中国历史时期土地覆被数据集地理空间重建进展评述

重建长时间序列具有空间属性的土地覆被数据集,对研究历史时期土地利用/土地覆被变化及其气候和生态效应具有重要意义。近年来,国内外学者就定量重建中国区域历史土地覆被数据集进行了积极探索。但由于历史时期土地利用数据来源多元、重建方法多样、验证方式各异等原因,不同学者的重建结果迥异,其中重建方法是导致差异形成的重要原因之一。本文从重建思路、假设和方法、结果验证等方面对覆盖中国区域的主要空间数据集进行了综合评述,结果表明：①基于历史记录的还原法和基于地理空间模型的重建法是历史土地覆被空间重建的主要方法,而根据建模过程,后者又可进一步分为“自上而下”的配置模型和“自下而上”的演化模型法。②基于数量重建进行空间重建是当前历史土地覆被数据集重建的主流,在缺少充分、客观历史数据的条件下,对基础数据、分布控制因素和限制因子进行合理假设是取得合理结果的重要条件。③为提高研究成果的解释力,需要对重建结果进行检验,直接验证法虽较为准确,但受时空尺度限制,具有显著的局限性,间接验证法可作为有效的补充。

Review on China’s spatially-explicit historical land cover datasets and reconstruction methods

Constructing a spatially-explicit time series of historical land cover dataset is of upmost importance for climatic and ecological studies that make use of land use and cover change (LUCC) data. Some scholars have made efforts to simulate and reconstruct quantitatively information on China's historical land use. Due to the multiple sources of land use data, diversity of reconstruction approaches, and different methods of verification, the reconstruction results of spatially-explicit historical land cover datasets significantly differ. To better understand China's historical land cover datasets and provide effective references for future reconstruction study, it is necessary to comprehensively summarize and discuss the reconstruction methods. By comparing the characteristics of different historical land cover datasets, this research thoroughly analyzed the conceptual constructs, assumptions, reconstruction methods, and validation of the models. The main conclusions are as follows: (1) The majority of the research methods starts with reconstruction of the quantity of historical land cover and then proceeds to spatial pattern reconstruction. Given the lack of adequate and reliable historical data, making reasonable hypotheses about basic data, factors controlling distribution, and limiting factors is an important condition for attaining sound reconstruction results. (2) Methodologically, the majority of the studies uses the reduction method based on historical records and the reconstruction method based on spatial models that consist of the "top-down" method of spatial distribution and the "bottom-up" method of evolutional reconstruction models. (3) In order to improve the explanatory power of the results, it is necessary to validate the reconstruction output of historical land cover datasets in detail. While direct validation of reconstruction results is a more precise method, it is often restricted by the spatiotemporal scales of the research and data sources. Indirect validation method provides an important alternative for accuracy evaluation of the reconstruction results.