Fiche du document
6 avril 2016
- handle: 10670/1.3b0go7
- isbn: 9780470684054 / 9781118648551
- halshs: halshs-01338510
- doi: 10.1002/9781118648551.ch21
Ce document est lié à :
info:eu-repo/semantics/altIdentifier/doi/10.1002/9781118648551.ch21
Mots-clés
Temporal patterns Bivariate statistics Event probabilities Fluvial geomorphology Multinomial models Multiple regression model Multivariate analysis Spatial patterns Statistical toolsSujets proches
Tools Hand tools HandtoolsCiter ce document
Hervé Piégay et al., « Statistics and fluvial geomorphology », HAL-SHS : géographie, ID : 10.1002/9781118648551.ch21
Métriques
Partage / Export
Résumé
This chapter reviews statistical tools and illustrates their use to answer geomorphological questions, and also overviews their advantages and limits. Application of statistical tools in fluvial geomorphology has the advantages of reducing subjectivity, eliminating assumptions, facilitating comparison between different spatial and temporal datasets of large sizes and refining data collection. Bivariate statistics, and regressions in particular, have been one of the most popular statistical tools in geomorphology. They focus on the relationship, or correlation, between two variables. Probabilities are useful to generate models in which the variables of interest are categorical, such as indicator variables of events (e.g. occurrence of peak flows). One of the most basic tools dealing with probabilities is the logistic or multinomial models. To develop more realistic descriptions of fluvial morphological systems, process‐response systems, time and space trends and size effects, requires the collection of sufficient data and more thought about their relevance.