Spatially penalized regression for dependence analysis of rare events: A study in precipitation extremes

Debasish Das, Auroop Ganguly, Snigdhansu Chatterjee, Vipin Kumar, Zoran Obradovic

Research output: Contribution to conferencePaper

1 Citation (Scopus)

Abstract

Discovery of dependence structure between precipitation extremes and other climate variables (covariates) within a smaller spatial and temporal neighborhood is an important step in better understanding the drivers of this complex phenomenon as well as short-term prediction of extremes occurrence. Apart from the inherent spatio-temporal variability of the dependence, it is further complicated by the availability of the covariates at different vertical levels. The above problem can be split into three different sub-problems. Firstly, a spatio-temporal neighborhood of influence has to be discovered, which can be different for different locations. Secondly, the dependence structure between the precipitation extremes and the covariates has to be discovered within this neighborhood and thirdly, it has to be investigated whether this dependence structure can be exploited for any predictive power. Climate scientists have already discovered some physics-based relations between some of the covariates (e.g. temperature, relative humidity, precipitable water etc.) and precipitation extremes. We are exploring data-dependent alternatives for these problems and any possibility of incorporating the physics-based relations into the resulting data model. In particular, we used elastic net-based sparse optimization technique which solves all three problems of neighborhood discovery, covariate dependence discovery and predictive modeling and at the same time maintains the interpretability of the resulting model. Preliminary results look promising and show potential for some interesting knowledge discovery. We are currently exploring non-linear correlations and the alternatives to combine the physics-based relationships into the data model.

Original languageEnglish (US)
Pages1948-1951
Number of pages4
DOIs
StatePublished - Dec 1 2012
Event2012 32nd IEEE International Geoscience and Remote Sensing Symposium, IGARSS 2012 - Munich, Germany
Duration: Jul 22 2012Jul 27 2012

Other

Other2012 32nd IEEE International Geoscience and Remote Sensing Symposium, IGARSS 2012
CountryGermany
CityMunich
Period7/22/127/27/12

Fingerprint

Physics
physics
Data structures
Precipitation (meteorology)
Data mining
precipitable water
Atmospheric humidity
climate
Availability
relative humidity
analysis
Water
prediction
modeling
Temperature
temperature

Keywords

  • One
  • five
  • four
  • three
  • two

Cite this

Das, D., Ganguly, A., Chatterjee, S., Kumar, V., & Obradovic, Z. (2012). Spatially penalized regression for dependence analysis of rare events: A study in precipitation extremes. 1948-1951. Paper presented at 2012 32nd IEEE International Geoscience and Remote Sensing Symposium, IGARSS 2012, Munich, Germany. https://doi.org/10.1109/IGARSS.2012.6351120

Spatially penalized regression for dependence analysis of rare events : A study in precipitation extremes. / Das, Debasish; Ganguly, Auroop; Chatterjee, Snigdhansu; Kumar, Vipin; Obradovic, Zoran.

2012. 1948-1951 Paper presented at 2012 32nd IEEE International Geoscience and Remote Sensing Symposium, IGARSS 2012, Munich, Germany.

Research output: Contribution to conferencePaper

Das, D, Ganguly, A, Chatterjee, S, Kumar, V & Obradovic, Z 2012, 'Spatially penalized regression for dependence analysis of rare events: A study in precipitation extremes' Paper presented at 2012 32nd IEEE International Geoscience and Remote Sensing Symposium, IGARSS 2012, Munich, Germany, 7/22/12 - 7/27/12, pp. 1948-1951. https://doi.org/10.1109/IGARSS.2012.6351120
Das D, Ganguly A, Chatterjee S, Kumar V, Obradovic Z. Spatially penalized regression for dependence analysis of rare events: A study in precipitation extremes. 2012. Paper presented at 2012 32nd IEEE International Geoscience and Remote Sensing Symposium, IGARSS 2012, Munich, Germany. https://doi.org/10.1109/IGARSS.2012.6351120
Das, Debasish ; Ganguly, Auroop ; Chatterjee, Snigdhansu ; Kumar, Vipin ; Obradovic, Zoran. / Spatially penalized regression for dependence analysis of rare events : A study in precipitation extremes. Paper presented at 2012 32nd IEEE International Geoscience and Remote Sensing Symposium, IGARSS 2012, Munich, Germany.4 p.
@conference{ff23085b4c4749d5b538870e61827037,
title = "Spatially penalized regression for dependence analysis of rare events: A study in precipitation extremes",
abstract = "Discovery of dependence structure between precipitation extremes and other climate variables (covariates) within a smaller spatial and temporal neighborhood is an important step in better understanding the drivers of this complex phenomenon as well as short-term prediction of extremes occurrence. Apart from the inherent spatio-temporal variability of the dependence, it is further complicated by the availability of the covariates at different vertical levels. The above problem can be split into three different sub-problems. Firstly, a spatio-temporal neighborhood of influence has to be discovered, which can be different for different locations. Secondly, the dependence structure between the precipitation extremes and the covariates has to be discovered within this neighborhood and thirdly, it has to be investigated whether this dependence structure can be exploited for any predictive power. Climate scientists have already discovered some physics-based relations between some of the covariates (e.g. temperature, relative humidity, precipitable water etc.) and precipitation extremes. We are exploring data-dependent alternatives for these problems and any possibility of incorporating the physics-based relations into the resulting data model. In particular, we used elastic net-based sparse optimization technique which solves all three problems of neighborhood discovery, covariate dependence discovery and predictive modeling and at the same time maintains the interpretability of the resulting model. Preliminary results look promising and show potential for some interesting knowledge discovery. We are currently exploring non-linear correlations and the alternatives to combine the physics-based relationships into the data model.",
keywords = "One, five, four, three, two",
author = "Debasish Das and Auroop Ganguly and Snigdhansu Chatterjee and Vipin Kumar and Zoran Obradovic",
year = "2012",
month = "12",
day = "1",
doi = "10.1109/IGARSS.2012.6351120",
language = "English (US)",
pages = "1948--1951",
note = "2012 32nd IEEE International Geoscience and Remote Sensing Symposium, IGARSS 2012 ; Conference date: 22-07-2012 Through 27-07-2012",

