The term “big data” first emerged in scientific communities in the mid-1990s and gained popularity in 2006, a year after the creation of the second generation of the World Wide Web . In general, big data refers to a tremendous and complicated dataset that is difficult to store, manage, and process using traditional processing tools. Importantly, big data is characterized by three dimensions known as the 3Vs. First is volume, an inherent characteristic of big data that comprises a huge volume of data from various sources which poses challenges for storage and analysis. Second is variety, because big data typically comes in various types and formats, which may come to the user already combined in some fashion and/or may need to be combined by the user for a specific purpose. Accordingly, tremendous efforts have been conducted to manipulate different data types with complex structures. The third is velocity, which deals with the unprecedented speed of data streams emerging from different sources. Over the past few years, big data analysis has drawn attention in different disciplines, such as business, health science, disaster management, and geoscience. The growth of geospatial data has changed our perception and interaction with the planet. Given the massive volumes of existing geospatial data, its variety of origins and formats, and growing diversity and accessibility, it can be defined as big data. Geo-big data is collected from different sources such as ground surveying, remote sensing, geo-located sensors, and mobile mapping. When it comes to remote sensing big data, special intrinsic and extrinsic characteristics can be determined. Dynamicstate, multi-scale, and non-linear features are intrinsic characteristics of remote sensing big data. In particular, remote sensing big data reflects a dynamic state, as the Earth’s surface changes continuously. Multi-scale features are related to resolution, time interval, spectral range, angle and polarization. In addition, remote sensing big data is nonlinear, since time series data are typically non-linear and noisy. On the other hand, the multi-source, high-dimensional and isomer characteristics are extrinsic characteristics of remote-sensing big data. The reason behind the first two characteristics is the existence of different sensors and spectral/ temporal dimensions of satellite data, respectively. The isomer characteristic reflects the variation in structure of available remote sensing data, such as raster or vector. These characteristics raise several challenges, including the acquisition, storage, searching, sharing, transferring, analysis, and visualization of big data. To overcome these difficulties, the need for novel methods is imperative.