Ecological-Niche Factor Analysis(ENFA) is a multivariate approach to study species geographic distribution on a large scale with only presence data. It was widely applied in many fields including wildlife management, habitat assessment and habitat prediction. In this paper, this approach was applied in habitat suitability assessment for giant pandas in Pingwu county, Sichuan province, China. With giant pandas presence data and remote sensing data, panda habitat suitability in this county was evaluated based on ENFA model, and spatial distribution pattern of nature reserves and conservation gaps were then evaluated. The results show that giant pandas in this county prefer the high elevation zones(>2128m)dominated by coniferous forest, mixed coniferous and broadleaf forest,and avoid deciduous broadleaf forest and shrubs. Pandas avoid staying at habitats with human disturbances. Farmland is a major factor threatening panda habitat. Panda habitat is mainly distributed in northand west of Pingwu with atotal area of 234033 ha(106345 hafor suitable and 127688 ha for marginally suitable).Three nature reserves located in Pingwu, cover 49.2% of total suitable habitat and 45.6% of total marginally suitable habitat. Although 47.2% of pandas′ habitat was under protection in reserves, the connectivity between reserves is weak and a conservation gap existed in the north part of Pingwu. Thus, a new nature reserve in Baima and Mupi should be established to link the isolated habitats.