Predicting drought using pattern recognition

V. K. Boken, C. E. Haque, G. Hoogenboom

Research output: Contribution to journalArticle

1 Citation (Scopus)

Abstract

Predicting droughts and their impacts upon overall agricultural production helps in drought management. Generally, statistical regression or time series techniques are employed to predict agricultural droughts quantitatively. Linear (error correction, linear discriminant) and nonlinear (k-Nearest Neighbor) techniques of pattern recognition were used for predicting agricultural droughts qualitatively. A total of five crop districts in the province of Saskatchewan in the Canadian prairies, were selected. Thirty two variables were derived for each district from the daily temperature and precipitation data for the period from 1975 to 2002 to develop pattern recognition models. The variables derived from the minimum or maximum temperatures were found to be more significant than the variables derived from the precipitation for predicting moderate-to-very severe agricultural droughts. The 1975-1997 data were used for model development while the 1998-2002 data were used for model testing. About 83% accuracy was achieved in predicting the non-drought category while 71% accuracy was achieved in predicting the drought category. It was concluded that pattern recognition techniques could be applied for predicting drought qualitatively, which would aid current methods of drought prediction.

Original languageEnglish (US)
Pages (from-to)133-144
Number of pages12
JournalAnnals of Arid Zone
Volume46
Issue number2
StatePublished - Jun 1 2007

Fingerprint

pattern recognition
drought
error correction
Saskatchewan
methodology
prairies
agricultural production
prairie
time series analysis
temperature
regression analysis
time series
agriculture
crop
prediction
crops

Keywords

  • Crop yield
  • Discriminant analysis
  • Prairies
  • Wheat

ASJC Scopus subject areas

  • Ecology, Evolution, Behavior and Systematics
  • Animal Science and Zoology
  • Agronomy and Crop Science

Cite this

Boken, V. K., Haque, C. E., & Hoogenboom, G. (2007). Predicting drought using pattern recognition. Annals of Arid Zone, 46(2), 133-144.

Predicting drought using pattern recognition. / Boken, V. K.; Haque, C. E.; Hoogenboom, G.

In: Annals of Arid Zone, Vol. 46, No. 2, 01.06.2007, p. 133-144.

Research output: Contribution to journalArticle

Boken, VK, Haque, CE & Hoogenboom, G 2007, 'Predicting drought using pattern recognition', Annals of Arid Zone, vol. 46, no. 2, pp. 133-144.
Boken VK, Haque CE, Hoogenboom G. Predicting drought using pattern recognition. Annals of Arid Zone. 2007 Jun 1;46(2):133-144.
Boken, V. K. ; Haque, C. E. ; Hoogenboom, G. / Predicting drought using pattern recognition. In: Annals of Arid Zone. 2007 ; Vol. 46, No. 2. pp. 133-144.
@article{bd9e7fc737f24f7a8e9da0bc97bd658c,
title = "Predicting drought using pattern recognition",
abstract = "Predicting droughts and their impacts upon overall agricultural production helps in drought management. Generally, statistical regression or time series techniques are employed to predict agricultural droughts quantitatively. Linear (error correction, linear discriminant) and nonlinear (k-Nearest Neighbor) techniques of pattern recognition were used for predicting agricultural droughts qualitatively. A total of five crop districts in the province of Saskatchewan in the Canadian prairies, were selected. Thirty two variables were derived for each district from the daily temperature and precipitation data for the period from 1975 to 2002 to develop pattern recognition models. The variables derived from the minimum or maximum temperatures were found to be more significant than the variables derived from the precipitation for predicting moderate-to-very severe agricultural droughts. The 1975-1997 data were used for model development while the 1998-2002 data were used for model testing. About 83{\%} accuracy was achieved in predicting the non-drought category while 71{\%} accuracy was achieved in predicting the drought category. It was concluded that pattern recognition techniques could be applied for predicting drought qualitatively, which would aid current methods of drought prediction.",
keywords = "Crop yield, Discriminant analysis, Prairies, Wheat",
author = "Boken, {V. K.} and Haque, {C. E.} and G. Hoogenboom",
year = "2007",
month = "6",
day = "1",
language = "English (US)",
volume = "46",
pages = "133--144",
journal = "Annals of Arid Zone",
issn = "0570-1791",
publisher = "Arid Zone Research Association of India",
number = "2",

}

TY - JOUR

T1 - Predicting drought using pattern recognition

AU - Boken, V. K.

AU - Haque, C. E.

AU - Hoogenboom, G.

PY - 2007/6/1

Y1 - 2007/6/1

N2 - Predicting droughts and their impacts upon overall agricultural production helps in drought management. Generally, statistical regression or time series techniques are employed to predict agricultural droughts quantitatively. Linear (error correction, linear discriminant) and nonlinear (k-Nearest Neighbor) techniques of pattern recognition were used for predicting agricultural droughts qualitatively. A total of five crop districts in the province of Saskatchewan in the Canadian prairies, were selected. Thirty two variables were derived for each district from the daily temperature and precipitation data for the period from 1975 to 2002 to develop pattern recognition models. The variables derived from the minimum or maximum temperatures were found to be more significant than the variables derived from the precipitation for predicting moderate-to-very severe agricultural droughts. The 1975-1997 data were used for model development while the 1998-2002 data were used for model testing. About 83% accuracy was achieved in predicting the non-drought category while 71% accuracy was achieved in predicting the drought category. It was concluded that pattern recognition techniques could be applied for predicting drought qualitatively, which would aid current methods of drought prediction.

AB - Predicting droughts and their impacts upon overall agricultural production helps in drought management. Generally, statistical regression or time series techniques are employed to predict agricultural droughts quantitatively. Linear (error correction, linear discriminant) and nonlinear (k-Nearest Neighbor) techniques of pattern recognition were used for predicting agricultural droughts qualitatively. A total of five crop districts in the province of Saskatchewan in the Canadian prairies, were selected. Thirty two variables were derived for each district from the daily temperature and precipitation data for the period from 1975 to 2002 to develop pattern recognition models. The variables derived from the minimum or maximum temperatures were found to be more significant than the variables derived from the precipitation for predicting moderate-to-very severe agricultural droughts. The 1975-1997 data were used for model development while the 1998-2002 data were used for model testing. About 83% accuracy was achieved in predicting the non-drought category while 71% accuracy was achieved in predicting the drought category. It was concluded that pattern recognition techniques could be applied for predicting drought qualitatively, which would aid current methods of drought prediction.

KW - Crop yield

KW - Discriminant analysis

KW - Prairies

KW - Wheat

UR - http://www.scopus.com/inward/record.url?scp=46449107274&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=46449107274&partnerID=8YFLogxK

M3 - Article

AN - SCOPUS:46449107274

VL - 46

SP - 133

EP - 144

JO - Annals of Arid Zone

JF - Annals of Arid Zone

SN - 0570-1791

IS - 2

ER -