By means of the Graphic Method, Gray Correlation Method and Frequency Method, the total amount of loose solid materials, the amount of loose solid materials per unit area in the source area, and the risk degree and critical rainfall of the study area are determined, respectively. On the basis of the calculation results, the total amount of loose solid materials is 57.2 × 10⁴ m³, the amount of loose solid materials per unit area is 0.13 m³/m² (greater than 0.1 m³/m²), which better meets the initiation conditions for dilute debris flow. The Hazard Evaluation Model of Debris Flow in earthquake areas is established by the gray correlation method and the hazard index H = 0.725 is determined. According to the hazard classification standards for debris flows in earthquake areas, the debris flow in Lecugou was moderate after the earthquake. The critical rainfall value of 1H, calculated by the frequency method, ranged from 17.45 to 22.21 mm (R_e = 25 mm) and 8.17 to 13.01 mm (R_e = 50 mm). The critical rainfall value at 10 min ranged from 6.23 to 8.44 mm (R_e = 25 mm) and 2.92 to 5.13 mm (R_e = 50 mm), respectively. The maximum rainfall intensity of 45.5 mm/h was reached between 10:00 and 11:00 on July 29, 2019 and the cumulative rainfall reached 144 mm, far exceeding the critical rainfall of Legugou. Under the influence of earthquakes and human activities, the loose solid source amount and loose solid source amount per unit area increased greatly, and the critical rainfall for debris flow after earthquakes decreased sharply, which induced debris flows under the influence of heavy rainfall. This research will be helpful for the establishment of monitoring and early warning systems based on artificial intelligence methods, and can greatly improve the effectiveness of disaster prevention and mitigation.

Many slope failures have been observed in deep-cutting gorges in Southwest China due to rainfall events. This study conducted model-based tests of the failure of a gently dipping accumulation slope caused by intermittent rainfall. The physical model was established by configuring soil samples based on similarity theory and direct shear test data. Intermittent rainfall was simulated by supplying surface runoff and bottom pressure water of the slope. The deformation, earth pressure, and pore pressure were observed during the test using a series of transducers. The results of this research showed that slope failure was first initiated at the toe of the slope. Staged sliding failure then occurred, which spread the unstable zone to the trailing edge of the slope. As infiltration of rainwater continued, the pore pressure increased and the matric suction and effective stress at the bedrock surface decreased, causing failure of the slope.

For a binary structure slope with a soil layer on the top and a rock layer on the bottom, during the rainfall process, surface runoff will cause soil and water loss on the slope surface and damage to the slope environment. When rainwater infiltrates into the slope, the pore water pressure in the soil gradually increases, the shear strength of the soil decreases, and a weak zone is formed at the soil–rock interface, which has a significant impact on the stability of the slope. Therefore, to study the soil and water loss on the slope surface and the stability of the slope under rainfall conditions, we used theoretical analysis, indoor model tests, and numerical simulations to conduct a comprehensive exploration of this issue, and the following conclusions were formed: the pore water pressure in the shallow layer is greater than that in the deep layer, and the pore water pressure at the toe of the slope is greater than that at the top of the slope; as the slope gradient increases, the time when the pore water pressure at the toe of the slope begins to respond gradually speeds up; the slope displacement first occurs at the lower part of the slope, then in the middle, and finally at the upper part; the time when the displacement at each point on the slope surface begins to respond gradually speeds up with the increase in the slope; the damage form at a small slope gradient is mainly flow sliding, and the damage process is continuous; the damage form at a large slope gradient is mainly flow sliding and overall sliding, and the damage process is continuous and sudden; when the binary structure slope fails, the sliding surface includes the internal sliding surface of the soil and the sliding surface at the soil–rock interface, but when the slope gradient is small, the relative sliding at the soil–rock interface is small, and a continuous sliding surface cannot be formed; and when the slope gradients are small (30° and 40°), the displacement decreases continuously from top to bottom, and no overall sliding surface is formed. The larger values of plastic strain mainly occur in the upper and middle parts of the slope, there is no formation of a continuous plastic strain zone, and the damage mode is flow sliding; when the slope gradients are large (50° and 60°), the displacement is the largest in the upper part, and a large displacement also occurs in the lower part, forming a sliding surface that penetrates through the soil–soil and rock–soil layers. The larger values of plastic strain occur in the upper, middle, and lower parts of the slope, a continuous plastic strain zone is formed, and the damage modes are flow sliding and overall sliding; numerical simulations were carried out on a typical actual slope, and consistent results were obtained.

