Clustering and upscaling of station precipitation records to regional patterns using self-organizing maps (SOMs)

Robert George Crane, Bruce C. Hewitson

Research output: Contribution to journalArticle

35 Citations (Scopus)

Abstract

Self-organizing maps (SOMs), a particular application of artificial neural networks, are used to proportionately combine precipitation records of individual stations into a regional data set by extracting the common regional variability from the locally forced variability at each station. The methodology is applied to a 100 yr record of precipitation data for 104 stations in the Mid-Atlantic/Northeast United States region. The SOM combines stations with common precipitation characteristics and identifies precipitation regions that are consistent across a range of spatial scales. A variation of the SOM application identifies the temporal modes of the regional precipitation record and uses them to fill missing data in the station observations to produce a regional precipitation record. A test of the methodology with a complete data set shows that the 'missing data' routine improves the regional signal when up to 80% of the data are missing from 80% of the stations. The improvement is almost as pronounced when there is a bias in the missing data for both high-precipitation and low-precipitation events.

Original languageEnglish (US)
Pages (from-to)95-107
Number of pages13
JournalClimate Research
Volume25
Issue number2
DOIs
StatePublished - Dec 5 2003

Fingerprint

upscaling
regional pattern
Self organizing maps
Neural networks
methodology
artificial neural network
station

All Science Journal Classification (ASJC) codes

  • Environmental Chemistry
  • Environmental Science(all)
  • Atmospheric Science

Cite this

@article{309b1595e1484205982777647107c53c,
title = "Clustering and upscaling of station precipitation records to regional patterns using self-organizing maps (SOMs)",
abstract = "Self-organizing maps (SOMs), a particular application of artificial neural networks, are used to proportionately combine precipitation records of individual stations into a regional data set by extracting the common regional variability from the locally forced variability at each station. The methodology is applied to a 100 yr record of precipitation data for 104 stations in the Mid-Atlantic/Northeast United States region. The SOM combines stations with common precipitation characteristics and identifies precipitation regions that are consistent across a range of spatial scales. A variation of the SOM application identifies the temporal modes of the regional precipitation record and uses them to fill missing data in the station observations to produce a regional precipitation record. A test of the methodology with a complete data set shows that the 'missing data' routine improves the regional signal when up to 80{\%} of the data are missing from 80{\%} of the stations. The improvement is almost as pronounced when there is a bias in the missing data for both high-precipitation and low-precipitation events.",
author = "Crane, {Robert George} and Hewitson, {Bruce C.}",
year = "2003",
month = "12",
day = "5",
doi = "10.3354/cr025095",
language = "English (US)",
volume = "25",
pages = "95--107",
journal = "Climate Research",
issn = "0936-577X",
publisher = "Inter-Research",
number = "2",

}

Clustering and upscaling of station precipitation records to regional patterns using self-organizing maps (SOMs). / Crane, Robert George; Hewitson, Bruce C.

In: Climate Research, Vol. 25, No. 2, 05.12.2003, p. 95-107.

Research output: Contribution to journalArticle

TY - JOUR

T1 - Clustering and upscaling of station precipitation records to regional patterns using self-organizing maps (SOMs)

AU - Crane, Robert George

AU - Hewitson, Bruce C.

PY - 2003/12/5

Y1 - 2003/12/5

N2 - Self-organizing maps (SOMs), a particular application of artificial neural networks, are used to proportionately combine precipitation records of individual stations into a regional data set by extracting the common regional variability from the locally forced variability at each station. The methodology is applied to a 100 yr record of precipitation data for 104 stations in the Mid-Atlantic/Northeast United States region. The SOM combines stations with common precipitation characteristics and identifies precipitation regions that are consistent across a range of spatial scales. A variation of the SOM application identifies the temporal modes of the regional precipitation record and uses them to fill missing data in the station observations to produce a regional precipitation record. A test of the methodology with a complete data set shows that the 'missing data' routine improves the regional signal when up to 80% of the data are missing from 80% of the stations. The improvement is almost as pronounced when there is a bias in the missing data for both high-precipitation and low-precipitation events.

AB - Self-organizing maps (SOMs), a particular application of artificial neural networks, are used to proportionately combine precipitation records of individual stations into a regional data set by extracting the common regional variability from the locally forced variability at each station. The methodology is applied to a 100 yr record of precipitation data for 104 stations in the Mid-Atlantic/Northeast United States region. The SOM combines stations with common precipitation characteristics and identifies precipitation regions that are consistent across a range of spatial scales. A variation of the SOM application identifies the temporal modes of the regional precipitation record and uses them to fill missing data in the station observations to produce a regional precipitation record. A test of the methodology with a complete data set shows that the 'missing data' routine improves the regional signal when up to 80% of the data are missing from 80% of the stations. The improvement is almost as pronounced when there is a bias in the missing data for both high-precipitation and low-precipitation events.

UR - http://www.scopus.com/inward/record.url?scp=0347765758&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=0347765758&partnerID=8YFLogxK

U2 - 10.3354/cr025095

DO - 10.3354/cr025095

M3 - Article

AN - SCOPUS:0347765758

VL - 25

SP - 95

EP - 107

JO - Climate Research

JF - Climate Research

SN - 0936-577X

IS - 2

ER -