Assessment of a new GeoAI foundation model for flood inundation mapping

TL;DR

This paper evaluates IBM-NASA's Prithvi, a novel GeoAI foundation model, for flood inundation mapping, demonstrating its transferability and identifying areas for further improvement in geospatial image analysis.

Contribution

It introduces and assesses the first geospatial foundation model, Prithvi, for flood mapping, comparing its performance with existing architectures using a benchmark dataset.

Findings

Prithvi shows good transferability to unseen regions.

It outperforms CNN and transformer models in flood segmentation accuracy.

Identifies potential improvements in multi-scale learning and input flexibility.

Abstract

Vision foundation models are a new frontier in Geospatial Artificial Intelligence (GeoAI), an interdisciplinary research area that applies and extends AI for geospatial problem solving and geographic knowledge discovery, because of their potential to enable powerful image analysis by learning and extracting important image features from vast amounts of geospatial data. This paper evaluates the performance of the first-of-its-kind geospatial foundation model, IBM-NASA's Prithvi, to support a crucial geospatial analysis task: flood inundation mapping. This model is compared with convolutional neural network and vision transformer-based architectures in terms of mapping accuracy for flooded areas. A benchmark dataset, Sen1Floods11, is used in the experiments, and the models' predictability, generalizability, and transferability are evaluated based on both a test dataset and a dataset that is completely unseen by the model. Results show the good transferability of the Prithvi model, highlighting its performance advantages in segmenting flooded areas in previously unseen regions. The findings also indicate areas for improvement for the Prithvi model in terms of adopting multi-scale representation learning, developing more end-to-end pipelines for high-level image analysis tasks, and offering more flexibility in terms of input data bands.