Applied Numerical Modelling of Landscape Evolution in Anthropogenic Landscapes: A Systematic Review

Understanding long‐term landscape evolution in anthropogenic contexts is critical as cumulative geomorphic change over extended timescales shapes hazard regimes and impacts the resilience of human communities. Conventional approaches have tended to prioritise short‐term events, leaving broader system dynamics underexplored. Landscape evolution models (LEMs) offer a robust numerical framework for modelling long‐term landscape change, yet their practical application remains insufficiently synthesised in the scholarly literature. The existing literature, although grounded in theoretical and methodological advances, offers limited insight into the practical application of these models for real‐world planning and management. This systematic review addresses this gap by examining how numerical modelling has been used to analyse the long‐term evolution of human‐modified landscapes. Adhering to Preferred Reporting Items for Systematic Reviews and Meta‐Analyses (PRISMA) guidelines, 55 peer‐reviewed journal articles and formally published conference papers were retrieved through an unrestricted global search of Scopus and Google Scholar across all publication years and languages, with a focus on original research to ensure analytical rigour. Studies were included only if they employed numerical, process‐based models that produced quantitative geomorphic outputs, rather than purely conceptual approaches. The review aims to evaluate the anthropogenic settings in which modelling has been applied, examine the methodologies employed, synthesise key findings, identify gaps and limitations and assess the potential of these models to guide long‐term decision‐making. The analysis revealed that modelling applications are predominantly focused on post‐mining and mine‐rehabilitation landscapes, whereas other anthropogenic land‐use contexts remain relatively underrepresented. The models are inherently parameter‐driven, making accurate parameter derivation a significant challenge in data‐scarce regions and a primary source of model uncertainty. Validation has been noted as a limitation in nearly all studies, as the long‐term nature of landscape evolution precludes the availability of observational data to rigorously test model predictions. This highlights the importance of fostering closer collaboration among researchers, infrastructure managers and local communities to enhance model reliability, ensure practical relevance and support informed decision‐making for landscape and risk management.