The River That Puts on Its Own Brakes ✎

Original: "Long-range properties and data validity for hydrogeological time series: the case of the Paglia river"
arXiv:1611.09167 · 2016-11-24 · arXiv non-exclusive · independent analysis · 1 min
Twenty-two years of measurements on the Paglia River show that after a strong surge, the river doesn't keep speeding up but quickly returns to its usual level.

The Paglia River behaves like a pendulum: give it a push, and instead of accelerating, it swings back. For twenty-two years, scientists recorded how much water passed through the river near the Italian town of Orvieto, every half hour. That yielded 377,399 measurements. They calculated how neighboring fluctuations were related and saw that after a large surge, the river quickly returns to its normal level.

The Hurst exponent for the Paglia is about 0.34. This means that after each step forward, the river takes half a step back.

Over time, the variations in this exponent revealed rhythms with periods of about 12, 6, 3, and 2.5 days. These resemble cycles of large-scale ocean and atmospheric phenomena, such as El Niño and the North Atlantic Oscillation. It seems that the distant ocean influences the river, which responds with its own fluctuations. The scientists also checked how often leading digits appear in the data and found an anomaly: some values repeat too frequently. This happens because water levels were measured to the nearest centimeter and then converted to discharge using a single formula. As a result, the records contain "steps." Such checks help assess flood risks and verify the reliability of measurements at hydrological stations.

The two most frequent streamflow values in the dataset each occur more than 10,000 times. This results from water levels being measured to the nearest centimeter and streamflow being computed via a fixed rating curve, creating 'steps' in the data.

Concepts
Pacific decadal oscillation Hurst exponent multifractal detrended fluctuation analysis nonlinear laplacian spectral analysis Bernoulli's principle