Accurately identifying soil class at a specific point-location or position within a landscape is critical for implementing sustainable soil management. Soil classes (e.g., soil components) are information carriers that allow land managers to infer a general range of soil behavior in response to management actions and disturbance effects. Recent advances in information technologies, in particular the global ubiquity of smart phones, has made it possible to create m.obile decision support tools that can inform rangeland management decisions. The Land Potential Knowledge System (LandPKS) is one such example, providing a complete mobile computing platform for assessing land potential and informing management activities. Here we present the development of a global soil identification modeling framework (SoillD) implemented within the LandPKS mobile app. SoillD leverages smartphone-based data acquisition and information delivery, with cloud-based computing to determine the most probable soil class at a user specified point. SoillD makes it possible for non-soil scientists to describe and identify soils in the field using limited, simple soil observations. Additionally, SoillD provides information on Ecological Sites based on the most probable soil class matched to the user's soil. Our presentation will describe the details of SoillD and its implementation in the LandPKS mobile app, and provide examples of its utility for rangeland management.