土地利用数据综合是从大型土地利用数据库中派生多尺度数据的基本操作，提高土地利用数据综合的客观性和可靠性的有效途径是解析数据自身特征、从数据特征导出影响综合操作的规则、遵循规则实施土地利用数据综合操作。本文主要探讨随地理区域不同而变化的区域依赖性数据综合规则的提取和应用。首先阐明了数据综合规则的实现目标和规则层次性；然后详细描述了利用空间关联关系分析和景观指数分析来确定用地类型的重要性级别、各用地类型地块的最小面积域值和空间分布状态等规则的提取方法和应用，并以湖北省九峰县1:10 000土地利用数据综合生成1:50 000数据为例检验上述规则的有效性。

Land-use data generalization is one of the basic operations for deriving multi-scale data from a comprehensive land-use database. Acquiring data characteristics and converting them into generalization rules used to implement data generalization is an effective way to improve the objectivity and validity for land-use data generalization. This paper focuses on the extraction and application of region-dependent rules for land-use data generalization. First, the objectives and hierarchy of the rules are clarified. Then, spatial association analysis and landscape indices are employed to set the importance ranks, minimum area thresholds and spatial distribution pattern of each land-use class. Finally, a case study was undertaken to show the application of these rules using 110 000 land-use data around Jiufeng county of Hubei, China.