Compare AlphaEarth Foundations vs. GeoThrust

AlphaEarth Foundations, a cutting-edge AI model developed by DeepMind, functions as a "virtual satellite" by synthesizing extensive and diverse Earth observation data, which includes optical and radar imagery, 3D laser mapping, and climate simulations, into a compact and unified embedding for every 10x10 meter area of land and coastal regions. This innovative approach allows for efficient, on-demand mapping of planet-wide terrains while significantly reducing storage requirements compared to earlier systems. By merging various data streams, it adeptly addresses issues of data overload and inconsistencies, resulting in summaries that are 16 times smaller than those generated by traditional methods, all while achieving a remarkable 24% reduction in error for tested tasks, even in scenarios where labeled data is limited. The annual collections of embeddings are made available as the Satellite Embedding dataset on Google Earth Engine, and they are already being utilized by various organizations to classify previously unmapped ecosystems and to monitor changes in agriculture and the environment, showcasing the practical applications of this groundbreaking technology. This model not only enhances our understanding of Earth’s complexities but also paves the way for future advancements in environmental monitoring and conservation efforts.

GeoThrust features a sophisticated workflow architecture aimed at generating precise earth models and images over time and depth, utilizing data captured with irregular geometries in challenging terrains characterized by complex near-surface and subsurface conditions, all while adhering to exceptionally high standards for data analysis and quality control, yet remaining user-friendly and accessible for new users. The estimation of the near-surface model is conducted through nonlinear tomography that leverages first-arrival times, effectively taking topographical variations into account and accurately delineating both lateral and vertical velocity changes. In addition to applying statics corrections, the near-surface adjustments are executed through wavefield dating, which is crucial for effective imaging in areas with irregular landscapes. Furthermore, GeoThrust adeptly carries out subsurface velocity estimation, modeling, and imaging based on topographical data rather than a flat datum, employing both rms and interval velocities determined at reflector positions instead of mere reflection points. This innovative approach ensures that GeoThrust is not only powerful but also versatile, enabling geoscientists to tackle a wide range of geological challenges with confidence.