High‐mountain rock slopes are increasingly controlled by ice-affected, discontinuous joints. This study quantifies how joint persistence and infill govern shear resistance and upscales the implications to slope stability. We use a three-dimensional discrete element method (DEM) to simulate fully persistent and discontinuous joints with empty, sand-filled, and ice-filled apertures. From contact forces and bond-break histories we measure a stress-partition factor (n)—the ratio of normal stress borne by rock bridges to that in infill—and propose a Generalized Jennings Criterion (GJC) in which cohesion mixes linearly with persistence (k), while friction is blended through n. DEM results show: (i) for fully persistent joints, cohesion increases with ice content whereas friction decreases; (ii) for discontinuous joints, low-stress cohesion scales ≈(1−k) times the intact value for all infills, while friction is nearly k-invariant for empty/sand but drops with k for ice; and (iii) highly persistent ice-filled joints exhibit diagnostic double-peak shear from sequential infill/interface degradation followed by bridge failure. Contact networks reveal n>1, larger for ice than for sand, and declining as k→1. Embedding the DEM-derived envelopes and GJC into a rigid-block slope model of the Punta Gnifetti east face reproduces the present morphology with two admissible states: empty joints (φ≈35–40°, c≈0.5–0.7 MPa) and ice-filled joints at −1 to −4 °C (φ≈11–13°, c≈1.3–1.4 MPa), both requiring high persistence (k≈96.5–99%). Scenario analyses indicate progressive retreat under increasing k for empty joints, but step-like recession under modest warming for ice-filled joints. The GJC thus offers a practical, temperature-aware framework for stability assessment in alpine permafrost rock slopes.
 

Discontinuous joints; Joint persistence; Discrete Element Method (DEM); Alpine permafrost rock slope; Ice-filled joints; Generalized Jennings Criterion (GJC)