A Markov Chain Model of Daily Rainfall
Abstract
The design of many water resources projects requires knowledge of possible long-term rainfall patterns. A stochastic model based on a first-order Markov chain was developed to simulate daily rainfall at a point. The model uses historical rainfall data to estimate the Markov transitional probabilities. A separate matrix is estimated for each month of the year. In this research, 7 × 7 transitional probability matrices were used. The model is capable of simulating a daily rainfall record of any length, based on the estimated transitional probabilities and frequency distributions of rainfall amounts. The simulated data have statistical properties similar to those of historical data.
- Publication:
-
Water Resources Research
- Pub Date:
- June 1976
- DOI:
- 10.1029/WR012i003p00443
- Bibcode:
- 1976WRR....12..443H