Relationship between landscape pattern and environmental factorsin Jiaodong mountain region -The case of Qixia City in Shandong Province

To determine the distribution characteristics of landscape patterns in mountain regions, the case of Qixia City was investigated and a geological database built for grid datum using the 2003/2010 land use/landscape pattern map, DEM map, soil map, etc. of Qixia City. These data were processed in ARCGIS and CANOCO environments on Windows 4.5 platform. CCA (canonical correlation analysis) ranking method was used in the study to comparatively analyze changes in landscape pattern and environmental factors in different years in Qixia City. The results showed that percent cumulative variances of landscape-environment relation were respectively 98.8% and 98.7% in 2003 and 2010, indicating a decreasing trend. However, correlation analysis between landscape and environment showed a higher coefficient for 2003 than for 2010. Meanwhile total eigen-values of the ordination axes were 0.572 and 0.808, respectively. Different environmental factors had different impacts on landscape pattern with time. There was little change in land use/landscape pattern between 2003 and 2010. In addition, the other six land use types were not obviously different, except for unused land. There were similar regularities in the relationship between landscape pattern and environmental factors for 2003 and 2010. The main factors with major impact on landscape were sunshine hours, slope gradient and altitude. Water surface and build-up land were in areas with higher sunshine hours, lower altitudes and gentler slopes. However, unused land and grassland were mainly in the other areas. Through land use/landscape pattern and environmental factor relationship, this paper determined land use/landscape pattern distribution characteristics in Qixia City. This study therefore provided theoretical basis for adjustments in agricultural land structure and in the general land use structure for optimized land use in the study area and beyond.