| A large amount of research has proven that the land use/cover change (LUCC) resulting from human activities can bring significant effect on regional and even global climatic/ecological changes. As technologies and productivity improve and the way of life and ideology of humans evolve, such effect has become much more conspicuous since the industrial revolution in western world. In history, particularly in the past 300 years, land use/cover change has been an important parameter used to conduct long-time-scale climate simulation study, diagnose climate forming mechanisms, identify the sensitivity of the climatic system to the nature and human forces and predict future climate change and so on. As the global change study continuingly intensifies, LUCC draws extensive attention from the international academic world. In the 1990s, a LUCC research project jointly initiated by the International Geosphere-Biosphere Programme (IGBP) and the International Human Dimensions Programme (IHDP) on Global Environmental Change stressed that the history of land use changes in the past must be reconstructed by all necessary means, which triggered a wave of researches into the land cover changes in history. Reconstruction of the data on land cover in history, particularly the high-precision spatial data, has drawn extensive attention of scholars. Representative achievements of the researches intended for the reconstruction of the spatial pattern of land use include the SAGE dataset of the Center forSustainability and the Global Environment and the Historical Database of the Global Environment (HYDE dataset) of the Netherlands Environment Assessment Agency. Some scholars have argued that these two datasets seem to be very rough when used in regional simulation researches, in terms of either arable land amount estimation or spatial distribution pattern, and that the data acquired can only be applicable in a global scale and cannot be used as the basis for regional researches. The traditional agricultural area refers to the region mostly located in the second-and third-level terrain ladders in the topographic pattern of China which is to the east of the Hu Huan-yong population line and to the south of Yanshan Mountain. In terms of administration organization system, this region refers to the 17 inland provinces as administrative divisions existing in the 25th year during the reign of Emperor Jiaqing in the Qing Dynasty, generally covering North China, regions south of the Yangtze River and southwest regions, with Zhili Province and Shanxi Province excluding the regions north of the Yangtze River and Fujian Province excluding Taiwan Province. In history, this region was the primary cropping area and socio-economic active area of China and the cultivated land areas in this region were rather concentrated. In the recent 300 years, despite the dynastic transitions and warfare during which farming had to be discontinued, the peripheral boundaries of the cultivated land areas did not change significantly. Spatial changes of the cultivated land were just temporary discontinued farming and restorative reclamation. This region is the key area of the research into reconstruction of the spatial pattern of cultivated land of China in history.Considering the concentrated distribution of cultivated land and various factors influenced cropland distribution, including the limitation of nature and human factors, this study developed a reconstructed model of historical cropland based on constrained CA of "bottom-up" and the partitioned& synchronous notion. Then, to inspect and calibrate these spatial patterns, author built an examination system from the view of the available cropland index of labor force. Taking historical cropland area and 1980’s cropland distribution as an external variable and maximum potential scope of historical cropland respectively, selecting altitude, slope, river system, average annual precipitation and distance to the settlement as the main influencing factors of land-use suitability degree, regarding population as testing factor, this paper applied the reconstruction model to traditional cultivated region in China, reconstructed its spatial distribution of cropland during 6 typical periods. The results are shown as follows:(1) Based on the modeling concept of the constrained CA model, this article takes into account the effects of factors such as cell state and cell neighborhood on land use units and treats the natural environment and cultural environment in regional land use as constraint conditions, to enable the model to simulate the spatial changes of cultivated land in history. It is properly suitable for applying constrained CA to simulate and reconstruct the spatial distribution of cropland in traditional cultivated region.(2) Taking the different factors affecting spatial pattern of cropland into consideration, the partitioned & synchronous notion have effectively reflected the spatial difference of cropland evolving rule and rate which is caused from different spatial scale.(3) Coupling with modern cultivated land spatial pattern, land suitability, neighborhood and random factor, under the control of the historical amount of cultivated land, the constrained CA model is used to reconstruct the 1km×1km cultivated land spatial pattern under the traditional cropland section, so as to provide a new angle or way for the research into reconstruction of the spatial pattern of cultivated land in history.(4) Reconstructiong historical cultivated land in the traditional cropland region is mainly located in the plain of north China, Yangtze plain, Guanzhong plain, Sichuan basin, Dongting lake plain, coastal beach in eastern china, hilly area in southeast china, which were the primary cropping area and socio-economic active area of China and the cultivated land areas in this region were rather concentrated in history, due to its flat terrain, rich in water resources, fertile soil, suitable agricultural farming temperature.(5) Conparied this paper’s result with HYDE3.1、SAGE2010, CHCD datasets, while the increasing amount and rate in the cropland varys, four datasets’amount of cultivated land have showed an upward trend without exception in historical traditional cropland region on recent 300 years, which agree with the population’s increasing trend perfectly. SAGE (2010) dataset was established by linear interpolation and the average annual growth rate was 0.51%from 1700 to 1950 and-0.34% after 1950. The normalized trend could not reflect historical traditional cropland region’s history of cultivation, and the estimation of cropland area of HYDE3.1 was obviously high compared with this paper’s dataset. Due to collect more regions’researchs, HYDE3.1 was better than SAGE2010 dataset in the area and trend of cropland. Because of collecting more regional research result and referencing true historical data, so, CHCD and this paper’s dataset truly reflect the objective reality in traditional cropland in china, dispite some difference, which have similarity and substitutability in the change trend and numerical change of cultivated land.(6) Compared with "HYDE datasets", this research has formed Boolean spatial distribution datasets of historical cropland with higher solution, which have more definitive concept of spatial pattern in term of fractional datasets. Reconstructed result is more closed to the actual change pattern of traditional cultivated region in China. |
PDF Full Text Request