Seismic swarm activity serves as a key indicator during the earthquake preparation process. Its spatio-temporal evolution is jointly controlled by subsurface fluid migration, tectonic stress loading, and physical heterogeneities in the crustal medium. During the preparation stage of large earthquakes, anomalous activities of moderate-to-small swarms （M_L≤5.2） often occur. Analyzing their focal parameters, spatial distribution, and temporal characteristics is crucial for capturing the dynamic adjustment of the crustal stress field and pre-instability signals in seismogenic zones. On June 26, 2024, an earthquake with M_L5.1 struck Yongsheng County, Lijiang City, Yunnan Province, providing a valuable case for investigating regional seismogenic mechanisms and tectonic stress regimes. This study presents a systematic analysis of the seismogenic structures and genesis of the 2024 Yongsheng M_L5.1 earthquake swarm. We first use broadband waveform data from the Yunnan Provincial Digital Seismic Monitoring Network, combined with regional velocity models and tectonic settings, as the basic data and geological framework. On this basis, we apply a suite of methods including double‑difference relocation, CAP focal mechanism inversion, and fluid diffusion modeling, together with statistical analyses of b‑values and aftershock decay index p‑values, to carry out a comprehensive investigation.

A total of 929 seismic events （M_L≥−0.5） between June 26 and November 3, 2024 were collected, with 876 events used for double-difference relocation. The hypoDD algorithm was applied with a local velocity model （model A, v_P/v_S＝1.629） for Yongsheng County, achieving horizontally and vertically positioning errors of 6.50 m and 5.23 m, respectively. For five events with M_L≥3.5, we used the CAP method for focal mechanism inversion, applying a high-resolution velocity model （model B） for the Sichuan-Yunnan region and using broadband waveform data from seismic stations at epicentral distances of 60−360 km. Nodal plane parameters and stress axis azimuths were obtained through frequency-band fitting of body waves and surface waves. Additionally, b-values （M_c＝0.5） were calculated using the G-R relationship; the modified Omori law was used to analyze the aftershock decay index p. We also compared fluid diffusion models with poroelastic stress transfer models to investigate the triggering mechanisms.

Double-difference relocation results showed that the swarm formed a rectangular strip 27 km long and 10 km wide, with strikes of NW （120°） and NE （28°）, and focal depths concentrated at 8−14 km. The root-mean-square （RMS） of travel-time residual decreased from 0.043 s to 0.004 s after relocation, indicating a significant improvement in location accuracy. The temporal evolution of the swarm comprised three phases: an initial phase （0−3.47 h） dominated by NE-directed migration; a main phase （3.47−70 h） during which seismic activity diffused toward NW and SE; and a decay phase （after 70 h） characterized by an exponential decline in seismicity rate （ p＝1.572）. Energy release was concentrated within the first three days, with the M_L5.1 and M_L4.9 events accounting for 97.45% of the total energy released, reflecting rapid stress adjustment.

CAP inversion results revealed that the focal mechanism of the main shock was dominated by left-lateral strike-slip with normal components （nodal plane Ⅱ : strike 28°±2°, dip −13°±5°）. The principal compressive stress axis （P-axis） was oriented at 344° （NNW）, and the principal tensile stress axis （T-axis） was oriented at 75° （SW−NE）, consistent with the regional stress field of the Sichuan-Yunnan rhombic block. The focal mechanisms of the five larger events were consistent, with nodal plane parameters differing by less than 10°, which indicates that these events were controlled by a unified tectonic stress field. The focal mechanism solutions are consistent with the left-lateral kinematics of the deep strike-slip segment of the Chenghai-Binchuan fault. Combined with crustal low-velocity anomalies revealed by seismic tomography, these solutions indicate that the deep segment of the fault represents the primary seismogenic structure.

The b-value of the swarm area （0.66±0.03） was significantly lower than the regional background b-value （0.83±0.01）, indicating a state of high-stress accumulation in the source area, with a reduced proportion of small earthquakes and an elevated potential for large events. The aftershock decay index p＝1.572 was higher than that of typical mainshock-aftershock sequences （ p≈1.1）, reflecting rapid energy release on strike-slip faults under high confining pressure. The discrepancy between the fluid diffusion model and the observed migration patterns suggests that the poroelastic stress transfer model better explains the linear migration characteristics of the swarm. This indicates that the sequence was dominated by regional tectonic stress, while fluids only locally promoted fault slip during the initial rupture stage.

Comprehensive analysis confirms that the deep strike-slip segment of the Chenghai-Binchuan fault was the primary seismogenic structure. It experienced left-lateral strike-slip rupture under NNW-directed principal compressive stress, triggering shallow normal faulting in secondary structures such as the Shangyangping fault, and forming a coupling model of “deep strike-slip control, shallow normal fault response”.

The low b-value, high p-value, and earthquake swarm migration patterns in the source area indicate a critical stress state, implying a high potential for future moderate-to-strong earthquakes. This study provides new constraints on the stress transfer mechanism of boundary faults within the Sichuan-Yunnan rhombic block, and has important implications for regional seismic hazard assessment and fault activity monitoring.