<p>Vegetated ecosystems are subject to abrupt regime shifts, influenced by external factors. These critical transitions threaten ecological and human societies, and have received a great deal of attention in the last decades. Despite extensive research, early detection of tipping points across diverse datasets remains a persistent challenge. In this study, we propose a novel spatial occupancy entropy (OE) indicator to detect impending regimes shifts by quantifying the complexity of vegetation spatial patterns. Specifically, we apply this indicator to three distinct types of vegetation pattern data: periodic, scale-free, and realistic vegetation transects. Our analysis reveals a consistent and reliable early warning criterion: a reduction in spatial OE that correlates with a decline in control parameters. Our results confirm the robustness of the spatial OE indicator with respect to variations in sampling window size and step length. Furthermore, when compared to traditional critical slowing down indicators, the spatial OE shows clear advantages in detecting impending critical transitions. Thus, our findings suggest that space OE as an additional metric for early detection of critical transitions in vegetated ecosystems.</p>