Incorporating Correlated Nugget Effects in Multivariate Spatial Models: An Application to Argo Ocean Data
Damilya Saduakhas, David Bolin, Xiaotian Jin, Alexandre B. Simas, Jonas Wallin · 2025
Read Incorporating Correlated Nugget Effects in Multivariate Spatial Models: An Application to Argo Ocean Data by Damilya Saduakhas, David Bolin, Xiaotian Jin, Alexandre B. Simas, Jonas Wallin (2025) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2506-03042.txt.
Description
Saduakhas and co-authors (2025) improve the statistical models used to map ocean temperature and salinity from the Argo float program. Such maps need models that capture tangled spatial patterns across several variables at once. Standard models treat the small-scale noise in each variable as independent. This team built hierarchical Matérn models, in plain and heavier-tailed versions, that let the noise terms correlate.
In simulations and on real Argo data, adding the correlated noise improved both parameter estimates and spatial predictions. On the global data, the new model estimated a lower correlation between the temperature and salinity fields than the older models did. The authors argue that assuming independent measurement errors makes the fields look more alike than they are. Sorting the noise from the signal lets fine-scale ocean patterns show through more cleanly.
Catalog record
| CREATOR | Damilya Saduakhas, David Bolin, Xiaotian Jin, Alexandre B. Simas, Jonas Wallin |
|---|---|
| YEAR | 2025 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 80 KB |
| VERSION | 1.0 |
| LICENSE | http://creativecommons.org/licenses/by/4.0/ |
| RIGHTS | CC BY |
| SOURCE | arXiv |
Source: https://arxiv.org/abs/2506-03042