The concept of historical range of variability (HRV) was initiated and developed by many ecologists for ecosystems management in 1990s, providing reference and goal conditions for managed forest ecosystems. The past decade has seen an increasing interest in forest management based on historical or natural disturbance dynamics. This review summarizes the recent trends in theories, methods and applications. The researches focused on initially characterizing historical fire regime indicated by frequent and low-intensity fire occurrence, and later analyzing impacted factors of fire regime involving topography, elevation and fuel loading etc.. Recently, more attention has been paid to natural characteristics of forest ecosystem at multiple spatial and temporal scales instead of a certain single scale in understanding the historical fire regime. The climate change creates up-bottom controls but the topography does bottom-up controls on fire spread. Meanwhile, the historical landscape dynamic has been investigated by more indicators in structure and even function instead of single structural indicator. Generally, the long-term unreasonable management moves the landscape away from the HRV and it required long time to toward and reach HRV. Methodologically, Scientists focus on the evaluation of errors in historical data, especially those data of Public Land Survey (PLS) used widely in estimating HRV in North American, novel methods for sampling or analyzing and multiple-scale characters of fire scar data. Also, they seek for the multiple historical data resources to better reconstruct the landscape dynamic. Practically, Natural Disturbance Emulation (NDE) is considered as one of important applications of HRV in forest ecosystem management and the research about NDE has been improved to the more precise emulation in multiple parameters including frequency, intensity and size of disturbance. Additionally, the heterogeneity of HRV is emphasized in emulating the natural disturbance, meaning varying historical parameter should be provided in different ecoregions. However,two problems were argued: 1) Will the HRV still be the best option in the context of growing climatic change? Despite of the opposed opinions, some of scientists consider the NDE as a useful concept before reducing the uncertainty of climate change. 2) Will those anthropogenic harvests and fire managements have the same impacts as the natural disturbance? More long-term researches are needed to answer those questions. Furthermore, this review elaborates the existing weak research in China. Most researches are conducted in Great Xing'an Mountains where the fire occurred frequently due mainly to dry climate to demonstrate the human-disturbed fire regime compared to historical range. Then this review seeks for reasons for the gaps with other countries on one hand, we are faced with poor understanding of the nature of the landscape and the absence of strong records of landscape conditions prior to intensive anthropogenic alteration, on the other hand, the forest in our country was explored for hundreds of years and less private forest in large area are maintained, which hinders greatly investigating the HRV. To close this gap, the research should be conducted at multiple scale and strengthen simulation by landscape model while seeking for better historical data.