Data-Driven Prediction of Thresholded Time Series of Rainfall and SOC models
arXiv:1411.2256 · doi:10.1103/PhysRevE.91.052808
Abstract
We study the occurrence of events, subject to threshold, in a representative SOC sandpile model and in high-resolution rainfall data. The predictability in both systems is analyzed by means of a decision variable sensitive to event clustering, and the quality of the predictions is evaluated by the receiver operating characteristics (ROC) method. In the case of the SOC sandpile model, the scaling of quiet-time distributions with increasing threshold leads to increased predictability of extreme events. A scaling theory allows us to understand all the details of the prediction procedure and to extrapolate the shape of the ROC curves for the most extreme events. For rainfall data, the quiet-time distributions do not scale for high thresholds, which means that the corresponding ROC curves cannot be straightforwardly related to those for lower thresholds.
arXiv comment19 pages, 10 figures
References in corpus (15)
- Long-Term Clustering, Scaling, and Universality in the Temporal Occurrence of Earthquakes
- Critical phenomena in atmospheric precipitation
- Power Laws in Solar Flares: Self-Organized Criticality or Turbulence?
- The abelian sandpile and related models
- A Complexity View of Rainfall
- Inter-occurrence Times in the Bak-Tang-Wiesenfeld Sandpile Model: A Comparison with the Turbulent Statistics of Solar Flares
- Universality of rain event size distributions
- Time-Decreasing Hazard and Increasing Time until the Next Earthquake
- Scaling behaviour in daily air humidity fluctuations
- Avalanche Merging and Continuous Flow in a Sandpile Model
- Scaling in the Timing of Extreme Events
- Comment on "Do Earthquakes Exhibit Self-Organized Criticality?"
- Rain, power laws, and advection
- Comment on: "Avalanches and Non-Gaussian Fluctuations of the Global Velocity of Imbibition Fronts"
- Testing Universality in Critical Exponents: the Case of Rainfall
Cited by in corpus (10)
- Scaling in the Timing of Extreme Events
- Predictability of Extreme Intensity Pulses in Optically Injected Semiconductor Lasers
- Short term prediction of extreme returns based on the recurrence interval analysis
- Universal avalanche statistics and triggering close to failure in a mean field model of rheological fracture
- Increasing power-law range in avalanche amplitude and energy distributions
- Fundamental interactions in self-organized critical dynamics on higher-order networks
- Spectrum of the Stress Dynamics with the Bak-Tang-Wiesenfeld Sandpile
- Some Properties of Sandpile Models as Prototype of Self-Organized Critical Systems
- Testing Universality in Critical Exponents: the Case of Rainfall
- Convergence of Rain Process Models to Point Processes