The methodology for extracting, integrating and mapping of massive geo-data is a new method combined by different disciplinary approaches in agriculture and environmental science, geo science, cartography, information and computer science. The methodology consists mainly in the data model design and the working flow design for processing, analyzing and mapping massive geo-info. By using the methodology, big data in heterogeneous format and structure originated from different resources and regions of agricultural and environmental research and working programs can be effectively extracted, analyzed and correlated for mapping and thematic expression. The method can be used to analyze point observation data, map data, remote sensing data in different spaces and times. So that it supplies a useful tool for providing information with higher accuracy, larger covering area and more system concerning elements, through which major factors and mechanism in agricultural and environmental system can be much more exactly qualified and quantified. Based on many years of research and practical experience, the author of this paper introduced the connotation, the application range, the basic and the concerning conceptions and the main contents of the methodology, with the purpose to provide a brief understanding for researchers and managers in agriculture and the environment sectors to apply the new method in their working fields. As a big data approach, the method can be widely used for evaluation of the quantity and quality of soil resources, climate change, environmental quality, evaluation of crop varieties suitable distribution area, agricultural non-point source pollution control，erosion control, drought and flood disaster mitigation, and other working and research areas. It can also be applied for precise and visualized expressing of soil fertility and environmental quality, so that farmers or other users from different sectors can access to research results and progress much easier and get benefits from it. It provides also a useful approach to establish scientific basis for developing and implementing incentive policy in agriculture and environment sectors. The core of methodology is to define the rules according to scientific target for classifying massive geo-data. Based on the rules defined, grouping, coding, extracting and mapping of massive geo-data can be then carried out. Because of horizontal and vertical features of data structure of a massive geo-dataset, the four component expression of massive geo-dataset should be applied. Through which both logical and physical structure differences of massive geo information originated from different sources can be distinguished and displayed clearly. After normalization of data logical structure, the extracting, integrating and mapping of massive geo-datasets are then followed. In agriculture and environmental research works, however, frequent difficulty of big data analysis approach is the weakening or even lost of the scientific target during data treatment. Therefore, scientific or specific target should be defined as precise as possible at the beginning of the data analysis working program. A five-level designing process should be applied for drafting working flow of data extracting, integrating and mapping. Checking and examining the realization of defined target should be done time to time during the data processing. With deep understanding of the target, researchers and professionals from agriculture and environment sectors should be responsible for designing data processing flow of high-levels and drafting the corresponding design documents according to specifications of the methodology.