Recent advances in robotic vision are permitting breakthrough solutions to various agriculture-related problems, including farming, field phenotyping, food security, etc. Among these technological advancements, autonomous data collection for field phenotyping enables plant scientists to further correlate plant behavior to environmental conditions using a large amount of high-quality data captured throughout the growing season. In that sense, the ability to process such a large amount of image data must be made possible via computer vision techniques that fuse raw data in order to extract relevant changes during the growing season. For instance, correlating 3D models of plants over time enables scientists to analyze phenotypical data and identify favorable traits in terms of maximum growth, yielding, response to abiotic stressors, etc. Furthermore, the fusion of 3D models of individual plants into larger and cohesive models of the entire crop over time can lead to important discoveries in plant genomics by biology researchers. In this work, we propose an end-to-end system to acquire, fuse, and visualize multimodal data collected over time from an agricultural field. The system consists of two field robots and a visualization tool. The robots can collect and process RGB, Thermal, and 3D information from close proximity or from above ground. The visualization tool allows users to select and process the same data in order to create Spatio-temporal, high dimensional models that describe the physical and temporal structure, color, and temperature of plants in crop fields. Once created, users can visualize the same N-dimensional models using a newly proposed GUI, named VisND. Technically speaking, the proposed system registers time-stamped 5D models (3D plus RGB and IR) of the field as observed by the robots into a single-registered Spatio-temporal model. This process requires a non-rigid (ie. deformable) registration of the 3D structures so that plant behavior (e.g. growth, leaf angle, response to stress, etc) can be visualized over time.