Ecological footprint (EF) analysis estimates the"load"imposed on the ecosphere by any specified human population or activity. It is already proved to be a good measure of the dependence and impact on nature of production and consumption. In recent years EF researchers have expanded their focus from macro-level accounts of the national, regional and global EF of human activities to both mesco-level EF accounts of industrial sectors and micro-level EF calculation of some defined products. But the deficiency of EF parameters has become the bottleneck. After air and water, food is the most essential resource people require to sustain ourselves. We therefore provide an EF accounts of producing key edible agricultural products with about 80 kinds of raw and processed agricultural products involved. And in the case of production, the EF of selected products represents the amount of bio-productive land required to produce the raw materials and to absorb corresponding waste, represented in global hectares per tonne of products. Life Cycle Assessment (LCA) was applied with supply-chain starting from the raw material extraction and ending with the formation of the product. Thus, the EF of a defined product exclude those impact during distribution, use, maintenance and disposal. The data for the ecological footprint of raw agricultural products (primary goods) is mainly excerpted from National Agricultural Costs and Returns Compilation of 2011. Derived agricultural products are related to primary goods by demand coefficient for primary goods, which is the mass ratio of primary input required to derive per unit of product. All the data on demand coefficient are from both published yearly statistics and investigation. Based on demand coefficient for primary goods and EF of primary goods, bio- capacity needed to support the production of processed food products could be calculated. For both primary and derived agricultural goods, its EF is shown in consumption-land use matrix. Using the ecological efficiency of production provided by this study, EF could be easily estimated required to support some consumption patterns at different levels, such as individual, community, organizational and regional.