Glaciers sustain downstream water resources by providing delayed meltwater inputs, but ongoing climate change is rapidly altering their dynamics and associated hydrological responses. Understanding how glacier variability propagates through hydrological systems remains a key challenge, particularly in data-scarce mountain regions. This study characterises glacier surface dynamics and quantifies their lagged influence on streamflow, and explores associated vegetation responses, across glacierised catchments in central Chile. Multi-decadal Landsat observations were used to derive physically based glacier proxies, including snow-covered fraction, albedo, land surface temperature (LST), and snow elevation. Principal component analysis was applied to decompose glacier behaviour into dominant modes of long-term change, short-term variability, and temporal persistence, while prewhitened cross-correlation analysis was used to assess lagged relationships with downstream responses. Physical consistency among glacier proxies was evaluated to ensure coherent representation of glacier surface conditions. Results reveal a pervasive signal of glacier degradation, with most glaciers exhibiting declining trends in snow-covered fraction and albedo alongside increasing LST. Variability shows strong scale dependence, with contrasting glacier-level and regional trends driven by the influence of large glaciers, whereas temporal persistence exhibits a consistent increase across scales. Lagged analyses show a transition from glacier-dominated headwaters, where glacier variability precedes streamflow, to downstream basins where hydrological responses become increasingly decoupled from cryospheric processes. A regression framework linking glacier-response modes to hydrological lag explains a substantial fraction of variability in streamflow timing (R² ~0.72). Vegetation responses are spatially heterogeneous and not significant at the system scale. These findings demonstrate the potential of satellite-based approaches to quantify the timing, persistence, and propagation of cryospheric signals in data-scarce mountain regions.