In order to analyse the dynamic evolution characteristics of soil arching during sliding processes, we conducted a series of model push-pile and direct shear tests. The tests’ results were used to design a relative displacement monitoring system and to introduce two dimensionless parameters (the push–compaction ratio (e) and the different degrees of push–compaction (t)) to investigate the degree of uneven soil deformation during the sliding processes. This innovative method was used to analyse the rational spacing between adjacent anti-slide piles. The results revealed that there was a push-to-compaction effect in sliding soil during the sliding process. Firstly, in terms of space, the gradual transfer characteristics of the landslide thrust and push-to-compaction effect, rather than a uniform deformation over the entire area, were revealed. In terms of time, the results demonstrated a law for the variation in push-to-compaction ratios: The expansion of e occurred earlier in the rear sliding body than in the front e, while the growth rate of front e was faster than that of e in the rear sliding body. The dynamic evolution process was divided into three stages: an elastic formation stage, a plastic development stage, and a failure stage. Secondly, during the sliding process, the shear strength parameters of the sliding soil did not have constant values but underwent a dynamic process of strengthening, and cohesion responded more efficiently than friction. Finally, the degree of mobilisation of the anti-sliding effect of the sliding soil can be used as a new means of quantitative analysis for rational spacing. The results indicated that the rational spacing between adjacent piles should be five times the width of the pile.

Particle size distribution of gravelly soil plays a crucial role in debris flow initiation. For better understanding the mechanism of debris flow formation, two crucial mechanical property parameters of the gravelly soil are required to be studied meticulously: hydraulic conductivity and strength. With the aim of measuring the composition of the gravelly soil, 182 soil samples were taken from debris flow prone areas. With the aid of a sieve test, the particle size distribution of the samples can be obtained and analyzed. Then fractal theory was employed to compute the fractal dimension of the soil samples. By analyzing the results of sieve test (particle size distribution curves) and the results of the fractal theory calculations, the relationship between fractal dimension and particle size distribution can be explored. The results illustrate that the particle compositions of the gravelly soil tends to remain uniform as the fractal dimension increases. Moreover, as the coarse particle content increases, the fractal dimension decreases. To better understand the formation mechanism of debris flows, direct shear tests were conducted. Subsequently the experimental results were analyzed. By analysis, the following conclusions can be drawn: the soil strength decreases as the fractal dimension increases, and for soils with lower moisture content and identical dry density, a linear relationship between fractal dimension and cohesion force was identified. Moreover, cohesion force and internal friction force both decrease as the fractal dimension increases, but the internal friction angle decreases slightly while the cohesion force decreases greatly. Therefore we concluded that soil strength decreased mainly due to the reduction in cohesion force.

In order to assess the impact of the M_s7.0 Jiuzhaigou earthquake that occurred on 8 August 2017 on vegetation, the Carnegie-Ames-Stanford Approach (CASA) model was adopted to estimate the vegetation net primary productivity (NPP) of Jiuzhaigou Valley, one of the World Heritage Sites, in July, August and September from 2015 to 2019. Then the characteristics of the impact of different earthquake-induced geohazards on vegetation were discussed, and a vulnerability-resilience assessment system concerning the seismic intensity was proposed. The results show that the NPP_max and NPP_mean values in Jiuzhaigou Valley first decreased and then increased and were 151.5–261.9 gC/m² and 54.6–116.3 gC/m², respectively. The NPP value of more than 70% area was 90–150 gC/m² in July. In August, the NPP_mean values decreased, and the areas with lower values became larger; the NPP_mean values of most areas affected by geohazards were 60–150 gC/m². During the earthquake, the NPP_mean values of areas hit by geohazards sharply declined by 27.2% (landslide), 22.4% (debris flow) and 15.7% (collapse) compared with those in the same month in 2016. Vegetation in debris flow zones showed a stronger recovery, with a maximum NPP value increase of about 23.0% in September 2017. The vegetation gradually recovered after the earthquake, as indicated by the uptrend of the NPP values in the corresponding period in 2018 and 2019. In general, the reduction magnitude of NPP values decreased year by year in comparison to that in 2015 and 2016, and the decrease slowed down after the earthquake. The vulnerability and resilience index corresponding to the three seismic intensity ranges were 0.470–0.669 and 0.642–0.693, respectively, and those of Jiuzhaigou Valley were 0.473 and 0.671, respectively. The impact coefficient defined to represent the impact of the earthquake on NPP was 0.146–0.213. This paper provides a theoretical reference and guidance for the impact assessment of earthquakes on the ecosystem.