Multimodal Earth Observation Workflow for Machine Learning (MEOW-ML): A Case Study in Canopy Height Model and Canopy Height Change Prediction
Description
Here we introduce an updated version of Multimodal Earth Observation Workflow for Machine Learning (MEOW-ML), an end-to-end data fusion and artificial intelligence (AI) and machine learning (ML) framework tailored for Earth Observation (EO). MEOW-ML supports the full AI/ML lifecycle, from data preparation to model training and evaluation.
Detailed example
As an example, we present a trained ML model with promising performance using this framework, for predicting forest productivity and degradation over time by using canopy height change (CHC) as a proxy. We conducted modeling at two spatial resolutions to illustrate the potential use of this framework for NOS design activities.
AI / analytics pattern
Classical/Predictive Machine Learning: Models trained on data to make predictions or classifications based on identified patterns or relationships.
Automation level / stage
a) Pre-deployment – The use case is in a development or acquisition status.
Expected benefit
It speeds up the iterative loops of data processing and model architecture development, and, critically, enables the integration of heterogeneous data sources. A primary intent for MEOW-ML is its application in the design of New Observing Strategies (NOS) which require very large data sets with data of diverse types acquired by a variety of sensors potentially aboard multiple sub-orbital and orbital platforms. MEOW-ML can be used to test various combinations of data types, qualities and resolutions to optimize the design of an NOS.
Controls / human review
ATO: Not reported; PIA: Not published