This study examines whether the level of community accessibility to green spaces varies as a function of socioeconomic factors, focusing on three underexplored aspects affecting green inequity assessment: implications of alternative green measures, spatial non-stationarity of socioeconomic impacts, and regional/local context. Using six green measures involving general greenness, tree canopy, developed open space, agricultural lands, parks, green trails, golf courses and sports fields, Geographically Weighted Regression models are developed, and coefficient surfaces are created and compared over two metropolitan areas, Columbus and Atlanta. Second-order effects among individual factors are tested to explore synergistic linkages among them. The results show that the identified patterns of green access inequity vary significantly with the selected green measure. In both metropolitan areas, access to parks, green trails, golf and sports fields turns out to be spatially independent from general greenness and tree canopies. GWR coefficient surfaces show that the socioeconomic-green relationships are spatially heterogeneous and context-dependent, affected by a complex web of forces, including urban heritage, racial and lifestyle diversity, and natural landscape. While natural landscape and income have greater effects in Columbus, racial disparity is dominant in Atlanta. Income-driven disparities are most visible in inner suburban areas and tend to abate in the urban core and exurbs. Urban greening programs need to diversify in terms of green type and location to address localized deficits over a wide urban spectrum.