Premium
STOCHASTIC MATRICES APPLIED TO THE PROBABILISTIC ANALYSIS OF RUNOFF EVENTS IN A SEMI‐ARID STREAM
Author(s) -
CONESAGARCÍA C.,
ALONSOSARRIÁ F.
Publication year - 1997
Publication title -
hydrological processes
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 1.222
H-Index - 161
eISSN - 1099-1085
pISSN - 0885-6087
DOI - 10.1002/(sici)1099-1085(19970315)11:3<297::aid-hyp451>3.0.co;2-p
Subject(s) - surface runoff , markov chain , probabilistic logic , arid , environmental science , runoff model , markov process , hydrology (agriculture) , stochastic modelling , stochastic process , mathematics , computer science , statistics , geology , ecology , geotechnical engineering , paleontology , biology
Stochastic models offer an objective and quantitative method of analysing the behaviour of discrete hydrological variables. For the Algeciras semi‐arid stream in south‐east Spain, homogeneous first‐ and second‐order Markov chains were applied to rainfall and runoff series in order to establish the length of the rainy sequences using probability as well as the beginning and duration of runoff. At the same time, a stochastic process was devised based on two subsystems, one for rainfall, modelled as a four‐state n ‐order Markov chain, and another for transition of runoff depending on the former subsystem. By considering the median rainfall obtained in each season as a class threshold, the possible number of states in the first subsystem is raised to nine, giving more complicated stochastic matrices which, without using an excessive number of parameters, allow better description and prediction of the transition from rainfall to runoff. © 1997 by John Wiley & Sons, Ltd.