中文摘要：

“城市群”是我国新型城镇化的主体形态，对推进国民经济发展具有重大意义，但其聚集连片的快速扩张模式对资源环境的压力持续增加，已经成为制约未来可持续的瓶颈。以我国经济发展最为活跃，但生态环境问题十分突出的京津冀城市群为例，基于CLUE—S模型，模拟分析了1990-2010年京津冀城市群景观格局的变化特征及其驱动机制，并预测了未来景观格局的变化趋势。结果表明，（1）1990-2010年京津冀城市群景观格局变化显著。其中，人工表面持续增加，耕地明显下降，林地和草地格局的变化也存在明显的时序差异；（2）京津冀城市群景观格局的变化主要受自然和社会经济要素的综合影响，且不同景观类型之间的驱动机制存在明显差异。其中，林地更易在地势较高、坡度较大的西部地区分布。而河流、人工表面等更易在平坦低洼的区域分布；此外，不同景观类型变化的驱动机制存在显著的时序差异，例如，人工表面受地形的影响程度逐步降低，呈现更加离散的分布，且其分布特征由较早时期的向市中心集聚分布发展为逐渐远离市中心并向铁路、高速路周边集聚的趋势；（3）经检验，CLUE-S模型能够较好地动态模拟京津冀城市群的土地覆盖格局的变化特征，模型的Kappa指数达0．84。模拟预测结果显示，未来（2020年）景观格局演变的显著特征是人工表面将持续增加，耕地将继续显著减少。北京、天津、唐山和石家庄等核心城市的景观格局变化将最为显著。

英文摘要：

Urban agglomeration is the major form of China＇s new urbanization, which has greatly advanced national economic development. However, rapid regional urbanization has posed serious pressures on ecosystems, and it restricts the sustainable development of cities. We studied the Beijing-Tianjin-Hebei Metropolitan Region （BTH）, which is undergoing rapid urbanization but is also facing many ecological environmental problems, as a case example to analyze the dynamics of landscape change and their driving factors during 1990--2010. To do this, we used the logistic model and simulated changes in land-cover by using the CLUE-S model. Then, we predicted the changing trends of the landscape patterns of BTH in 2020. The results were as follows： （ 1 ） there was a marked change in the landscape patterns of BTH during 1990-- 2010. The proportion of the artificial surface clearly continued to increase, whereas that of the cultivated land significantly decreased. The spatial configurations of forest land and grassland showed obvious temporal changes; （2） The major driversof land-cover changes of BTH included both topographic and social-economic factors. Moreover, for different landscape types, the driving dynamics varied. For instance, the forested lands were more likely to occur northeast of BTH at high elevation and steep slopes while the water body, artificial surface and other landscape types were more likely to occur in plains and low-lying areas. The driving forces of landscape change also changed temporally. An example of this is how the distribution of artificial surface changed from being close to the central area in 1990-2000 to being away from the center, thus closer to railways and highways; （3） The CLUE-S model was robust in its application for the simulations of land-use dynamic changes of BTH, and the kappa index was 0.84. By 2020, the artificial surface of BTH was predicted to increase whereas cultivated land would continue to decrease. The most stark changes in landscape patterns should occur i