}

TY - CONF

T1 - Spatially penalized regression for dependence analysis of rare events

T2 - A study in precipitation extremes

AU - Das, Debasish

AU - Ganguly, Auroop

AU - Chatterjee, Snigdhansu

AU - Kumar, Vipin

AU - Obradovic, Zoran

PY - 2012/12/1

Y1 - 2012/12/1

N2 - Discovery of dependence structure between precipitation extremes and other climate variables (covariates) within a smaller spatial and temporal neighborhood is an important step in better understanding the drivers of this complex phenomenon as well as short-term prediction of extremes occurrence. Apart from the inherent spatio-temporal variability of the dependence, it is further complicated by the availability of the covariates at different vertical levels. The above problem can be split into three different sub-problems. Firstly, a spatio-temporal neighborhood of influence has to be discovered, which can be different for different locations. Secondly, the dependence structure between the precipitation extremes and the covariates has to be discovered within this neighborhood and thirdly, it has to be investigated whether this dependence structure can be exploited for any predictive power. Climate scientists have already discovered some physics-based relations between some of the covariates (e.g. temperature, relative humidity, precipitable water etc.) and precipitation extremes. We are exploring data-dependent alternatives for these problems and any possibility of incorporating the physics-based relations into the resulting data model. In particular, we used elastic net-based sparse optimization technique which solves all three problems of neighborhood discovery, covariate dependence discovery and predictive modeling and at the same time maintains the interpretability of the resulting model. Preliminary results look promising and show potential for some interesting knowledge discovery. We are currently exploring non-linear correlations and the alternatives to combine the physics-based relationships into the data model.

AB - Discovery of dependence structure between precipitation extremes and other climate variables (covariates) within a smaller spatial and temporal neighborhood is an important step in better understanding the drivers of this complex phenomenon as well as short-term prediction of extremes occurrence. Apart from the inherent spatio-temporal variability of the dependence, it is further complicated by the availability of the covariates at different vertical levels. The above problem can be split into three different sub-problems. Firstly, a spatio-temporal neighborhood of influence has to be discovered, which can be different for different locations. Secondly, the dependence structure between the precipitation extremes and the covariates has to be discovered within this neighborhood and thirdly, it has to be investigated whether this dependence structure can be exploited for any predictive power. Climate scientists have already discovered some physics-based relations between some of the covariates (e.g. temperature, relative humidity, precipitable water etc.) and precipitation extremes. We are exploring data-dependent alternatives for these problems and any possibility of incorporating the physics-based relations into the resulting data model. In particular, we used elastic net-based sparse optimization technique which solves all three problems of neighborhood discovery, covariate dependence discovery and predictive modeling and at the same time maintains the interpretability of the resulting model. Preliminary results look promising and show potential for some interesting knowledge discovery. We are currently exploring non-linear correlations and the alternatives to combine the physics-based relationships into the data model.

KW - One

KW - five

KW - four

KW - three

KW - two

UR - http://www.scopus.com/inward/record.url?scp=84873180292&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84873180292&partnerID=8YFLogxK

U2 - 10.1109/IGARSS.2012.6351120

DO - 10.1109/IGARSS.2012.6351120

M3 - Paper

AN - SCOPUS:84873180292

SP - 1948

EP - 1951

ER -