Abstract Context Forest fragmentation, defined as the spatial configuration of habitat within landscapes, has been quantified using an expanding range of metrics and analytical workflows. Although methodological diversity has increased rapidly with advances in remote sensing and computational capacity, comparability and ecological interpretability remain uneven. Objectives This review advances forest fragmentation analysis by systematically tracking the evolution of methodological families from 1990 to 2025 and identifying structural constraints that limit transferability across regions and scales. Methods We synthesized 138 studies and quantitatively analyzed an operational subset of 127 methodological and hybrid papers. Studies were grouped into twelve methodological families, and their proportional representation was evaluated across four temporal periods. Study level classification and complete search details are provided in the supplementary materials to ensure transparency and reproducibility. Results Patch–mosaic metrics remain the analytical backbone of fragmentation research. Over time, analytical approaches have expanded from early patch-based measures toward connectivity-oriented, density-based, and emerging three-dimensional formulations. This trajectory reflects cumulative methodological expansion rather than paradigm replacement. Across methodological families, recurring constraints include sensitivity to spatial support, context-dependent parameterization, uneven validation of automated or global products, limited linkage to biological responses, and inconsistent reporting of methodological settings. Conclusions Progress in fragmentation analysis is likely to depend less on introducing new indices and more on strengthening comparability through explicit documentation of habitat definitions, spatial resolution, connectivity rules, edge settings, window supports, and change detection parameters. By clarifying methodological trajectories and emphasizing transparent reporting, this review provides a structured foundation for more transferable and ecologically grounded fragmentation research.
Dutt et al. (2026) studied this question.