|
Activity Number:
|
504
- Spatial Modeling on the Sphere and Over Large Domains
|
|
Type:
|
Contributed
|
|
Date/Time:
|
Thursday, August 6, 2020 : 10:00 AM to 2:00 PM
|
|
Sponsor:
|
Section on Statistics and the Environment
|
|
Abstract #313459
|
|
|
Title:
|
Test for Isotropy on a Sphere Using Spherical Harmonic Coefficients
|
|
Author(s):
|
Indranil Sahoo* and Joseph Guinness and Brian Reich
|
|
Companies:
|
Virginia Commonwealth University and Cornell University and North Carolina State University
|
|
Keywords:
|
spatial statistics;
Anisotropy;
spherical harmonics
|
|
Abstract:
|
Analyses of geostatistical data are often based on the assumption that the spatial random field is isotropic. Since many applications consider global data, it is necessary to check the assumption of isotropy on a sphere. This talk covers a testing procedure for spatial isotropy on a sphere. The data are first projected onto the set of spherical harmonic functions. Under isotropy, the spherical harmonic coefficients are uncorrelated, but are correlated if the underlying fields are not isotropic. This motivates a test based on the sample correlation matrix of the spherical harmonic coefficients. Our method requires temporal replication in the data and hence, is applicable to many data sets in Earth sciences. We show how temporal correlation affects the test and provide methods for handling such correlation. Extensive simulations show that the test proposed has correct Type-I errors under different scenarios and is powerful in detecting deviations from isotropy. The method is applied to near-surface air temperature fields which are clearly not isotropic. We propose a sequence of anisotropic models with increasing complexity and test how well they capture the anisotropy in the fields.
|
Authors who are presenting talks have a * after their name.
