Spatiotemporal dynamics of wind-speed volatility

2026-05-08

AI Summary

This study addresses the lack of systematic modeling of wind speed volatility dynamics across space and time, particularly the spatial dependence within monitoring networks and vertical variations. Leveraging daily wind speed observations from 141 stations in northern Italy over 2016–2021, the authors propose a volatility model integrating a spatiotemporal GARCH structure, wherein the conditional variance is formulated as a function of local historical shocks and spatially weighted information from neighboring sites. A novel dual weighting scheme incorporating both distance and direction captures spatial dependence more effectively. The work innovatively couples a structured mean model with the volatility framework, revealing pronounced volatility persistence and inter-height dependencies. Results demonstrate that appropriately specified mean models enhance residual properties; once spatial interactions are adequately accounted for, a parsimonious distance-weighted specification yields robust out-of-sample forecasts, with volatility persistence increasing monotonically with height.

Abstract

Wind-speed processes exhibit substantial temporal variability and spatial dependence, yet volatility dynamics across monitoring networks remain relatively unexplored. This study investigates the spatiotemporal behaviour of wind-speed volatility using daily observations from 141 stations in Northern Italy over 2016--2021, with measurements at 10 m and 100 m enabling the analysis of spatial and vertical dependence. We adopt a parsimonious spatiotemporal volatility framework based on GARCH-type dynamics, in which conditional variance depends on past local shocks and spatially aggregated information from neighbouring stations. The approach combines a spatial mean specification with structured volatility models using distance-based and directionally informed weight matrices. Results show that properly modelling spatial dependence in the mean is essential for well-behaved residuals and reliable inference. Forecast performance is strongly driven by the mean specification: flexible structures perform better when residual spatial dependence remains, while parsimonious distance-based models yield robust out-of-sample forecasts once spatial interactions are captured. Persistence increases with height, and a multivariate extension reveals cross-height dependence.

Problem

• wind-speed volatility
• spatiotemporal dynamics
• spatial dependence
• temporal variability
• monitoring networks

Innovation

• spatiotemporal volatility
• GARCH
• spatial dependence
• wind-speed dynamics
• directional weight matrix