gregory A.
United States, Gainesville
University of Florida
Linkov, Igor
United States, Vicksburg
U.s. Army Engineer Research and Development Center
United States, Pittsburgh
Carnegie Mellon University
Statistics
Citations: 148
Authors: 5
Affiliations: 5
Identifiers
Doi:
10.1016/j.envsoft.2013.10.001
ISSN:
13648152
Research Areas
Environmental
Study Approach
Quantitative