# Biomazon: A Multimodal Dataset for 3D Forest Structure and Biomass Modeling in the Amazon Basin

arXiv: https://arxiv.org/abs/2606.05368 · HTML: https://mlsanity.com/paper/2606.05368/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 25 of its 96 references that are in the map.
- First posted on arXiv 3 Jun 2026.

## Builds on (most cited first, 25 of 25)

- [An Image is Worth 16x16 Words: Transformers for Image Recognition at Scale](https://mlsanity.com/paper/2010.11929/) (2020)
- [SGDR: Stochastic Gradient Descent with Warm Restarts](https://mlsanity.com/paper/1608.03983/) (2016)
- [Gaussian Error Linear Units (GELUs)](https://mlsanity.com/paper/1606.08415/) (2016)
- [FiLM: Visual Reasoning with a General Conditioning Layer](https://mlsanity.com/paper/1709.07871/) (2017)
- [Vision Transformers for Dense Prediction](https://mlsanity.com/paper/2103.13413/) (2021)
- [Delving into Deep Imbalanced Regression](https://mlsanity.com/paper/2102.09554/) (2021)
- [Prithvi-EO-2.0: A Versatile Multi-Temporal Foundation Model for Earth Observation Applications](https://mlsanity.com/paper/2412.02732/) (2024)
- [AlphaEarth Foundations: An embedding field model for accurate and efficient global mapping from sparse label data](https://mlsanity.com/paper/2507.22291/) (2025)
- [GEO-Bench: Toward Foundation Models for Earth Monitoring](https://mlsanity.com/paper/2306.03831/) (2023)
- [A high-resolution canopy height model of the Earth](https://mlsanity.com/paper/2204.08322/) (2022)
- [Very high resolution canopy height maps from RGB imagery using self-supervised vision transformer and convolutional decoder trained on Aerial Lidar](https://mlsanity.com/paper/2304.07213/) (2023)
- [TESSERA: Temporal Embeddings of Surface Spectra for Earth Representation and Analysis](https://mlsanity.com/paper/2506.20380/) (2025)
- [Channel Vision Transformers: An Image Is Worth 1 x 16 x 16 Words](https://mlsanity.com/paper/2309.16108/) (2023)
- [Open-Canopy: Towards Very High Resolution Forest Monitoring](https://mlsanity.com/paper/2407.09392/) (2024)
- [Non-decreasing Quantile Function Network with Efficient Exploration for Distributional Reinforcement Learning](https://mlsanity.com/paper/2105.06696/) (2021)
- [AGBD: A Global-scale Biomass Dataset](https://mlsanity.com/paper/2406.04928/) (2024)
- [High-resolution canopy height map in the Landes forest (France) based on GEDI, Sentinel-1, and Sentinel-2 data with a deep learning approach](https://mlsanity.com/paper/2212.10265/) (2022)
- [Unified Deep Learning Model for Global Prediction of Aboveground Biomass, Canopy Height and Cover from High-Resolution, Multi-Sensor Satellite Imagery](https://mlsanity.com/paper/2408.11234/) (2024)
- [DUNIA: Pixel-Sized Embeddings via Cross-Modal Alignment for Earth Observation Applications](https://mlsanity.com/paper/2502.17066/) (2025)
- [Interpolation of GEDI Biomass Estimates with Calibrated Uncertainty Quantification](https://mlsanity.com/paper/2601.16834/) (2026)
- [Inferring Height from Earth Embeddings: First insights using Google AlphaEarth](https://mlsanity.com/paper/2602.17250/) (2026)
- [Forest aboveground biomass estimation using GEDI and earth observation data through attention-based deep learning](https://mlsanity.com/paper/2311.03067/) (2023)
- [Super-Resolved Canopy Height Mapping from Sentinel-2 Time Series Using Airborne LiDAR HD Reference Data across Metropolitan France](https://mlsanity.com/paper/2512.11524/) (2025)
- [PrediTree: A Multi-Temporal Sub-meter Dataset of Multi-Spectral Imagery Aligned With Canopy Height Maps](https://mlsanity.com/paper/2509.01202/) (2025)
- [SERA-H: Super-Resolution of Sentinel Time Series for Fine-Scale Canopy Height Mapping](https://mlsanity.com/paper/2512.18128/) (2025)

## Abstract (arXiv)

Accurate, spatially explicit characterization of tropical forest structure is essential for carbon accounting and ecosystem monitoring, yet most ML pipelines predict canopy-top height proxies (e.g., RH95/RH98) or AGBD as separate scalar targets, rather than learning the forest vertical structure as an ordered profile. The community lacks a ML-ready multimodal benchmark for predicting the entire GEDI RH profile jointly with AGBD, or for evaluating methods that enforce physically consistent ordering across RH percentiles. We address this with Biomazon, a 20 m multimodal benchmark dataset over the Amazon Basin that pairs GEDI RH and AGBD targets with multi-sensor predictors (Sentinel-1/2, ALOS-2 PALSAR-2, Copernicus DEM, Dynamic World LULC, and AlphaEarth embeddings) under standardized spatial splits and evaluation protocols. Using a shared encoder-decoder with task-specific heads as a baseline framework, we conduct a comprehensive ablation study of (i) backbone/model scale, (ii) modality contributions, and (iii) the use of auxiliary embeddings under standalone and fusion settings, and we report both single-target and joint-target results to quantify tradeoffs under a unified training protocol. Finally, we contextualize baseline performance through regionally aligned comparisons against existing gridded products, including GEDI L4D RH10-RH98 and AGBD, at matching temporal scale. Biomazon, together with the accompanying protocols and baseline results, establishes a reference benchmark for future work on structurally consistent RH-profile prediction and structure-biomass modeling in tropical forests.