Statistical and optimization techniques in climate modeling

Srinivasa Raju, K and Nagesh Kumar, D (2017) Statistical and optimization techniques in climate modeling. [Book Chapter]

PDF spr_cli_107-135_2018.pdf - Published Version Restricted to Registered users only Download (975kB) | Request a copy

Abstract

This chapter presents data compression techniques, namely, cluster and fuzzy cluster analysis, Kohonen neural networks for clustering GCMs and principal component analysis for transforming a set of observations of possible correlation into a set of linearly uncorrelated variables applying an orthogonal transformation. F--statistic test which can be used as the basis for finding optimal clusters is also discussed. Trend detection techniques, namely, Kendall’s rank correlation and turning point test along with mathematical background are also briefed with the objective to ascertain the quality of the hydrological or climatological records. In addition, optimization techniques, namely, linear and non-linear programming and genetic algorithms along with mathematical description are also discussed. The reader is expected to understand various statistical and optimization techniques along with their applicability by studying this chapter.

Item Type:	Book Chapter
Publication:	Springer Climate
Series.:	Springer Climate
Publisher:	Springer
Additional Information:	The copyright for this article belongs to the Springer Nature Singapore Pte Ltd.
Keywords:	Cluster; Data compression techniques; Fuzzy; Genetic algorithms; Kohonen neural networks; Linear; Non-linear; Optimization; Principal component analysis; Trend detection
Department/Centre:	Division of Mechanical Sciences > Centre for Earth Sciences Division of Mechanical Sciences > Divecha Centre for Climate Change Division of Interdisciplinary Sciences > Interdisciplinary Centre for Water Research Division of Mechanical Sciences > Civil Engineering
Date Deposited:	14 Aug 2022 06:14
Last Modified:	14 Aug 2022 06:14
URI:	https://eprints.iisc.ac.in/id/eprint/75746

Actions (login required)

View Item