At landscape scale, the normalized difference vegetation index (NDVI) can be used to indicate the vegetation’s dynamic characteristics and has been widely employed to develop correlated and dependent relationships with the climatic and environmental factors. However, studies show that NDVI-environment relationships always emerge with complex features such as nonlinearity, scale dependency, and nonstationarity, especially in highly heterogeneous areas. In this study, we used geographically weighted regression (GWR), a local modeling technique to estimate regression models with spatially varying relationships, to investigate the spatially nonstationary relationships between NDVI and climatic factors at multiple scales in North China. The results indicate that all GWR models with appropriate bandwidth represented significant improvements of model performance over the ordinary least squares (OLS) models. The spatial relationships between NDVI and climatic factors varied significantly over space and were more significant and sensitive in the ecogeographical transition zone. Clear spatial patterns of slope parameters and local coefficient of determination (R ²) were found from the results of the GWR models. Moreover, the spatial patterns of the local R ² of NDVI-precipitation are much clearer than the R ² of NDVI-temperature in the semi-arid and subhumid areas, which mean that precipitation has more significant influence on vegetation in these areas. In conclusion, the study revealed detailed site information on the variable relationships in different parts of the study area, especially in the ecogeographical transition zone, and the GWR model can improve model ability to address spatial, nonstationary, and scale-dependent problems in landscape ecology.