|
|||||
|
|
| 改进的案例推理CA模型及土地覆盖变化模拟 | |
| 引用本文： | 张琴,张友静,张滔.改进的案例推理CA模型及土地覆盖变化模拟[J].地理与地理信息科学,2012,28(3):59-62,113. |
| 作者姓名： | 张琴 张友静 张滔 |
| 作者单位： | 1. 河海大学地球科学与工程学院,江苏南京,210098 2. 河海大学水文水资源与水利工程科学国家重点实验室,江苏南京210098;河海大学地球科学与工程学院,江苏南京210098 3. 上海数慧系统技术有限公司,上海,201210 |
| 基金项目： | 国家自然科学基金，国家"973"计划项目 |
| 摘 要： | 对基于案例推理的元胞自动机模型(CBR-CA)进行改进,将各类别的宏观转移概率添加到目标函数中,体现各类别的转变特征,并增加时间权重来确定转移概率,实现时间尺度上的模拟;由于土地覆盖变化的多样性和空间结构的复杂性,利用Monte Carlo(M-C)法确定土地覆盖的最终转换类别。选择黄河源区为试验区,利用1977年、1985年土地覆盖数据建立原始案例库,模拟了该区域1995年、2000年和2006年的土地覆盖变化,模拟的各类别转换的数量精度与实际相吻合,各年份的总体误差分别为0.002%、0.012%和0.005%,空间位置精度总体在70%以上,并进行未来土地覆盖情景预测。该模型可用于多类别、长时间序列区域土地覆盖变化的模拟与预测。 |
| 关 键 词： | 案例推理 元胞自动机 土地覆盖变化模拟 黄河源区 |
Improved Case-Based Reasoning Based Cellular Automaton for Simulating Land Cover Change |
|
| ZHANG Qin , ZHANG You-jing , ZHANG Tao.Improved Case-Based Reasoning Based Cellular Automaton for Simulating Land Cover Change[J].Geography and Geo-Information Science,2012,28(3):59-62,113. | |
| Authors: | ZHANG Qin ZHANG You-jing ZHANG Tao |
| Institution: | 1.School of Earth Science and Engineering,Hohai University,Nanjing 210098;2.State Key Laboratory of Hydrology-Water Resource and Hydraulic Engineering,Hohai University,Nanjing 210098; 3.Shanghai Shuhui Technology Co.Ltd,Shanghai 201210,China) |
| Abstract: | This paper presents an improved Case-Based Reasoning based Cellular Automata model(CBR-CA) in modeling multiple land covers changes.Currently,various research communities have tried to build and apply mathematic models with CBR or CA as systemic method to study concerning specific geo-scientific problems.However,original CBR-CA model has problems in modeling numerous class changes,especially in modeling land cover changes with large region and long time.In this paper,an improved CBR-CA is designed to calculate conversion probabilities for computing multiple land cover.Firstly,the macro transition probability of each class is added to the objective function,which can express the characteristic of multiple class changes.Then,considering the time weight,which can reflect the space-time changes of land cover,the final probability function is derived by both.In addition,due to variation of land cover types and their spatial relationships in complex geography changes,Monte Carlo is made use of deciding the final transition classes.This model has been applied to simulate the land cover changes of the source of the Yellow River in 1995,2000 and 2006,and has been applied to predict this region in 2012 and 2018 by building the historical case-based database with the land cover data in 1977 and 1985.The simulated and actual patterns are basic consistent in amounts,accounting for 0.002%,0.012% and 0.005% of the total error,respectively,the space accuracies totally beyond 70%.The result indicates that this model can simulate and predict the land cover changes which are with multiple classes and with long time. |
| Keywords: | Case-Based Reasoning Cellular Automata land cover changes modeling source of the Yellow River |
| 本文献已被 CNKI 万方数据 等数据库收录！ | |