Modeling two-dimensional Stress Field evolution and its role in reservoir-induced seismicity of Koyna Warna region and general seismicity of Northwest Himalayas

Abstract

Reservoir-triggered seismicity (RTS) has emerged as a critical subject in earthquake science following the devastating 1967 Koyna-Warna earthquake (Mw 6.3), underscoring the complex interplay between anthropogenic reservoir loading and natural fault systems. This thesis presents a comprehensive investigation into RTS, primarily focusing on two geologically distinct regions in India: the Koyna-Warna region in Maharashtra and the Tehri region in Uttarakhand. A detailed static stress modeling approach, employing analytical solutions and numerical simulations, has been developed to evaluate stress perturbations induced by reservoir water level fluctuations. Our findings reveal distinct responses based on fault mechanisms— strike-slip faulting near the Koyna reservoir exhibits in-phase Coulomb stress changes with sinusoidal water loading. In contrast, normal faulting near the Warna reservoir shows out-ofphase behavior. These theoretical insights are supported by earthquake catalogue analysis, including temporal patterns and variations in seismicity. A comparative study of the b-value, a statistical measure of earthquake magnitude distribution, has been conducted over four decades (1980–2022) for both the Koyna and Tehri regions. Results show a significant stress release in the Koyna region in the most recent decade, with comparatively subdued seismicity in the Tehri region, where the RTS potential remains under speculative observation. The spatial extent of RTS has also been assessed by categorizing seismic responses into local (within 30–50 km of the dam) and remote (beyond 50 km) zones, revealing notable differences in the triggering mechanisms between the two regions. Further, the thesis delves into the broader seismicity of the Indian plate boundaries, characterized by high tectonic activity and complex earthquake behavior. Special emphasis is given to the North-West Himalayas, where an upward trend in seismic activity has been observed between 2018 and 2023. An innovative analytical technique—Chaos Game Representation (CGR) using driven iterated function systems—has been introduced to better understand the temporal evolution of seismicity in this region. This fractal-based approach enables visualization of nonlinear and self-similar patterns in seismic time series data, offering insights beyond traditional methods.