A MODERN STATISTICAL STUDY OF MONTHLY LEVELS OF THE ORINOCO RIVER

A mathematical model is built for monthly river flows of the Orinoco river. The model consists of a cyclic part which explains up to 93% of the total variance of the series of monthly water levels, and a stochastic part which is shown to follow a 1st order autoregressive scheme with a primary variable or random component that behaves as white noise and appears to have near 60% chance of coming from a normal population. A time series of flow anomalies is obtained from long term monthly means. Statistical techniques are applied to the series of flow anomalies in order to obtain the mean number of crossings at an arbitrary level during an arbitrary time. There are also studied the mean length of an upward excursion over an arbitrary level and the mean time between successive upcrossings. The actual results for the Orinoco river are in good agreement with the statistical theory showing that this kind of analysis can be extremely useful in the design and planning of reservoir operations. (Key words: hydrology, runoff, statistical analysis.). 1968 Taylor & Francis Group, LLC.

A MODERN STATISTICAL STUDY OF MONTHLY LEVELS OF THE ORINOCO RIVER Academic Article