Tectonic regionality and geophysical controls of the fractal dimension and b-value of seismicity in the Chinese mainland

Abstract Seismicity exhibits complex spatiotemporal patterns that are critical for hazard assessment. Using the Chinese mainland earthquake catalog (1970–2025), we analyze seismicity parameters across 20 seismic belts. After estimating the completeness magnitude ( M c ) for each belt, we apply nearest-neighbor distance declustering and calculate b -value and spatial fractal dimension ( D c ) before and after declustering using the maximum likelihood and correlation integral methods. We examine their spatial distributions and relationships with heat flow and maximum shear strain rate. Declustering increases D c -value (from 1.30 to 1.65) while decreasing a - and b -values, confirming that aftershock removal diffuses spatial patterns and reduces the proportion of small events. The relationship between b -value and D c -value varies regionally: negative in the Tibetan seismic region, positive in the South China seismic region and Xinjiang seismic region, and weakly negative in the North China seismic region. Heat flow modulates seismicity parameters in a consistent manner. The b -value is positively correlated with heat flow both before and after declustering, whereas the D c -value is negatively correlated with heat flow before declustering but not after. These results suggest that thermal weakening promotes microcracking, raising the b -value, and facilitates strain localization along major faults, lowering the D c -value. The D c -value and maximum shear strain rate relationship is positive in eastern extensional regions and negative in western compressional regions, reflecting contrasting strain energy partitioning mechanisms. These findings provide quantitative constraints on deep controls of seismicity complexity and offer new criteria for seismic hazard assessment.