Graph-Guided Regularized Regression of Pacific Ocean Climate Variables to Increase Predictive Skill of Southwestern U.S. Winter Precipitation

Understanding the physical drivers of seasonal hydroclimatic variability and improving predictive skill remains a challenge with important socioeconomic and environmental implications for many regions around the world. Physics-based deterministic models show limited ability to predict precipitation...

وصف كامل

محفوظ في:
التفاصيل البيبلوغرافية
المؤلف الرئيسي: Tejedor, Alejandro (author)
مؤلفون آخرون: Stevens, Abby (author), Willett, Rebecca (author), Mamalakis, Antonios (author), Foufoula-Georgiou, Efi (author), Randerson, James T. (author), Smyth, Padhraic (author), Wright, Stephen (author)
منشور في: 2021
الوصول للمادة أونلاين:https://depot.sorbonne.ae/handle/20.500.12458/1390
الوسوم: إضافة وسم
لا توجد وسوم, كن أول من يضع وسما على هذه التسجيلة!
_version_ 1857415062784835584
author Tejedor, Alejandro
author2 Stevens, Abby
Willett, Rebecca
Mamalakis, Antonios
Foufoula-Georgiou, Efi
Randerson, James T.
Smyth, Padhraic
Wright, Stephen
author2_role author
author
author
author
author
author
author
author_facet Tejedor, Alejandro
Stevens, Abby
Willett, Rebecca
Mamalakis, Antonios
Foufoula-Georgiou, Efi
Randerson, James T.
Smyth, Padhraic
Wright, Stephen
author_role author
dc.creator.none.fl_str_mv Tejedor, Alejandro
Stevens, Abby
Willett, Rebecca
Mamalakis, Antonios
Foufoula-Georgiou, Efi
Randerson, James T.
Smyth, Padhraic
Wright, Stephen
dc.date.none.fl_str_mv 2021
2023-03-09T10:58:17Z
2023-03-09T10:58:17Z
dc.identifier.none.fl_str_mv 0894-8755
1520-0442
https://depot.sorbonne.ae/handle/20.500.12458/1390
10.1175/JCLI-D-20-0079.1
dc.language.none.fl_str_mv en
dc.relation.none.fl_str_mv Journal of Climate
dc.title.none.fl_str_mv Graph-Guided Regularized Regression of Pacific Ocean Climate Variables to Increase Predictive Skill of Southwestern U.S. Winter Precipitation
dc.type.none.fl_str_mv Controlled Vocabulary for Resource Type Genres::text::periodical::journal::contribution to journal::journal article
description Understanding the physical drivers of seasonal hydroclimatic variability and improving predictive skill remains a challenge with important socioeconomic and environmental implications for many regions around the world. Physics-based deterministic models show limited ability to predict precipitation as the lead time increases, due to imperfect representation of physical processes and incomplete knowledge of initial conditions. Similarly, statistical methods drawing upon established climate teleconnections have low prediction skill due to the complex nature of the climate system. Recently, promising data-driven approaches have been proposed, but they often suffer from overparameterization and overfitting due to the short observational record, and they often do not account for spatiotemporal dependencies among covariates (i.e., predictors such as sea surface temperatures). This study addresses these challenges via a predictive model based on a graph-guided regularizer that simultaneously promotes similarity of predictive weights for highly correlated covariates and enforces sparsity in the covariate domain. This approach both decreases the effective dimensionality of the problem and identifies the most predictive features without specifying them a priori. We use large ensemble simulations from a climate model to construct this regularizer, reducing the structural uncertainty in the estimation. We apply the learned model to predict winter precipitation in the southwestern United States using sea surface temperatures over the entire Pacific basin, and demonstrate its superiority compared to other regularization approaches and statistical models informed by known teleconnections. Our results highlight the potential to combine optimally the space–time structure of predictor variables learned from climate models with new graph-based regularizers to improve seasonal prediction.
id sorbonner_1f9ac7007dcf8b2d12ddadd145a7f4e8
identifier_str_mv 0894-8755
1520-0442
10.1175/JCLI-D-20-0079.1
language_invalid_str_mv en
network_acronym_str sorbonner
network_name_str Sorbonne University Abu Dhabi repository
oai_identifier_str oai:depot.sorbonne.ae:20.500.12458/1390
publishDate 2021
repository.mail.fl_str_mv
repository.name.fl_str_mv
repository_id_str
spelling Graph-Guided Regularized Regression of Pacific Ocean Climate Variables to Increase Predictive Skill of Southwestern U.S. Winter PrecipitationTejedor, AlejandroStevens, AbbyWillett, RebeccaMamalakis, AntoniosFoufoula-Georgiou, EfiRanderson, James T.Smyth, PadhraicWright, StephenUnderstanding the physical drivers of seasonal hydroclimatic variability and improving predictive skill remains a challenge with important socioeconomic and environmental implications for many regions around the world. Physics-based deterministic models show limited ability to predict precipitation as the lead time increases, due to imperfect representation of physical processes and incomplete knowledge of initial conditions. Similarly, statistical methods drawing upon established climate teleconnections have low prediction skill due to the complex nature of the climate system. Recently, promising data-driven approaches have been proposed, but they often suffer from overparameterization and overfitting due to the short observational record, and they often do not account for spatiotemporal dependencies among covariates (i.e., predictors such as sea surface temperatures). This study addresses these challenges via a predictive model based on a graph-guided regularizer that simultaneously promotes similarity of predictive weights for highly correlated covariates and enforces sparsity in the covariate domain. This approach both decreases the effective dimensionality of the problem and identifies the most predictive features without specifying them a priori. We use large ensemble simulations from a climate model to construct this regularizer, reducing the structural uncertainty in the estimation. We apply the learned model to predict winter precipitation in the southwestern United States using sea surface temperatures over the entire Pacific basin, and demonstrate its superiority compared to other regularization approaches and statistical models informed by known teleconnections. Our results highlight the potential to combine optimally the space–time structure of predictor variables learned from climate models with new graph-based regularizers to improve seasonal prediction.2023-03-09T10:58:17Z2023-03-09T10:58:17Z2021Controlled Vocabulary for Resource Type Genres::text::periodical::journal::contribution to journal::journal article0894-87551520-0442https://depot.sorbonne.ae/handle/20.500.12458/139010.1175/JCLI-D-20-0079.1enJournal of Climateoai:depot.sorbonne.ae:20.500.12458/13902023-06-14T09:41:18Z
spellingShingle Graph-Guided Regularized Regression of Pacific Ocean Climate Variables to Increase Predictive Skill of Southwestern U.S. Winter Precipitation
Tejedor, Alejandro
title Graph-Guided Regularized Regression of Pacific Ocean Climate Variables to Increase Predictive Skill of Southwestern U.S. Winter Precipitation
title_full Graph-Guided Regularized Regression of Pacific Ocean Climate Variables to Increase Predictive Skill of Southwestern U.S. Winter Precipitation
title_fullStr Graph-Guided Regularized Regression of Pacific Ocean Climate Variables to Increase Predictive Skill of Southwestern U.S. Winter Precipitation
title_full_unstemmed Graph-Guided Regularized Regression of Pacific Ocean Climate Variables to Increase Predictive Skill of Southwestern U.S. Winter Precipitation
title_short Graph-Guided Regularized Regression of Pacific Ocean Climate Variables to Increase Predictive Skill of Southwestern U.S. Winter Precipitation
title_sort Graph-Guided Regularized Regression of Pacific Ocean Climate Variables to Increase Predictive Skill of Southwestern U.S. Winter Precipitation
url https://depot.sorbonne.ae/handle/20.500.12458/1390